- Gpc - gnu.de

FAQ v0.2
by J.J. van der Heijden 24 Oct '96

24 Oct '96

Ok folks, I updated my first attempt at a FAQ a bit, and did some very crude Texification, so a html version exists too, which I sent to Peter for the homepages at agnes.dida.physik.uni-essen.de Let me know if you find any errors or sections that need clarification. JanJaap This is Info file FAQ, produced by Makeinfo-1.64 from the input fil faq.texi aq.texi. fo file FAQ, produced by Makeinfo-1.64 from the input file OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Revision 0.2 of 24 October 199 Revision 0.2 of 24 October 1996 Makeinfo-1.64 from the input file OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Author: J.J. van der Heijden <j.j.vanderheijden(a)student.utwente.nl Author: J.J. van der Heijden <j.j.vanderheijden(a)student.utwente.nl> OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & What is GNU Pascal ****************** ****************** der Heijden <j.j.vanderheijden(a)student.utwente.nl> OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & The purpose of the GNU Pascal project is to produce a Pascal compile (called GNU Pascal or GPC) whic called GNU Pascal or GPC) which project is to produce a Pascal compiler OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & * combines the clarity of Pascal with powerful tools suitable fo real-life programming real-life programming, Pascal with powerful tools suitable for ler OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & * supports both the Pascal standard and the Extended Pascal standar as defined by ISO, ANSI and IEEE. (ISO 7185:1990, ISO/IE 10206:1991, ANSI/IEEE 770X3.160-1989 10206:1991, ANSI/IEEE 770X3.160-1989) 7185:1990, ISO/IEC standard OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & * supports other Pascal standards (UCSD Pascal, Borland Pascal Pascal-SC) in so far as this serves the goal of clarity an usability usability, in so far as this serves the goal of clarity and , dard OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & * may be distributed under GNU license condition * may be distributed under GNU license conditions clarity and , dard OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & * can generate code and run on any computer for which the GNU compiler can generate code and run compiler can generate code and run. puter for which the GNU C dard OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Pascal was originally designed for teaching. GNU Pascal provides smooth way to proceed to challenging programming tasks without learnin a completely different language completely different language. ing programming tasks without learning OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & What is the current version of GNU Pascal ========================================= ========================================= mming tasks without learning OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & The last official release is GPC 1.1, based on GCC version 2.6.3 GPC version 1.2, based on GCC 2.7.2 is currently in beta testing PC version 1.2, based on GCC 2.7.2 is currently in beta testing. 3. ng OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Where is the GNU Pascal FTP site? WWW ===================================== ===================================== currently in beta testing. 3. ng OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & The master FTP site for GNU Pascal is kampi.hu.fi GNU Pascal relate files can be found in iles can be found in: for GNU Pascal is kampi.hu.fi GNU Pascal related OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & ftp://ftp.kampi.hut.fi/jtv/gnu-pascal ftp://ftp.kampi.hut.fi/jtv/gnu-pascal/ pi.hu.fi GNU Pascal related OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & This site is mirrored on This site is mirrored on: tv/gnu-pascal/ pi.hu.fi GNU Pascal related OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & ftp://sunsite.doc.ic.ac.uk/gnu/pascal ftp://sunsite.doc.ic.ac.uk/gnu/pascal/ pi.hu.fi GNU Pascal related OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & The latest developer releases can be downloaded from The latest developer releases can be downloaded from: Pascal related OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & ftp://agnes.dida.physik.uni-essen.de ftp://agnes.dida.physik.uni-essen.de/ nloaded from: Pascal related OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Also, we have a homepage on the web Also, we have a homepage on the web: e/ nloaded from: Pascal related OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & http://agnes.dida.physik.uni-essen.de/~gnu-pascal http://agnes.dida.physik.uni-essen.de/~gnu-pascal/ Pascal related OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Is it compatible with Turbo Pascal (R) ======================================= ======================================= e/~gnu-pascal/ Pascal related OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & GPC is currently *not* a drop-in replacement for Borland's Turb Pascal (R). It supports a number, but not all of the Borlan extensions to the Pascal language. There is no replacement for most o the Borland runtime library functions. GNU Pascal is part of the GN project, so portability is one of its primary goals. For this reason non-portable features of Borland Pascal will probably not be include into GNU Pascal. More information can be found in the section "Borlan Pascal" of the GPC manual ascal" of the GPC manual. mation can be found in the section "Borland OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Installation related question ***************************** ***************************** on can be found in the section "Borland OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & This section discusses some common problems with the installation o GNU Pascal NU Pascal. ion discusses some common problems with the installation of OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Which platforms are supported by GNU Pascal =========================================== =========================================== s with the installation of OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & GPC uses the GCC backend, so it should run on any system that i supported by GNU CC. This includes a large variety of Unix systems MS-DOS, OS/2 and Win32. A full list of platforms supported by GCC ca be found in the file "INSTALL" of the GCC distribution. Not all o these have actually been tested, but the gpc-1.2 pre-release is know t run on these platforms un on these platforms: tested, but the gpc-1.2 pre-release is know to OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & i486-linux (Linux 2.x, ELF i486-linuxaou i486-linuxoldl i486-freebsd2. i386-cygwin32 (Windows95/Windows NT i386-djgppv2 (MS-DOS i386-emx (OS/2, MS-DOS mips-sgi-irix5. mips-sgi-irix5.3 (OS/2, MS-DOS) ows NT) -release is know to OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & >>> Ok people - send us your success stories, with canonical machin name! << ame! <<< people - send us your success stories, with canonical machine OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Which components do I need to compile Pascal source code ======================================================== ======================================================== nical machine OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & A complete Pascal compiler system should at least have A complete Pascal compiler system should at least have: ical machine OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & * The actual compiler, GPC * The actual compiler, GPC. ystem should at least have: ical machine OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & * Assembler, linker, librarian and friends * Assembler, linker, librarian and friends. least have: ical machine OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & * A C library * A C library. nker, librarian and friends. least have: ical machine OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & * A debugger, if you want to debug your programs * A debugger, if you want to debug your programs. have: ical machine OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & You don't need a C compiler to compile your Pascal programs, but yo do need it to build the GNU Pascal compiler itself. GNU Pascal versio 1.1 is based on GCC 2.6.3, GPC v1.2 is based on GCC v2.7.2 Any attemp to build GPC with the wrong version of GCC is bound to fail o build GPC with the wrong version of GCC is bound to fail. y attempt OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & For most people, the GNU binutils and GNU debugger (gdb) are a goo choice, although some may prefer to use vendor specific tools hoice, although some may prefer to use vendor specific tools. a good OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Help! linking gpc1 fails: _emit_string_move undefined (and more =============================================================== =============================================================== good OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & If linking `gpc1' bombs out with an error message that looks lik this his: linking `gpc1' bombs out with an error message that looks like d OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & ld: Undefined symbo _emit_string_mov _emit_string_pa _maybe_find_function_dat _dbxout_set_type_statu _version_fla *** Error code make: Fatal error: Command failed for target `gpc1 make: Fatal error: Command failed for target `gpc1' ks like d OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & you probably suffer from a VPATH make problem. A few GPC sourc files have counterparts with identical name in the GCC sourc directory. When you have built GCC in the GCC source directory and yo are not using a recent version of GNU make this problem may occur There are three solutions here are three solutions: ion of GNU make this problem may occur. you OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & * Get a recent version of GNU make. Version 3.74 or better is know to work to work. ent version of GNU make. Version 3.74 or better is known OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & * Build GCC in a seperate directory instead of using the GCC sourc directory directory. n a seperate directory instead of using the GCC source OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & * Manually delete these files from the GCC object directory * Manually delete these files from the GCC object directory: source OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & `stor-layout.o dbxout.o expr.o fold-const.o optabs.o convert.o `function.o setop.o toplev.o `function.o setop.o toplev.o' fold-const.o optabs.o convert.o' e OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & then resume `make' then resume `make'. toplev.o' fold-const.o optabs.o convert.o' e OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & When I build libgpc.a, rts-rt0.c says: SIGXCPU undefined (and more ================================================================== ================================================================== e OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Compilation of the runtime system may fail in rts-rt0.c with message simular to this essage simular to this: time system may fail in rts-rt0.c with a = e OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & rts-rt0.c: `SIGXCPU' undeclared (first use this function or rts-rt0.c: storage size of `sv' isn't know rts-rt0.c: storage size of `osv' isn't know rts-rt0.c: storage size of `osv' isn't known is function) a = e OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & If this happens to you, you have to edit the rts/rts-config.h fil and comment out the last line to nd comment out the last line to: e to edit the rts/rts-config.h file OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & /* #undef HAVE_SIGSYS * /* #undef HAVE_SIGSYS */ to: e to edit the rts/rts-config.h file OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & This will be fixed in a future release This will be fixed in a future release. the rts/rts-config.h file OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & I'm using unix, and all my Pascal programs have linking problems ================================================================ ================================================================ ile OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & A number of unix configurations use their system's linker instead o GNU ld. Usually, GPC and GCC need a program called `collect2' befor calling the system's `ld'. `collect2' is installed by GCC, and if yo only install GPC, it will not find collect2, and use the system linke directly, which will result in various linker errors. The solution i to copy `collect2' by hand from the GCC directory to the location wher `gpc1' lives gpc1' lives. ct2' by hand from the GCC directory to the location where OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & I do I debug my Pascal programs =============================== =============================== he GCC directory to the location where OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & To debug your programs, (a) GNU Pascal must be able to generat executables with debug info for your platform, and (b) you must have debugger which understands this ebugger which understands this. our platform, and (b) you must have a OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & * If `gpc -g -o hello hello.p' says: "gpc: -g not supported for thi platform", then GPC is unable to generate debug info. Usually installing GAS instead of your system's assembler can overcom this. When you configure the GCC used for GPC, specif "-with-gnu-as", and possibly "-with-gnu-ld" and/or "-with-stabs" More information can be found in "INSTALL" file in the GNU C source directory source directory. an be found in "INSTALL" file in the GNU CC s". OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & * Your system's debugger may not understand the debug info generate by GNU tools. In this case, installing GDB may help by GNU tools. In this case, installing GDB may help. nfo generated OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & The bottom line: if you can debug GCC compiled programs, you shoul be able to do this with GPC too e able to do this with GPC too. bug GCC compiled programs, you should OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & The GNU debugger (GDB) currently does not have a "Pascal" mode, s it is unable to display certain Pascal structures etc. When debugging please note that the Initial Letter In Each Identifier Is In Upper Cas And The Rest Are In Lower Case. If you want to display variable 'foo in the debugger, type `show Foo' or `display Foo' instead n the debugger, type `show Foo' or `display Foo' instead. able 'foo' e OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Can you recommend an IDE ======================== ======================== w Foo' or `display Foo' instead. able 'foo' e OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Users of Borland Pascal may wonder if there's a replacement for th IDE (Integrated Development Environment). Here's a few suggestions DE (Integrated Development Environment). Here's a few suggestions: he OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & * *(X)Emacs*. Some people think it's the answer to the question o Life, the Universe and Everything, others decide it's uGNUsable Available from your friendly GNU mirror Available from your friendly GNU mirror. decide it's uGNUsable. OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & * *XWPE* is an immitation of the Borland IDE, so users of Borlan Pascal may find it a good alternative Pascal may find it a good alternative. IDE, so users of Borland OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Although GDB is an excellent debugger, it's user interface is no very attractive. Refer to the comp.windows.x FAQ: "Where can I get a X-based debugger?" at -based debugger?" at: to the comp.windows.x FAQ: "Where can I get an OF Z tY 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & http://www.cis.ohio-state.edu/hypertext/faq/usenet/x-faq/part6/faq-doc-2.htm http://www.cis.ohio-state.edu/hypertext/faq/usenet/x-faq/part6/faq-doc-2.ht… 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Some useful frontends include: XXGDB, tGDB and XWPE. see Some useful frontends include: XXGDB, tGDB and XWPE. see: /part6/faq-doc-2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & http://www.ee.ryerson.ca:8080/~elf/xapps/Q-IV.htm http://www.ee.ryerson.ca:8080/~elf/xapps/Q-IV.html see: /part6/faq-doc-2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Very nice, but resource consuming is the Motif based DDD Very nice, but resource consuming is the Motif based DDD: /part6/faq-doc-2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & http://sol.ibr.cs.tu-bs.de/softech/ddd http://sol.ibr.cs.tu-bs.de/softech/ddd/ otif based DDD: /part6/faq-doc-2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Do we have a binary for you =========================== =========================== de/softech/ddd/ otif based DDD: /part6/faq-doc-2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Currently, we have binaries for these platforms Currently, we have binaries for these platforms: sed DDD: /part6/faq-doc-2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & i486-linux (ELF i486-linuxaou i486-linuxoldl djgpp V2 (msdos emx 0.9B (OS/2, msdos cygwin32 beta16 (Windows95, Windows NT mips-sgi-irix5. mips-sgi-irix5.3 (Windows95, Windows NT) ms: sed DDD: /part6/faq-doc-2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & New binaries may have been added after the release of this FAQ New binaries may have been added after the release of this FAQ. /faq-doc-2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & GNU Pascal and your system librarie *********************************** *********************************** fter the release of this FAQ. /faq-doc-2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & This section discusses common problems people have when they try t access their system's libraries ccess their system's libraries. problems people have when they try to -doc-2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & How do I use <insert_your_function_here> from the C library =========================================================== =========================================================== ey try to -doc-2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & GNU Pascal can use every function of your C library, but it may b up to you to write declaration of an external function, before you ca use it. Consider the function `sleep'. man(3) sleep reveals se it. Consider the function `sleep'. man(3) sleep reveals: you can -doc-2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & -------------------------------------------------------- NAM sleep - Sleep for the specified number of second SYNOPSI #include <unistd.h unsigned int sleep(unsigned int seconds) -------------------------------------------------------- --------------------------------------------------------- you can -doc-2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & This small demo program shows how to use `sleep' in a Pascal program This small demo program shows how to use `sleep' in a Pascal program: oc-2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & -------------------------------------------------------- program SleepDemo typ word = __unsigned__ integer function sleep(seconds: word): word; C va result : word begi result := sleep(10) end -------------------------------------------------------- --------------------------------------------------------- program: oc-2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & What's this confusion about Pascal and C style strings ====================================================== ====================================================== ------ program: oc-2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & It is important not to confuse Pascal and C string types It is important not to confuse Pascal and C string types. - program: oc-2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & * The Pascal string schema, as defined in section 6.4.3.3.3 of th ISO-10206:1990 Extended Pascal standard, is a record ISO-10206:1990 Extended Pascal standard, is a record: 3.3 of the m: oc-2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & type string = recor Capacity : integer length : integer string : packed array [ 1..Capacity ] of char end end; ing : packed array [ 1..Capacity ] of char; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & 'string' is not 'string(256)', unlike Turbo Pascal. The capacict must be declared must be declared: ing(256)', unlike Turbo Pascal. The capacicty ar; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & typ MyString = string(256) MyString = string(256); unlike Turbo Pascal. The capacicty ar; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & before it can be used, i.e. before it can be used, i.e.: ; unlike Turbo Pascal. The capacicty ar; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & function MyFunction: MyString function MyFunction: MyString; Turbo Pascal. The capacicty ar; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & * A C string ( "char *" ) is an array of char, terminated with NULL char NULL char. ( "char *" ) is an array of char, terminated with a ty ar; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & C library functions require C, not Pascal style string arguments Consider this code snippet to convert Pascal style strings to C styl and vice versa nd vice versa: de snippet to convert Pascal style strings to C style ar; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & -------------------------------------------------------- typ word = __unsigned__ integer TString = string(256); { Pascal string schema CString = __cstring__; { C style string { Convert a "C" string to a "Pascal" string function StrPas(Src: CString): TString va S : TString begi S := '' if (Src <> NIL then while ( (Src^ <> chr(0)) AND (length(S) < S.capacity)) d begi S := S + Src^ inc(Word(Src)) end StrPas := S end { Convert a "Pascal" string to a "C" string function StrPCopy(Dest: CString; Src: String): CString va c: integer p : CString begi p := Dest for c:=1 to length(Src) d begi p^ := Src[c] inc(word(p)) end p^ := chr(0) StrPCopy := Dest end -------------------------------------------------------- --------------------------------------------------------- y)) do ar; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Then this small example will print the PATH Then this small example will print the PATH: -------------- y)) do ar; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & -------------------------------------------------------- Program EnvDemo { include the above StrPas() and StrPCopy() snippet here { C library function prototype: char *getenv(const char *name); function GetEnv(name : CString): CString; C va pName: CString begi getmem(pName, 256) pName := StrPCopy(pName, 'PATH') writeln('Your PATH is: ', StrPas(GetEnv(pName))) freemem(pName, 256) end -------------------------------------------------------- --------------------------------------------------------- ame); } ar; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & And this is how you access the 'system()' call in you C library And this is how you access the 'system()' call in you C library: } ar; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & -------------------------------------------------------- program SysCall { include the above StrPas() and StrPCopy() snippet here function system(name : CString): integer; C va pName: CString result : integer begi getmem(pName, 256) pName := StrPCopy(pName, 'ls -l') result := system(pName) writeln('system() call returned : ', result) freemem(pName, 256) end -------------------------------------------------------- --------------------------------------------------------- ry: } ar; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & There may be other ways to do the same thing; you could declare type PChar instead of CString ype PChar instead of CString: o the same thing; you could declare a } ar; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & type PChar = ^char type PChar = ^char; ring: o the same thing; you could declare a } ar; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & and replace all references to `CString' with `PChar'. Do NOT pass "C" style string as a var-argument if the C prototype says "const cha *" or you will get a coredump " or you will get a coredump. ent if the C prototype says "const char ar; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & You are right if you think this stuff belongs in a library, to b distributed with GPC. Have patience, or start coding istributed with GPC. Have patience, or start coding! library, to be r ar; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Are GNU Pascal objects compatible with GNU C++ classes ======================================================= ======================================================= rary, to be r ar; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & No. (This may change in a future version No. (This may change in a future version) =========== rary, to be r ar; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Where are the Turbo Pascal (R) replacement units ================================================ ================================================ ====== rary, to be r ar; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & They don't exist (yet). Most of their fuctionality can easily b implemented, some things are very x86/msdos dependant and would b meaningless on any other platform eaningless on any other platform. 86/msdos dependant and would be r ar; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & How do I build/use a shared library =================================== =================================== /msdos dependant and would be r ar; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & (topic under construction (topic under construction) ====== /msdos dependant and would be r ar; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & GNU Pascal on the djgpp (MS-DOS) platfor **************************************** **************************************** s dependant and would be r ar; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & This chapter discusses some potential problems with GNU Pascal o MS-DOS, using djgpp S-DOS, using djgpp. sses some potential problems with GNU Pascal on r ar; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & If you need more informatio =========================== =========================== e potential problems with GNU Pascal on r ar; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & GPC/djgpp is a djgpp V2 application, and most of the djgp documentation applies for GPC too. A great source of information is th djgpp FAQ jgpp FAQ: on applies for GPC too. A great source of information is the r; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & http://www.delorie.com/djgpp/v2faq/faq202b.zi http://www.delorie.com/djgpp/v2faq/faq202b.zip information is the r; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Another place to look for DJGPP documentation is the DJGPP Knowledg Base, at this URL ase, at this URL: look for DJGPP documentation is the DJGPP Knowledge r; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & http://www.delorie.com/djgpp/doc/kb http://www.delorie.com/djgpp/doc/kb/ tation is the DJGPP Knowledge r; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & What do I download ================== ================== rie.com/djgpp/doc/kb/ tation is the DJGPP Knowledge r; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & As discussed in section 2.2, other than GPC itself, you need a assembler, linker and friends, a C library and possibly a debugger >From your local djgpp mirror, you can get these as rom your local djgpp mirror, you can get these as: bly a debugger. dge r; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & v2/djdev200.zip (C library v2gnu/bnu252b.zip (assembler, .... v2gnu/gdb412b.zip (debugger v2gnu/gdb412b.zip (debugger) ....) as: bly a debugger. dge r; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & The rest is up to you; 'make' (v2gnu/mak372b.zip) can be useful RHIDE (an IDE with Borland-look) is nice, but still under development Future releases of RHIDE may have better GPC support uture releases of RHIDE may have better GPC support. der development. r; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & How do I install the compiler ============================= ============================= ve better GPC support. der development. r; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & If you don't have djgpp installed on your harddisk, create directory for GNU pascal ("c:\gpc"), and unzip the archives. Make sur you preserve the directory structure (use "pkunzip -d"). Now, add th directory where gpc.exe lives ("c:\gpc\bin") to your path and set th DJGPP environment variable to point to your djgpp.env file JGPP environment variable to point to your djgpp.env file: d set the r; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & set DJGPP=c:\gpc\djgpp.en set DJGPP=c:\gpc\djgpp.env int to your djgpp.env file: d set the r; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Then, add this to your djgpp.env file Then, add this to your djgpp.env file: r djgpp.env file: d set the r; -2.html 'J& pY L& < 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & -------------------------------------------------------- [gpc-cpp C_INCLUDE_PATH=%/>;C_INCLUDE_PATH%%DJDIR%/lang/pascal;%DJDIR%/include;%DJDIR%/contrib/grx20/includ [gpc COMPILER_PATH=%/>;COMPILER_PATH%%DJDIR%/bi LIBRARY_PATH=%/>;LIBRARY_PATH%%DJDIR%/lib;%DJDIR%/contrib/grx20/li -------------------------------------------------------- --------------------------------------------------------- grx20/lib e;%DJDIR%/contrib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & The binary distribution should come with a `djgpp.env' which i already modified, so you may not have to do this lready modified, so you may not have to do this. pp.env' which is 0/lib e;%DJDIR%/contrib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Specific information for low-memory conditions and more can be foun in the djgpp FAQ and documentation n the djgpp FAQ and documentation. ry conditions and more can be found e;%DJDIR%/contrib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & GPC says: no DPM ================ ================ nd documentation. ry conditions and more can be found e;%DJDIR%/contrib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & You don't have a DPMI server installed, and DJGPP v2 requires it t run. You can either use one of the commercial DPMI servers (e.g., ru `gpc' in a DOS box under Windows) or download and install CWSDPM (`csdpmi3b.zip') which is a free DPMI server written for DJGPP `csdpmi3b.zip') which is a free DPMI server written for DJGPP. I run e;%DJDIR%/contrib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & I cannot read the info pages ============================ ============================ ee DPMI server written for DJGPP. I run e;%DJDIR%/contrib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & To read the info pages, you need the `info' program from txi360b.zip At least for some of the pre-releases of gpc-1.2, the gpc.info file i invalid: it refers to the "gpc.i*" sections as "gpc.info-*". This can b fixed by loading gpc.inf in an editor and replacing "gpc.info-*" wit "gpc.i* gpc.i*" loading gpc.inf in an editor and replacing "gpc.info-*" with be e;%DJDIR%/contrib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & I have troubles with assembly cod ================================= ================================= or and replacing "gpc.info-*" with be e;%DJDIR%/contrib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & The GNU Assembler (`as.exe'), or `gas', called by GCC accepts "AT&T syntax which is different from "Intel" syntax. Differences ar discussed in section 17.1 of the djgpp FAQ iscussed in section 17.1 of the djgpp FAQ. x. Differences are s "AT&T" e;%DJDIR%/contrib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & A guide is available which was written by Brennan Mr. Wacko Underwoo <brennan(a)mack.rt66.com> and describes how to use inline assembl programming with DJGPP, at this URL rogramming with DJGPP, at this URL: how to use inline assembly derwood e;%DJDIR%/contrib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & http://www.rt66.com/~brennan/djgpp/djgpp_asm.htm http://www.rt66.com/~brennan/djgpp/djgpp_asm.html assembly derwood e;%DJDIR%/contrib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Section 17.3 of the djgpp FAQ discusses some methods to conver "Intel" syntax to "AT&T" Intel" syntax to "AT&T". pp FAQ discusses some methods to convert rwood e;%DJDIR%/contrib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Tell me how to do DPMI, BIOS and other DOS related things ========================================================= ========================================================= convert rwood e;%DJDIR%/contrib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & DPMI, BIOS and other functions are no different than other syste functions. Refer to section 3.1 how to access your system's C-library This small example shows how to use DPMI, copying some structures an function prototypes of <dpmi.h> unction prototypes of <dpmi.h>: se DPMI, copying some structures and d e;%DJDIR%/contrib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & -------------------------------------------------------- program dpmitest {$X+ typ word = __unsigned__ integer short = __short__ integer byte = __byte__ integer typ PDpmiVersionRet = ^TDpmiVersionRet TDpmiVersionRet = recor major : byte minor : byte flags : short cpu : byte master_pic : byte slave_pic : byte end typ PDpmiFreeMemInfo = ^TDpmiFreeMemInfo TDpmiFreeMemInfo = recor largest_available_free_block_in_bytes : word maximum_unlocked_page_allocation_in_pages : word maximum_locked_page_allocation_in_pages : word linear_address_space_size_in_pages : word total_number_of_unlocked_pages : word total_number_of_free_pages : word total_number_of_physical_pages : word free_linear_address_space_in_pages : word size_of_paging_file_partition_in_pages : word reserved1 : byte reserved2 : byte reserved3 : byte end function DpmiGetVersion(ret: PDpmiVersionRet): integer asmname '__dpmi_get_version' function DpmiGetFreeMemoryInformation(info: PDpmiFreeMemInfo): integer asmname '__dpmi_get_free_memory_information' va version: TDpmiVersionRet meminfo: TDpmiFreeMemInfo begi DpmiGetVersion(@version) writeln('CPU type : ', version.cpu, '86') writeln('DPMI major : ', version.major) writeln('DPMI minor : ', version.minor) DpmiGetFreeMemoryInformation(@meminfo) writeln('Free DPMI memory : ', meminfo.total_number_of_free_pages, ' pages.') end -------------------------------------------------------- --------------------------------------------------------- ree_pages, ' pages.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & I'm accessing an "array[1..4000000] of byte" and I got an exception =================================================================== =================================================================== ges, ' pages.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Per default, the maximum stack size of a djgpp application is 256K If you need more, you have to adjust it with the stubedit program, i.e. f you need more, you have to adjust it with the stubedit program, i.e.: ' pages.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & stubedit your_app.exe minstack=5000 stubedit your_app.exe minstack=5000K th the stubedit program, i.e.: ' pages.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Still, it might be a good idea to use pointers for this kind o structures, and allocate the memory at runtime tructures, and allocate the memory at runtime. s for this kind of i.e.: ' pages.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Getting Hel *********** *********** nd allocate the memory at runtime. s for this kind of i.e.: ' pages.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & This section discusses ways to get help with GNU Pascal. Please rea the documentation (info files, readme's) that come with GPC, plus othe docs that might help (the djgpp FAQ if you use djgpp etc.) before yo send email to the maintainers or mailing list end email to the maintainers or mailing list. djgpp etc.) before you r ' pages.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Help! the compiler crashed ========================== ========================== s or mailing list. djgpp etc.) before you r ' pages.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & If the compiler crashes, you have discovered a bug. A reliabl compiler never crashes. To help the maintainers fix this bug, it i important that you send us a problem report mportant that you send us a problem report. rs fix this bug, it is u r ' pages.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & I think I found a bug - now what ================================ ================================ em report. rs fix this bug, it is u r ' pages.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Bugs are best reported to the GPC mailinglist, gpc(a)hut.fi. Tha way, they always reach the maintainers. Try to give as much informatio as possible, plus a short code snippet that triggers the compiler bug If you're on unix, you can find out where the compiler crashed if yo enable coredumps, then load the compiler (gpc1) plus the core file i the debugger ("gdb /your_path_here/gpc1 core"), then type `backtrace to get a stacktrace. Include this stacktrace in your bug report o get a stacktrace. Include this stacktrace in your bug report. ace' ' pages.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Which newsgroup(s) are suited for GPC related problems ====================================================== ====================================================== report. ace' ' pages.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & There are several Pascal related newsgroups, but none is dedicate just to GNU Pascal. This one may be useful ust to GNU Pascal. This one may be useful: ps, but none is dedicated ' pages.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & comp.lang.pascal.misc Pascal in general and ungrouped Pascals comp.lang.pascal.misc Pascal in general and ungrouped Pascals. pages.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Pascal syntax related questions may be appropriate in Pascal syntax related questions may be appropriate in: ngrouped Pascals. pages.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & comp.lang.pascal.ansi-iso Pascal according to ANSI and ISO standards comp.lang.pascal.borland Borland Pascal questions comp.lang.pascal.borland Borland Pascal questions. nd ISO standards. es.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & How to post to the mailing lis ============================== ============================== Borland Pascal questions. nd ISO standards. es.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & You can send a message to the GPC mailing list by sending email t the list address <gpc(a)hut.fi> as if it were a person he list address <gpc(a)hut.fi> as if it were a person. ending email to andards. es.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & How to become a subscriber to the mailing lis ============================================= ============================================= erson. ending email to andards. es.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & You can join the mailing list by sending a message t <gpc-request(a)hut.fi> (NOT <gpc(a)hut.fi> !) with your request to be adde to the list. Maintenance is done by hand, so some delay is possible o the list. Maintenance is done by hand, so some delay is possible. ed dards. es.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & How to unsubscribe from the mailing lis ======================================= ======================================= so some delay is possible. ed dards. es.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & To leave the mailing list, send a message to <gpc-request(a)hut.fi> To leave the mailing list, send a message to <gpc-request(a)hut.fi>. d dards. es.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Miscellaneou ************ ************ e mailing list, send a message to <gpc-request(a)hut.fi>. d dards. es.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & I want to contribute; where do I start ====================================== ====================================== sage to <gpc-request(a)hut.fi>. d dards. es.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & A list of jobs which should be done for GNU-Pascal can be found i the section "How to contribute of the Texinfo deocumentation. In case where somebody is already working on a topic, the name of that perso is written behind the job's description s written behind the job's description. pic, the name of that person s dards. es.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & This does not mean that you shouldn't do that but just that yo should get in contact with that person if you would like to contribut to that field o that field. ontact with that person if you would like to contribute dards. es.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & About this FAQ ============== ============== ntact with that person if you would like to contribute dards. es.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Maintainer: J.J. van der Heijde <j.j.vanderheijden(a)student.utwente.nl j.j.vanderheijden(a)student.utwente.nl> if you would like to contribute dards. es.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & This is the first incarnation of the GNU Pascal FAQ list. Comment about, suggestions for, or corrections to this FAQ list are welcome bout, suggestions for, or corrections to this FAQ list are welcome. dards. es.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Please make sure to include in your mail the version number of th document to which your comments apply (you can find the version at th beginning of this FAQ list) eginning of this FAQ list). ts apply (you can find the version at the dards. es.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & Much of the info in this FAQ list was taken from the gpc mailin list traffic, so you may have (unbeknownst to you) contributed to thi list ist. raffic, so you may have (unbeknownst to you) contributed to this dards. es.'); trib/grx20/include 'J& pYd' : 'J&J& > 'J& (J& 3 ' :8 *J& (J&J& WriteString &4 : 'J& pY ' :8 'J& p p p p p p p p p p p p p p p p p p p p & --- "Nothing shocks me, I'm a scientist", Indiana Jones

1 0

No more trouble with GPC (and a solution)
by Niels Heirbaut 24 Oct '96

24 Oct '96

Hi, Hereby I would like to thank everybody who helped me getting GPC-2.7.2 started. Their hints and tips really helped me getting started. The biggest problems occured because there were no installation instructions in the binary *.ZIP file for DJGPP. People who answered my request for help pointed out to me that I had to add a few lines to DJGPP.ENV, but there was no clue about this in the files I got. It was also not until I got the source files that I found out there had to be a SPECS.GPC in my library-directory. This file was also not included in the binary distribution. So if someone would add the (revised?) installation instruction and the SPECS.GPC file to the binary *.ZIP file, or if someone could tell me how (I'm also a newbie at uploading to FTP-sites :-)) and where to upload the revised gpc272b.zip file, the problems I had will be forever in the past. Regards --- When I am grown to man's estate I shall be very proud and great. And tell the other girls and boys Not to meddle with my toys. --- Stevenson --- Niels Heirbaut ----> heirbaut(a)pi.net

1 0

Re: Trouble with GPC
by J.J. van der Heijden 23 Oct '96

23 Oct '96

On Sun, 20 Oct 1996 17:26:58 +0100, Niels Heirbaut <heirbaut(a)pi.net> wrote: > > > >Hi, > >I want to thank everybody who has helped me getting GPC started, but >another problem came up: > >I get the following messages when I try to compile my program: > >_main.c(.text+0x1a): undefined reference to `djgpp_last_ctor' >_main.c(.text+0x1f): undefined reference to `djgpp_first_ctor' >_main.c(.text+0x2f): undefined reference to `djgpp_first_ctor' >exit.c(.text+0x29): undefined reference to `djgpp_last_dtor' >exit.c(.text+0x2e): undefined reference to `djgpp_first_dtor' >exit.c(.text+0x3f): undefined reference to `djgpp_first_dtor' > >I couldn't find any reference to these error messages in the GPC info files >or DJGPP FAQ. Could some please help me (again)? > Reread the faq. especially section 8.13 (18.3?) which is all about this problem. You have probably edited your djgpp.env file and screwed up the library path. This is my djgpp.env, which works. The sections [gpc] and [gpc-cpp] are important here. They are copied from [gcc] and [cpp] resp. Hope this helps, JanJaap ----------------- djgpp.env --------------- #= Don't edit this line unless you move djgpp.env outside #= of the djgpp installation directory. If you do move #= it, set DJDIR to the directory you installed DJGPP in. #= DJDIR=%:/>DJGPP% +USER=dosuser +TMPDIR=%TEMP% +EMU387=%DJDIR%/bin/emu387.dxe +LFN=n [bison] BISON_HAIRY=%DJDIR%/lib/bison.hai BISON_SIMPLE=%DJDIR%/lib/bison.sim [cpp] CPLUS_INCLUDE_PATH=%/>;CPLUS_INCLUDE_PATH%%DJDIR%/lang/cxx;%DJDIR%/lang/cxx/tvision;%DJDIR%/include;%DJDIR%/contrib/grx20/include C_INCLUDE_PATH=%/>;C_INCLUDE_PATH%%DJDIR%/include;%DJDIR%/contrib/grx20/include OBJCPLUS_INCLUDE_PATH=%/>;OBJCPLUS_INCLUDE_PATH%%DJDIR%/include;%DJDIR%/lang/objc OBJC_INCLUDE_PATH=%/>;OBJC_INCLUDE_PATH%%DJDIR%/include;%DJDIR%/lang/objc [gcc] COMPILER_PATH=%/>;COMPILER_PATH%%DJDIR%/bin LIBRARY_PATH=%/>;LIBRARY_PATH%%DJDIR%/lib;%DJDIR%/contrib/grx20/lib [info] INFOPATH=%/>;INFOPATH%%DJDIR%/info [gpc-cpp] C_INCLUDE_PATH=%/>;C_INCLUDE_PATH%%DJDIR%/lang/pascal;%DJDIR%/include;%DJDIR%/contrib/grx20/include [gpc] COMPILER_PATH=%/>;COMPILER_PATH%%DJDIR%/bin LIBRARY_PATH=%/>;LIBRARY_PATH%%DJDIR%/lib;%DJDIR%/contrib/grx20/lib ------------------------------------------- --- "Nothing shocks me, I'm a scientist", Indiana Jones

1 0

getenv - any solutions?
by Peter Gerwinski 23 Oct '96

23 Oct '96

According to John Michael Black: > There's been a lot of getenvs going around here lately. :) Has anyone > found a method that works yet? The last thing posted about it gave this: > > Function GetEnv ( Name: __CString__ ): PChar; C; > > GPC does not recognize PChar -- at least ours doesn't. You must *define* PChar first: Type PChar = ^Char; Perhaps you should try that Borland compatible GetEnv Function in the Tools Unit of BO5. See ftp://kampi.hut.fi/jtv/gnu-pascal/turbo-alpha/bo5.zip Greetings, Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

Re: 2.7.2 compiler
by Peter Gerwinski 23 Oct '96

23 Oct '96

According to Jim Brander: > gpc-typeck.c:626 > >626 for (r =3D rhs; r; r =3D TYPE_LANG_SPECIFIC (r)->elts [1]) > >(gdb) l > >621 tree lhs, rhs; > >622 { > >623 register tree r; > >624 if (TREE_CODE (lhs) =3D=3D RECORD_TYPE > >625 && TREE_CODE (rhs) =3D=3D RECORD_TYPE) > >626 for (r =3D rhs; r; r =3D TYPE_LANG_SPECIFIC (r)->elts [1]) > >627 { > >628 if (comptypes (lhs, r)) > >629 return 1; > >630 } Please try the following patch: replace if (TREE_CODE (lhs) == RECORD_TYPE && TREE_CODE (rhs) == RECORD_TYPE) by if (IS_OBJECT_TYPE (lhs) && IS_OBJECT_TYPE (rhs)) and let us know the result. You discovered a severe bug. Thanks. Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

getenv - any solutions?
by John Michael Black 23 Oct '96

23 Oct '96

There's been a lot of getenvs going around here lately. :) Has anyone found a method that works yet? The last thing posted about it gave this: Function GetEnv ( Name: __CString__ ): PChar; C; GPC does not recognize PChar -- at least ours doesn't. This could go on for weeks.... :) -john jm_black(a)csunix1.lvc.edu

1 0

2.7.2 compiler
by Jim Brander 23 Oct '96

23 Oct '96

Ran binary distribution 2.7.2 elf format compiler on three files which compile cleanly on 2.6.3. Each time gpc1 failed with segmentation fault in comp_object_pointer_types () but with dissimilar record structures being compared. Here is gdb trace of one failure - file is 20000 lines, failure at 8000 lines >er_end_main Away Away_or_both Towards_or_both Towards Direq Single_direq Makelist Makelistup Duplicate_list Duplicate_node Getnumvalue Getlists Makelistx Makesave Getxset Getxset_from_listarr Find_create_var Add_inconsistency Check_inconsistency Setfalsevar Check_where Partset Makepossle Equalxset Compare_members Compare_lists Listequality String_equality Test_equality Attach_link Patmatch Mesg Ztrim Subs2 Subs3 All_numeric Bnch Bran Create_operator New_link Cut_link Findsubhead Create_sub Buildsub Writej Plusminus Makeset Setpoints Getnum Debug_state Listkill Add_mber_to_list Solution Undoable Remove_member_from_list Add_delete_list Goal Add_data Cleaned_path Add_search Add_variable Highsearch Check_searcFirst_rec New_job_rec First New_job Remove Remove_shuffle Remove_q Remove_search Remove_data Pointing Point_link Pointed_at Setconsistentlink A_link_job Get_link Clear Remove_entry Clear_obsolete Killdata Next_job Killlist Kill_entry Kill_inference Kill_invoc Kill_branch Check_level Adush Check_path Check_out Check_any Check_zero Rate_action List_unlink Breaklink Destroy_constant_list Kill_fun2 Kill_fun3 Another Kill_externalhange_environment Clear_environment Killnet Add_clean Purge_cleanjob_of Add_match Nosearch Cleanupjobs Cleanup Cleanvar Setlinkdata Linksearch t_first_link Uke_out Check_list Link_open Path_open Deadend Reconnoitre Check_paths Contingency_plan Check_tuf Search_for_link Getrange Setvarfmlink Build_list Same_pred Side_link Funcop Symmetric Searches_in Equality Valid Find_search_in Search_in Find_info Revive Start_up_job Clean_uDo_cleaning Find_link Traverse_network Mirror_image Nan Shortnan Pushdata Find_environment Purge_search Gettrueset Unsetlist Readfilename Read_ion_list Next_nextch Getchar Skipblanks Lex_string Lex_id >Program received signal SIGSEGV, Segmentation fault. >0x8011c33 in comp_object_pointer_types (lhs=0x817b3e4, rhs=0x8590dc0) at gpc-typeck.c:626 >626 for (r = rhs; r; r = TYPE_LANG_SPECIFIC (r)->elts [1]) >(gdb) l >621 tree lhs, rhs; >622 { >623 register tree r; >624 if (TREE_CODE (lhs) == RECORD_TYPE >625 && TREE_CODE (rhs) == RECORD_TYPE) >626 for (r = rhs; r; r = TYPE_LANG_SPECIFIC (r)->elts [1]) >627 { >628 if (comptypes (lhs, r)) >629 return 1; >630 } >(gdb) p r >$1 = (union tree_node *) 0x8590dc0 >(gdb) p rhs >$2 = (union tree_node *) 0x0 >(gdb) p lhs >$3 = (union tree_node *) 0x817b3e4 >(gdb) where >#0 0x8011c33 in comp_object_pointer_types (lhs=0x817b3e4, rhs=0x8590dc0) at gpc-typeck.c:626 >#1 0x8016e76 in convert_for_assignment (type=0x8208ac8, rhs=0x859b31c, errtype=0x0, fundecl=0x817b9f0, funname=0x8178e20, parmnum=1) > at gpc-typeck.c:4960 >#2 0x80139ae in convert_arguments (typelist=0x817b90c, values=0x859b2f4, name=0x8178e20, fundecl=0x817b9f0) at gpc-typeck.c:2359 The record being compared, and the offending function >procedure read_orion_list(fs: string300; transient: Boolean; > anchor: net_index; var ls:logstate); >label > error_exit; (* When abandoning hope *) >const > maxlevels = 50; (* Maximum depth of lists within lists *) > max_string = 63; (* maximum length identifier string *) >type > symtype = (uninit, curly_left, curly_right, lit_string, comma, > number, logcon, identifier, endfile, dot_dot, from_to, > maybe,ellip); > lexical_symbol = record > case sym: symtype of > lit_string: (str_val: string(maxstring)); > number: (val: real); > logcon: (log: logstate); > identifier: (name: string(max_string); id_type: ltype; > use: usage); > end; (* lexical_symbol *) > >var > edfile: text; (* File to read from *) > b : BindingType; > string_pp: integer; (* position in command *) > file_io: Boolean; (* Reading from file? *) > readlist_nextch: char; (* next input char *) > eoln1, > eof1: Boolean; > s1 : string300; > > s2,s3: string(maxstring); (* used for opening file *) > i, > max1: integer; > set1: listarr; > prevprevsym, > prevsym, > now_sym, > nextsym: lexical_symbol; > filebuf: string(maxlinelen); > >function next_nextch(far:integer;file_or_string:boolean): char; >(* Returns character beyond nextch *) >var ch,nnch:char; >begin > if file_or_string > then begin > if length(filebuf) < far > then nnch := ' ' > else nnch := filebuf[far]; > end > else begin > if string_pp+far > length(system_data.p^.command) > then nnch := ' ' > else begin > ch := system_data.p^.command[string_pp+far]; > if ch = chr(10) > then nnch := ' ' > else nnch := ch; > end; > end; >next_nextch := nnch; >end; (* func next_nextch *) > >procedure getchar(file_or_string:boolean); >(* input is managed by this routine, and the following variables: > * file_io, edfile, system_data.p^.command, nextch, eoln1, eof1. > *) >begin >while true do begin > if file_or_string > then begin > if filebuf = '' > then begin > if eoln1 > then begin > if eof(edfile) > then begin eof1 := true; readlist_nextch := ' '; eoln1 := false; end > else begin > readln(edfile, filebuf); eoln1 := false; > filebuf := trim(ztrim(filebuf)); > continue; > end > end > else begin > eoln1 := true; readlist_nextch := ' '; > end > end > else begin (* filebuf <> '' *) > readlist_nextch := filebuf[1]; > filebuf := delete(filebuf, 1, 1); > end; > end > else begin > string_pp := string_pp + 1; > if string_pp > length(system_data.p^.command) > then begin > readlist_nextch := ' '; > if eoln1 > then begin > eoln1 := false; eof1:= true; > end > else eoln1 := true; > end > else begin > readlist_nextch := system_data.p^.command[string_pp]; > if readlist_nextch = chr(10) > then begin > eoln1 := true;readlist_nextch := ' '; > end > else eoln1 := false; > end; > end; > break; >end; { while true } >end; (* proc getchar *) > > procedure skipblanks; > const > blank = ' '; > exclam = '!'; > begin (* skip blanks and comments in edfile *) > while not eoln1 and_then not eof1 and_then > ((readlist_nextch = blank) or_else (readlist_nextch = exclam)) do > if readlist_nextch = exclam > then repeat getchar(file_io) > until eoln1 > else getchar(file_io); > end; > > procedure lex_string; > var > termch: char; > begin > with nextsym do begin > termch := readlist_nextch; getchar(file_io); > sym := lit_string; str_val := ''; > while not eoln1 and_then (readlist_nextch <> termch) and_then > (length(str_val) < maxstring) do begin > str_val := str_val + readlist_nextch; > getchar(file_io); > end; > if readlist_nextch = termch > then getchar(file_io) > else begin status := 818; goto error_exit; (* fast escape *) end; > end > end; (* proc lex_string *) > > procedure lex_id; > const > max_name = 31; > var s:string(80); > begin > with nextsym do begin > name := ''; > sym := identifier; > use := normal; > while (readlist_nextch in alphanumund) and_then > (length(name) < max_name) do begin > name := name + readlist_nextch; getchar(file_io); > end; > if (readlist_nextch in ['?', '$', '%', '@']) > then begin > if length(name) < max_name > then begin > case readlist_nextch of > '?': id_type := multi; > '$': id_type := string_p; > '%': id_type := list_p; > '@': id_type := logical; > end; (* esac *) > name := name + readlist_nextch; getchar(file_io); > end > else begin status := 857; goto error_exit; end; > end > else id_type := numeric; (* move from end of routine *) > s := name; > uppercaseproc(s); > if (readlist_nextch in ['a'..'z','A'..'Z','_']) (* too long *) > or_else (s = 'NYK') > then begin status := 862; goto error_exit; end; > if (s = 'TRUE') or_else (s = 'FALSE') or_else (s = 'UKE') > or_else (s = 'ERROR') > then begin > log := logstate(index('UEFT', substr(s, 1, 1))); > sym := logcon; > end; > end; (* with *) > end; (* proc lex_id *) Another trace, from compiling another file with a different record structure being compared >Starting program: /tmp/gpc1 externfunc.i > >Program received signal SIGSEGV, Segmentation fault. >0x8011c33 in comp_object_pointer_types (lhs=0x817ab08, rhs=0x817af64) > at gpc-typeck.c:626 >626 for (r = rhs; r; r = TYPE_LANG_SPECIFIC (r)->elts [1]) >(gdb) where >#0 0x8011c33 in comp_object_pointer_types (lhs=0x817ab08, rhs=0x817af64) > at gpc-typeck.c:626 >#1 0x8016e76 in convert_for_assignment (type=0x8a060b4, rhs=0x8a05e54, > errtype=0x0, fundecl=0x822cd78, funname=0x822caa4, parmnum=2) > at gpc-typeck.c:4960 >#2 0x80139ae in convert_arguments (typelist=0x822cc80, values=0x8a05e40, > name=0x822caa4, fundecl=0x822cd78) at gpc-typeck.c:2359 >#3 0x8013244 in build_function_call (function=0x822cd78, params=0x8a05e40) > at gpc-typeck.c:1896 >#4 0x8007223 in yyparse () at ../gpc-1.2-2.7.2/gpc-parse.y:4364 >#5 0x805ab6e in compile_file (name=0xbffffe61 "externfunc.i") > at ./toplev.c:2288 >#6 0x805d25d in main (argc=2, argv=0xbffffda4, envp=0xbffffdb0) > at ./toplev.c:3977 >#7 0x800126b in _start () >(gdb) up 1 >#1 0x8016e76 in convert_for_assignment (type=0x8a060b4, rhs=0x8a05e54, > errtype=0x0, fundecl=0x822cd78, funname=0x822caa4, parmnum=2) > at gpc-typeck.c:4960 >4960 if (TYPE_MAIN_VARIANT (ttl) == void_type_node >(gdb) p *funname >$1 = {common = {chain = 0x818f334, type = 0x0, code = IDENTIFIER_NODE, > side_effects_flag = 0, constant_flag = 0, permanent_flag = 1, > addressable_flag = 0, volatile_flag = 0, readonly_flag = 0, > unsigned_flag = 0, asm_written_flag = 0, used_flag = 0, raises_flag = 0, > static_flag = 0, public_flag = 1, private_flag = 0, protected_flag = 0, > lang_flag_0 = 0, lang_flag_1 = 0, lang_flag_2 = 0, lang_flag_3 = 0, > lang_flag_4 = 0, lang_flag_5 = 0, lang_flag_6 = 0}, int_cst = { > common = "4 \030\b\000\000\000\000\001\004\b", int_cst_low = 13, > int_cst_high = 136497872}, real_cst = { > common = "4 \030\b\000\000\000\000\001\004\b", rtl = 0xd, real_cst = {r = { > 136497872, 136498552, 0}}}, string = { > common = "4 \030\b\000\000\000\000\001\004\b", rtl = 0xd, > length = 136497872, pointer = 0x822cd78 " \"\b\024 \"\b\035\004\t"}, > complex = {common = "4 \030\b\000\000\000\000\001\004\b", rtl = 0xd, > real = 0x822cad0, imag = 0x822cd78}, identifier = { > common = "4 \030\b\000\000\000\000\001\004\b", length = 13, > pointer = 0x822cad0 "Dollarsubproc"}, decl = { > common = "4 \030\b\000\000\000\000\001\004\b", > filename = 0xd <Address 0xd out of bounds>, linenum = 136497872, > size = 0x822cd78, uid = 0, mode = VOIDmode, external_flag = 0, > nonlocal_flag = 0, regdecl_flag = 0, inline_flag = 0, bit_field_flag = 0, > virtual_flag = 0, ignored_flag = 0, abstract_flag = 0, > in_system_header_flag = 0, common_flag = 0, defer_output = 0, > transparent_union = 0, static_ctor_flag = 0, static_dtor_flag = 0, > artificial_flag = 0, weak_flag = 0, lang_flag_0 = 0, lang_flag_1 = 0, > lang_flag_2 = 0, lang_flag_3 = 0, lang_flag_4 = 0, lang_flag_5 = 0, > lang_flag_6 = 0, lang_flag_7 = 0, name = 0x0, context = 0x0, > arguments = 0x0, result = 0x6c6c6f44, initial = 0x75737261, > abstract_origin = 0x6f727062, assembler_name = 0x63, section_name = 0x0, > machine_attributes = 0x0, rtl = 0x403, frame_size = {i = 0, u = 0, > f = NOT_BUILT_IN}, saved_insns = {r = 0x817bdf4, i = 135773684}, > vindex = 0x0, lang_specific = 0x0}, type = { > common = "4 \030\b\000\000\000\000\001\004\b", values = 0xd, > size = 0x822cad0, attributes = 0x822cd78, uid = 0, precision = 0 '\000', > mode = VOIDmode, string_flag = 0, no_force_blk_flag = 0, > needs_constructing_flag = 0, transparent_union_flag = 0, packed_flag = 0, > lang_flag_0 = 0, lang_flag_1 = 0, lang_flag_2 = 0, lang_flag_3 = 0, > lang_flag_4 = 0, lang_flag_5 = 0, lang_flag_6 = 0, align = 0, > pointer_to = 0x0, reference_to = 0x0, symtab = {address = 1819045700, > pointer = 0x6c6c6f44 ""}, name = 0x75737261, minval = 0x6f727062, > maxval = 0x63, next_variant = 0x0, main_variant = 0x0, binfo = 0x403, > noncopied_parts = 0x0, context = 0x817bdf4, obstack = 0x0, > lang_specific = 0x0}, list = { > common = "4 \030\b\000\000\000\000\001\004\b", purpose = 0xd, > value = 0x822cad0}, vec = {common = "4 \030\b\000\000\000\000\001\004\b", > length = 13, a = {0x822cad0}}, exp = { > common = "4 \030\b\000\000\000\000\001\004\b", complexity = 13, > operands = {0x822cad0}}, block = { > common = "4 \030\b\000\000\000\000\001\004\b", handler_block_flag = 1, > abstract_flag = 0, vars = 0x822cad0, type_tags = 0x822cd78, > subblocks = 0x0, supercontext = 0x0, abstract_origin = 0x0, > end_note = 0x0}} >(gdb) >$2 = {common = {chain = 0x818f334, type = 0x0, code = IDENTIFIER_NODE, > side_effects_flag = 0, constant_flag = 0, permanent_flag = 1, > addressable_flag = 0, volatile_flag = 0, readonly_flag = 0, > unsigned_flag = 0, asm_written_flag = 0, used_flag = 0, raises_flag = 0, > static_flag = 0, public_flag = 1, private_flag = 0, protected_flag = 0, > lang_flag_0 = 0, lang_flag_1 = 0, lang_flag_2 = 0, lang_flag_3 = 0, > lang_flag_4 = 0, lang_flag_5 = 0, lang_flag_6 = 0}, int_cst = { > common = "4 \030\b\000\000\000\000\001\004\b", int_cst_low = 13, > int_cst_high = 136497872}, real_cst = { > common = "4 \030\b\000\000\000\000\001\004\b", rtl = 0xd, real_cst = {r = { > 136497872, 136498552, 0}}}, string = { > common = "4 \030\b\000\000\000\000\001\004\b", rtl = 0xd, > length = 136497872, pointer = 0x822cd78 " \"\b\024 \"\b\035\004\t"}, > complex = {common = "4 \030\b\000\000\000\000\001\004\b", rtl = 0xd, > real = 0x822cad0, imag = 0x822cd78}, identifier = { > common = "4 \030\b\000\000\000\000\001\004\b", length = 13, > pointer = 0x822cad0 "Dollarsubproc"}, decl = { > common = "4 \030\b\000\000\000\000\001\004\b", > filename = 0xd <Address 0xd out of bounds>, linenum = 136497872, > size = 0x822cd78, uid = 0, mode = VOIDmode, external_flag = 0, > nonlocal_flag = 0, regdecl_flag = 0, inline_flag = 0, bit_field_flag = 0, > virtual_flag = 0, ignored_flag = 0, abstract_flag = 0, > in_system_header_flag = 0, common_flag = 0, defer_output = 0, > transparent_union = 0, static_ctor_flag = 0, static_dtor_flag = 0, > artificial_flag = 0, weak_flag = 0, lang_flag_0 = 0, lang_flag_1 = 0, > lang_flag_2 = 0, lang_flag_3 = 0, lang_flag_4 = 0, lang_flag_5 = 0, > lang_flag_6 = 0, lang_flag_7 = 0, name = 0x0, context = 0x0, > arguments = 0x0, result = 0x6c6c6f44, initial = 0x75737261, > abstract_origin = 0x6f727062, assembler_name = 0x63, section_name = 0x0, > machine_attributes = 0x0, rtl = 0x403, frame_size = {i = 0, u = 0, > f = NOT_BUILT_IN}, saved_insns = {r = 0x817bdf4, i = 135773684}, > vindex = 0x0, lang_specific = 0x0}, type = { > common = "4 \030\b\000\000\000\000\001\004\b", values = 0xd, > size = 0x822cad0, attributes = 0x822cd78, uid = 0, precision = 0 '\000', > mode = VOIDmode, string_flag = 0, no_force_blk_flag = 0, > needs_constructing_flag = 0, transparent_union_flag = 0, packed_flag = 0, > lang_flag_0 = 0, lang_flag_1 = 0, lang_flag_2 = 0, lang_flag_3 = 0, > lang_flag_4 = 0, lang_flag_5 = 0, lang_flag_6 = 0, align = 0, > pointer_to = 0x0, reference_to = 0x0, symtab = {address = 1819045700, > pointer = 0x6c6c6f44 ""}, name = 0x75737261, minval = 0x6f727062, > maxval = 0x63, next_variant = 0x0, main_variant = 0x0, binfo = 0x403, > noncopied_parts = 0x0, context = 0x817bdf4, obstack = 0x0, > lang_specific = 0x0}, list = { > common = "4 \030\b\000\000\000\000\001\004\b", purpose = 0xd, > value = 0x822cad0}, vec = {common = "4 \030\b\000\000\000\000\001\004\b", > length = 13, a = {0x822cad0}}, exp = { > common = "4 \030\b\000\000\000\000\001\004\b", complexity = 13, > operands = {0x822cad0}}, block = { > common = "4 \030\b\000\000\000\000\001\004\b", handler_block_flag = 1, > abstract_flag = 0, vars = 0x822cad0, type_tags = 0x822cd78, > subblocks = 0x0, supercontext = 0x0, abstract_origin = 0x0, > end_note = 0x0}} >(gdb) p *fundecl >$3 = {common = {chain = 0x822c9f0, type = 0x822cd14, code = FUNCTION_DECL, > side_effects_flag = 0, constant_flag = 0, permanent_flag = 1, > addressable_flag = 0, volatile_flag = 0, readonly_flag = 0, > unsigned_flag = 0, asm_written_flag = 0, used_flag = 1, raises_flag = 0, > static_flag = 0, public_flag = 1, private_flag = 0, protected_flag = 0, > lang_flag_0 = 0, lang_flag_1 = 0, lang_flag_2 = 0, lang_flag_3 = 0, > lang_flag_4 = 0, lang_flag_5 = 0, lang_flag_6 = 0}, int_cst = { > common = " \"\b\024 \"\b\035\004\t", int_cst_low = 135759440, > int_cst_high = 8539}, real_cst = {common = " \"\b\024 \"\b\035\004\t", > rtl = 0x8178650, real_cst = {r = {8539, 0, 2549}}}, string = { > common = " \"\b\024 \"\b\035\004\t", rtl = 0x8178650, length = 8539, > pointer = 0x0}, complex = {common = " \"\b\024 \"\b\035\004\t", > rtl = 0x8178650, real = 0x215b, imag = 0x0}, identifier = { > common = " \"\b\024 \"\b\035\004\t", length = 135759440, > pointer = 0x215b <Address 0x215b out of bounds>}, decl = { > common = " \"\b\024 \"\b\035\004\t", > filename = 0x8178650 "externfunc.pas", linenum = 8539, size = 0x0, > uid = 2549, mode = QImode, external_flag = 1, nonlocal_flag = 0, > regdecl_flag = 0, inline_flag = 0, bit_field_flag = 0, virtual_flag = 0, > ignored_flag = 0, abstract_flag = 0, in_system_header_flag = 0, > common_flag = 1, defer_output = 0, transparent_union = 0, > static_ctor_flag = 0, static_dtor_flag = 0, artificial_flag = 0, > weak_flag = 0, lang_flag_0 = 0, lang_flag_1 = 0, lang_flag_2 = 0, > lang_flag_3 = 0, lang_flag_4 = 0, lang_flag_5 = 0, lang_flag_6 = 0, > lang_flag_7 = 0, name = 0x822caa4, context = 0x0, arguments = 0x0, > result = 0x0, initial = 0x0, abstract_origin = 0x0, > assembler_name = 0x822cde0, section_name = 0x0, machine_attributes = 0x0, > rtl = 0x822ce24, frame_size = {i = 0, u = 0, f = NOT_BUILT_IN}, > saved_insns = {r = 0x0, i = 0}, vindex = 0x0, lang_specific = 0x89fda98}, > type = {common = " \"\b\024 \"\b\035\004\t", values = 0x8178650, > size = 0x215b, attributes = 0x0, uid = 2549, precision = 1 '\001', > mode = QImode, string_flag = 0, no_force_blk_flag = 1, > needs_constructing_flag = 0, transparent_union_flag = 0, packed_flag = 0, > lang_flag_0 = 0, lang_flag_1 = 0, lang_flag_2 = 0, lang_flag_3 = 0, > lang_flag_4 = 0, lang_flag_5 = 0, lang_flag_6 = 0, align = 136497828, > pointer_to = 0x0, reference_to = 0x0, symtab = {address = 0, > pointer = 0x0}, name = 0x0, minval = 0x0, maxval = 0x822cde0, > next_variant = 0x0, main_variant = 0x0, binfo = 0x822ce24, > noncopied_parts = 0x0, context = 0x0, obstack = 0x0, > lang_specific = 0x89fda98}, list = {common = " \"\b\024 \"\b\035\004\t", > purpose = 0x8178650, value = 0x215b}, vec = { > common = " \"\b\024 \"\b\035\004\t", length = 135759440, a = {0x215b}}, > exp = {common = " \"\b\024 \"\b\035\004\t", complexity = 135759440, > operands = {0x215b}}, block = {common = " \"\b\024 \"\b\035\004\t", > handler_block_flag = 0, abstract_flag = 0, vars = 0x215b, type_tags = 0x0, > subblocks = 0x9f5, supercontext = 0x20101, abstract_origin = 0x822caa4, > end_note = 0x0}} >(gdb) p *rhs >$4 = {common = {chain = 0x0, type = 0x825b390, code = CONVERT_EXPR, > side_effects_flag = 0, constant_flag = 0, permanent_flag = 1, > addressable_flag = 0, volatile_flag = 0, readonly_flag = 0, > unsigned_flag = 0, asm_written_flag = 0, used_flag = 0, raises_flag = 0, > static_flag = 0, public_flag = 0, private_flag = 0, protected_flag = 0, > lang_flag_0 = 0, lang_flag_1 = 0, lang_flag_2 = 0, lang_flag_3 = 0, > lang_flag_4 = 0, lang_flag_5 = 0, lang_flag_6 = 0}, int_cst = { > common = "\000\000\000\000\220 %\bi\004\000", int_cst_low = 0, > int_cst_high = 136644252}, real_cst = { > common = "\000\000\000\000\220 %\bi\004\000", rtl = 0x0, real_cst = {r = { > 136644252, 0, 135769552}}}, string = { > common = "\000\000\000\000\220 %\bi\004\000", rtl = 0x0, > length = 136644252, pointer = 0x0}, complex = { > common = "\000\000\000\000\220 %\bi\004\000", rtl = 0x0, real = 0x825069c, > imag = 0x0}, identifier = {common = "\000\000\000\000\220 %\bi\004\000", > length = 0, pointer = 0x825069c "á\a%\b \005%\b\"L\001"}, decl = { > common = "\000\000\000\000\220 %\bi\004\000", filename = 0x0, > linenum = 136644252, size = 0x0, uid = 135769552, mode = 39, > external_flag = 0, nonlocal_flag = 0, regdecl_flag = 1, inline_flag = 0, > bit_field_flag = 0, virtual_flag = 0, ignored_flag = 0, abstract_flag = 0, > in_system_header_flag = 0, common_flag = 0, defer_output = 0, > transparent_union = 0, static_ctor_flag = 0, static_dtor_flag = 0, > artificial_flag = 0, weak_flag = 0, lang_flag_0 = 0, lang_flag_1 = 0, > lang_flag_2 = 0, lang_flag_3 = 0, lang_flag_4 = 0, lang_flag_5 = 0, > lang_flag_6 = 0, lang_flag_7 = 0, name = 0x0, context = 0x8a05e54, > arguments = 0x0, result = 0x0, initial = 0x403, abstract_origin = 0x0, > assembler_name = 0x8a05e68, section_name = 0x0, > machine_attributes = 0x822cd14, rtl = 0x440d, frame_size = {i = 0, u = 0, > f = NOT_BUILT_IN}, saved_insns = {r = 0x815aa68, i = 135637608}, > vindex = 0x0, lang_specific = 0x906}, type = { > common = "\000\000\000\000\220 %\bi\004\000", values = 0x0, > size = 0x825069c, attributes = 0x0, uid = 135769552, precision = 39 '\'', > mode = SImode, string_flag = 0, no_force_blk_flag = 0, > needs_constructing_flag = 0, transparent_union_flag = 0, packed_flag = 0, > lang_flag_0 = 0, lang_flag_1 = 0, lang_flag_2 = 0, lang_flag_3 = 0, > lang_flag_4 = 0, lang_flag_5 = 0, lang_flag_6 = 0, align = 0, > pointer_to = 0x8a05e54, reference_to = 0x0, symtab = {address = 0, > pointer = 0x0}, name = 0x403, minval = 0x0, maxval = 0x8a05e68, > next_variant = 0x0, main_variant = 0x822cd14, binfo = 0x440d, > noncopied_parts = 0x0, context = 0x815aa68, obstack = 0x0, > lang_specific = 0x906}, list = { > common = "\000\000\000\000\220 %\bi\004\000", purpose = 0x0, > value = 0x825069c}, vec = {common = "\000\000\000\000\220 %\bi\004\000", > length = 0, a = {0x825069c}}, exp = { > common = "\000\000\000\000\220 %\bi\004\000", complexity = 0, operands = { > 0x825069c}}, block = {common = "\000\000\000\000\220 %\bi\004\000", > handler_block_flag = 0, abstract_flag = 0, vars = 0x825069c, > type_tags = 0x0, subblocks = 0x817add0, supercontext = 0x427, > abstract_origin = 0x0, end_note = 0x8a05e54}} >(gdb) p *lhs >(gdb) >(gdb) p *type >$5 = {common = {chain = 0x0, type = 0x817ab08, code = POINTER_TYPE, > side_effects_flag = 0, constant_flag = 0, permanent_flag = 1, > addressable_flag = 0, volatile_flag = 0, readonly_flag = 0, > unsigned_flag = 1, asm_written_flag = 0, used_flag = 0, raises_flag = 0, > static_flag = 0, public_flag = 0, private_flag = 0, protected_flag = 0, > lang_flag_0 = 0, lang_flag_1 = 0, lang_flag_2 = 0, lang_flag_3 = 0, > lang_flag_4 = 0, lang_flag_5 = 0, lang_flag_6 = 0}, int_cst = { > common = "\000\000\000\000\b½\027\b\rD\000", int_cst_low = 0, > int_cst_high = 135637608}, real_cst = { > common = "\000\000\000\000\b½\027\b\rD\000", rtl = 0x0, real_cst = {r = { > 135637608, 0, 2311}}}, string = { > common = "\000\000\000\000\b½\027\b\rD\000", rtl = 0x0, > length = 135637608, pointer = 0x0}, complex = { > common = "\000\000\000\000\b½\027\b\rD\000", rtl = 0x0, real = 0x815aa68, > imag = 0x0}, identifier = {common = "\000\000\000\000\b½\027\b\rD\000", > length = 0, pointer = 0x815aa68 ""}, decl = { > common = "\000\000\000\000\b½\027\b\rD\000", filename = 0x0, > linenum = 135637608, size = 0x0, uid = 2311, mode = 32, external_flag = 0, > nonlocal_flag = 0, regdecl_flag = 1, inline_flag = 0, bit_field_flag = 0, > virtual_flag = 0, ignored_flag = 0, abstract_flag = 0, > in_system_header_flag = 0, common_flag = 0, defer_output = 0, > transparent_union = 0, static_ctor_flag = 0, static_dtor_flag = 0, > artificial_flag = 0, weak_flag = 0, lang_flag_0 = 0, lang_flag_1 = 0, > lang_flag_2 = 0, lang_flag_3 = 0, lang_flag_4 = 0, lang_flag_5 = 0, > lang_flag_6 = 0, lang_flag_7 = 0, name = 0x20, context = 0x0, > arguments = 0x0, result = 0x0, initial = 0x0, abstract_origin = 0x0, > assembler_name = 0x0, section_name = 0x0, machine_attributes = 0x8a060b4, > rtl = 0x0, frame_size = {i = 0, u = 0, f = NOT_BUILT_IN}, saved_insns = { > r = 0x0, i = 0}, vindex = 0x8151654, lang_specific = 0x0}, type = { > common = "\000\000\000\000\b½\027\b\rD\000", values = 0x0, > size = 0x815aa68, attributes = 0x0, uid = 2311, precision = 32 ' ', > mode = SImode, string_flag = 0, no_force_blk_flag = 0, > needs_constructing_flag = 0, transparent_union_flag = 0, packed_flag = 0, > lang_flag_0 = 0, lang_flag_1 = 0, lang_flag_2 = 0, lang_flag_3 = 0, > lang_flag_4 = 0, lang_flag_5 = 0, lang_flag_6 = 0, align = 32, > pointer_to = 0x0, reference_to = 0x0, symtab = {address = 0, > pointer = 0x0}, name = 0x0, minval = 0x0, maxval = 0x0, > next_variant = 0x0, main_variant = 0x8a060b4, binfo = 0x0, > noncopied_parts = 0x0, context = 0x0, obstack = 0x8151654, > lang_specific = 0x0}, list = {common = "\000\000\000\000\b½\027\b\rD\000", > purpose = 0x0, value = 0x815aa68}, vec = { > common = "\000\000\000\000\b½\027\b\rD\000", length = 0, a = {0x815aa68}}, > exp = {common = "\000\000\000\000\b½\027\b\rD\000", complexity = 0, > operands = {0x815aa68}}, block = { > common = "\000\000\000\000\b½\027\b\rD\000", handler_block_flag = 0, > abstract_flag = 0, vars = 0x815aa68, type_tags = 0x0, subblocks = 0x907, > supercontext = 0x420, abstract_origin = 0x20, end_note = 0x0}} Please let me know what you would like to see in terms of output. Regards Jim

1 0

GPC WWW home page modified
by Peter Gerwinski 23 Oct '96

23 Oct '96

Hello, friends of GNU Pascal, I did some modifications to the GNU Pascal home page, http://home.pages.de/~GNU-Pascal/ or http://agnes.dida.physik.uni-essen.de/~gnu-pascal/ Please post your comments here; especially let me know if I missed to mention you and/or your work in the "development team" page. Yours, Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

Re: Asmname Directive
by J.J. van der Heijden 21 Oct '96

21 Oct '96

On Sun, 20 Oct 96 22:30:09 -0400, Kevin A. Foss <kfoss(a)atlsysnet.com> wrote: >Hello, > >I am having trouble getting gpc to call external function names the way I >would like. According to builtin.texi, there is a directive called 'Asmname' >which should allow for case sensitive function names. > >However when I declare: > >function WinInitialize(fsoptions : u_long) : u_long; Asmname; >[u_long is a type defined elsewhere] > >gpc responds with a parse error on the line. What is the correct format for >using the 'Asmname' directive? -- I couldn't find it mentioned anywhere else >in the docs. > >Using C or EXTERN in place of Asmname it produces calls to _wininitialize or >_Wininitialize, neither of which I want. > You must give asmname an argument -- the name you would like to have in the object code. So, if you would like to have '_MyMixedCaseName', you do: function MyMixedCaseName: integer; asmname 'MyMixedCaseName'; It also allows this: function MyOtherName: integer; asmname 'MyMixedCaseName'; Hope this helps, JanJaap --- "Nothing shocks me, I'm a scientist", Indiana Jones

1 0

Asmname Directive
by Kevin A. Foss 21 Oct '96

21 Oct '96

Hello, I am having trouble getting gpc to call external function names the way I would like. According to builtin.texi, there is a directive called 'Asmname' which should allow for case sensitive function names. However when I declare: function WinInitialize(fsoptions : u_long) : u_long; Asmname; [u_long is a type defined elsewhere] gpc responds with a parse error on the line. What is the correct format for using the 'Asmname' directive? -- I couldn't find it mentioned anywhere else in the docs. Using C or EXTERN in place of Asmname it produces calls to _wininitialize or _Wininitialize, neither of which I want. -Kevin --- Kevin A. Foss -- kfoss(a)atlsysnet.com --- --- Kevin A. Foss -- kfoss(a)atlsysnet.com ---

1 0

Main Program Name decapitalization
by Peter Gerwinski 20 Oct '96

20 Oct '96

Hi everybody, I was reported several errors resulting from the decapitalization of the main program's name, e.g. Program Log; begin writeln ( ln ( 2.7184 ) ); end. crashed because the program's name was decapitalized, i.e. "Log" was mapped to "log" which did hide the function in libm.a ... One could think about letting the program name like it is, say "Log". But (as Juki writes in gpc-util.c :-) the standard says that the program's name has to go into a separate name space i.e. mustn't be visible inside the program's declaration part and body. I solved the problem by mapping "Name" to "program_Name" instead of "name". But this is not the end of the story. Borland Pascal defines so-called qualified identifiers: Program Foo; Function writeln: Real; begin (* writeln *) writeln:= 2.7184; end (* writeln *); begin System.writeln ( Foo.writeln ); end. So the program's name *must* go into its declaration part and body -- and we have a classical tragedy. How to proceed? I think Borland Pascal's qualified identifiers are important and must be implemented, one day. Introduce another command line switch? My own suggestion follows. ;-) We should "clean up" Pascal standards in the future by making the (existing) command line switches to force one standard: --standard-pascal --extended-pascal --borland-pascal ... Without any switch, GPC will try to support *all* standards and warn about misuse (e.g. Borland's typed constants as initialized variables). The default for this case would be to *have* qualified identifiers (i.e. "Program Log" would yield "Log", not "log" and not "program_Log"); with "--standard-pascal" or "--extended-pascal" it would separate the program's name from the rest of the world (and create "program_Log"). Yours, Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

Re: Trouble with GPC
by Niels Heirbaut 20 Oct '96

20 Oct '96

Hi, I want to thank everybody who has helped me getting GPC started, but another problem came up: I get the following messages when I try to compile my program: _main.c(.text+0x1a): undefined reference to `djgpp_last_ctor' _main.c(.text+0x1f): undefined reference to `djgpp_first_ctor' _main.c(.text+0x2f): undefined reference to `djgpp_first_ctor' exit.c(.text+0x29): undefined reference to `djgpp_last_dtor' exit.c(.text+0x2e): undefined reference to `djgpp_first_dtor' exit.c(.text+0x3f): undefined reference to `djgpp_first_dtor' I couldn't find any reference to these error messages in the GPC info files or DJGPP FAQ. Could some please help me (again)? here are my env.lst and messages with the -v option on: C:\>type env.lst TMP=C:\WINDOWS\TEMP winbootdir=C:\WINDOWS COMSPEC=C:\WINDOWS\COMMAND.COM PROMPT=$p$g PATH=C:\WINDOWS;C:\WINDOWS\COMMAND;C:\DOS;C:\DJGPP\BIN;C:\DJGPP\LIB;C:\PHOTO ;C:\PKWARE TEMP=C:\DOS DJGPP=C:\DJGPP\DJGPP.ENV windir=C:\WINDOWS BLASTER=A220 I10 D3 T4 CMDLINE=gpc -v test.pas -o test.exe C:\>gpc -v test.pas -o test.exe gpc version 1.2(2.7.2) gpc-cpp -lang-pascal -v -nocharescape -undef -D__GNUC__=1 -D__GPC__=1 -D__GNUC_MINOR__=2(2 -D__unix__ -D__i386__ -D__GO32__ -D__MSDOS__ -D__unix -D__i386 -D__GO32 -D__MSDOS -Asystem(unix) -Asystem(msdos) -Acpu(i386) -Amachine(i386) test.pas c:/djgpp/tmp\ccbaaaaa.i GNU CPP version 1.2(2.7.2) (80386, BSD syntax) #include "..." search starts here: #include <...> search starts here: c:/djgpp/include /usr/local/include /usr/local/go32/include /usr/local/lib/gcc-lib/go32/2.7.2/include /usr/include End of search list. c:/djgpp/bin\gpc1.exe c:/djgpp/tmp\ccbaaaaa.i -quiet -dumpbase test.pas -version -o c:/djgpp/tmp\cccaaaaa.s GNU Pascal version 1.2(2.7.2) (80386, BSD syntax) compiled by GNU C version 2.7.2. c:/djgpp/bin\as.exe -o c:/djgpp/tmp\ccdaaaaa.o c:/djgpp/tmp\cccaaaaa.s c:/djgpp/bin\ld.exe c:/djgpp/lib\crt0.o -Lc:/djgpp/lib c:/djgpp/tmp\ccdaaaaa.o -lgpc -lgcc -lm -lc -lgcc _main.c(.text+0x1a): undefined reference to `djgpp_last_ctor' _main.c(.text+0x1f): undefined reference to `djgpp_first_ctor' _main.c(.text+0x2f): undefined reference to `djgpp_first_ctor' exit.c(.text+0x29): undefined reference to `djgpp_last_dtor' exit.c(.text+0x2e): undefined reference to `djgpp_first_dtor' exit.c(.text+0x3f): undefined reference to `djgpp_first_dtor' I also would like to know where I could get the latest version of GPC. Mine came from ftp://kampi.hut.fi/. But there was no installation manual or DJGPP.ENV included with the binary *.ZIP file. Thanks in advance again, Niels Heirbaut

1 0

Re: System.pas module
by Peter Gerwinski 20 Oct '96

20 Oct '96

Hello Berend, hello everybody! The system.pas module triggers one bug in GPC concerning String schemas as parameters: > procedure GetDir(D: byte; var s: string); declares a string schema without specified length as a formal parameter. When trying to use it with an actual parameter which must have a specified length, the compiler crashes. To work around, always specify string length, e.g. procedure GetDir(D: byte; var s: string255); According to Berend de Boer: > This program is free software; you can redistribute it and/or modify > it under the terms of the GNU General Public License as published by > the Free Software Foundation; either version 2 of the License, or > (at your option) any later version. Berend, you know that the normal GNU GPL implies that a program linked with your module automatically becomes free software? If you want to enable commercial use of your module, either state explicitely that the above does *not* hold, or use the GNU LGPL (GNU Library General Public License) instead of the normal GPL. > type > shortint = __byte__ integer; > byte = __byte__ integer; That's incompatible with Borland's definition of "byte". They have byte = __unsigned__ shortint; Same holds for "word". > TChar = array[0..MaxInt] of char; > PChar = TChar; > pointer = void; It must be PChar = ^TChar; pointer = ^void; with adaptions below when "PChar" is used. On my machine, your module compiled (?!;), but "GetDir" always returned empty strings for this reason. > string255= string(255); > > var > {?ExitProc: Pointer; { Exit procedure } Can be implemented using "to end do" and GPC's pointers to procedures. > export > System = (shortint, byte, word, longint, PChar, pointer, string255, > HeapError, ExitCode, PrefixSeg, > Assign, BPChDir => ChDir, Close, Copy, Dec, Delete, Erase, > GetDir, GetMem, > Inc, IOResult, > MaxAvail, MemAvail, BPMkDir => MkDir, > ParamCount, ParamStr, > BPRename => Rename, BPRmDir => RmDir, > UpCase); "Inc", "dec", "GetMem" are in GPC (and couldn't be implemented in a module anyway) - and there is no implementation of them in the body. Yours, Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

Re: How to implement writeln()
by The African Chief 18 Oct '96

18 Oct '96

On 17 Oct 1996 18:08:00 +0200 Andreas Bauer <andi.tio(a)hit.handshake.de> wrote: >I'm currently programming a bp-like CRT unit. >But i don't know how to do writeln(), which is not limited, >neither in the number of arguments nor in the type of this. This type of thing has to be catered for internally by the compiler itself. This is how Borland does it. I don't see how you can do it without changing the internals of the compiler to handle it specifically. Warmest regards, Dr A. Olowofoyeku (The African Chief, and the Great Elephant) Senior Lecturer, School of Law, Keele University, England. Author of: Chief's Installer Pro v2.91 for Win16 and Win32. Email: laa12(a)cc.keele.ac.uk Home Page: http://ourworld.compuserve.com/homepages/African_Chief/

1 0

Re: proto-FAQ -- reactions please!
by Peter Gerwinski 17 Oct '96

17 Oct '96

According to Sven Hilscher: > Would be a nice thing - but it dos'nt work ... > If you call this GetEnv you get an error: > "incompatible type for argument 1 of 'GetEnv'" It does work with me, on Linux as well as with DJGPP. EMX compiles it without complain but doesn't read the environment variable :-( I have to investigate this ...) > The other way : > function GetEnv(name : PChar): PChar; C; > > This also not work with the string-conversion from the faq ;-( > Every time I get a empty string from "GetEnv". (I tried it under DOS.) Same as what I get with __CString__ and EMX. > Moment please ... I think I've found a solution: > > Function GetEnv (var Name: __CString__ ): pChar; C; > ^^^ > You get a warning about incompatible pointers, but it works ... > Please remove this stupid warning and I'm happy. Yes, I know - I can use > {$W-}, but it's not so elegant ;-) It's getting more and more strange ... this (with "var") should *not* work: __CString__ already denotes a pointer as required by the C library function, and with "var" you pass a pointer to that pointer ... no, the latter was wrong. Your function *expects* a pointer to the pointer, but you pass the pointer itself, so you get the warning, but you pass the correct thing. It works. ;-) However the correct thing would still be Function GetEnv ( Name: __CString__ ): pChar; C; without "var". You are using DJGPP, correct? There it worked on my computer. Try the "GetEnv" from BO5s (no apostrophy: it's a German abbreviation ;) "Tools" Unit. (Must probably isolate it because BO5 is not (yet!) ready to work with DJGPP.) It worked here also. Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

Re: Call for installation patches and hacks.
by Kevin A. Foss 17 Oct '96

17 Oct '96

On Thu, 17 Oct 96 11:47:26 CST, J.J. van der Heijden wrote: >On Thu, 17 Oct 1996 00:05:21 +0100 (MET), >Peter Gerwinski <peter(a)agnes.dida.physik.uni-essen.de> wrote: > >>According to Kevin A. Foss: >>> Specifically, even though gcc_bcmp.c is copied into the >>> RTS directory, it is never referenced in the makefile. (MakeHPFS.RTS) >> >>You are right! =:-O >> >I never understood what gcc_bcmp.[c|o] did in the RTS. It comes from >libgcc2.a which ends up in libgcc.a. Every app is linked to libgcc.a, so >why duplicate it in libgpc.a? This doesn't seem to be the case in EM's port of gcc. All that exists is the bcmp() based on the BSD implementation. There is no gcc_bcmp in any library that I can find. Peter said later: >I don't understand either, but I got error messages when trying to >compile Pascal programs without it, at least in 2.6.3. Perhaps the >problems have vanished with 2.7.2 (: should check ;). I have both 2.7.2 and 2.7.2.1 installed of EM's gcc (on different machines) and neither has gcc_bcmp in any libraries, including gcc.a. So it -should- definately complain if gcc_bcmp isn't included in gpc.a when linking. This entire discussion reminds me of my other problem about the OS/2 MakeXXX.RTS files: If you run 'make install' -- it doesn't rename libgpc.a to gpc.a when it is copied and the user has to manually rename it before he can compile anything. -Kevin --- Kevin A. Foss -- kfoss(a)atlsysnet.com ---

1 0

Re: Call for installation patches and hacks.
by J.J. van der Heijden 17 Oct '96

17 Oct '96

On Thu, 17 Oct 1996 00:05:21 +0100 (MET), Peter Gerwinski <peter(a)agnes.dida.physik.uni-essen.de> wrote: >According to Kevin A. Foss: >> I did have to make a couple changes to the OS/2(EMX) makefile in order to get >> everything working. > [...] > >> Specifically, even though gcc_bcmp.c is copied into the >> RTS directory, it is never referenced in the makefile. (MakeHPFS.RTS) > >You are right! =:-O > I never understood what gcc_bcmp.[c|o] did in the RTS. It comes from libgcc2.a which ends up in libgcc.a. Every app is linked to libgcc.a, so why duplicate it in libgpc.a? JanJaap --- "Nothing shocks me, I'm a scientist", Indiana Jones

1 0

How to implement writeln()
by andi.tio＠hit.handshake.de 17 Oct '96

17 Oct '96

Hi I'm currently programming a bp-like CRT unit. But i don't know how to do writeln(), which is not limited, neither in the number of arguments nor in the type of this. Bye Andreas

1 0

proto-FAQ -- reactions please!
by J.J. van der Heijden 17 Oct '96

17 Oct '96

Hello, I gathered some of the recently asked questions, together with some usefull information from various readmes etc. and called it a 'FAQ'. It needs some expanding, especially about the differences between various dialects of the Pascal language. Also, I'm not a native english speaker so there are probably lots of errors. Any feedback is appreciated. It will be texified in the future, so it can be put on a www page. JanJaap =========================================================================== This is the GNU Pascal Frequently-Asked Questions List. Author: J.J. van der Heijden <j.j.vanderheijden(a)student.utwente.nl> Revision: 0.1, Oct 14, 1996 1 What is GNU Pascal? 1.1 What is the current version of GNU Pascal? 1.2 Where is the GNU Pascal FTP site? WWW? 1.3 Is it compatible with Turbo Pascal (R) ? 2 Installation related questions 2.1 Which platforms are supported by GNU Pascal? 2.2 Which components, other than the GNU Pascal compiler do I need to compile Pascal source code? 2.3 Help! linking gpc1 fails: _emit_string_move undefined (and more) 2.4 When I build libgpc.a, rts-rt0.c says: SIGXCPU undefined (and more) 2.5 I do I debug my Pascal programs? 2.6 Can you recommend an IDE? 2.7 Do we have a binary for you? 3 GNU Pascal and your system libraries 3.1 How do I use <insert_your_function_here> from the C library? 3.2 What's this confusion about Pascal and C style strings? 3.3 Are GNU Pascal objects compatible with GNU C++ classes ? 3.4 Where are the Turbo pascal (R) replacement units? 3.5 How do I build/use a shared library? 4 GNU Pascal on the djgpp (MS-DOS) platform 4.1 If you need more information 4.2 What do I download? 4.3 How do I install the compiler? 4.4 GPC says: no DPMI 4.5 I cannot read the info pages! 4.6 I have troubles with inline assembly 4.7 Tell me how to do DPMI, BIOS and other DOS related things. 4.8 I'm accessing an "array[1..4000000] of byte" and I got an exception. 5 Getting Help 5.1 Help! the compiler crashed! 5.2 I think I found a bug -- now what? 5.3 Which newsgroup(s) are suited for GPC related problems? 5.2 How to post to the mailing list 5.3 How to become a subscriber to the mailing list 5.4 How to unsubscribe from the mailing list 6 Miscellany 6.1 I want to contribute, where do I start? 6.2 About this FAQ. 1 What is GNU Pascal? ********************* The purpose of the GNU Pascal project is to produce a Pascal compiler (called GNU Pascal or GPC) which * combines the clarity of Pascal with powerful tools suitable for real-life programming, * supports both the Pascal standard and the Extended Pascal standard as defined by ISO, ANSI and IEEE. (ISO 7185:1990, ISO/IEC 10206:1991, ANSI/IEEE 770X3.160-1989) * supports other Pascal standards (UCSD Pascal, Borland Pascal, Pascal-SC) in so far as this serves the goal of clarity and usability, * may be distributed under GNU license conditions * can generate code and run on any computer for which the GNU C compiler can generate code and run. Pascal was originally designed for teaching. GNU Pascal provides a smooth way to proceed to challenging programming tasks without learning a completely different language. 1.1 What is the current version of GNU Pascal? ============================================== The last official release is GPC 1.1, based on GNU CC version 2.6.3. GPC version 1.2, based on GCC 2.7.2 is currently in beta testing. 1.2 Where is the GNU Pascal FTP site? WWW? ========================================== The master FTP site for GNU Pascal is kampi.hu.fi GNU Pascal related files can be found in: ftp://ftp.kampi.hut.fi/jtv/gnu-pascal/ This site is mirrored on: ftp://sunsite.doc.ic.ac.uk/gnu/pascal/ The latest developer releases can be downloaded from: ftp://agnes.dida.physik.uni-essen.de/ Also, we have a homepage on the web: http://home.pages.de/~GNU-Pascal/ 1.3 Is it compatible with Turbo Pascal (R) ? ============================================ GPC is currently *not* a drop-in replacement for Borland's Turbo Pascal (R). It supports a number, but not all of the Borland extensions to the Pascal language. There is no replacement for most of the Borland runtime library functions. GNU Pascal is part of the GNU project, so portability is one of its primary goals. For this reason, non-portable features of Borland Pascal will probably not be included into GNU Pascal. More information can be found in the file "bpqstart.txt" 2 Installation related questions ******************************** This section discusses some common problems with the installation of GNU Pascal. 2.1 Which platforms are supported by GNU Pascal? ================================================ GPC uses the GCC backend, so it should run on any system that is supported by GNU CC. This includes a large variety of Unix systems, MS-DOS, OS/2 and Win32. A full list of platforms supported by GCC can be found in the file "INSTALL" of the GCC distribution. Not all of these have actually been tested, but the gpc-1.2 pre-release is know to run on these platforms: i486-linux (Linux 2.x, ELF) i486-linuxaout i486-linuxoldld i486-freebsd2.1 i386-cygwin32 (Windows95/Windows NT) i386-djgppv2 (MS-DOS) i386-emx (OS/2, MS-DOS) mips-sgi-irix5.3 >>> Ok people - send us your success stories, with canonical machine name! <<< 2.2 Which components, other than the GNU Pascal compiler do I need to compile Pascal source code? ===================================================================== A complete Pascal compiler system should at least have: * The actual compiler, GPC. * Assembler, linker, librarian and friends. * A C library. * A debugger, if you want to debug your programs. You don't need a C compiler to compile your Pascal programs, but you do need it to build the GNU Pascal compiler itself. GNU Pascal version 1.1 is based on GCC 2.6.3, GPC v1.2 is based on GCC v2.7.2 Any attempt to build GPC with the wrong version of GCC is bound to fail. For most people, the GNU binutils and GNU debugger (gdb) are a good choice, although some may prefer to use vendor specific tools. 2.3 Help! linking gpc1 fails: _emit_string_move undefined (and more) ==================================================================== If linking gpc1 bombs out with an error message that looks like this: ld: Undefined symbol _emit_string_move _emit_string_pad _maybe_find_function_data _dbxout_set_type_status _version_flag *** Error code 1 make: Fatal error: Command failed for target `gpc1' you probably suffer from a VPATH make problem. A few GPC source files have counterparts with identical name in the GCC source directory. When you have built GCC in the GCC source directory and you are not using a recent version of GNU make this problem may occur. There are three solutions: * Get a recent version of GNU make. Version 3.74 or better is known to work. * Build GCC in a seperate directory instead of using the GCC source directory. * Manually delete these files from the GCC object directory: stor-layout.o dbxout.o expr.o fold-const.o optabs.o convert.o function.o setop.o toplev.o then resume `make'. 2.4 When I build libgpc.a, rts-rt0.c says: SIGXCPU undefined (and more) ======================================================================== Compilation of the runtime system may fail in rts-rt0.c with a message simular to this: rts-rt0.c: `SIGXCPU' undeclared (first use this function) or: rts-rt0.c: storage size of `sv' isn't known rts-rt0.c: storage size of `osv' isn't known If this happens to you, you have to edit the rts/rts-config.h file and comment out the last line to: /* #undef HAVE_SIGSYS */ This will be fixed in a future release. 2.5 I do I debug my Pascal programs? ==================================== To debug your programs, (a) GNU Pascal must be able to generate executables with debug info for your platform, and (b) you must have a debugger which understands this. * If `gpc -g -o hello hello.p' says: "gpc: -g not supported for this platform", then GPC is unable to generate debug info. Usually, installing GAS instead of your system's assembler can overcome this. When you configure the GCC used for GPC, specify "--with-gnu-as", and possibly "--with-gnu-ld" and/or "--with-stabs". More information can be found in "INSTALL" file in the GNU CC source directory. * Your system's debugger may not understand the debug info generated by GNU tools. In this case, installing GDB may help. The bottom line: if you can debug GCC compiled programs, you should be able to do this with GPC too. The GNU debugger (GDB) currently does not have a "Pascal" mode, so it is unable to display certain Pascal structures etc. When debugging, please note that the Initial Letter In Each Identifier Is In Upper Case And The Rest Are In Lower Case. If you want to display variable 'foo' in the debugger, type `show Foo' or `display Foo' instead. 2.6 Can you recommend an IDE? ============================= Users of Borland Pascal may wonder if there's a replacement for the IDE (Integrated Development Environment). Here's a few suggestions: * (X)Emacs. Some people think it's the answer to the question of Life, the Universe and Everything, others decide it's uGNUsable. Available from your friendly GNU mirror. * XWPE XWPE is an immitation of the Borland IDE, so users of Borland Pascal may find it a good alternative. Although GDB is an excellent debugger, it's user interface is not very attractive. Refer to the comp.windows.x FAQ: "Where can I get an X-based debugger?" at: http://www.cis.ohio-state.edu/hypertext/faq/usenet/x-faq/part6/faq-doc-2.ht… Some usefull frontends include: XXGDB, tGDB and XWPE. see: http://www.ee.ryerson.ca:8080/~elf/xapps/Q-IV.html Very nice, but resource consuming is the Motif based DDD: http://sol.ibr.cs.tu-bs.de/softech/ddd/ 2.7 Do we have a binary for you? ================================ Currently, we have binaries for these platforms: i486-linux (ELF) i486-linuxaout i486-linuxoldld djgpp V2 (msdos) emx 0.9B (OS/2, msdos) cygwin32 beta16 (Windows95, Windows NT) mips-sgi-irix5.3 3 GNU Pascal and your system libraries ************************************** This section discusses common problems people have when they try to access their system's libraries. 3.1 How do I use <insert_your_function_here> from the C library? ================================================================ GNU Pascal can use every function of your C library, but it may be up to you to write declaration of an external function, before you can use it. Consider the function `sleep'. man(3) sleep reveals: --------------------------------------------------------- NAME sleep - Sleep for the specified number of seconds SYNOPSIS #include <unistd.h> unsigned int sleep(unsigned int seconds); --------------------------------------------------------- This small demo program shows how to use `sleep' in a Pascal program: --------------------------------------------------------- program SleepDemo; type word = __unsigned__ integer; function sleep(seconds: word): word; C; function sleep(seconds: word): word; external; var result : word; begin result := sleep(10); end. --------------------------------------------------------- 3.3 What's this confusion about Pascal and C style strings? =========================================================== It is important not to confuse Pascal and C string types. * The Pascal string schema is a record: STRING = RECORD Capacity : integer; length : integer; string : packed array [ 1..Capacity ] of char; END; 'string' is not 'string(256)', unlike Turbo Pascal. The capacicty must be declared: type MyString = string(256); function MyFunction: MyString; * A C string is an array of char, terminated with a NULL char. C library functions require C, not Pascal style string arguments! Consider this code snippet to convert Pascal style strings to C style and vice versa: --------------------------------------------------------- type word = __unsigned__ integer; TString = string[256]; PChar = ^char; function StrPas(Src: PChar): TString; var S : TString; begin S := ''; if (Src <> NIL) then while ( (Src^ <> chr(0)) AND (length(S) < S.capacity)) do begin S := S + Src^; inc(Word(Src)); end; StrPas := S; end; function StrPCopy(Dest: PChar; Src: String): PChar; var c: integer; p : PChar; begin p := Dest; for c:=1 to length(Src) do begin p^ := Src[c]; inc(word(p)) end; p^ := chr(0); StrPCopy := Dest; end; --------------------------------------------------------- Then this small example will print the PATH: --------------------------------------------------------- function GetEnv(name : PChar): PChar; C; function GetEnv(name : PChar): PChar; external; var pName: PChar; begin getmem(pName, 256); pName := StrPCopy(pName, 'PATH'); writeln('PATH=', StrPas(GetEnv(pName))); freemem(pName, 256); end. --------------------------------------------------------- And this is how you access the 'system()' call in you C library: --------------------------------------------------------- program SysCall; function system(name : PChar): integer; C; function system(name : PChar): integer; external; var pName: PChar; result : integer; begin getmem(pName, 256); pName := StrPCopy(pName, 'ls -l'); result := system(pName); writeln('system() call returned : ', result); freemem(pName, 256); end. --------------------------------------------------------- 3.4 Are GNU Pascal objects compatible with GNU C++ classes ? ============================================================ No. 3.5 Where are the Turbo Pascal (R) replacement units? ===================================================== They don't exist (yet). Most of their fuctionality can easily be implemented, some things are very x86/msdos dependant and would be meaningless on any other platform. 3.6 How do I build/use a shared library? ======================================== (topic under construction) 4 GNU Pascal on the djgpp (MS-DOS) platform ******************************************* This chapter discusses some potential problems with GNU Pascal on MS-DOS, using djgpp. 4.1 If you need more information ================================ GPC/djgpp is a djgpp V2 application, and most of the djgpp documentation applies for GPC too. A great source of information is the djgpp FAQ: http://www.delorie.com/djgpp/v2faq/faq201b.zip Another place to look for DJGPP documentation is the DJGPP Knowledge Base, at this URL: http://www.delorie.com/djgpp/doc/kb/ 4.2 What do I download? ======================= As discussed in section 2.2, other than GPC itself, you need an assembler, linker and friends, a C library and possibly a debugger. From your local djgpp mirror, you can get these as: v2/djdev200.zip (C library) v2gnu/bnu252b.zip (assembler, ....) v2gnu/gdb412b.zip (debugger) The rest is up to you; 'make' (v2gnu/mak372b.zip) can be usefull, RHIDE (an IDE with Borland-look) is nice, but still under development. Future releases of RHIDE may have better GPC support. 4.3 How do I install the compiler? ================================== If you don't have djgpp installed on your harddisk, create a directory for GNU pascal ("c:\gpc"), and unzip the archives. Make sure you preserve the directory structure (use "pkunzip -d"). Now, add the directory where gpc.exe lives ("c:\gpc\bin") to your path and set the DJGPP environment variable to point to your djgpp.env file: set DJGPP=c:\gpc\djgpp.env Specific information for low-memory conditions and mre can be found in the djgpp FAQ and documentation. 4.4 GPC says: no DPMI ===================== You don't have a DPMI server installed, and DJGPP v2 requires it to run. You can either use one of the commercial DPMI servers (e.g., run `gpc' in a DOS box under Windows) or download and install CWSDPMI (`csdpmi3b.zip') which is a free DPMI server written for DJGPP. 4.5 I cannot read the info pages! ================================= To read the info pages, you need the `info' program from txi360b.zip. At least for some of the pre-releases of gpc-1.2, the gpc.info file is invalid: it refers to the "gpc.i*" sections as "gpc.info-*". This can be fixed by loading gpc.inf in an editor and replacing "gpc.info-*" with "gpc.i*" 4.6 I have troubles with assembly code ====================================== The GNU Assembler (`as.exe'), or `gas', called by GCC accepts "AT&T" syntax which is different from "Intel" syntax. Differences are discussed in section 17.1 of the djgpp FAQ. A guide is available which was written by Brennan Mr. Wacko Underwood <brennan(a)mack.rt66.com> and describes how to use inline assembly programming with DJGPP, at this URL: http://www.rt66.com/~brennan/djgpp/djgpp_asm.html Section 17.3 of the djgpp FAQ discusses some methods to convert "Intel" syntax to "AT&T". 4.7 Tell me how to do DPMI, BIOS and other DOS related things. ============================================================== DPMI, BIOS and other functions are no different than other system functions. Refer to section 3.1 how to access your system's C-library. This small example shows how to use DPMI, copying some structures and function prototypes of <dpmi.h>: --------------------------------------------------------- program dpmitest; {$X+} type word = __unsigned__ integer; short = __short__ integer; byte = __byte__ integer; type PDpmiVersionRet = ^TDpmiVersionRet; TDpmiVersionRet = record major : byte; minor : byte; flags : short; cpu : byte; master_pic : byte; slave_pic : byte; end; type PDpmiFreeMemInfo = ^TDpmiFreeMemInfo; TDpmiFreeMemInfo = record largest_available_free_block_in_bytes : word; maximum_unlocked_page_allocation_in_pages : word; maximum_locked_page_allocation_in_pages : word; linear_address_space_size_in_pages : word; total_number_of_unlocked_pages : word; total_number_of_free_pages : word; total_number_of_physical_pages : word; free_linear_address_space_in_pages : word; size_of_paging_file_partition_in_pages : word; reserved1 : byte; reserved2 : byte; reserved3 : byte; end; function DpmiGetVersion(ret: PDpmiVersionRet): integer; asmname '__dpmi_get_version'; function DpmiGetFreeMemoryInformation(info: PDpmiFreeMemInfo): integer; asmname '__dpmi_get_free_memory_information'; function DpmiGetVersion(ret: PDpmiVersionRet): integer; external; function DpmiGetFreeMemoryInformation(info: PDpmiFreeMemInfo): integer; external; var version: TDpmiVersionRet; meminfo: TDpmiFreeMemInfo; begin DpmiGetVersion(@version); writeln('CPU type : ', version.cpu, '86'); writeln('DPMI major : ', version.major); writeln('DPMI minor : ', version.minor); DpmiGetFreeMemoryInformation(@meminfo); writeln('Free DPMI memory : ', meminfo.total_number_of_free_pages, ' pages.'); end. --------------------------------------------------------- 4.8 I'm accessing an "array[1..4000000] of byte" and I got an exception. ======================================================================== Per default, the maximum stack size of a djgpp application is 256K. If you need more, you have to adjust it with the stubedit program, i.e.: stubedit your_app.exe minstack=5000K Still, it might be a good idea to use pointers for this kind of structures. 5 Getting Help ************** This section discusses ways to get help with GNU Pascal. Please read the documentation (info files, readme's) that come with GPC, plus other docs that might help (the djgpp FAQ if you use djgpp etc.) before you send email to the maintainers or mailing list. 5.1 Help! the compiler crashed! =============================== If the compiler crashes, you have discovered a bug. A reliable compiler never crashes. To help the maintainers fix this bug, it is important that you send us a problem report. 5.2 I think I found a bug -- now what? ====================================== Bugs are best reported to the GPC mailinglist, gpc(a)hut.fi. That way, they always reach the maintainers. Try to give as much information as possible, plus a short code snippet that triggers the compiler bug. If you're on unix, you can find out where the compiler crashed if you enable coredumps, then load the compiler (gpc1) plus the core file in the debugger ("gdb /your_path_here/gpc1 core"), then type `backtrace' to get a stacktrace. Include this stacktrace in your bug report. 5.3 Which newsgroup(s) are suited for GPC related problems? =========================================================== There are several Pascal related newsgroups, but none is dedicated just to GNU Pascal. These may be usefull: comp.lang.pascal.ansi-iso Pascal according to ANSI and ISO standards. comp.lang.pascal.misc Pascal in general and ungrouped Pascals. 5.2 How to post to the mailing list =================================== You can send a message to the GPC mailing list by sending email to the list address <gpc(a)hut.fi> as if it were a person. 5.3 How to become a subscriber to the mailing list ================================================== You can join the mailing list by sending a message to <gpc-request(a)hut.fi> (NOT <gpc(a)hut.fi> !) with your request to be added to the list. Maintenance is done by hand, so some delay is possible. 5.4 How to unsubscribe from the mailing list ============================================ To leave the mailing list, send a message to <gpc-request(a)hut.fi>. 6 Miscellany ************ 6.1 I want to contribute, where do I start? =========================================== A list of jobs which should be done for GNU-Pascal can be found in the file `how_to_contrib.txt'. In cases where somebody is already working on a topic, the name of that person is written behind the job's description. This does not mean that you shouldn't do that but just that you should get in contact with that person if you would like to contribute to that field. 6.2 About this FAQ. ================== Maintainer: J.J. van der Heijden <j.j.vanderheijden(a)student.utwente.nl> This is the first incarnation of the GNU Pascal FAQ list. Comments about, suggestions for, or corrections to this FAQ list are welcome. Please make sure to include in your mail the version number of the document to which your comments apply (you can find the version at the beginning of this FAQ list). Much of the info in this FAQ list was taken from the gpc mailing list traffic, so you may have (unbeknownst to you) contributed to this list. EOF --- "Nothing shocks me, I'm a scientist", Indiana Jones

5 4

Re: Call for installation patches and hacks.
by Peter Gerwinski 17 Oct '96

17 Oct '96

> >According to Kevin A. Foss: > >> I did have to make a couple changes to the OS/2(EMX) makefile in order to get > >> everything working. It seems to be an incompatibility between EMX 0.9b and EMX0.9c. According to J.J. van der Heijden: > I never understood what gcc_bcmp.[c|o] did in the RTS. It comes from > libgcc2.a which ends up in libgcc.a. Every app is linked to libgcc.a, so > why duplicate it in libgpc.a? I don't understand either, but I got error messages when trying to compile Pascal programs without it, at least in 2.6.3. Perhaps the problems have vanished with 2.7.2 (: should check ;). Greetings, Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

DJGPP installation problems
by Peter Gerwinski 16 Oct '96

16 Oct '96

Newsgroups: comp.os.msdos.djgpp References: <01bbbab6$1e0cf340$LocalHost@default> Organization: Universitaet Essen, Germany Reply-To: peter.gerwinski(a)uni-essen.de Distribution: Niels Heirbaut (heirbaut(a)pi.net) wrote: > I don't know if this is the right group to post to (I haven't found an > alternative), so please forgive me if this post is off topic. Since this is a DJGPP-specific problem, I think this NewsGroup is appropriate. However, there is a GNU Pascal mailing list, gpc(a)hut.fi. To subscribe, write to gpc-request(a)hut.fi. (This followup is also posted to the GPC mailing list.) > I downloaded the binary files for GPC (version 2.7.2) and extracted them to > my DJGPP directory. But when I tried to compile the program hello.pas (see > below) with: C:\>gpc hello.pas -o hello.exe > > I got the next message: > C:\DJGPP\BIN/ld.exe: cannot open crt0.o: No such file or directory (ENOENT) > > I checked if crt0.o existed (it did). I also tried to put the directory > (C:\DJGPP\LIB) in my PATH-variable, but still the same response. In the > info file I read something about setting the variable GPC_EXEC_PREFIX. I > tried this (in DJGPP.ENV and AUTOEXEC.BAT) but it still didn't work. > Here are my program, my AUTOEXEC.BAT and my DJGPP.ENV files: > > [...] > > DJGPP.ENV: > > [...] > > [gcc] > COMPILER_PATH=%/>;COMPILER_PATH%%DJDIR%/bin > LIBRARY_PATe=%/>;LIBRARY_PATH%%DJDIR%/lib;%DJDIR%/contrib/grx20/lib > > [...] > > [gpc] > GPC_EXEC_PREFIX=%/>;GPC_EXEC_PREFIX%%DJDIR%/lib >From the GPC installation instructions (you might have an older version which does not yet contain this): 8< ------------ Finally, add these lines to your \djgpp\djgpp.env file: --------- cut here ------- cut here --------- [gpc-cpp] C_INCLUDE_PATH=%/>;C_INCLUDE_PATH%%DJDIR%/include [gpc] COMPILER_PATH=%/>;COMPILER_PATH%%DJDIR%/bin LIBRARY_PATH=%/>;LIBRARY_PATH%%DJDIR%/lib 8< ------------ Hope this helps, Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

Word and LongWord
by Peter Gerwinski 16 Oct '96

16 Oct '96

According to Andreas Bauer: > JJvdH> type > JJvdH> word = __unsigned__ integer; > You should call it longword. > I had some trouble, because I declared a word like this and i wondered > that the entries after flags had some errors :) > Perhaps longword... A matter of taste and self-discipline while counting bits. ;-) I prefer "word" to stand for the "natural" unsigned integer - which has 32 bits with GNU Pascal. (But I also had trouble with this being a Borland Pascal programmer for many years ... ;-) Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

GPC Bugs ....
by Phil Nelson 16 Oct '96

16 Oct '96

Hello GPC fans, Recently a gnats database was set up for GNU Pascal. Gnats is the GNU bugtracking system. This will allow the developers of GNU Pascal follow bugs and bug fixes much better. It will also allow users of GNU Pascal to see if a bug has been reported before and/or fixed. You many both search the bug database and submit new bug reports at the URL http://fawn.cs.wwu.edu/~phil/gpc. A link to this site will be added to the GNU Pascal home page soon. (The GNU Pascal home page is http://agnes.dida.physik.uni-essen.de/~gnu-pascal.) Currently, a report of a bug to the GNU Pascal gnats site will send a message to gpc(a)hut.fi so we can all see which bugs have been reported. -- Phil Nelson NetBSD: http://www.netbsd.org e-mail: phil(a)cs.wwu.edu LPF: http://www.lpf.org http://www.cs.wwu.edu/~phil !gifs: http://www.lpf.org/Patent/Gif/Gif.html

1 0

Re: Call for installation patches and hacks.
by Peter Gerwinski 16 Oct '96

16 Oct '96

According to Kevin A. Foss: > I did have to make a couple changes to the OS/2(EMX) makefile in order to get > everything working. :-( That's strange because I made *a lot* of tests with the EMX version on FAT partitions as well as on HPFS partitions to ensure that you just type [I:\emx\gnu\gpc-272] configure (* configure.cmd *) and [I:\emx\gnu\gpc-272] make (* make.cmd - calls dmake *) and GPC compiles through (assumed that you have compiled GCC before). Are you sure that you aren't using the 2.6.3 EMX patch? (Of course. ;-) Please send me your modified Makefile. > Specifically, even though gcc_bcmp.c is copied into the > RTS directory, it is never referenced in the makefile. (MakeHPFS.RTS) You are right! =:-O > It was > only after I tried to compile some code that apparantly needed _gcc_bcmp() > that I noticed the problem. So I had to go back and edit the Makefile. > > Or did I do something wrong? I couldn't find any reference to gcc_bcmp.c/.o > in the MakeFAT.RTS file either. Again: Please send me your modified Makefile. > I also experienced confusion over which version of Make to use. INSTALL says > to use nmake, but README.EMX says to use dmake. `dmake' is correct. > (Sorry if this has been addressed before, I just recently joined the list..) No, it's a new bug. :-)-: Thank you for the report. Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

extend bug
by Jim Brander 16 Oct '96

16 Oct '96

Running gpc 2.6.3 on ELF format Linux 1.2.13 _p_extend() in rts/rts-file.c has following problem. Program is writing out 3000 line file, extending file for each line (don't grumble, it works this way). File extends OK for several hundred lines (a line or 2 each time), then the contents of the file buffer is spilled, say 20 lines written twice but mucked up on line endings, then extends OK for several hundred more lines, then does it again. The file buffer changes to a different address when spill occurs - that is, while it happens to be the same buffer address used each time, no problem. Doesn't display this behaviour on gpc 2.6.3 on Linux 1.1.59 with a.out format. Is fixed by commenting out line in rts/rts-file.c in _p_extend() int ch = getc(m_FILNUM(File)); File pointer is already pointing at end of file without this line. Regards Jim

1 0

Re: Call for installation patches and hacks.
by Kevin A. Foss 15 Oct '96

15 Oct '96

On Tue, 15 Oct 96 15:33:43 CST, J.J. van der Heijden wrote: >We are finishing up preparations for the "final" 1.2 GPC release, and now I >would like to know, did any of you have to apply special hacks to get GPC >up and running? Need to edit a config file, Makefile etc., add special >CFLAGS? I did have to make a couple changes to the OS/2(EMX) makefile in order to get everything working. Specifically, even though gcc_bcmp.c is copied into the RTS directory, it is never referenced in the makefile. (MakeHPFS.RTS) It was only after I tried to compile some code that apparantly needed _gcc_bcmp() that I noticed the problem. So I had to go back and edit the Makefile. Or did I do something wrong? I couldn't find any reference to gcc_bcmp.c/.o in the MakeFAT.RTS file either. I also experienced confusion over which version of Make to use. INSTALL says to use nmake, but README.EMX says to use dmake. (Sorry if this has been addressed before, I just recently joined the list..) -Kevin --- Kevin A. Foss -- kfoss(a)atlsysnet.com ---

1 0

STABS format -- solaris users around?
by J.J. van der Heijden 15 Oct '96

15 Oct '96

Hello, I have started a project to add a "Pascal" mode to GDB. This means GDB can demangle Pascal symbols, and evaluate Pascal expressions. The debug format for most platforms is stabs, and exploit all possibilities, GPC and GDB have to support a number of Pascal specific stabs. These are rumoured to be ducumented by SUN for their SUN pascal compiler, and this document might be on the Solaris SDK CD-ROM. So, if any of you have this CD-ROM around, and could pass me this document, maybe I can follow an existing standard, instead of reinventing the wheel. I don't care about the format, I can handle most formats between raw text and PostScript. Sorry for my vagueness, JanJaap --- "Nothing shocks me, I'm a scientist", Indiana Jones

1 0

Call for installation patches and hacks.
by J.J. van der Heijden 15 Oct '96

15 Oct '96

Hello, We are finishing up preparations for the "final" 1.2 GPC release, and now I would like to know, did any of you have to apply special hacks to get GPC up and running? Need to edit a config file, Makefile etc., add special CFLAGS? We know about the problems with stor-layout and friends (undefined symbols when linking gpc1) and BSD/SYSV confusion with signals in rts/rts-rt0.c We would like to hear about any other hacks you had to do, and this is your chance to get it fixed before the release. In the ideal world, `configure ; make' should work. I have heard people failing to build GPC on Solaris. Is this still true, and if yes, can you give me a detailed description what goes wrong? Greetings, JanJaap --- "Nothing shocks me, I'm a scientist", Indiana Jones

1 0

getenv() and system()
by Peter Gerwinski 15 Oct '96

15 Oct '96

According to John Michael Black: > IMHO, I would think that the function getenv() and the procedure system() > would be two widely used routines for a lot of people's programs. Instead > of expecting everyone to "create" them each time, what possibility is > there of including them in a standard library? My BO5 project is intended to become a standard library for GNU Pascal. Its Unit "Tools" already contains a Borland- compatible version of getenv(). > I know they're not > standard or extended pascal, GetEnv is in Borland's DOS Unit, so it's de-facto-standard on DOS-like platforms. Berend already ported some features of DOS to an Extended Pascal module, other things are in BO5, so it should be straightforward to write (at least parts of) a true Borland-compatible DOS Unit. > but like I said it seems like a lot of people > would need to use them. If there isn't an include for operating system > functions, perhaps this could start one. Includes are not part of ISO-Pascal, modules are. With Borland or UCSD Pascal, you can include a file via (*$I foo *) (supported by GPC, too). However in my opinion, a Module or Unit would be preferable. If somebody would like to write such a thing (or already has written), please let me know. Greetings, Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

Bug report :-)
by Jukka Virtanen 15 Oct '96

15 Oct '96

Reported by: Juki <jtv(a)hut.fi> GPC version: 1.2-2.7.2.1 Platform : Digital Alpha, DEC unix 4.0 Description: ------------ The attached program does not compile. narya gpc-1.2-2.7.2.1 58 % gpc fail.p fail.p: In function `zap': fail.p:5: parse error before `Absolute' Attached program: ----------------- program zap; var enumvar : (absolute); begin end.

1 0

getenv() and system()
by John Michael Black 15 Oct '96

15 Oct '96

Thanks so much, JJ, for the sample code for these two functions in the FAQ. In all I thought the faq was good and informative. ---------- IMHO, I would think that the function getenv() and the procedure system() would be two widely used routines for a lot of people's programs. Instead of expecting everyone to "create" them each time, what possibility is there of including them in a standard library? I know they're not standard or extended pascal, but like I said it seems like a lot of people would need to use them. If there isn't an include for operating system functions, perhaps this could start one. For example, right now I've taken these two external makeshift routines (along with some others) and put them in a file "unix.h", so all I do is #include <unix.h> in programs I know will use them. (I'm considering renaming it to <op-sys.h>, since most of them could probably work on DOS as well.) I would reccommend using the literal names "getenv" and "system" since they are familiar from C. Any thoughts? -john

1 0

Shared images, was: Re: OS/2 DLL's
by J.J. van der Heijden 14 Oct '96

14 Oct '96

On Sun, 13 Oct 1996 11:06:07 +0100 (MET), Peter Gerwinski <peter(a)agnes.dida.physik.uni-essen.de> wrote: >> After some experimenting, I produced an OS/2 dynamic link library (DLL) >> with GNU Pascal that I could call from another application. Perhaps this >> is common knowledge but I do not see it documented anywhere. > >Thank you. I didn't know this either. > >I planned to implement Borland's style of creating >DLLs into GPC. So with EMX, GPC will have to create >a correct .def file as well. > >Probably somebody here knows how to compile a shared >library for Linux or other UNIXes with GPC? > Shared libraries are very platform dependant. 1) Win32 and OS/2 have the .def files, and need seperate import libraries to link an app to a shared image (.dll) 2) ELF unices don't have all of the above, you just use gcc -shared -o libfoo.so.X.Y your_obj_files_here to build a shared image. ELF shared images export everything so no .def is needed. If you have a 'libfoo.so.X.Y' shared image, you link it simply with -lfoo. Several very nifty features exist, to help object oriented languages, to execute initialization code when a shared image is loaded and finalized (unloaded). Look at the sources of the Objective C base library if you want to know more. 3) Some COFF or A.OUT systems (BSD, older Linux flavours) can use shared images too, but at least Linux needed special tools to do so. The Win32 dll's must be `pure' (no global vars), but this restriction doesn't exist for ELF unix. I think you should just leave the DLL stuff as it is, enforcing a Borland PC standard would cripple the possibilities of the ELF for instance. The winapi-0.1.2.tar.gz archive in the GNU archives has a few smart `sed' scripts to generate .def files from a DLL. For cygwin32, the dlltool.exe program can build the import libraries. I have done some attempts to build the GNU Pascal RTS (libgpc.a) as a shared library. For ELF unix, this a very easy. A `hello world' program is reduced to 2K. For win32, several global variables ('input', 'output', gpc_argv and more) make this impossible. I have some plans to rework the RTS, split it in 'ISO', 'ISO extended' and add 'Borland System.tpu' to it. Purify the code. Add documentation fragments to the code, so a `chew' program can generate a rts.texi documentation file. Someday... Just my HFL 0.02, JanJaap --- "Nothing shocks me, I'm a scientist", Indiana Jones

1 0

gpc under SunOS or Solaris
by Friedrich Fahnert 14 Oct '96

14 Oct '96

Has anybody gotten gpc to run on sunOS or on Solaris? I managed to get version 1.2-2.7.2 compiled on both, but neither works very well. On Solaris gpc1 gets a fatal signal, and an internal compiler error, when compiling certain programs. On SunOS, it seems to compile properly, but it dumps core when linking. Is there any platform where gpc coud be considered to be stable? I'd consider getting another machine, if I could just get gpc to run correctly on it. Friedrich

2 2

Re: External, C, AsmName, ...
by Berend de Boer 14 Oct '96

14 Oct '96

> Where is it? > Since GPC is essentially BP-compatible now, perhaps we can > just use it? Good question. I'm not sure. It probably is on CompuServe. I think the author is Bob Swart. It could also be it is only for Delphi. I see if I can get some more info. Groetjes, Berend.

1 0

Re: gpc for Solaris 2.5.1
by Peter Gerwinski 14 Oct '96

14 Oct '96

> Is there a version of gpc built for Solaris 2.5? Or do you > have any hints on getting the lastest source to compile > under Solaris 2.5? The 2.7.2-pre-release should (;-) be straightforward to compile under any system. First, compile the C compiler gcc-2.7.2 and let the object files in place. Then, unarchive gpc-1.2-2.7.2.tar.gz, cd to the gpc directory, type "./configure" and "make" -- voil`a. If you are successful, please let me know, so I can include the binaries into the distribution. Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

Re: External, C, AsmName, ...
by Peter Gerwinski 14 Oct '96

14 Oct '96

According to Arcadio Alivio Sincero: > Hey ... did y'all know that a Linux version of FPKPascal (or is it > just FPPascal??) is now out (currently only a.out but and ELF version is FPKPascal. The author's name is Florian Klaempfl, therefore the FK. What the P stands for, I don't know. > expected soon). One thing I like about FPKPascal is that it has > function overloading. Any plans of adding this to GPC? This would be a > nice thing to throw in, IMHO. I have planned to add this feature together with qualified identifiers not too far in the future. Nice to see that I am not the only one who likes it. Somebody who would like to help implementing it? Or to solve some other tasks (e.g. write documentation), so I am free for it? Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

OS/2 DLL's
by Peter Gerwinski 13 Oct '96

13 Oct '96

> After some experimenting, I produced an OS/2 dynamic link library (DLL) > with GNU Pascal that I could call from another application. Perhaps this > is common knowledge but I do not see it documented anywhere. Thank you. I didn't know this either. I planned to implement Borland's style of creating DLLs into GPC. So with EMX, GPC will have to create a correct .def file as well. Probably somebody here knows how to compile a shared library for Linux or other UNIXes with GPC? Somebody willing to write a documentation section about this? Greetings, (-; Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

OS/2 DLL's
by phy0a0＠spf1.power.uni-essen.de 12 Oct '96

12 Oct '96

Hello, After some experimenting, I produced an OS/2 dynamic link library (DLL) with GNU Pascal that I could call from another application. Perhaps this is common knowledge but I do not see it documented anywhere. The following seems to work compiled with the latest emx distribution of GPC using "gpc -Zomf -Zdll cube.pas cube.def": unit cube; { a simple function in a DLL} interface function cubit(x:integer):integer; implementation function cubit(x:integer):integer; begin cubit := x*x*x; end; end. You have to be careful in the module definition file to list the function with only the first letter capitalized or LINK386.EXE will complain: /* module definition file for Cube dll */ LIBRARY CUBE EXPORTS Cube @1 Allan D. Derickson -- Derickson Forestry Services -- Freeland, WA USA ----------------------------------------------------------- alland(a)whidbey.com ----------------------------------------------------------- ******************************** ACHTUNG: Neue e-mail-Adresse! Ab dem 22. Juni 1996 bitte an peter.gerwinski(a)uni-essen.de schreiben. ******************************** ATTENTION: New e-mail address! >From 22. June 1996 on please write to peter.gerwinski(a)uni-essen.de ******************************** -------------------------------------------------------------------------------- Dipl. Phys. Peter Gerwinski Fachbereich Physik Universitaet-GH Essen Phone: +49-201-183-2763 D-45117 Essen Fax: +49-201-183-2120 Germany e-mail: peter.gerwinski(a)uni-essen.de --------------------------------------------------------------------------------

1 0

Re: it's running :-)
by Peter Gerwinski 12 Oct '96

12 Oct '96

According to Andreas Bauer: > Diese Nachricht antwortet auf einen Text, > den j.j.vanderheijden(a)student.utwente.nl am Freitag, den 11.10.96, > durch die Datennetze geschossen hat: > > JJvdH> This was a remark by Peter, but he was probably confusing unix and= > dos. > JJvdH> Not sure you can redirect stderr in DOS. > Yes, i read something it's impossible in DOS. > :-( No, that was no confusion; I really meant DOS. It actually works with a DOS box in OS/2. However, it seems *not* to work with MS-DOS nor with Novell DOS. :-( In the djgpp NewsGroup somebody said there would be a recepy how to redirect stderr with DOS in the djgpp FAQ. > If it is possible to disable those ugly underscore-warnings... ?! Yes, with (*$W-*) ... (*$W+*) you can switch off/on warnings for a certain section of the code. (However I don't like __underscored_identifiers__ either. ;-) > No, i'll try to write my own system-unit and I'll try to access > the LFB of an VESA card. Did you try Berend's System unit? (And don't forget to send us your useful units so we can publish it. ;-) Bye, Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

Re: it's running :-)
by andi.tio＠hit.handshake.de 12 Oct '96

12 Oct '96

Diese Nachricht antwortet auf einen Text, den j.j.vanderheijden(a)student.utwente.nl am Freitag, den 11.10.96, durch die Datennetze geschossen hat: Hallo J.J. JJvdH> >I must include djgpp.lnk manually, though. JJvdH> When I said 'specs' I meant 'djgpp.env' JJvdH> The 'specs' and 'specs.gpc' file should not be touched! I copied the [gcc] section of *djgpp.env* to [gpc]. And the linker ld cannot find something with ..._ctor_... and _dtor_. until i include djgpp.lnk manually into linking. JJvdH> This was a remark by Peter, but he was probably confusing unix and dos. JJvdH> Not sure you can redirect stderr in DOS. Yes, i read something it's impossible in DOS. :-( JJvdH> I only heard rumours, but if you watch the djgpp newsgroup JJvdH> (comp.os.msdos.djgpp) or it's archive at http://www.delorie.com you may JJvdH> find it. Maybe it's in the contrib section of the djgpp archive at JJvdH> ftp.delorie.com I'll try it. JJvdH> ==================================== program dpmitest; JJvdH> JJvdH> It's a little C-ish style (ulong's etc) but if you look at <dpmi.h> JJvdH> from djgpp's C-lib you will see how I did it. I noticed that the GNU-integer is *not* the same as the BP-integer. :-) JJvdH> BTW optimization kill's this sample, maybe because GPC aligns the JJvdH> byte size variables in the records to 32bit boundaries for speed then JJvdH> (??!) It seems that the «ret»-instruction at the end of the program causes the exception. if you use a halt() as the last command in the source-file, it will work correctly. A bug in the compiler ? JJvdH> The 'asmname' could have been just a 'C' directive, but then you would JJvdH> have had functions called __dpmi_.... in your pascal program. Why not? If it is possible to disable those ugly underscore-warnings... ?! JJvdH> crazy memory issues in DOS. I'm not sure you need all this DPMI stuff JJvdH> in the first place. Unless you want to write protected mode interrupt JJvdH> handlers, in which case you'd better take a good look in the djgpp JJvdH> docs..... No, i'll try to write my own system-unit and I'll try to access the LFB of an VESA card. JJvdH> JanJaap Bye Andreas

1 0

Re: it's running :-)
by J.J. van der Heijden 11 Oct '96

11 Oct '96

On Tue, 8 Oct 1996 21:04:16 +0100 (MET), Peter Gerwinski <peter(a)agnes.dida.physik.uni-essen.de> wrote: >According to Andreas Bauer: >> - I've installed GPC into H:\GPC including DJGPP and the BINUTILS, >> I think it's all v2. When running gpc.exe says: >> ld.exe: cannot open -lgpc: No such file or directory (ENOENT) >> Do i have to pass params to gpc or to set an environment-var ? > >DJGPP (including GPC) searches for libraries and such things in a >specific directory structure. Normally, an environment variable >points to the DJGPP directory, e.g. "H:\DJGPP", where there is a >configuration file for the rest of the story. If you set it up >in another directory tree structure, you have to specify library >pathes either manually (-L) or by another environment variable >which should be in the DJGPP documentation. > There's a more subtle way to do this: Edit the "specs" file in your main djgpp directory. Locate the section [gcc] and copy it into a new section [gpc] The variables in each section are passed to this program as environment variables, but they don't clutter your real environment. You have to set one env. var "DJGPP" (case sensitive if you're in win95!) to point to the directory where "specs" lives. This is all documented in the DJGPP FAQ >> - I noticed that the output of GPC can't be redirected to a file. >> Is there any posibility to write the compiler-output into a file ? > >Yes: gpc foo.pas -o foo.exe 2>foo.err > >Errors go to the "error device", therefore ">foo.err" is not sufficient. > >> - Can you tell me if it's possible to use standard x86-assembly with >> gpc? Or is there at least a complete documentation of the gnu-assembler >> anywhere ? I can hardly get used to it :( > >I have the same problem. Look into the GNU C documentation, section >"Extensions to ASM" or similar. There is a description, but I had >problems to understand it. Anybody out there who could explain >to both of us how to use that inline assembler syntax? > It's AT&T syntax, not intel. Essentially, you have to read the arguments to an instruction in reverse. I think there was some explanation of this in the djgpp docs too (not sure), and there's a intel->at&t asm converter around somewhere (but probably won't do inline GPC asm). >> - What must be done so that i can access the DPMI-interface through pascal >> without using assembly language ? > >I think you *must* use assembly language. Interrupts and such things >are not in Pascal's language standard. However it would be nice to have >a Unit with a function (written in assembler) which solves this task ... > There are a number of dpmi functions in the djgpp's C library. You _should_ be able to use these, if you declare them properly as external C style functions. These are the functions who's names start with __dpmi. I say _should_ because I have never tried this. >> Is there any docu about that topic ? > >I don't know of such a thing. There could be some hints in the DJGPP >and/or EMX documentation. > Yes, djgpp's DPMI functions are all documented in the djgpp C lib reference. (libc.inf) Please send us any working snapshots of code, then we have documentation! JanJaap --- "Nothing shocks me, I'm a scientist", Indiana Jones

2 1

Re: gpc-1.2-2.7.2
by J.J. van der Heijden 10 Oct '96

10 Oct '96

On Thu, 10 Oct 1996 12:15:33 +0200, schuh(a)informatik.uni-siegen.de <schuh(a)informatik.uni-siegen.de> wrote: >Hello, > >I have a Problem with the gpc-1.2-2.7.2 under Sun's SunOS 4.1.4. > >The first Problem was to compile gpc: >------------------------------------- >... >../gcc-2.7.2/obstack.o `if [ -f "../gcc-2.7.2/alloca.o" ]; then echo >"../gcc-2.7.2/alloca.o" ; else true ; fi` `if [ -f "../gcc-2.7.2/malloc.o" ]; then echo >"../gcc-2.7.2/malloc.o" ; else true ; fi` >collect2: ld returned 2 exit status >ld: Undefined symbol > _emit_string_move > _emit_string_pad > _maybe_find_function_data > _dbxout_set_type_status > _version_flag >*** Error code 1 >make: Fatal error: Command failed for target `gpc1' > > > >Now I delete > stor-layout.o dbxout.o expr.o fold-const.o optabs.o convert.o function.o toplev.o >from the gcc-source and I can build the Pascal-Compiler. > These are the symptoms of a VPATH problem of your `make' utility. You have two options: 1) ugrade to a recent version of GNU make 2) use different directories for GCC sources and object code when you build GCC > > >My second Problem ist to test the Compiler from the 'gpc/test'-directory: >-------------------------------------------------------------------------- >[schuh@pi20]test make >for x in head.pas nohead.pas cstparam.pas switches.pas bitmanip.pas incdec.pas minimax.pas mem.pas abso.pas asmnames.pas openarr.pas varrec.pas ; do \ > rm -f a.out 2>/dev/null ; \ > ../gpc -B../ -L../rts -g -O3 --automake="-g -O3 -B../" $x 2>/dev/null ; \ > echo -n "$x: " ; \ > if [ -f "a.out" ] ; then ./a.out ; else echo "failed" ; fi ; \ >done ; \ >rm myunit.o myunit.gpi a.out 2>/dev/null ; >*** Error code 1 >make: Fatal error: Command failed for target `bptest' > > >What make I wrong? >I hope that anybody can help me to fix this problem! Some more output would be nice, so we know which test(s) failed. Please add "-v" to the compiler flags in the Makefile, then rerun the tests. --- "Nothing shocks me, I'm a scientist", Indiana Jones

1 0

getenv()
by John Michael Black 10 Oct '96

10 Oct '96

OK, I got the declarations to compile perfectly, but how do I use the function? It's not compatible with string types, so I can't do any assigning or string-testing. Here is how I declare the function: function getenv(s: __CString__): __CString__; C; Is there a way to assign the result into a string? -john

1 0

gpc-1.2-2.7.2
by schuh＠informatik.uni-siegen.de 10 Oct '96

10 Oct '96

Hello, I have a Problem with the gpc-1.2-2.7.2 under Sun's SunOS 4.1.4. The first Problem was to compile gpc: ------------------------------------- ... ../gcc-2.7.2/obstack.o `if [ -f "../gcc-2.7.2/alloca.o" ]; then echo "../gcc-2.7.2/alloca.o" ; else true ; fi` `if [ -f "../gcc-2.7.2/malloc.o" ]; then echo "../gcc-2.7.2/malloc.o" ; else true ; fi` collect2: ld returned 2 exit status ld: Undefined symbol _emit_string_move _emit_string_pad _maybe_find_function_data _dbxout_set_type_status _version_flag *** Error code 1 make: Fatal error: Command failed for target `gpc1' Now I delete stor-layout.o dbxout.o expr.o fold-const.o optabs.o convert.o function.o toplev.o from the gcc-source and I can build the Pascal-Compiler. My second Problem ist to test the Compiler from the 'gpc/test'-directory: -------------------------------------------------------------------------- [schuh@pi20]test make for x in head.pas nohead.pas cstparam.pas switches.pas bitmanip.pas incdec.pas minimax.pas mem.pas abso.pas asmnames.pas openarr.pas varrec.pas ; do \ rm -f a.out 2>/dev/null ; \ ../gpc -B../ -L../rts -g -O3 --automake="-g -O3 -B../" $x 2>/dev/null ; \ echo -n "$x: " ; \ if [ -f "a.out" ] ; then ./a.out ; else echo "failed" ; fi ; \ done ; \ rm myunit.o myunit.gpi a.out 2>/dev/null ; *** Error code 1 make: Fatal error: Command failed for target `bptest' What make I wrong? I hope that anybody can help me to fix this problem! bye Frank _\\|//_ (' O-O ') ---------------------------ooO-(_)-Ooo--------------------------- < email: schuh(a)informatik.uni-siegen.de >

1 0

getting/setting environment variables
by Peter Gerwinski 10 Oct '96

10 Oct '96

According to John Michael Black: > Okay, here's another question. Is there a standard in Pascal that > retrieves and/or sets environment variables, like C's getenv()? If not, I > suppose I would declare and external procedure just like the others. I don't know of a de-jure Pascal standard for getenv(). Borland uses the C standard: Function GetEnv ( Index: String ): String; The "Tools" Unit of BO5 implements Borland's GetEnv by using C's one, Greetings, Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

external system() call
by Peter Gerwinski 10 Oct '96

10 Oct '96

According to John Michael Black: > procedure system(s: string); C; Use __CString__ instead of String. String is a Schema type, __CString__ a pointer to chars. Greetings, Peter

1 0

getting/setting environment variables
by John Michael Black 10 Oct '96

10 Oct '96

Okay, here's another question. Is there a standard in Pascal that retrieves and/or sets environment variables, like C's getenv()? If not, I suppose I would declare and external procedure just like the others. -john

1 0

it's running :-)
by Peter Gerwinski 09 Oct '96

09 Oct '96

According to Andreas Bauer: > - I've installed GPC into H:\GPC including DJGPP and the BINUTILS, > I think it's all v2. When running gpc.exe says: > ld.exe: cannot open -lgpc: No such file or directory (ENOENT) > Do i have to pass params to gpc or to set an environment-var ? DJGPP (including GPC) searches for libraries and such things in a specific directory structure. Normally, an environment variable points to the DJGPP directory, e.g. "H:\DJGPP", where there is a configuration file for the rest of the story. If you set it up in another directory tree structure, you have to specify library pathes either manually (-L) or by another environment variable which should be in the DJGPP documentation. > - I noticed that the output of GPC can't be redirected to a file. > Is there any posibility to write the compiler-output into a file ? Yes: gpc foo.pas -o foo.exe 2>foo.err Errors go to the "error device", therefore ">foo.err" is not sufficient. > - Can you tell me if it's possible to use standard x86-assembly with > gpc? Or is there at least a complete documentation of the gnu-assembler > anywhere ? I can hardly get used to it :( I have the same problem. Look into the GNU C documentation, section "Extensions to ASM" or similar. There is a description, but I had problems to understand it. Anybody out there who could explain to both of us how to use that inline assembler syntax? > - What must be done so that i can access the DPMI-interface through pascal > without using assembly language ? I think you *must* use assembly language. Interrupts and such things are not in Pascal's language standard. However it would be nice to have a Unit with a function (written in assembler) which solves this task ... > Is there any docu about that topic ? I don't know of such a thing. There could be some hints in the DJGPP and/or EMX documentation. > - I created a program accessing an "array[1..4000000] of byte". > All i got was an exception. How large must the stack be for this > application and why the stack ? I don't know why the stack is used, and I will change it (and use the data segment instead) unless somebody can tell me why the stack is preferable for the main program. However, if you put the array into a Unit (or Module), it will be accepted. Greetings, Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

2 1

external system() call
by John Michael Black 09 Oct '96

09 Oct '96

Perhaps I am not familiar enough with external calls, but I cannot get the following program to do anything: -------- program test_system; procedure system(s: string); C; begin system('ls -l'); end. -------- It is not a question of writing output to the screen, because even commands like "mkdir" do nothing. Any suggestions? -john

1 0

it's running :-)
by andi.tio＠hit.handshake.de 08 Oct '96

08 Oct '96

Hi I'm a former (experienced) BP programmer and i'd like to move to GPC. Here are a few questions: - I've installed GPC into H:\GPC including DJGPP and the BINUTILS, I think it's all v2. When running gpc.exe says: ld.exe: cannot open -lgpc: No such file or directory (ENOENT) Do i have to pass params to gpc or to set an environment-var ? - I noticed that the output of GPC can't be redirected to a file. Is there any posibility to write the compiler-output into a file ? - Can you tell me if it's possible to use standard x86-assembly with gpc? Or is there at least a complete documentation of the gnu-assembler anywhere ? I can hardly get used to it :( - What must be done so that i can access the DPMI-interface through pascal without using assembly language ? Is there any docu about that topic ? - I created a program accessing an "array[1..4000000] of byte". All i got was an exception. How large must the stack be for this application and why the stack ? Thank you for any hints. Bye Andreas

1 0

Re: Known GPC bugs
by JanJaap van der Heijden 07 Oct '96

07 Oct '96

At 12:57 PM 9/23/96 +0200, you wrote: >Hello folks, > >I updated the "known bugs" section of the Texinfo documentation >of the (experimental) gpc-1.2-2.7.2 pre-release. The new file >"bugs.texi" can be found at > Maybe we should get one of those systems where people can file bug reports in HTML forms. I have seen these before, but can't remember ever having seen a ready-to-go distribution. Anybody got any ideas? JanJaap -- The latest & greates in software, hardware and manswear -- Bono Vox (U2)

2 1

OS/2 PM programs and DLL's
by Peter Gerwinski 07 Oct '96

07 Oct '96

According to Allan Derickson: > Is it possible at this point to make OS/2 Presentation Manager programs > with GNU Pascal? Yes, just as with GNU C in the EMX environment. Look into the EMX documentation and example programs (written in C). Use the "asmname" directive to access OS/2 API functions. I will release a version of BO5 which runs with the OS/2 pm not too far in the future. Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

OS/2 PM programs and DLL's
by alland＠whidbey.com 06 Oct '96

06 Oct '96

Hello, Is it possible at this point to make OS/2 Presentation Manager programs with GNU Pascal? Same question with OS/2 DLL's. If so, what is the format of the code? I have been totally unsuccessful with both. Allan D. Derickson -- Derickson Forestry Services -- Freeland, WA USA ----------------------------------------------------------- alland(a)whidbey.com -----------------------------------------------------------

1 0

subscribe sixconse@sc.ehu.es
by Ikasle Kontseilua 05 Oct '96

05 Oct '96

subscribe sixconse(a)sc.ehu.es

1 0

Double and uses...
by Peter Gerwinski 04 Oct '96

04 Oct '96

According to Alessandro Graizzaro: > I'm new with gnu pascal, and with Linux too. > I decided to try it, because I need a lot of memory for my computations. > 1) Can anoybody of tell me how can I define DOUBLE type ? "Real" *is* what you know as "Double" from Borland Pascal. Borland's 6-byte software real does not exist with GNU; "Single" is "__short__ Real", "Extended" is "__long__ Real". Known bug: you cannot writeln __short__ or __long__ Reals at the moment, but have to convert them to "ordinary" Reals, first. > 2) Is it possible to use the USES clause as in TP ? Yes. See README.TURBO for details (2.6.3 version) or the "From Borland to GNU Pascal" section of the INFO documentation of the 2.7.2-beta version. > Where to find something > like CRT, GRAPH ...? Graph is worked on for the DOS platform. For Linux, you can use the VGAlib C library with "external", "C", and "asmname" Procedures and Functions. CRT does not yet exist -- we are looking for somebody to implement it. However, have a look at BO5 (in the turbo-alpha directory). Its "Events" and "Displays" Units provide important parts of CRT's functionality. Hope this helps, Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

Double and uses...
by Alessandro Graizzaro 04 Oct '96

04 Oct '96

-- [ From: Alessandro Graizzaro * EMC.Ver #2.5.02 ] -- I'm new with gnu pascal, and with Linux too. I decided to try it, because I need a lot of memory for my computations. 1) Can anoybody of tell me how can I define DOUBLE type ? 2) Is it possible to use the USES clause as in TP ? Where to find something like CRT, GRAPH ...?

1 0

BUG: gpc and type modifiers.
by J.J. van der Heijden 03 Oct '96

03 Oct '96

I discovered a bug in the way GPC treats type modifiers: Consider this snippet, for a 32bit architecture: ------------------------------------------- type uint = __unsigned__ integer; { unsigned 32-bit } const c1 : uint = $7FFFFFFF ; { maxint } c2 : uint = $80000000 ; ------------------------------------------- The declaration of c2 gives this error message: `integer constant does not fit in integer type' while this faulty snippet is accepted: ------------------------------------------- type short = __short__ integer; { signed 16-bit } const c3 : short = $FFFF ; c4 : short = $00010000 ; ------------------------------------------- Although c4 gives a warning: `warning: overflow in implicit constant conversion' Obviously, GPC doesn't treat the type modifiers the right way when it evaluates whether a constant is valid or not. The same is true for __longlong__ (64bit), which cannot be initialized with a value > $7FFFFFFF. Workaround ========== With this silly trick it's still possible to assign an unsigned 32bit value: ------------------------------------------- type uint = __unsigned__ integer; { unsigned 32-bit } const c1 : uint = $7FFFFFFF + 1; ------------------------------------------- This gives again: `warning: overflow in implicit constant conversion' which can be suppressed by surrounding the declaration with {$W-} .. {$W+} JanJaap --- "Nothing shocks me, I'm a scientist", Indiana Jones

1 0

patch for linking-dumps-core problems
by J.J. van der Heijden 02 Oct '96

02 Oct '96

Somebody asked for it, so here it is, the patch for the `linking only dumps core' problem. It was a strcpy() with NULL source problem. Apply it, then rebuild the `gpc' driver program. Greetings, JanJaap *** gpc-1.2-2.7.2/gcc.c Tue Aug 20 21:59:57 1996 --- gpc/gcc.c Tue Sep 24 01:08:44 1996 *************** *** 4976,4980 **** putenv (obstack_finish (&collect_obstack)); ! #ifdef __EMX__ /* Substitute the basename of the first input file for %b in link_command_spec. */ --- 4978,4982 ---- putenv (obstack_finish (&collect_obstack)); ! #ifdef GPC /* Substitute the basename of the first input file for %b in link_command_spec. */ *************** *** 4992,4998 **** if (*p == '.' && p != input_basename) basename_length = p - input_basename; - #endif /* __EMX__ */ - #ifdef GPC explicit_link_files = add_automake_files (explicit_link_files); #endif /* GPC*/ --- 4994,4998 ---- --- "Nothing shocks me, I'm a scientist", Indiana Jones

1 0

ANNOUNCE: new win32 GPC binary available.
by J.J. van der Heijden 02 Oct '96

02 Oct '96

The latest cygwin32-compatible GNU Pascal binary is now available from: ftp://agnes.dida.physik.uni-essen.de/home/janjaap/cygwin32/ I keep this seperate from the main release directory because some things have to be done before the next all-platform release. Still, the win32 GPC has improved enough that I think you should all try it out. Files in this directory: gpc-960930-2.7.2.i386-cygwin32-b16.zip Sep.30 snapshot binary of cygwin32 GPC. For a cygwin32-beta16 toolchain. devtools.i386-cygwin32-b16.zip For those who only want to do Pascal, it's a waist to download a complete toolchain from ftp.cygnus.com. This archive has all the tools (as, ld, libraries) needed to complete GPC. If you already have cygwin32-beta16 installed, you don't need this. Changes ------- Since the last release (gpc-1.2-2.7.2.i386-cygwin32-b14.zip), these things have changed: * Now beta16 compatible. * Fixed internal fork() problems, now --automake works. * Pipes (`gpc -pipe ...') supported. * Binary/text file IO problem fixed. * Some new test programs. * Support for function attributes and multiple directives per function declaration. Effectively, this means this GPC can access the Win32 API and build both GUI and console applications! Thanks, Peter. See the demos in the "test\windows" subdirectory. Installation ------------ * create a subdirectory of your choice (`c:\gpcwin32') and unzip gpc-960930-2.7.2.i386-cygwin32-b16.zip and devtools.i386-cygwin32-b16.zip from there. Use a version of `unzip' that can handle long filenames, i.e. not PKunzip. * Add the directory where `gpc.exe' lives (`c:\gpcwin32\bin') to your path. * Set GCC_EXEC_PREFIX=C:\gpcwin32\lib\gcc-lib\ Beware of the trailing slash! * Set LIBRARY_PATH=c:\gpcwin32\lib;c:\gpcwin32\i386-cygwin32\lib Not yet done ------------ GPC has incomplete support for the `stabs' debugging format. This affects the win32 platform (and all ELF unices). Although you can step your program in GDB, it may lose track of where you really are, be unable to display variables etc. I'm working on this. The function attributes and asmname directives seem to get lost in the GPI mechanism, so you will have to #include `units' for now, if they define WinAPI type functions. Other problems -------------- These things often give problems: * Do NOT mix binaries from different beta's -- they are incompatible. * Be sure to have only one `cygwin.dll' in your path and it must be beta16. The symptoms of these are crashes, stack/register dumps and Windows telling you it cannot load a dll. Read the cygwin32 FAQ (http://www.cygnus.com/ or this directory) Most of the information also applies for GNU Pascal. And please, do not bother the people of the cygwin32 mailinglist if you have problems with GPC, send them to the gpc(a)hut.fi list instead. Have fun, JanJaap <j.j.vanderheijden(a)student.utwente.nl> --- "Nothing shocks me, I'm a scientist", Indiana Jones

1 0

Subscription to your mailing list
by premiere＠askesis.it 02 Oct '96

02 Oct '96

Hi, I'm a Student in Computer Programming at University (Milan,Italy). I,ve downloaded Gnu Pascal Compiler and I've given it to my Teacher. Their comments are Enthusiastic! Please , add my E-mail Address to your mailing list. Thanx a lot. HardPile.

1 0

Re: gpc under SunOS or Solaris
by J.J. van der Heijden 01 Oct '96

01 Oct '96

On Mon, 30 Sep 1996 09:38:13 -0700, Friedrich Fahnert <fritz(a)engg2.mobinfo.com> wrote: >J.J. van der Heijden wrote: >> >> On Sat, 21 Sep 1996 15:57:22 +0200 (MET DST, >> Peter Gerwinski <peter(a)agnes.dida.physik.uni-essen.de> wrote: >> >> >According to Friedrich Fahnert: >> >> I managed to get version 1.2-2.7.2 compiled on both, but neither works >> >> very well. On Solaris gpc1 gets a fatal signal, and an internal >> >> compiler error, when compiling certain programs. On SunOS, it seems >> >> to compile properly, but it dumps core when linking. >> > >> >> I fixed the linking problems. Yes Peter, they _were_ there :-( They just >> didn't show in EMX. >> >> JanJaap >> > >Did you make some patches? How about new binaries? > > I cannot build new binaries for every platform each time I change something -- I would spend all my time building binaries. And I don't have access to a Solaris box anyway. But I can publish the patch, then you can build a new binary yourself. There are some more things that have to be done before a new "all-platform" release. Groeten, JanJaap --- "Nothing shocks me, I'm a scientist", Indiana Jones

1 0

Re: System calls
by J.J. van der Heijden 30 Sep '96

30 Sep '96

On Mon, 30 Sep 1996 02:03:07 -0400 (EDT), John Michael Black <jm_black(a)csunix1.lvc.edu> wrote: >This may sound like a dumb question. Is there a predefined routine for >executing O/S commands built into gpc? (like C's universal "system()" >routine.) If not, what would be the easiest way to go about this in unix? > >(I suppose I could use the C routine itself as an external procedure, as a >last resort...) > There's no builtin system() call because this is not part of the Pascal standard. But there is nothing to stop you from using your C-lib's system() call. In C: int system (const char *string); In Pascal: function system(s: cstring): integer; C; function system; external; (I didn't try this, but have done simular things with fork() and others in the past) Someday, we should get some units/modules to implement this ... oh, well, keep dreaming... JanJaap --- "Nothing shocks me, I'm a scientist", Indiana Jones

1 0

System calls
by John Michael Black 30 Sep '96

30 Sep '96

This may sound like a dumb question. Is there a predefined routine for executing O/S commands built into gpc? (like C's universal "system()" routine.) If not, what would be the easiest way to go about this in unix? (I suppose I could use the C routine itself as an external procedure, as a last resort...) -j

1 0

Re: DosFile.pas workaround for the CrLF-bug under old DOS
by JanJaap van der Heijden 27 Sep '96

27 Sep '96

At 03:40 PM 9/20/96 +0200, you wrote: > >{ BP-like file operations for binary files } >{ simple workaround for the CRLF-bug in GCC/GPC } >{ Sven Hilscher - sven(a)rufus.central.de } > > I can see the use of functions with a borland-compatible syntax, like assign(), but fail to see the link with CRLF problems. I think I fixed those. var f: text is now opened in text mode. A line is terminated with CRLF in DOS, while type byte = __byte__ integer; bytefile = file of byte; var g: bytefile; is opened in binary mode and $0A is not silently converted to $0A$0D. I think this solves the problems. If you think otherwise, let me know. JanJaap -- The latest & greates in software, hardware and manswear -- Bono Vox (U2)

2 1

Re: gpc under SunOS or Solaris
by J.J. van der Heijden 24 Sep '96

24 Sep '96

On Sat, 21 Sep 1996 15:57:22 +0200 (MET DST, Peter Gerwinski <peter(a)agnes.dida.physik.uni-essen.de> wrote: >According to Friedrich Fahnert: >> I managed to get version 1.2-2.7.2 compiled on both, but neither works >> very well. On Solaris gpc1 gets a fatal signal, and an internal >> compiler error, when compiling certain programs. On SunOS, it seems >> to compile properly, but it dumps core when linking. > I fixed the linking problems. Yes Peter, they _were_ there :-( They just didn't show in EMX. JanJaap --- Someone once asked me if they could use Open Windows on a Mac. I said, "I don't know. Let's find out." I went to the Mac, opened the window and threw out the Mac. "Yep, You can use Open Windows on a Mac." bofh(a)RedDragon.Empire.Net (Psycho Sysadmin)

1 0

MAN page?
by John Michael Black 23 Sep '96

23 Sep '96

Has there ever been written a man page for the unix versions? Just basic command syntax & options would be enough to start. I'd offer to do it if I knew how. :( -john

2 2

Functions/strings
by Peter Gerwinski 23 Sep '96

23 Sep '96

According to John Michael Black: > I remember reading in an older version's GUIDE that functions returning > strings would not work. Was this corrected with 2.7.2? Yes. Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

MAN page?
by Peter Gerwinski 23 Sep '96

23 Sep '96

According to John Michael Black: > Has there ever been written a man page for the unix versions? Just basic > command syntax & options would be enough to start. No. > I'd offer to do it if I knew how. :( That would be great. Unfortunately, I don't know either how to do this. Anybody here who knows? Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

Functions/strings
by John Michael Black 23 Sep '96

23 Sep '96

I remember reading in an older version's GUIDE that functions returning strings would not work. Was this corrected with 2.7.2? -john

1 0

Re: some questions ...
by JanJaap van der Heijden 23 Sep '96

23 Sep '96

At 11:15 PM 9/22/96 +0200, you wrote: >Hi Sven, hi everybody, Hello :) > > >> Why is the boolean type internal so big (Integer) ? > >I could not reproduce this bug. Var B: Boolean; writeln ( Sizeof ( B ) ) >yields 1 on my machine. > >> Boolean should only use a bit - yes thats stupid, > This is not so stupid. A 32bit CPU accesses memory in 32bit chunks. It is possible to access a byte, but this is considerably slower. Accessing a bit would require accessing a 32bit word first, then bitshifting. All compilers make this speed v.s. memory-consumption tradeof in favor of speed, and GPC is no exception. For a 64 bit CPU, it's even worse: 63bits per boolean are "waisted". JanJaap -- The latest & greates in software, hardware and manswear -- Bono Vox (U2)

1 0

Known GPC bugs
by Peter Gerwinski 23 Sep '96

23 Sep '96

Hello folks, I updated the "known bugs" section of the Texinfo documentation of the (experimental) gpc-1.2-2.7.2 pre-release. The new file "bugs.texi" can be found at ftp://agnes.dida.physik.uni-essen.de/home/peter/gpc-2.7.2/bugs.texi Other doc files got some minor changes, too. I included a link to the GPC documentation into the GPC home page, http://home.pages.de/~GNU-Pascal/ which is actually an alias to http://agnes.dida.physik.uni-essen.de/~gnu-pascal/ See you, Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

Re: |||||||||| Free Pascal Compiler? ||||||||||
by Peter Gerwinski 23 Sep '96

23 Sep '96

According to Louis Cappitelli: > > On 22 Sep 1996 12:02:40 GMT, in comp.lang.pascal.misc you wrote: > > >Louis Cappitelli (lcappite(a)sprynet.com) wrote: > >> Does anybody know if there is one for DOS or win95? I am taking a > >> course in High School that uses Turbo pascal, but because it's illegal > >> to copy software, I can't work on my program projects at home and test > >> them. Is there something like djgpp c++ compiler or something better? > > > >Get GNU Pascal. It runs on the DJGPP platform for DOS and Win95, > >on the EMX platform for DOS and OS/2 and on any other platform > >supported by the GNU compiler family (e.g. Linux, FreeBSD). > > I got a question. Does it require DJGPP? Yes, you need at least the assembler (as), the linker (ld) and some libraries (libc.a, libm.a, crt0.o, ...) to run the DJGPP version (for DOS) of the GNU Pascal compiler. The C compiler is - of course - not needed. However, djgpp is not really a big packet. You do *not* need DJGPP to run a program *compiled with* GPC. With EMX, you need the EMX development tools (except, again, the C compiler) to compile programs; to run a compiled program you also need the EMX run time system, either an extender program (for DOS) or a DLL (for OS/2). But that's not a problem since the EMX license allows you to distribute the EMX runtime system along with your program. By the way, programs compiled with Borland Pascal also need Borland's runtime system (files RTM.EXE and DPMI16BI.OVL) to run in protected mode. Fortunately, Borland allows its licensed users to distribute them along with their programs. (Also posted to the GPC mailing list.) Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

some questions ...
by Peter Gerwinski 22 Sep '96

22 Sep '96

Hi Sven, hi everybody, > Somebody have found the C2PAS converter? I have never seen such a thing - only the p2c. > Why is the boolean type internal so big (Integer) ? I could not reproduce this bug. Var B: Boolean; writeln ( Sizeof ( B ) ) yields 1 on my machine. > Boolean should only use a bit - yes thats stupid, It's not stupid at all. One day, GPC will actually pack `packed arrays' of Boolean or `packed records' containing Booleans and Integer subranges, then each Boolean will only get a single bit. > Why crashed the compiler very often? If we knew the answer, the compiler wouldn't crash any more, sigh. Please collect the source which makes the compiler crash, so we can correct the bug. > If the compiler found a syntax fault, I think in some cases it's better to > show the fault position in the line. I agree, but I have no idea how to implement this with a Bison parser (I am not so experienced with Bison). Somebody has? > How I can _create_ functions with free type parameters, > _without get a warning_ while using with different typed variabels? This is a bug in GPC and must be fixed. For the moment, use (*$W-*) to switch off annoying warnings, use (*$W+*) to switch them on again. By the way, Sven, your DosFile Unit does not only work around a bug, but it also implements some missing functions such as BlockRead, BlockWrite, and IOresult. Gut gemacht! :-) :-) Shall I put it into a gpc-2.7.2/contrib directory? Somebody out there willing to make it portable? It should not be too difficult to translate the DOS system calls to UNIX system calls. Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

BUG: String schemas as function parameters
by Peter Gerwinski 22 Sep '96

22 Sep '96

Subject: Re: gnu pascal unit problem Newsgroups: comp.os.msdos.djgpp References: <Dxx3Bv.2G5(a)bercos.de> Organization: Universitaet Essen, Germany Reply-To: peter.gerwinski(a)uni-essen.de Distribution: Helge Kruse (hk(a)bercos.de) wrote: > I downloaded the version 1.2-2.7.2 of gpc and tried a unit > sample. But the code won't work. It's so small, that Id like > to post the code here: > program p1; > (* This program does NOT work correctly with GPC 2.7.2 *) > (* compiled with 'gpc --automake p1.pas' the a.out will *) > (* be compiled. But the function u1 in the unit u1 *) > (* does not prints the parameter. *) > (* Renaming the unit, so that unit/function name differ *) > (* gives no success. *) The function u1 accepts a formal parameter of type "String". Other than in Borland Pascal, this does not default to a String of maximum length 255 but stands for a String Schema type which is something more complicated. Indeed, you discovered a bug. Thank you for the report. To work around, use Strings of known maximum length instead of generic String Schemas, i.e. Type WrkString = String ( 80 ); Function u1 ( S: WrkString ): whatever; (Also posted to the GNU Pascal mailing list.) Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

Re: gpc under SunOS or Solaris
by Peter Gerwinski 22 Sep '96

22 Sep '96

According to Friedrich Fahnert: > I found a solution for the problem I was having under SunOS. I can get > the program compiled if I specify the source files, to gpc, but not if I > compile the files, and link them separately. Should this be considered > to be a bug? Yes. :-( Thank you for the report. I was reported a similar bug for Linux but could not reproduce it. Strange. Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

some questions ...
by Sven＠rufus.central.de 22 Sep '96

22 Sep '96

Hallo gpc, Somebody have found the C2PAS converter? Why is the boolean type internal so big (Integer) ? Boolean should only use a bit - yes thats stupid, but a byte should work fast and easy !? Why crashed the compiler very often? Yes, I know syntax errors are my faults - but other compilers (you know what I mean) are not so touchy. If the compiler found a syntax fault, I think in some cases it's better to show the fault position in the line. How I can _create_ functions with free type parameters, _without get a warning_ while using with different typed variabels? working example in BP : function SendBuffer(var Buffer; BufferSize: Integer): Integer; My suggestion: function SendBuffer(var Buffer: Void; BufferSize: Integer): Integer; should not generate a warning in using of SendBuffer ... Greetings, Sven ...

1 0

DosFile.pas workaround for the CrLF-bug under old DOS
by Sven＠rufus.central.de 22 Sep '96

22 Sep '96

{ BP-like file operations for binary files } { simple workaround for the CRLF-bug in GCC/GPC } { Sven Hilscher - sven(a)rufus.central.de } unit DosFile; { version 0.01Beta } { provisional arrangement } Interface Type WrkStr = String(255); Pointer = ^void; NameT = array[1..13] of char; BinFile = record Name : NameT; Handle : Integer; BSize : Integer; end; Var IOResultTemp : Integer; procedure Assign(var F: BinFile; Name: WrkStr); procedure Reset(var F: BinFile; RecSize: Integer); procedure Rewrite(var F: BinFile; RecSize: Integer); procedure BlockWrite(var F: BinFile; var Buf: void; Count: Integer; var Result: Integer); procedure BlockRead(var F: BinFile; var Buf: Void; Count: Integer; var Result: Integer); procedure Close(var F: BinFile); function IOResult: Integer; Implementation procedure Assign(var F: BinFile; Name: WrkStr); var i : Integer; begin for i := 1 to 12 do if length(Name) >= i then F.Name[i] := Name[i] else F.Name[i] := chr(0); F.Name[13] := chr(0); F.Handle := 0; end; { functions converted from "dos.h" } function DosOpen(n: Pointer; i: Integer; h: pointer):Integer; AsmName '_dos_open'; function DosCreate(var n: NameT; i: Integer; h: pointer):Integer; AsmName '_dos_creat'; function DosClose(h: Integer): Integer; AsmName '_dos_close'; function DosWrite(h: Integer; b: Pointer; c: Integer; var r: Integer): Integer; AsmName '_dos_write'; function DosRead(h: Integer; b: Pointer; c: Integer; var r: Integer): Integer; AsmName '_dos_read'; procedure Reset(var F: BinFile; RecSize: Integer); begin if F.handle <> 0 then IOResultTemp := DosClose(F.Handle); IOResultTemp := DosOpen((a)F.Name, 0, @F.Handle); F.BSize := RecSize; end; procedure Rewrite(var F: BinFile; RecSize: Integer); begin if F.Handle <> 0 then IOResultTemp := DosClose(F.Handle); IOResultTemp := DosCreate(F.Name, 0, @F.Handle); F.BSize := RecSize; end; procedure Close(var F: BinFile); begin IOResultTemp := DosClose(F.Handle); F.Handle := 0; F.BSize := 0; end; function IOResult: Integer; begin IOResult := IOResultTemp; IOResultTemp := 0; end; procedure BlockWrite(var F: BinFile; var Buf: void; Count: Integer; var Result: Integer); begin IOResultTemp := DosWrite(F.Handle, @Buf, Count * F.BSize, Result); Result := Result div F.BSize end; procedure BlockRead(var F: BinFile; var Buf: Void; Count: Integer; var Result: Integer); begin IOResultTemp := DosRead(F.Handle, @Buf, Count * F.BSize, Result); Result := Result div F.BSize end; begin IOResultTemp := 0; end.

1 0

gpc under SunOS or Solaris
by Peter Gerwinski 21 Sep '96

21 Sep '96

According to Friedrich Fahnert: > I managed to get version 1.2-2.7.2 compiled on both, but neither works > very well. On Solaris gpc1 gets a fatal signal, and an internal > compiler error, when compiling certain programs. On SunOS, it seems > to compile properly, but it dumps core when linking. It would be nice to fix the bugs. If possible, I would like to have a minimum source which still triggers the error. Perhaps we can reproduce the error on other platforms we have access to. > Is there any platform where gpc could be considered to be stable? I tested GPC on various iX86-based platforms, so I'd consider it to be stable on DOS (DJGPP and EMX), OS/2 (EMX), Linux and FreeBSD. Good luck, Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

dos and binary file IO: fixed
by J.J. van der Heijden 20 Sep '96

20 Sep '96

Hello folks, I fixed the problems of gpc/djgpp with binary file IO. An updated libgpc.a plus the patch can be downloaded from: ftp://agnes.dida.physik.uni-essen.de/home/janjaap/updates/rtsdjgpp.zip The patch also applies for win32 and emx who also treat text and binary files differently, unlike unix. (I have no libgpc.a for them though). Happy hacking, JanJaap --- Someone once asked me if they could use Open Windows on a Mac. I said, "I don't know. Let's find out." I went to the Mac, opened the window and threw out the Mac. "Yep, You can use Open Windows on a Mac." bofh(a)RedDragon.Empire.Net (Psycho Sysadmin)

1 0

Re: how to write binary datas in a file
by Peter Gerwinski 20 Sep '96

20 Sep '96

Hi Sven, hi everybody, > I can use only the "write" function - the result is all $0A will be > converted into "$0D $0A" - Yes, thats the CrLf - Yes, that's even true for "File of Integer"! Congratulations - you have discovered a serious bug in the Run Time Library. :-( > I think it's no problem > under UNIX, Linux, Ponyx, Limbux etc. :-) Yes, it isn't. Your diagnostics was right. > How can I write binary datas in a file? > It must be work in MS-DOS !!! I'd like to fix it as soon as possible, but I am out of time right now. :-( :-( Anybody out there to do it? The error must be in rts/rts-write.c, something with fopen() ... Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

2 1

Re: how to write binary datas in a file
by JanJaap van der Heijden 19 Sep '96

19 Sep '96

At 02:18 PM 9/17/96 +0200, you wrote: >Hi Sven, hi everybody, > >> I can use only the "write" function - the result is all $0A will be >> converted into "$0D $0A" - Yes, thats the CrLf - > >Yes, that's even true for "File of Integer"! Congratulations - you >have discovered a serious bug in the Run Time Library. :-( > >> I think it's no problem >> under UNIX, Linux, Ponyx, Limbux etc. :-) > >Yes, it isn't. Your diagnostics was right. > >> How can I write binary datas in a file? >> It must be work in MS-DOS !!! > >I'd like to fix it as soon as possible, but I am out of time right now. >:-( :-( > >Anybody out there to do it? The error must be in rts/rts-write.c, >something with fopen() ... > I fixed this problem with the MS-DOS files, but it broke some the "standard" files "Input" and "Output". I will think of something more elegant soon. Probably change the BSD style 'fopen()' calls to 'open()' too because then I can easily add an #ifdef MSDOS flags = flags | O_BINARY; #endif once, instead of #iddef-ing every instance of fopen which eats "r" , "w" arguments (which become "wb" etc for non-text files in dos & co.) greetings, JanJaap -- The latest & greates in software, hardware and manswear -- Bono Vox (U2)

1 0

Problems with DJGPP info files
by Peter Gerwinski 19 Sep '96

19 Sep '96

Subject: Re: gnu pascal info problems Newsgroups: comp.os.msdos.djgpp References: <323E879C.3047(a)blackmagic.tait.co.nz> Organization: Universitaet Essen, Germany Reply-To: peter.gerwinski(a)uni-essen.de Distribution: Bill Currie (billc(a)blackmagic.tait.co.nz) wrote: > I'm having problems getting info to browse the info files from gpc 2.7.2. > Info can't 'Top'. I've tried dtou, re-makeing the files with and without > splitting, and dtou-ing again, to no avail (trying before and after > dtou). This is a bug in the GPC info. Thank you for the report. To fix it, you can re-make the info with djgpp's "makeinfo", but you have to edit "gpc.txi" before and replace all "@include foo.texi" by "@include foo.txi". (Fortunately, the other files do not contain "@include" commands. :-) (Also posted to the GPC mailing list.) Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

Which version/location
by John Michael Black 18 Sep '96

18 Sep '96

Okay, we've upgraded to Linux 2.0 with gcc 2.7.2. We can use either ELF or a.out, (although we prefer ELF.) Which version would be best for this configuration? [At one time we had one problem with GPC lookng for "crt0.o". Our configuration doesn't even have this file, but perhaps (as Peter suggested) we had the wrong compiler version.] -john jm_black(a)csunix1.lvc.edu

1 0

Re: how to write binary datas in a file
by J.J. van der Heijden 18 Sep '96

18 Sep '96

On 17 Sep 1996 02:00:00 +0200, Sven Hilscher <Sven(a)rufus.central.de> wrote: > >Hi folks, > >another stupid problem: > >I tryed to write binary datas in a file. >I can use only the "write" function - the result is all $0A will be >converted into "$0D $0A" - Yes, thats the CrLf - I think it's no problem >under UNIX, Linux, Ponyx, Limbux etc. :-) > Yep - a bug. djgpp opens all files in textmode unless instructed otherwise, which the RTS _should_ do for all files not of type TEXT. Unfortunately, a quick fix broke the standard "Input" and "Output" so please wait a few more days 'till I have time to do this the right way. >How can I write binary datas in a file? >It must be work in MS-DOS !!! > Win32 affected too probably. Hey Peter, does OS/2 (EMX) discriminate between text and binary files like DOS, or not, like unix? JanJaap --- Someone once asked me if they could use Open Windows on a Mac. I said, "I don't know. Let's find out." I went to the Mac, opened the window and threw out the Mac. "Yep, You can use Open Windows on a Mac." bofh(a)RedDragon.Empire.Net (Psycho Sysadmin)

1 0

how to write binary datas in a file
by Sven＠rufus.central.de 17 Sep '96

17 Sep '96

Hi folks, another stupid problem: I tryed to write binary datas in a file. I can use only the "write" function - the result is all $0A will be converted into "$0D $0A" - Yes, thats the CrLf - I think it's no problem under UNIX, Linux, Ponyx, Limbux etc. :-) How can I write binary datas in a file? It must be work in MS-DOS !!! ___, ((__ o ,____)) V E N I

1 0

Re: SYSV vs BSD signal blues :-(
by J.J. van der Heijden 12 Sep '96

12 Sep '96

On 11 Sep 96 16:01:47 EDT, Berend de Boer <100120.3121(a)CompuServe.COM> wrote: >> I think I'll just rewrite the signal related stuff in rts-rt0.c, implement >> some BSD/SYSV independant routines that ONLY mask signals that are actually >> present on the target system. > >I don't have handson experience with signals, but it seems to me that POSIX >already did this. I suggest using signals according to the POSIX standard. > >And every unix box should come with some decent posix library these days. > POSIX describes a "minimum set" of signals that must be present in a system. It's the non-POSIX, system specific signals that caused the confusion here. Of course I could just remove all non-POSIX stuff from GPC, but I don't think that's necessary. It also shouldn't be hard to implement a runtime switch to turn signal catching off altogether to enable coredumps, like somebody else suggested. JanJaap --- The latest & greatest in software, hardware and manswear. Bono Vox (U2)

1 0

Re: I don't unterstand __byte__ :-(
by Peter Gerwinski 12 Sep '96

12 Sep '96

Hi all, > const > arHex : array[0..$0f] of char > = '0123456789abcdef'; Nice that this works. I was sure it wouldn't work. ;-) > a8 : array[0..max] of __byte__ Integer absolute a32; > > write(a8[i]: 10); > 1 1 34561 > > write(hexa(a8[i])); > 1 $00000001 $00000001 Symptoms: the value of a __byte__ Integer is written correctly via the "hexa" function, it's wrong when we simply "write" it. This is due to a BUG in GPC's run tims system (RTS) which doesn't (yet) know about type modifiers like __byte__ at all. :-( For the same reason, you get run time errors when you try to writeln __longlong__ (64-bit) Integers (\approx Comp in BP) or __long__ Reals (Extended in BP). To work around, do an explicit type cast: > write ( Integer ( a8 [ i ] ) : 10 ); > 1 1 1 Bye, Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

New win32 binaries almost done. please read.
by J.J. van der Heijden 11 Sep '96

11 Sep '96

Hello, I'm almost done building a cygwin32-beta16 compatible GPC. It fixes a number of problems, like the hanging of GPC1 in "automake" (fork() related problems) Also, the -pipe option is now supported. The current cygwin32 b16 toolchain is a 25Meg ZIP, eating up >80Meg disk when installed. For those who only want to do Pascal, this is a complete waist because you only need binutils and some libraries from it. So, shall I wrap these components in a seperate ZIP (windev.zip or whatever) and upload them too? JanJaap --- The latest & greatest in software, hardware and manswear. Bono Vox (U2)

1 0

Re: SYSV vs BSD signal blues :-(
by Berend de Boer 11 Sep '96

11 Sep '96

> I think I'll just rewrite the signal related stuff in rts-rt0.c, implement > some BSD/SYSV independant routines that ONLY mask signals that are actually > present on the target system. I don't have handson experience with signals, but it seems to me that POSIX already did this. I suggest using signals according to the POSIX standard. And every unix box should come with some decent posix library these days. Groetjes, Berend.

1 0

Re: New win32 binaries almost done. please read.
by Berend de Boer 11 Sep '96

11 Sep '96

> So, shall I wrap these components in a seperate ZIP (windev.zip or > whatever) and upload them too? Great idea. Groetjes, Berend.

1 0

SYSV vs BSD signal blues :-(
by J.J. van der Heijden 11 Sep '96

11 Sep '96

Hello, We've seen some signal related troubles when building the RTS lately. This is about what's happening: * GPC's RTS installs a signal handler to give pretty error messages when a signal is caught. This is ONLY implemented for BSD like systems, probably because Juki didn't have access to a SYSV style box. * SYSV and BSD differ in the number of signals and their meaning. * Some BSD boxes don't implement all signals > 16. (SIGXCPU etc.) * Some SYSV boxes implement (a range of) BSD-only signals, making it hard to dectect a true BSD box. I'm getting tired of writing ever more clever system identification routines that break on the next system I encounter. I think I'll just rewrite the signal related stuff in rts-rt0.c, implement some BSD/SYSV independant routines that ONLY mask signals that are actually present on the target system. How 'bout it? JanJaap --- The latest & greatest in software, hardware and manswear. Bono Vox (U2)

2 1

I don't unterstand __byte__ :-(
by Sven＠rufus.central.de 11 Sep '96

11 Sep '96

Hi all, the following example I've used to test the __byte__ and __short__ types. There was some trouble while using the GRX20-library (I'm working on a interface-unit) with conversion of different structures. Look at my sample and say me why I can't get the correct value from Bytes. program ByteTest; const arHex : array[0..$0f] of char = '0123456789abcdef'; max = 15; type WrkStr = String(255); var a32 : array[0..max] of Integer; a16 : array[0..max] of __short__ Integer absolute a32; a8 : array[0..max] of __byte__ Integer absolute a32; a_c : array[0..max] of char absolute a32; i : Integer; function hexa(number: Integer): WrkStr; var i : Integer; s : WrkStr; begin s := ''; for i := 1 to 8 do begin s := arHex[number and $0f] + s; number := number shr 4; end; hexa := ' $'+s; end; begin for i := 0 to max do a_c[i] := chr(i); for i := 0 to max do begin write(i: 3); write(ord(a_c[i]): 5); write(a8[i]: 10); if i mod 2 = 0 then write(a16[i div 2]: 10); if i mod 4 = 0 then write(a32[i div 4]: 16); writeln; end; for i := 0 to max do begin write(i: 3); write(hexa(ord(a_c[i]))); write(hexa(a8[i])); if i mod 2 = 0 then write(hexa(a16[i div 2])); if i mod 4 = 0 then write(hexa(a32[i div 4])); writeln; end; end. I'm using the DJGPP-GPC on MS-DOS and that's the output: 0 0 34560 256 50462976 1 1 34561 2 2 34562 770 3 3 34563 4 4 34564 1284 117835012 5 5 34565 6 6 34566 1798 7 7 34567 8 8 34568 2312 185207048 9 9 34569 10 10 34570 2826 11 11 34571 12 12 34572 3340 252579084 13 13 34573 14 14 34574 3854 15 15 34575 0 $00000000 $00000000 $00000100 $03020100 1 $00000001 $00000001 2 $00000002 $00000002 $00000302 3 $00000003 $00000003 4 $00000004 $00000004 $00000504 $07060504 5 $00000005 $00000005 6 $00000006 $00000006 $00000706 7 $00000007 $00000007 8 $00000008 $00000008 $00000908 $0b0a0908 9 $00000009 $00000009 10 $0000000a $0000000a $00000b0a 11 $0000000b $0000000b 12 $0000000c $0000000c $00000d0c $0f0e0d0c 13 $0000000d $0000000d 14 $0000000e $0000000e $00000f0e 15 $0000000f $0000000f Please don't quote the whole text !!! ___, ((__ o ,____)) V E N I

1 0

Re: External, C, AsmName, ...
by Peter Gerwinski 11 Sep '96

11 Sep '96

Hi everybody, According to Berend de Boer: > There is a translator for BP, mabe it's easy to adapt this one? Where is it? Since GPC is essentially BP-compatible now, perhaps we can just use it? Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

Re: External, C, AsmName, ...
by Berend de Boer 11 Sep '96

11 Sep '96

> But it would be a good idea to write a C->pas header converter. Right now, > you have to translate all headers before you can use them. There is a translator for BP, mabe it's easy to adapt this one? Groetjes, Berend.

1 0

Re: gpc for Solaris 2.5.1
by J.J. van der Heijden 10 Sep '96

10 Sep '96

On Tue, 10 Sep 1996 11:12:16 -0400 (EDT), M BAILEY <mbailey(a)cc.cst.cmich.edu> wrote: > >If someone sends me a merged source tree of GCC and GPC i would be happy to >compile them into a finished binary product and possible build solaris pkg's >out the resulting answers to that you can install them with out needing any >other utilities... > If you can build a ready-to-run compiler we would be happy to distribute it. There is no merged source tree: you put gpc and gcc in their own directory. Because I have not updated the INSTALL file for a while, I shall shortly explain what to do: (I assume solaris is properly supported bu GCC) If you extract gcc-2.7.2.tar.gz and gpc-1.2-2.7.2.tar.gz from the same location, you get gcc-2.7.2 and gpc-1.2-2.7.2 subdirs, which is just fine. 1) configure and build gcc: ./configure make CFLAGS=-O2 LDFLAGS=-s LANGUAGS=c If this fails, build gcc with -O0 and build a stage2 compiler with this the stage1 gcc, then go on (this is all explained in GCC manuals) 2) configure gpc: ./configure configure should be able to find the gcc directory if it is reasonably "close". If not, run configure --help and see what options you must give it. 3) build gpc: make CFLAGS=-O2 LDFLAGS=-s Read earlier notes about what to do if VPATH troubles strike (below). BSD/SYSV signal confusion (RTS won't build) *may* be avoided by using a newer configure (message posted this list a few minutes ago) If not, I would like to know whether configure decides you have SYSV or BSD (and if you know if this is incorrect, possibly see your <signal.h> and <sys/signal.h>. 4) A binary distribution can be created with the "make bindist" command. I'm not familiar with the habbits of solaris pkgtools, so you may have to tweak it by hand (you should know this). Hope this helps, JanJaap >Matthew S. Bailey > > >On Tue, 10 Sep 1996, J.J. van der Heijden wrote: > >> On Mon, 9 Sep 1996 21:44:50 +0200 (MET DST), >> Peter Gerwinski <peter(a)agnes.dida.physik.uni-essen.de> wrote: >> >> >Hi Phil, Brad, and Jan Jaap, >> > >> > stor-layout & friends ARE BACK ... >> > >> >> I wish that was true. Here is the results on my machine. >> >> maybe_find_function_data gpc-parse.o >> >> [...] >> >> dbxout_set_type_status gpc-decl.o >> > >> >I know that problem. It is a bug in make, Jan Jaap and I had >> >hoped to have worked around. The problem is that there are >> >language independend source files which exist in the GPC source >> >as well as in the GCC source. In Makefile, look out for >> >stor-layout.o, dbxout.o, ..., toplev.o. For some reason, >> >they are not recompiled correctly. >> > >> >> What happens precisely is that "make" has a VPATH: it looks for object code >> in the gcc object directory too. When your gcc object directory is >> identical to your gcc source directory, make thinks $gccbin/foo.o is >> uptodate because it depends $gccbin/foo.o on $gccsrc/foo.c (which it can >> see because $gccsrc=$gccbin) Then it doesn't build ./foo.o based on >> $gpcsrc/foo.c (mind $gpcsrc vs $gccsrc) >> >> To make it more complicated, VPATH behaviour of various "make" programs >> is different and some versions of GNU make have VPATH bugs. >> >> >To work around, you can create these *.o files and copy them >> >into the gcc directory, so "make gpc" will find and use them. >> > >> >> Use different directories for $gccsrc and $gccbin. >> Or delete the ones in $gccbin to force make to rebuild them. >> >> /janjaap >> --- The latest & greatest in software, hardware and manswear. Bono Vox (U2)

1 0

SYSV vs BSD confusion
by J.J. van der Heijden 10 Sep '96

10 Sep '96

Hello, There has been some confusion about detecting BSD style signals versus SYSV style. I have updated the configure script to be less fascist about BSD and hopefully this fixes the problems. Get it from: ftp://agnes.dida.physik.uni-essen.de/home/janjaap/updates/ Now, hopefully, FreeBSD is considered BSD again (Peter?) I think SunOS and Solaris are BSD variants too, right? If this still doesn't work, I would like to see your <signal.h> and <sys/signal.h> header files. The problem is that some SYSV systems #define these BSD signals but don't support them unless you link -lbsd, so I made BSD detection more strict (so strict in fact that FreeBSD was no longer considered BSD ;-) Hope this helps, JanJaap --- The latest & greatest in software, hardware and manswear. Bono Vox (U2)

1 0

Re: gpc for Solaris 2.5.1
by J.J. van der Heijden 10 Sep '96

10 Sep '96

On Mon, 9 Sep 1996 21:44:50 +0200 (MET DST), Peter Gerwinski <peter(a)agnes.dida.physik.uni-essen.de> wrote: >Hi Phil, Brad, and Jan Jaap, > > stor-layout & friends ARE BACK ... > >> I wish that was true. Here is the results on my machine. >> maybe_find_function_data gpc-parse.o >> [...] >> dbxout_set_type_status gpc-decl.o > >I know that problem. It is a bug in make, Jan Jaap and I had >hoped to have worked around. The problem is that there are >language independend source files which exist in the GPC source >as well as in the GCC source. In Makefile, look out for >stor-layout.o, dbxout.o, ..., toplev.o. For some reason, >they are not recompiled correctly. > What happens precisely is that "make" has a VPATH: it looks for object code in the gcc object directory too. When your gcc object directory is identical to your gcc source directory, make thinks $gccbin/foo.o is uptodate because it depends $gccbin/foo.o on $gccsrc/foo.c (which it can see because $gccsrc=$gccbin) Then it doesn't build ./foo.o based on $gpcsrc/foo.c (mind $gpcsrc vs $gccsrc) To make it more complicated, VPATH behaviour of various "make" programs is different and some versions of GNU make have VPATH bugs. >To work around, you can create these *.o files and copy them >into the gcc directory, so "make gpc" will find and use them. > Use different directories for $gccsrc and $gccbin. Or delete the ones in $gccbin to force make to rebuild them. /janjaap --- The latest & greatest in software, hardware and manswear. Bono Vox (U2)

2 1

Re: External, C, AsmName, ...
by J.J. van der Heijden 09 Sep '96

09 Sep '96

On Sun, 8 Sep 1996 00:34:02 +0200 (MET DST), Peter Gerwinski <peter(a)agnes.dida.physik.uni-essen.de> wrote: >According to Berend de Boer: >> Can we not simply see a C .h file as an interface-only module? I mean I can >> simply import a c .h file which gpc parses and all symbols defined in that >> header file are exported symbols? > >This would mean that GPC would have to parse C source. > >Remember that #include "foo.h" or (*$include "foo.h" *) just means >that GPC treads the file `foo.h' just if its contents were written >in our source file. A Pascal program using this mechanism would >in fact look like the following: > > Program Test; > > extern int foo; > > extern void ioctl (int fildes, int cmd, ...); > > begin > [...] > end. > >You don't actually want this. ;-) > No, I don't :) But it would be a good idea to write a C->pas header converter. Right now, you have to translate all headers before you can use them. janjaap --- The latest & greatest in software, hardware and manswear. Bono Vox (U2)

1 0

External, C, AsmName, ...
by Peter Gerwinski 08 Sep '96

08 Sep '96

Hello folks, (* This mail is somehow long. In short: I want to take the directives `Extern', `C', and `AsmName' *out* of GPC and extend the syntax of `External' instead. Okay? :*) as you know (?;-), we have the directives `External', `Extern' (same as `External'), `C', and `AsmName' in GPC for specifying external procedures. This is especially useful when using libraries written in C, for example: Procedure Foo; External; (* yields "Foo" *) Procedure ioCtl ( FilDes, Cmd: Integer; ... ); C; (* yields "ioctl" *) Procedure XOpenDisplay ( Name: __CString__ ); AsmName 'XOpenDisplay'; This turned out not to be enough; we need the same for variables. Actually we already have Var Foo: __external__ Integer; (* yields "foo", not "Foo"! *) but this does not cover cases where the external variable needs a name with a capital letter. As an intermediate solution, I implemented Var Foo: __asmname__ 'foO' Integer; (patched source is available on demand), but I think it is not the best way to implement these things. Now I am looking out for reasonable ideas. Does somebody know what the ISO/ANSI Standards say about this? My own thoughts follow. I would suggest to define as few directives as possible and to drop everything but `External' but extend its syntax to meet our requirements. A first approach could be Procedure foo; external; (* yields "Foo" *) Var x: Integer; external; (* yields "X" *) and Procedure foo; external 'foO'; Var x: Integer; external 'my_x'; but it would crash with Borland Pascal's use of `External' where the latter declaration would mean that Procedure foo is contained in a shared library named `foO'. Okay, so let's stay for the moment at Borland's syntax. In the same context they also define Procedure foo; external 'MySharedLib' index 42; and Procedure foo; external 'MySharedLib' name 'foO'; where `foO' would be the external name as defined in the shared library. My current idea now would be to define Procedure foo; external; (* yields "Foo" *) Var x: Integer; external; (* yields "X" *) plus Procedure foo; external name 'foO'; Var x: Integer; external name 'my_x'; to replace "AsmName 'foO'". Later we can extend this to allow a name of a shared library between `external' and `name'. Once doing this, it could also be nice to make directives out of the other C-style storage qualifiers, too. I mean I'd like to replace Var x: __static__ Integer; by x: Integer; static; y: __volatile__ Integer; by y: Integer; volatile; etc.. My plan is to do it in a way that things like Var static: Integer; would still work. I think it would *not* be a good idea to make Procedure foO; external; yield "foO", because (i) it would be difficult to implement and (ii) Pascal is a case-insensitive language, so everything outside string constants *must not* depend on casing at all. Please tell me your opinion about this. Yours, Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter

2 1

gpc for Solaris 2.5.1
by brad＠shiloh.nimh.nih.gov 08 Sep '96

08 Sep '96

Hi, Is there a version of gpc built for Solaris 2.5? Or do you have any hints on getting the lastest source to compile under Solaris 2.5? Thanks Brad Zoltick NIH Email: brad(a)codon.nih.gov

1 0

Re: Compilation error
by Peter Gerwinski 08 Sep '96

08 Sep '96

> When I try to compile anything, even the simplest program, I get: > /usr/i486-linux/bin/ld: cannot open crt0.o: No such file or directory Perhaps the versions of GCC and GPC you are using don't coincide. Please check the following: gpc -c myprogram.pas gcc myprogram.o -o myprogram i.e. compile with GPC, but link with GCC. Did you already check that crt0.o is in a directory where GPC can find it? I remember some problems when I had crt0.o in the directory /usr/lib/gcc-lib/i486-unknown-linux/2.6.3. The problems disappeared when I moved the file to /usr/lib or /usr/local/lib. Good luck, Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter/

1 0

Re: External, C, AsmName, ...
by Peter Gerwinski 07 Sep '96

07 Sep '96

According to Berend de Boer: > Can we not simply see a C .h file as an interface-only module? I mean I can > simply import a c .h file which gpc parses and all symbols defined in that > header file are exported symbols? This would mean that GPC would have to parse C source. Remember that #include "foo.h" or (*$include "foo.h" *) just means that GPC treads the file `foo.h' just if its contents were written in our source file. A Pascal program using this mechanism would in fact look like the following: Program Test; extern int foo; extern void ioctl (int fildes, int cmd, ...); begin [...] end. You don't actually want this. ;-) > I don't know how far this is feasible. And it doesn't cover a.out modules > creating using other languages (GNAT, ...). But it does cover the majority of > cases you need externals. You can write a C header file for everything you want to import, independently of the language it was originally written in. Our point is that it must also be possible to write a *Pascal* interface for everything we want to import. Peter e-mail: peter.gerwinski(a)uni-essen.de home address: D\"usseldorfer Str. 35, 45145 Essen, Germany WWW: http://agnes.dida.physik.uni-essen.de/~peter

1 0

Re: External, C, AsmName, ...
by Berend de Boer 07 Sep '96

07 Sep '96

> Now I am looking out for reasonable ideas. Can we not simply see a C .h file as an interface-only module? I mean I can simply import a c .h file which gpc parses and all symbols defined in that header file are exported symbols? I don't know how far this is feasible. And it doesn't cover a.out modules creating using other languages (GNAT, ...). But it does cover the majority of cases you need externals. Groetjes, Berend.

1 0

Accessing external functions (was: Newbie question)
by Peter Gerwinski 06 Sep '96

06 Sep '96

Hello Patricio, > I currently teaching Operative System to students who only know Pascal, > so I'd like to know if we can use > Unix System calls in pascal programs wirh GPC ? I mean stuff like > 'fork()', 'exec()', etc... You can access any function using the "external", "C", and "asmname" directives: Procedure fork; external; means: there is an external Procedure "Fork" (first letter uppercase), Procedure fork; C; means: there is an external Procedure "fork" (everything lowercase), and Procedure fork; asmname 'foO'; means: there is an external Procedure with the funny name "foO" (last letter is uppercase in this example) which can be accessed from Pascal via the name "fork". Good luck, Peter

1 0

Newbie question
by Patricio Domingues 06 Sep '96

06 Sep '96

Hi, I currently teaching Operative System to students who only know Pascal, so I'd like to know if we can use Unix System calls in pascal programs wirh GPC ? I mean stuff like 'fork()', 'exec()', etc... Thanks in advance, Patricio.

1 0

Subscribe GPC's mail list
by Patricio Domingues 06 Sep '96

06 Sep '96

Thanks.

1 0

Compilation error
by John Michael Black 03 Sep '96

03 Sep '96

Hi all. I've had this error ever since I installed gpc, and I've never figured out why. I'm not all that experienced at unix. When I try to compile anything, even the simplest program, I get: /usr/i486-linux/bin/ld: cannot open crt0.o: No such file or directory Sorry if this seems like such a stupid question. Any ideas? -john

1 0

Problem with constant strings in gpc 1.2(2.7.2)#7 ???
by Peter Gerwinski 30 Aug '96

30 Aug '96

According to Harry Reed: > In the users' manual chapter "About Pascal and Extended Pascal > languages" > there is a small example on how to initialize constant strings. When I > try > compiling the fragment with gpc 1.2(2.7.2)#7 I get the error messages > as showm below. Is this a valid bug or am I just doing something wrong? Extended Pascal initialized *structured* variables are not (yet) implemented into GPC, only (parts of) Borland Pascal style initializers. However some fragments of Extended Pascal initializers *do* exist in gpc-parse.y, but I don't know when they will work. Everybody be invited to help implementing this! > MyStrings : array [1..MyStringsCount] of Ident value [ Square brackets do not work. Use parantheses instead. ^ > 1:'EXPORT'; 2:'IMPLEMENTATION'; 3:'IMPORT'; You may specify Indices etc., but they are ignored completely. Even worse: Initialization of Strings does not work at all. Only "simpler" types, i.e. array/record combinations containing Integers, Reals, etc. work. This is a known bug, but I forgot to document it. Thank you for pointing me to it. > PS: Great work on GPC! Keep it up!!!! Thanks a lot. :-) :-) I will continue unless somebody makes big efforts to stop me. ;-) Peter

1 0

Problem with constant strings in gpc 1.2(2.7.2)#7 ???
by Harry Reed 30 Aug '96

30 Aug '96

Hi, In the users' manual chapter "About Pascal and Extended Pascal languages" there is a small example on how to initialize constant strings. When I try compiling the fragment with gpc 1.2(2.7.2)#7 I get the error messages as showm below. Is this a valid bug or am I just doing something wrong? Cheers, Harry Reed doon(a)hrc.nevada.edu PS: Great work on GPC! Keep it up!!!! source program --- cut here --- 8< --- cut here --- 8< --- cut here --- program test; const MyStringsCount = 5; type Ident = string(20); var MyStrings : array [1..MyStringsCount] of Ident value [ 1:'EXPORT'; 2:'IMPLEMENTATION'; 3:'IMPORT'; 4:'INTERFACE'; 5:'MODULE']; begin end. compiler output --- cut here --- 8< --- cut here --- 8< --- cut here --- test.pas: In function `test': test.pas:9: parse error before `:' test.pas:9: Set constructor elements must be of ordinal type test.pas:9: missing comma test.pas:9: parse error before `;' test.pas:9: missing comma test.pas:9: parse error before `:' test.pas:9: Set constructor elements must be of ordinal type test.pas:9: missing comma test.pas:9: parse error before `;' test.pas:9: missing comma test.pas:9: parse error before `:' test.pas:9: Set constructor elements must be of ordinal type test.pas:9: missing comma test.pas:9: parse error before `;' test.pas:10: missing comma test.pas:10: parse error before `:' test.pas:10: Set constructor elements must be of ordinal type test.pas:10: missing comma test.pas:10: parse error before `;' test.pas:10: missing comma test.pas:10: initial value is of wrong type

1 0

gpc-2.7.2 for Linux-elf and Win32
by Peter Gerwinski 29 Aug '96

29 Aug '96

Hello folks, GPC-2.7.2 pre-release binaries for Linux-elf and Win32 (with CygWin beta 14) are now available on kampi and on agnes. Peter

2 1

gpc-2.7.2 on kampi.hut.fi
by Peter Gerwinski 27 Aug '96

27 Aug '96

Hello folks, the GPC 1.2 (2.7.2) pre-release is now available on kampi.hut.fi, too. The directory is jtv/gnu-pascal/gpc-2.7.2. Peter

1 0

ANNOUNCE: GPC 1.2 (2.7.2) pre-release
by Peter Gerwinski 27 Aug '96

27 Aug '96

Dear readers of the GPC list, we herewith announce a pre-release of GPC version 1.2, based on GNU C 2.7.2. This is a beta test release. We ask you to try the new GPC version on as many different platforms as possible, isolate bugs, correct the documentation, etc.. Once we know it is stable enough, we will do the "official" 1.2-2.7.2 release. New features: * easier to compile and install, also for cross-compilation * Texinfo documentation * More stable (Borland-style) objects * "String" bug fixed (see the old `PROBLEMS' file) * Extended Pascal "export foo = all" extension * PXSC operator fragments (redefineable symbols) * Precompiled Module/Unit interfaces (needn't #include the interface any longer) * AutoMake facility (automatically compile Modules/Units which have been changed) Source and binaries for DJGPP, EMX, Linux, FreeBSD, and IRIX5 are available per anonymous ftp from agnes.dida.physik.uni-essen.de in the directory gpc-2.7.2 (They also will be available on kampi.hut.fi, soon.) Have fun, Jan Jaap <J.J.vanderHeijden(a)student.utwente.nl> Peter <peter.gerwinski(a)uni-essen.de>

1 0

part of system.pas unit ready
by Berend de Boer 14 Aug '96

14 Aug '96

Hello All, I promised to create a System.pas unit, here is a large part of it (see next message). It seems to work on gnu-win32 systems (I tested it in a NT workstation). It probably also works on BSD unix systems. I used gpc 2.6.3 (which was more useful than I expected and have said so in the past )-: ), but I still encountered some strange things. So here my comments: 1. how do you use the __unsigned__ prefix? 2. Instead of using __byte__, etc. ranges should be better as Juki already suggested. 3. nil assignment to void pointer unacceptable? (var p: ^void; p := nil; ) Things I found missing: - schema types ( type t(max:integer) = array[0..max] of double; ) - function result variables ( function f(p: integer) = Result: integer; ) - pre-initialized variables - renaming in export clause I like to here comments, Berend.

1 0

System.pas module
by Berend de Boer 14 Aug '96

14 Aug '96

{ Created: 1996-07-26 System module for emulating the Borland Pascal System unit. This one is specific for the GNU Extended Pascal implementation and provides a System.Pas interface specifically to emulate the 16-bit Borland Pascal compilers. Copyright (C) 1996 Berend de Boer <berend(a)pobox.com> This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. Notes: - This one has hardcoded errorcodes!! It probably does not work on every machine. One really need to convert errno.h and stat.h first to make it portable. But I'll do that when I complete the Extended Pascal Posix interface for GNU Pascal. - assumes all Extended Pascal compilers have the local addition to provide an additional parameter to the Halt procedure. - certain procedures are almost untranslatable as Addr, Move and Str. They are therefore not translated, but require a change to the source code - there is no Assign or Close procedure for a file type variable, only for a text type variable. - Rename does not work for some reason. To do: - Append - FilePos - FileSize - Flush - Rename - Seek - Truncate $Date: 96-08-14 9:45 $ $Revision: 1 $ $Log: /Extended Pascal/Borland/GPC/system.pas $ * * 1 96-08-14 9:45 Berend * Borland Pascal's system unit for Extended Pascal, as far as applicable. } module System interface; export System = (shortint, byte, word, longint, PChar, pointer, string255, HeapError, ExitCode, PrefixSeg, Assign, BPChDir => ChDir, Close, Copy, Dec, Delete, Erase, GetDir, GetMem, Inc, IOResult, MaxAvail, MemAvail, BPMkDir => MkDir, ParamCount, ParamStr, BPRename => Rename, BPRmDir => RmDir, UpCase); type shortint = __byte__ integer; byte = __byte__ integer; word = __short__ integer; longint = integer; TChar = array[0..MaxInt] of char; PChar = TChar; pointer = void; string255= string(255); var HeapError: pointer; { Heap error function } {?ExitProc: Pointer; { Exit procedure } ExitCode: Integer; { Exit code } {?ErrorAddr: Pointer; { Runtime error address } PrefixSeg: Word; { Program segment prefix } InOutRes : integer; { I/O result buffer } procedure Assign(var t: text; protected Name: string); procedure BPChDir(protected s: string); procedure Close(var t: text); function Copy(protected s: string; Index: integer; Count: integer): string255; procedure Dec(var i: integer); procedure Delete(var s: string255; Index: integer; Count: integer); procedure Erase(var f: text); procedure GetDir(D: byte; var s: string); procedure Inc(var i : integer); function IOResult : integer; function MaxAvail: longint; function MemAvail : longint; procedure BPMkDir(protected s: string); function ParamCount: word; function ParamStr(Index: word): string255; procedure BPRename(var f: text; protected Newname: string); procedure BPRmDir(protected s: string); function UpCase(Ch: char): char; end. module System implementation; import StandardOutput; { support routines } const MaxPath = 4096; type TPath = array[0..MaxPath] of char; function StrPas(Str: PChar): String255; var i: integer; s: string255; begin i := 0; s := ''; while Str[i] <> Chr(0) do begin s := s + Str[i]; i := i + 1; end; StrPas := s; end; { StrPas } function unlink(path: __cstring__): integer; c; function chdir(path: __cstring__): integer; c; function getcwd(var buf: TPath; size: integer): PChar; c; function mkdir(path: __cstring__; mode: integer): integer; c; function _p_paramcount: integer; c; function _p_paramstr(num: integer; var str: string): Boolean; c; function rename(old: __cstring__; new: __cstring__): integer; c; function rmdir(path: __cstring__): integer; c; function MapUnixErrorToDosError(ErrorOccured: integer): integer; const EPERM = 1; { Not super-user } ENOENT = 2; { No such file or directory } EACCES = 13; { Permission denied } EBUSY = 16; { Mount device busy } EEXIST = 17; { File exists } ENOTDIR = 20; { Not a directory } ENOSPC = 28; { No space left on device } EROFS = 30; { Read only file system } EMLINK = 31; { Too many links } ENAMETOOLONG= 91; { File or path name too long } var Result: integer; UnixError: integer; begin if ErrorOccured then begin UnixError := EACCES; { should be errno... } case UnixError of ENOENT : Result := 2; ENOTDIR : Result := 3; EPERM, EACCES, EEXIST, ENOSPC, EROFS, EMLINK, ENAMETOOLONG: Result := 5; EBUSY : Result := 152; otherwise Result := UnixError; end; { case } end else begin Result := 0; end; MapUnixErrorToDosError := Result; end; { MapUnixErrorToDosError } { the system routines itself } procedure Assign; var b : BindingType; begin unbind(t); b := binding(t); b.Name := Name; bind(t, b); b := binding(t); end; procedure BPChDir(s: string); begin InOutRes := MapUnixErrorToDosError(chdir(s)); end; { ChDir } procedure Close; begin unbind(t); end; function Copy; begin if Index+Count > length(s) then Copy := SubStr(s, Index) else Copy := SubStr(s, Index, Count); end; procedure Dec; begin i := i - 1; end; procedure Delete; begin if Index = 1 then begin if 1+Count > length(s) then s := '' else s := s[1+Count..length(s)]; end else begin if Index+Count > length(s) then s := s[1..Index-1] else s := s[1..Index-1] + SubStr(s, Index+Count); end; end; procedure Erase; var bt: BindingType; begin bt := Binding(f); unbind(f); InOutRes := MapUnixErrorToDosError(unlink(bt.name)); end; procedure GetDir(D: byte; var s: string); var Buffer: TPath; pc: PChar; i: integer; begin pc := getcwd(Buffer, MaxPath); s := ''; i := 0; while pc[i] <> Chr(0) do begin s := s + pc[i]; i := i + 1; end; { call to StrPas does not work???? s := StrPas(pc);} end; { GetDir } procedure Inc; begin i := i + 1; end; function IOResult : integer; begin IOResult := InOutRes; InOutRes := 0; end; function MaxAvail; begin MaxAvail := MaxInt; end; function MemAvail; begin MemAvail := MaxInt; end; procedure BPMkDir; begin InOutRes := MapUnixErrorToDosError(mkdir(s, 8#0700)); end; { BPMkDir } function ParamCount; begin ParamCount := _p_paramcount - 1; end; { ParamCount } function ParamStr; var Str : string255; Success: Boolean; begin Success := _p_paramstr(Index, Str); if Success then ParamStr := Str else ParamStr := ''; end; { ParamStr } procedure BPRename; var bt: BindingType; begin bt := Binding(f); unbind(f); InOutRes := MapUnixErrorToDosError(rename(bt.name, Newname)); bt.Name := Newname; bind(f, bt); end; { BPRename } procedure BPRmDir; begin InOutRes := MapUnixErrorToDosError(rmdir(s)); end; { RmDir } function UpCase; begin if Ch in ['a'..'z'] then UpCase := Chr(Ord('A') + (Ord(Ch) - Ord('a'))) else UpCase := Ch; end; { we need an initialization section because gpc doesn't yet have initialized variables } to begin do begin {HeapError := nil;} InOutRes := 0; end; end.

1 0

Program to test system.pas unit
by Berend de Boer 14 Aug '96

14 Aug '96

Here a small program which 'demonstrates' the use of the system.pas unit. -------------------------test.pas------------------------ #include "system.pas" program Test(Output); import System; var f: text; path: string(1024); i: integer; begin Assign(f, 'tmp.tmp'); Erase(f); writeln('IOResult = ', IOResult); Assign(f, 'q.pas.bak'); BPRename(f, 'renamed.file'); writeln('IOResult = ', IOResult); GetDir(0, path); writeln('Current dir = ', path); BPMkDir('mydir'); writeln('IOResult = ', IOResult); BPRmDir('mydir'); writeln('IOResult = ', IOResult); writeln('ParamCount = ', ParamCount); for i := 1 to ParamCount do begin writeln('parameter ', i, ' = ', ParamStr(i)); end; end. -------------------------test.pas------------------------ Groetjes, Berend. (-:

1 0

HELP
by OESY＠HSOCIETY.berlinet.de 11 Aug '96

11 Aug '96

HELP Bye OESY -- OS/2: Never stop that feeling... Using Speed-Pascal/2 (SP/2)! Questions? PM me! ## CrossPoint v3.1 ##

1 0

GPC autoconf-ed
by J.J. van der Heijden 29 Jul '96

29 Jul '96

I mentioned my rts-autoconf adventures on the mailinglist; I dumped a modified RTS (rts-autoconf.tar.gz) in my homedir on agnes.dida.... Usage is straightforward: cd to your $(gpcbin)/rts directory and run ../../$(gpcsrc)/rts/configure ; make Autoconf is the way to go for GNU software and I think GNU Pascal should use it too. This is not finished: I have to do GPC itself, plus the result of a number of tests is not yet used in the RTS makefile. But it deals with all the -DSYSV -DNO_CASECMP -DMY_VERY_WEIRD_SYSTEM flags. It can also do crosscompilation: run CC=i386-go32-gcc AR=i386-go32-ar RANLIB=i386-go32-ranlib \ ../../$(gpcsrc)/rts/configure --target=i386-go32 ; make to get a djgpp RTS (on Linux) I also added a definition in Makefile.in to build libgpc as a shared lib. A 'make libgpc.so.2.8 CFLAGS="-O2 -fPIC"' results in: lrwxrwxrwx 1 root root 13 Jul 29 02:06 /usr/lib/libgpc.so -> libgpc.so.2.8* -rwxr-xr-x 1 root root 48960 Jul 29 02:06 /usr/lib/libgpc.so.2.8* This reduces the size of executables: -rwxr-xr-x 1 janjaap users 2856 Jul 29 02:09 hello* -rw-r--r-- 1 janjaap users 63 Jul 29 02:07 hello.pas Statically linked, the minimum is ~33K. BTW: this will work only on ELF un*x systems and I have only tested it on Linux. "ldd hello" shows: libgpc.so.2.8 => /usr/lib/libgpc.so.2.8 libm.so.5 => /lib/libm.so.5.0.5 libc.so.5 => /lib/libc.so.5.2.18 I will add this as an "--enable-shared" option to "configure" As for your desire to release 2.7.2: I can have this finished by friday (I think ...), then it should be tested on a number of different platforms. I'm especially curious about BSD unix systems, because I developed this on Linux, which is essentially SysV. I created a "configure.bat" for the dos world where there's no bash. Peter, you have the same with OS/2 so maybe you can take a look at it and create something simular for the EMX thing. (I ripped of binutils and GCC to create "configure.bat") /janjaap --- Thus spake the master programmer: "After three days without programming, life becomes meaningless." [The Tao Of Programming]

1 0

Re: release?
by J.J. van der Heijden 29 Jul '96

29 Jul '96

On Sat, 27 Jul 1996 23:00:20 +0200 (MET DST, Peter Gerwinski <peter(a)agnes.dida.physik.uni-essen.de> wrote: >Hello again, > >there was a bug in my GPI stuff: File types did not >survive transport through a GPI file. I am sure this >was not the last bug, but now a very useful physics >program works, so I am optimistic. > >gpc-pg-1.2-2.7.2.tar.gz has been updated. > >Release? > I know of two good reasons not to release 2.7.2 (yet): 1) Remaining bug(s). Last time I looked, that "functions returning a string are called three times" bug was still there. That's a serious problem for anything bigger than "Hello world". Also (I mentioned before): gpc fails two of the "paranoia" Pascal compiler float tests because it loses significant digits. I'd be carefull with that physics program, Peter. 2) Almost all messages / email I read about GPC are not about "compiler bug X keeps me from using it" or "feature Y is missing" but "HOW DO I INSTALL THE THING ON MY SYSTEM ?" "Where's the docs" is a popular one too. That's why I'm busy rewriting configuration scripts using GNU autoconf and have created .bat files for the dos version. Then, building it is reduced to "configure ; make" I have it done for the RTS, am working on GPC itself now. I think it should be able to locate gcc sources and objects by itself (if they are in a reasonable location). It works great for the RTS already. Arguments in favour of release: 1) New features (These have to be mentioned in a "NEWS" file. I think we have (a) better BPascal Objects (b) Improved module support, but there must be other user-visible changes as well. 2) Releasing 2.7.2 because gcc-2.6.3 is obsolete is an argument, but I don't think we'll have to wait much longer for gcc 2.8 and from what I've seen so far, it's not radically different from gcc-2.7.x (internally). So porting gpc to 2.8 should take too long. If this thing is to be released in one or two weeks I think we should make a "release candidate" so everybody can check wether his patches are in and binaries for all platforms can be built. /janjaap --- Thus spake the master programmer: "After three days without programming, life becomes meaningless." [The Tao Of Programming]

1 0

GNU Pascal WWW home page
by Peter Gerwinski 26 Jul '96

26 Jul '96

Hello folks, I have constructed a GNU Pascal WWW home page: http://agnes.dida.physik.uni-essen.de/~gnu-pascal Comments, suggestions, bug reports welcome. Peter <peter.gerwinski(a)uni-essen.de>

1 0

forwarded Usenet article
by operator＠civpczaal.utwente.nl 26 Jul '96

26 Jul '96

[article forwarded from Student (operator(a)civpczaal.utwente.nl)] Path: news.nic.utwente.nl!surfnet.nl!howland.reston.ans.net!newsfeed.internetmci.com!btnet!news.enterprise.net!usenet From: eurgain(a)enterprise.net (Alistair Hamilton) Newsgroups: comp.os.msdos.djgpp Subject: Re: GPC112B.ZIP & GPT112B.ZIP Date: Tue, 23 Jul 1996 23:30:15 GMT Organization: Ty Eurgain Lines: 45 Message-ID: <4t3nq5$ih8(a)news.enterprise.net> References: <960718165325_74617.365_EHH119-1(a)CompuServe.COM> NNTP-Posting-Host: max02-063.enterprise.net X-Newsreader: Forte Free Agent 1.0.82 "Troy D. Van Horn" <74617.365(a)CompuServe.COM> wrote: >Hi all, > I just last night discovered these files for a PASCAL compiler last night >on a simtelnet mirror, and I was wondering where I could get some documentation >on how to use either compiler? The source file is really large (>8Meg) so I >would really prefer not to download it. Where should I look for documentation >on GPC and the differences between GPC and GPT version? (I already know >PASCAL). >Thanks, Troy Van Horn 74617.365(a)compuserve.com >Please E-mail all replys direct to my address, Thank you. PROGRAM GetSomeHelp (Input, Output|) CONST PascalExists = TRUE; VAR CIsBetter : BOOLEAN; BEGIN WRITELN('I would really like this info too. OK, as an ex Unix/C/X nerd, I cannot dispute that C cannot be beaten (except , perhaps, by C++). However, age has revealed that Pascal really is useful sometimes! Typing B E G I N and E N D sure as hell is! (I have programmed for accountancy apps ZZZzzzz in Pascal - but better P than C in that case).'); CIsBetter := TRUE; WRITELN ('I also have the GNUPascal binaries, and am held back from using them by lack of documentation. So come on, folks! At least you might tell us to b****r off and consult another really tesious group specially for Pascal bores, but if you do, it least you might tell us waht the group is called.'); WRITELN ('Regards'); WRITELN ('Alistair') END. :-) :-) :-)

1 0

Re: Programmers wanted: CRT, Graph, etc. for GNU Pascal
by Peter Gerwinski 24 Jul '96

24 Jul '96

Berend de Boer wrote: > The point of having an extended pascal compiler > is being portable, at least that's the goal. I agree to the goal of being portable. But I don't think that we can reach this goal just by obeying the Extended Pascal standard. (More about this in a separate mail.) > Assembler and Dos are not the way to go. The point of having a > System, Crt, Printer of Graph unit is entirely uninteresting if it is not > portable except if you want to port Dos Borland Pascal programs to Extended > Pascal. I want to port DOS Borland programs to *GNU Pascal*, and I am sure that many people intend the same. System, Crt, Printer and Graph for GNU Pascal would be *highly interesting* for those programmers who are accustomed to Borland Pascal and think about changing to GNU Pascal (instead of changing to Delphi or C++) in the future. To have those Units on the DOS platform only would be a starting point -- portable versions would be preferred, of course. > Why should anyone want to do this? I assume everyone is programming for > Windows 95/Windows NT or Unix. Are programmers really running Dos today?? Yes, I am for example. More concrete, I am working in an OS/2 DOS box and in the Linux DOSemu (with Novell DOS 7). I do not intend to use Windows 95 or Windows NT, and it will take a long time to get my clients using UNIX (Linux). Up to that day, I will prefer writing good DOS programs to writing poor Windows programs (with a lot of special effects, but slow, uncomfortable, and unstable). In the long term, I intend to become completely platform independent with my BO5 library, i.e. my programs will run under Windows 95/NT, too, just be recompiling. But for the moment, I am stuck with DOS. > Quite a few calls in System or Crt do simply not apply in non-Dos > environments. There are really few of them, but since the source exists, I suggest that somebody ports *everything* to the DOS version of GPC, and we can make the "potentially portable" parts portable in a second stage. > A portable Printer library?? What's the problem? It would just be a Unit (Module) which "knows" the file name of the printer device for the actual operating system and opens a text file "lst", so the application program doesn't need to care about how the printer is named on the operating system. > The calls from System that are portable can for 90% be written in plain > Extended Pascal. However, it would be nice to have a compatibility Unit (Module) which simplifies the port of Borland Pascal programs to GNU Pascal. > We should do better by following existing standards and provide EP > interfaces for that. Borland's Units *are* an existing standard. There may be other standards, but I think these Units are very popular. Furthermore, we *have* the source for them, so I ask again: ******************************************************* We have free sources of Borland Units (CRT, Graph, ...) which cry for being ported to GNU Pascal. Anybody wants to do the job? ******************************************************* Greetings, Peter

2 2

Re: The purpose of GNU Pascal
by Berend de Boer 24 Jul '96

24 Jul '96

> By the way: Where are those many different Extended Pascal compilers? You have Prospero's compilers for Dos/Windows and OS/2, DEC has a Pascal compiler on VAX which implements almost all of the Extended Pascal standard. I'm not sure of the Macintosh, there is a Pascal compiler on the Mac from Warriar, but if that one implements Extended Pascal?? I'm going to research that today. > You don't lose portability by simply including bit-shift operators, objects, etc. into > the Extended Pascal standard. But they didn't include them. No idea, why. Objects are in another standard. bit-shift operators have to depend upon memory layout and byte ordering. But I agree with you that they should be available. > And what about, say, user-defined operators which are almost a *must* for scientific calculations? I completely agree with you that user-defined operators are a must in a field where those operators defined. In most other areas they are not. We can probably atttach a dozen explanations to the following construct: House := PersonA + PersonB; The example you gave had exactly one meaning clear to everyone. > But I don't intend to spend my time with trying to satisfy standard specifications I don't realize their > sense. I can quite agree with you. But don't forget: there are millions, literally millions of code written in some form of Standard Pascal out there in the world. For such people an alternative in the form of GNU Pascal could really be interesting. > How would you define "real Pascal spirit"? Let the compiler do as much as possible for you: basically that comes down to no untyped pointers. I *have* to use much more untyped pointers in Borland Pascal (in obtaining addresses from procedures, strings, etc.) than would be necessary. > Extended Pascal may be good, but GPC will be even better. It is time to set up the next standard! Completely agreed. I can accept the modifications to GPC.GUIDE you presented. Of course a compiler which accepts Borland Pascal syntax is a major advantage. Makes porting much easier. And if you really have good alternatives for Borland's standard libraries... Groetjes, Berend.

2 1

Re: Extended Pascal
by Berend de Boer 24 Jul '96

24 Jul '96

>> Last time I really looked at gpc (beginning 1995) it missed simply too much. > What is missing? If you publish a list, perhaps somebody would like > to do the job of implementing it. OK, I hope to present a list of missing features to this list within a few days. Groetjes, Berend.

1 0

Re: Programmers wanted: CRT, Graph, etc. for GNU Pascal
by Berend de Boer 24 Jul '96

24 Jul '96

>> What use does it have when you already have a good Dos compiler, i.e. Borland >> Pascal? It fullfills all my needs for the Dos environment. > Not mine, and when you look into comp.lang.pascal.borland you will find Yes, I did see your problem, but I never encountered these things. My units are small, so I don't have codesegment faults, and for large arrays I have a few objects of my own which either use XMS/EMS or conv memory to present large arrays. But I see. I make a Posix interface for GPC first and present any difficulties I encounter to you so you can make the necessary changes to the compiler ok? > Note, I HAVE the source of the Graph unit!!! Ah, I now remember you saying this, but shouldn't this be: I have the source of a Graph unit? >> - qualifed imports available? > What's that? To reference a symbol from these modules you always need to add the module name as prefix. Maybe I don't need this now, I'll see. OK, I see if I can write something with gpc. I'll tackle the Posix interface and the Dos unit first. I start using djgpp's port. Which one should I take? Groetjes, Berend.

1 0

Re: The purpose of GNU Pascal
by Berend de Boer 23 Jul '96

23 Jul '96

> Furthermore, I could not write my real-world programs > just with Extended Pascal; I need, for example, bit shift operators, GetMem, > FreeMem, and Objects. You don't need GetMem and FreeMem as I did show you :-) But you need the other two. > and integrate them in one powerful compiler which will set up the next standard. I certainly don't agree with this if the compiler does not have switches to accept only Extended Pascal or only ISO Pascal or only ... Programs conforming to the Extended Pascal will probably not only work in many different environments but also with many different compilers. I don't want to loose that feature/possibility. Extending a compiler with more features is not a problem, but IMO, one should complete the standard parts first. > It is okay to support Extended Pascal (why not?), but in my opinion, > Extended Pascal is not clear and powerful enough to serve as "the new > Pascal standard", so programmers are not motivated to switch from > Borland Pascal to Extended Pascal. I think, this is the reason why > computer vendors do not provide Extended Pascal compilers. Hmm, I don't think that Extended Pascal is not powerful enough. It's true that it does not contain everything one needs, but that's mainly due to the goal of creating a portable language. And the new object standard also rectififies some omissions. I can create everything I want in Extended Pascal programs with only a few routines accessing non-standard parts, mainly concerning bit-operators (you can do this with Extended Pascal, but it's a bit more tedious), and system call access (assembly). A compiler which accepts every Pascal standard and does everything would be nice of course. But there is quite some work to do. So I would like to add that we first focus on the ISO Pascal and Extended Pascal part. Even the ISO Pascal standard, according to Scott Moore (have email address if anyone wants it), needs much looking after. Don't overlook the Pascal standards boys (no girls here I assume). They do much more than you think, but different than Borland Pascal which has quite a C feeling and lost too much of it's real Pascal spirits. Groetjes, Berend.

2 1

Re: Programmers wanted: CRT, Graph, etc. for GNU Pascal
by Berend de Boer 23 Jul '96

23 Jul '96

> There are really few of them, but since the source exists, I suggest > that somebody ports *everything* to the DOS version of GPC, and we can > make the "potentially portable" parts portable in a second stage. What use does it have when you already have a good Dos compiler, i.e. Borland Pascal? It fullfills all my needs for the Dos environment. Besides, what do you mean by 'porting', i.e. should do exactly the same with every quirk inherited or just similar enough? The system unit can be implemented easily, I've done already that for 90%. A few things you can't do like Inc, and some functions which have overloaded parameters (Assign, Close, ...). If similar is good enough, I suggest we define a POSIX interface first, starting from the one I wrote for Dos, and build the Dos unit on top of that. We can do the Crt unit in the same manner, building on top of the curses library. > We have free sources of Borland Units (CRT, Graph, ...) Note, you don't have the source of the Graph unit. >From my viewpoint, Borland Pascal does everything, so spending time for a Dos only solution is not a thing I want. And it is not necessary. And is gpc already for such projects? I see some problems with gpc as I know it: - no renaming, necessary to avoid clashing with c libraries. - qualifed imports available? - how to link with C libraries? You have the c; construct, but exactly how does it work? What does it do? Groetjes, Berend.

2 1

Re: GPC: "how to contribute" file -- job list
by J.J. van der Heijden 23 Jul '96

23 Jul '96

On Fri, 5 Jul 1996 11:01:03 +0200 (MET DST), Peter Gerwinski <peter(a)agnes.dida.physik.uni-essen.de> wrote: [snip] > > List of jobs > ------------ > [snip] > DJ version (DOS) JanJaap works I came across a new DJGPP app called "rhide" It's a Borland-clone IDE that works with djgpp (C & C++) I'll see if I can hack it to work with GPC. Would make a neat Turbo Pascal replacement ;) > > Installation instructions Larry ? > Documentation Larry ? > Larry, if you're there: I rewrote much of the installation notes. Also, I took a "Using and porting GCC" manual, did a "search GCC -- replace with GPC" action, TeXi-fied much of the existing readme's and put it all together. Let me know how to send this stuff to you. > > Run time system Juki works I have preliminary, new "autoconf" scripts for RTS. Should reduce configuration to "configure ; make". Handles all CFLAGS, platform dependant includes etc. I'll see if I can do something simular for GPC itself. Some other RTS related remarks: *) GNU is going international. How about adding localization to RTS? Won't work if you're not on un*x, but can be buildtime selectable (flag to autoconf script) *) How 'bout RTS as an ELF shared library? *) Documentation again: RTS needs documentation. I looked at the Cygnus and DJGPP C libraries and they use Texinfo fragments in the source, extract it with some utility to one big texinfo file. That way it's easy to keep documentation up to date. > Replacement for (parts of) CRT Peter beta (BO5) > Replacement for Graph Peter planned > Replacement for Turbo Vision Peter planned > Now, if somebody would port TVision to un*x (use termcap / info instead of direct screen writes) it would probably be not too much work to port this RHIDE thing to unix. Then you have an IDE. (sorry guys, I'll never be a fan of emacs) /janjaap --- Thus spake the master programmer: "After three days without programming, life becomes meaningless." [The Tao Of Programming]

1 0

Extended Pascal
by Peter Gerwinski 23 Jul '96

23 Jul '96

Berend de Boer wrote: > I would do that when the Extended Pascal port is completed and the object > extensions are added (I think it's not wise to use Borland style objects). Why not? Borland style objects have some advantages with respect to those of Object Pascal (Delphi). Concerning portability, GNU Pascal itself is portable, and I don't know any Object Pascal compiler except Delphi. (I have heared about Object Pascal for the Apple Macintosh, but never seen it.) So there is no compiler around (except those of Borland) to preserve compatibility to. > Last time I really looked at gpc (beginning 1995) it missed simply too much. What is missing? If you publish a list, perhaps somebody would like to do the job of implementing it. Or do it by yourself -- it should be relatively easy since Borland style objects work in GNU Pascal. Yours, Peter ******************************** ATTENTION: New e-mail address! ******************************** -------------------------------------------------------------------------------- Dipl. Phys. Peter Gerwinski Fachbereich Physik Universitaet-GH Essen Phone: +49-201-183-2763 D-45117 Essen Fax: +49-201-183-2120 Germany e-mail: peter.gerwinski(a)uni-essen.de --------------------------------------------------------------------------------

1 0

Turbo Vision, IDE
by Peter Gerwinski 23 Jul '96

23 Jul '96

Jan Jaap van der Heijden wrote: > Now, if somebody would port TVision to un*x (use termcap / info instead of > direct screen writes) it would probably be not too much work to port this > RHIDE thing to unix. Then you have an IDE. (sorry guys, I'll never be a fan > of emacs) I am working on something similar to a port of Turbo Vision to UNIX and any other operating system I can access. The project is called BO5; version 0.0.1 was released on 19. December 1995. For more information about BO5 look at ftp://kampi.hut.fi/jtv/gnu-pascal/turbo-alpha/README.BO5. By the way: There is a Borland-style IDE for Linux called XWPE. Look at the Debian distribution of Linux in the subdirectory source/devel. Yours, Peter ******************************** ATTENTION: New e-mail address! ******************************** -------------------------------------------------------------------------------- Dipl. Phys. Peter Gerwinski Fachbereich Physik Universitaet-GH Essen Phone: +49-201-183-2763 D-45117 Essen Fax: +49-201-183-2120 Germany e-mail: peter.gerwinski(a)uni-essen.de --------------------------------------------------------------------------------

1 0

Re: GPC: "how to contribute" file -- job list
by Berend de Boer 23 Jul '96

23 Jul '96

> Now, if somebody would port TVision to un*x (use termcap / info instead of > direct screen writes) it would probably be not too much work to port this > RHIDE thing to unix. Then you have an IDE. (sorry guys, I'll never be a fan > of emacs) I would do that when the Extended Pascal port is completed and the object extensions are added (I think it's not wise to use Borland style objects). Last time I really looked at gpc (beginning 1995) it missed simply too much. Groetjes, Berend.

1 0

The purpose of GNU Pascal
by Peter Gerwinski 23 Jul '96

23 Jul '96

Hello folks, I think it's time to discuss about the purpose of GNU Pascal as stated by Juki (jtv) in GPC.GUIDE. More concrete: I disagree with some points and I would suggest to modify them. jtv> The purpose of the GNU Pascal project is to produce a jtv> Pascal compiler (called GNU Pascal or gpc) which jtv> - may be distributed under normal GNU license conditions jtv> - can genarate code and run on any computer for which the GNU C jtv> compiler can genarate code and run. Concerning the points above, I agree in complete. jtv> - supports both the Pascal standard and the Extended Pascal jtv> standard as defined by ISO and ANSI and IEEE. jtv> (ISO 7185:1990, ISO/IEC 10206:1991, ANSI/IEEE770X3.160-1989) It is okay and desireable to support these standards, but I wouldn't agree to obey the standards "come what may". As far as I would guess, 80% of all Pascal programmers are working with Borland Pascal or Delphi or a compatible system like Virtual Pascal, Speed Pascal, FPK Pascal, GNU Pascal or one of those ShareWare Pascal compilers written for the DOS platform. On the other side I know of only two Extended Pascal compilers: Prospero's Extended Pascal and GNU Pascal. So I call in question that Extended Pascal is more portable than Borland Pascal. Furthermore, I could not write my real-world programs just with Extended Pascal; I need, for example, bit shift operators, GetMem, FreeMem, and Objects. I think the way to go is to collect the good features of all the existing standards, ISO 7185, ISO 10206, Borland Pascal, Object Pascal, Pascal-SC, and integrate them in one powerful compiler which will set up the next standard. This is necessary, in my opinion, because *none* of the standards mentioned above is what I would call an "ideal" Pascal compiler by itself. I would be satisfied only with the *combination* of them. Apart from some bugs, the current version 1.1-2.6.3 of GNU Pascal is the best Pascal compiler I have ever seen, and we are not too far from making it even better. jtv> One reason for this is the desire to promote the use of Extended jtv> Pascal, which combines the clarity of Pascal with powerful jtv> tools (e.g. modules and string manipulation) suitable for real-life jtv> programming. Pascal was originally designed for teaching. jtv> Extended Pascal provides a smooth way to proceed to challenging jtv> programming tasks without learning a completely different language. The same holds for Borland Pascal. (In my opinion, the goals stated above are even better realized in Borland Pascal, but that's a matter of taste.) And there are other standards around, for example Pascal-SC (also named PXSC -- Pascal eXtensions for Scientific Calculations) with very useful features such as user-definable operators and procedure overloading. I would suggest to fix the goal of combining the clarity of Pascal with powerful real-life tools -- independently of a concrete standard. So my suggestion for the "Purpose" section of GPC.GUIDE would begin with: The purpose of the GNU Pascal project is to produce a Pascal compiler (called GNU Pascal or gpc) which - combines the clarity of Pascal with powerful tools suitable for real-life programming, - supports both the Pascal standard and the Extended Pascal standard as defined by ISO, ANSI and IEEE (ISO 7185:1990, ISO/IEC 10206:1991, ANSI/IEEE 770X3.160-1989), - supports other Pascal standards (UCSD Pascal, Borland Pascal, Pascal-SC) in so far as this serves the goal of clarity and usability, - may be distributed under normal GNU license conditions, - can generate code and run on any computer for which the GNU C compiler can generate code and run. Pascal was originally designed for teaching. GNU Pascal provides a smooth way to proceed to challenging programming tasks without learning a completely different language. jtv> GNU Pascal compiler is part of the GNU Compiler family jtv> combining a language independent part of the GNU Compiler with jtv> a Pascal specific front end. jtv> jtv> Other compilers of the family currently include compilers for jtv> the C, C++ and Objective C languages. Since this is true, we should compare the purpose of GNU Pascal with that of GNU C. In the section "Warning Options" of the GNU C info documentation (gcc), in the context of the "--pedantic" option which tells GCC to follow the ANSI standard as strictly as possible, we can find the following: gcc> A feature to report any failure to conform to ANSI C might be gcc> useful in some instances, but would require considerable gcc> additional work and would be quite different from `-pedantic'. We gcc> recommend, rather, that users take advantage of the extensions of gcc> GNU C and disregard the limitations of other compilers. Aside gcc> from certain supercomputers and obsolete small machines, there is gcc> less and less reason ever to use any other C compiler other than gcc> for bootstrapping GNU CC. In the same spirit I would suggest: Implement all reasonable extensions into GNU Pascal, encourage their use, and forget about the limitations of other compilers (such as UCSD or Borland Pascal) or standard definitions (such as Standard Pascal or Extended Pascal). jtv> ABOUT THE PASCAL AND EXTENDED PASCAL LANGUAGES jtv> ---------------------------------------------- jtv> jtv> [...] jtv> jtv> Extended Pascal is a standardized language which contains so jtv> significant extensions to Pascal that it is best regarded as jtv> a new language. It is currently not very well known, and computer jtv> vendors do not seem to be eager to provide compilers for it. jtv> Thus, there is social need for GNU Pascal supporting Extended jtv> Pascal. It is okay to support Extended Pascal (why not?), but in my opinion, Extended Pascal is not clear and powerful enough to serve as "the new Pascal standard", so programmers are not motivated to switch from Borland Pascal to Extended Pascal. I think, this is the reason why computer vendors do not provide Extended Pascal compilers. As far as I have understood, the committee which defined the standard knew Borland Pascal, so I simply cannot understand how they could define a standard which lacks some important features of existing compilers, for example the bit-shift operators or dynamical memory allocation with user-specified size (GetMem and FreeMem). There is social need for a *new* standard which integrates ISO 10206 Extended Pascal and other standards such as Borland Pascal and Pascal-SC. With GNU Pascal, we can *create* that new standard. For those who like standards: I suggest to implement compiler switches into GNU Pascal which force GPC to behave (inasmuch as possible) like a compiler of the specific standard, for example --standard-pascal for an ISO 7185 Standard Pascal compiler --extended-pascal for an ISO 10206 Extended Pascal compiler (which rejects Borland extensions) --borland-pascal for emulating the Borland Pascal compiler (which rejects Extended Pascal extensions) --pascal-sc for emulating a Pascal-SC compiler (not yet possible) but the default should be to accept *all* of the Pascal dialects and any combination of them, for example a Borland-style Unit which deals with Extended Pascal complex numbers and exports a Pascal-SC-style operator for matrix multiplication. (You actually can write such a thing with the current version of GNU Pascal!) jtv> Please refer to the document borland2ep.doc to find out more about jtv> the differences between Turbo Pascal and Extended Pascal. jtv> That document is written by Berend de Boer. I wrote another document about the same topic. It is somehow obsolete since GNU Pascal now understands most of the Borland Pascal language, but nevertheless you can find it at ftp://agnes.dida.physik.uni-essen.de/gpc-2.6.3/bp2ep.doc I am looking forward for reactions. Yours, Peter

1 0

Re: Programmers wanted: CRT, Graph, etc. for GNU Pascal
by Berend de Boer 22 Jul '96

22 Jul '96

Hay guys, Seems part of my message was lost, here the remaing part: > The calls from System that are portable can for 90% be written in plain Extended > Pa The calls from System that are portable can for 90% be written in plain Extended Pascal. We should do better by following existing standards and provide EP interfaces for that. Groetjes, Berend. (-:

1 0

Re: Programmers wanted: CRT, Graph, etc. for GNU Pascal
by Berend de Boer 20 Jul '96

20 Jul '96

First of all, does anyone know if there is a gpc archive? I went on holiday with my computer stuck and my compuserve mailbox overflooded (it seems to have a max of 100msgs). I miss about 1 week. > Florian has got free sources > for Borland compatible Units like SYSTEM, CRT, PRINTER, and GRAPH > for FPK Pascal which cry for being ported to GNU Pascal. > This should not be too difficult at least for the DOS platform because > FPK Pascal uses the same assembler syntax as GNU Pascal, and both > compilers can run with the GO32 DOS extender. The point of having an extended pascal compiler is being portable, at least that's the goal. Assembler and Dos are not the way to go. The point of having a System, Crt, Printer of Graph unit is entirely uninteresting if it is not portable except if you want to port Dos Borland Pascal programs to Extended Pascal. Why should anyone want to do this? I assume everyone is programming for Windows 95/Windows NT or Unix. Are programmers really running Dos today?? Quite a few calls in System or Crt do simply not apply in non-Dos environments. A portable Printer library?? The calls from System that are portable can for 90% be written in plain Extended Pa

1 0

Programmers wanted: CRT, Graph, etc. for GNU Pascal
by Peter Gerwinski 07 Jul '96

07 Jul '96

Hello, everybody! I am in contact with Florian Klaempfl, the author of FPK Pascal, another free 32-bit Pascal compiler. Florian has got free sources for Borland compatible Units like SYSTEM, CRT, PRINTER, and GRAPH for FPK Pascal which cry for being ported to GNU Pascal. This should not be too difficult at least for the DOS platform because FPK Pascal uses the same assembler syntax as GNU Pascal, and both compilers can run with the GO32 DOS extender. (GNU Pascal for GO32 can be downloaded as ftp://kampi.hut.fi/jtv/incoming/turbo11b.zip - see the recent mail by Jan Jaap. :-) Anybody wants to do the job? Yours, Peter -------------------------------------------------------------------------------- Dipl. Phys. Peter Gerwinski Fachbereich Physik Universitaet-GH Essen Phone: +49-201-183-2763 D-45117 Essen Fax: +49-201-183-2120 Germany e-mail: peter.gerwinski(a)uni-essen.de --------------------------------------------------------------------------------

1 0

NewsGroup announcement (from rs1)
by Peter Gerwinski 07 Jul '96

07 Jul '96

According to Peter Gerwinski: >From gpc-request(a)santra.hut.fi Sun Jul 7 02:54:19 1996 From: Peter Gerwinski <peter(a)agnes.dida.physik.uni-essen.de> Message-Id: <199607070038.CAA14344(a)agnes.dida.physik.uni-essen.de> Subject: NewsGroup announcement To: gpc(a)hut.fi Date: Sun, 7 Jul 1996 02:38:12 +0200 (MET DST) X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Hello, folks! Did somebody already announce GPC 2.6.3 with Borland extensions to the NewsGroups? If not, I am planning to crosspost the following to comp.lang.pascal.misc, comp.lang.pascal.borland, and comp.lang.pascal.iso-ansi: 8<------------------------------------------------------------------------------ NewsGroups: comp.lang.pascal.misc,comp.lang.pascal.borland,comp.lang.pascal.ansi-iso Followup-To: comp.lang.pascal.misc This is the announcement of version 1.1-2.6.3 of GNU Pascal (The announcement is somehow late since the actual release was on 19. December 1995. This won't happen again with the release of the next version.) -------- GNU Pascal is, as the name says, the Pascal compiler from the GNU family. This means: - 32 bit compiler, no limits, highly optimizing - runs on most operating systems (including DOS, OS/2, Linux, or any other UNIX) - FreeWare according to the GNU General Public License - compatible to other GNU tools such as GNU C and GNU debugger The compiler accepts the following language standards: - ISO 7185 Standard Pascal - ISO 10206 Extended Pascal (90%) - Borland Pascal (80%) Some highlights: - from Extended Pascal: complex numbers, initialized variables, structured function return values, modules - from Borland Pascal: inc, dec, shl, shr, absolute variables, units, objects - GNU extensions: min, max, (PXSC-style) user-definable operators -------- The source for the ISO compiler (and more information) is available from ftp://kampi.hut.fi/jtv/gnu-pascal Additional source for Borland Pascal compatible (and other) extensions is at ftp://kampi.hut.fi/jtv/gnu-pascal/turbo-alpha Binaries for DOS and OS/2 (with the EMX extender) and for Linux (a.out) are in ftp://kampi.hut.fi/jtv/gnu-pascal/binary Binaries for DOS (with the GO32 extender) and for Linux (elf) are in ftp://kampi.hut.fi/jtv/incoming ;-) If you want to join the gpc mailing list, write to gpc-request(a)hut.fi. The address of the list is gpc(a)hut.fi. -------- Peter Gerwinski <peter.gerwinski(a)uni-essen.de> 8<------------------------------------------------------------------------------ Is this okay, or too long, or whatever? If you have any improvements, please let me know. Bug reports for my English are also most welcome. :-) Yours, Peter ******************* NEW E-MAIL ADDRESS! ******************* -------------------------------------------------------------------------------- Dipl. Phys. Peter Gerwinski Fachbereich Physik Universitaet-GH Essen Phone: +49-201-183-2763 D-45117 Essen Fax: +49-201-183-2120 Germany e-mail: peter.gerwinski(a)uni-essen.de -------------------------------------------------------------------------------- ******************************** ACHTUNG: Neue e-mail-Adresse! Ab dem 22. Juni 1996 bitte an peter.gerwinski(a)uni-essen.de schreiben. ******************************** ATTENTION: New e-mail address! >From 22. June 1996 on please write to peter.gerwinski(a)uni-essen.de ******************************** -------------------------------------------------------------------------------- Dipl. Phys. Peter Gerwinski Fachbereich Physik Universitaet-GH Essen Phone: +49-201-183-2763 D-45117 Essen Fax: +49-201-183-2120 Germany e-mail: peter.gerwinski(a)uni-essen.de --------------------------------------------------------------------------------

1 0

Add me to the GPC list please
by Harry W. Reed 20 Jun '96

20 Jun '96

Hi, Please add me to the GPC mailing list. Thanks, Harry Reed doon(a)hec.nevada.edu This Message Was Sent With An UNREGISTERED Version Of PMMail. Please Encourage Its Author To Register Their Copy Of PMMail. For More Information About PMMail And SouthSide Software's Other Products, Contact http://www.southsoft.com.

1 0

Re: GNU pascal compiler uploaded
by JanJaap van der Heijden 13 Jun '96

13 Jun '96

Look what I found in the mail this morning: >Date: Wed, 12 Jun 1996 22:25:39 -0400 >From: dj(a)delorie.com (DJ Delorie) >To: j.j.vanderheijden(a)student.utwente.nl >Subject: Re: GNU pascal compiler uploaded > > >I finally got around to moving the GPC files to the outgoing simtelnet >directory. They'll be in the v2gnu directory. > >-rw-r--r-- 1 dj user 291 Apr 26 08:03 gpc112b.txt >-rw-r--r-- 1 dj user 685430 May 6 07:10 gpc112b.zip >-rw-r--r-- 1 dj user 8621302 May 6 07:39 gpc112s.zip >-rw-r--r-- 1 dj user 372 Apr 26 08:06 gpt112b.txt >-rw-r--r-- 1 dj user 698287 May 6 07:41 gpt112b.zip > For those of you who don't know: this is about the DOS (djgpp) GPC. djgpp is very very popular, and Simtel must be mirrored in just about every country in the world. This must generate some interest for GPC... /janjaap -- Science is what I'm doing when I don't know what I'm doing Werner von Braun

1 0

Units
by None 04 Jun '96

04 Jun '96

Hello, everybody! A known bug in GPC is that it does not automatically read Unit (or Module) interfaces when a Unit is used (or a Module is imported). For that reason, something like gpc -c myunit.pas gpc myprog.pas myunit.o -o myprog does only work when the main program (myprog.pas) (*$includes *) or #includes the Module interface. This is bearable with Extended Pascal where you can have the interface in a separate file; with Borland style Units you include the whole Unit, so it is recompiled each time. However, there is a workaround which is just another bug in GPC:-) A Borland style Unit with an empty implementation part but declaring some procedures in the interface part is not rejected by the compiler, only by the linker. Therefore, we can use the following trick: ---- file: myprog.pas ---- (*$I MyUnit *) (* Include the *whole* file MyUnit.pas *) Program MyProg; uses MyUnit; begin Hello; end. ---- file: myunit.pas ---- Unit MyUnit; Interface Procedure Hello; Implementation (*$ifdef IMPLEMENTATION *) (* suppress re-compilation of implementation part *) Procedure Hello; begin (* Hello *) writeln ( 'Hello!' ); end (* Hello *); (*$endif *) end. Now you can compile the Unit separately with gpc -D IMPLEMENTATION -c myunit.pas and then link it when compiling the main program via gpc myprog.pas myunit.o -o myprog Okay? Try it. Good luck! Peter

1 0

new binaries uploaded
by J.J. van der Heijden 20 May '96

20 May '96

Hello all, Today I uploaded a couple of new gpc-binaries to ftp://kampi.hut.fi/jtv/incoming. Hopefully, Juki will soon move them to an appropriate location. All of these are based on the public (2.6.3) snapshot of gpc. "turbo" flavours are based on Peter Gerwinski's "turbo-alpha" sources. The files are: gpc-1.1p2.linux.tar.gz turbo-1.1.linux.tar.gz Linux (ELF) gpc binary. I had to do some patching to gcc because gcc-2.6.3 ELF support was experimental and incompatible with the current (stable) state of ELF. Creates ELF executables. I think I uploaded these to sunsite.unc.edu (/pub/Linux/devel/lang/pascal) some time ago, although the "turbo" may have been version 1.0 Binaries were built on a late 1.3.x kernel and require libc-5.2.18 gpc112b.zip turbo11b.zip GPC binaries that fit into a djgpp (V2) distribution. djgpp is the "official" GCC release for MS-DOS, it should not be confused with Peter's EMX binaries, although they do roughly the same thing. Again, some patching to gcc because djgpp was v1.12 for gcc-2.6.3 and v2 for 2.7.2. gpc-1.1p2.cygwin32-beta14.zip turbo-1.1.cygwin32-beta14.zip GPC binaries that work ONLY with the current (beta14) release of the cygwin32 project. (Every new release seems to break compatibility with the previous one). The cygwin32 project is porting GNU software to Windows95 and NT (ix86 & ppc). This gpc creates PE/i386 executables. This is a work in progress: the compiler is restricted to console applications because it cannot access the Win32 API. The Win32 API requires a different way of stack management from the compiler (subroutine cleans stack), the rest of the world, including gpc, uses the C-style method (caller cleans stack). Hopefully, I can deal with this problem before gpc-1.2 is released, but I'm still interested in your experiences with this baby. More info on cygwin32 is available at: ftp://ftp.cygnus.com/pub/gnu-win32 I also have beta12 and beta13 win32 binaries around, but I don't think anybody is using b12 or 13 anymore so I didn't upload them. Last (but not least?), I have some Linux/ELF based gpc-crosscompilers They can be usefull if you want to develop from a single source tree, but support multiple target platforms. If anybody is interested in this stuff, he/she should contact me. You also need extra cross-binutils to make this work. I have linux->linuxaout, linux->djgpp & linux->win32 available. Happy hacking, J.J. van der Heijden

1 0

BO5 for Linux: bug report and workaround
by Peter Gerwinski 04 May '96

04 May '96

Hello, everybody! I was reported a bug in the BO5 screen drivers for Linux console (something like a CRT for GPC): on some machines the program stops with a runtime error "cannot access video memory", even if compiling and executing as root. To work around, use the TERMCAP version instead; the only visible differences are that you (1) have to specify the termcap library when compiling: gpc -I /usr/src/bo5 -D TERMCAP hallo.pas -ltermcap -o hallo and that (2) you have to set the CL_COLOR environment variable in order to get colors: export CL_COLOR=YES The next version will fix this bug (I will probably rewrite the Linux drivers in complete). Yours, Peter

1 0

Re: Q: Is there a gpc/emx for emx09b / gcc 2.7.2
by sad＠utkux.utcc.utk.edu 21 Mar '96

21 Mar '96

Hi Jukka, hi Peter, hi world (in that order :->), the post by Kevin Foss from the emx mailing list got me to experiment a bit more, and as it happens, with the following explicit linker statements at least my simple `Hallo world' programme is back to life: gpc test.pas -lgpc -lc_alias So, with this the binaries of the emx/gpc built by Peter Gerwinski for/with emx09a/gcc2.6.3 can be used under emx09b/gcc2.7.2 just fine. That is a minor adjustment to a properly structured makefile and much nicer than gpc -c test.pas gcc test.o which also worked. Late night cheers! Stefan ========================================================================== Stefan A. Deutscher, sad(a)utk.edu, (001)-423-[522-7845|974-7838|574-5897] home^ UTK^ ORNL^ ========================================================================== If there is software you'd like to have in a native version, visit the: OS/2 E-mail Campaign Page http://www.andrews.edu/~boyko/email.html --------------------------------------------------------------------------

1 0

Re: Q: Is there a gpc/emx for emx09b / gcc 2.7.2
by sad＠utkux.utcc.utk.edu 19 Mar '96

19 Mar '96

> > Hi again! > > > I don't find it (libgcc.a). the closest I find is gcc.lib on > > my gcc 2.6.3 setup at work. I am not at home right now, > > so I can't check. > > That's it. Since DOS restricts file names to 8 letters, emx names > libraries gcc.* rather than libgcc.*. But I thought that the suffix > would be .a, not .lib, or there are both versions, denoting different > lib file formats (perhaps elf and a.out?). Hu? libgcc.* is still well within the DOS / FAT conventions, isn't it :-) But, anyway, I have both gcc.lib and gcc.a, one (*.lib) is, as I understand it, for linking with *.obj files using link386 and -Zomf, the other one (*.a) for the traditional a.out and emx format. > > > Well, I have indeed had success with gcc 2.6.3, but not with 2.7.2 anymore. > > Being completely stupid I didn't just upgrade emx rts from 09a to 09b but > > did the wole works (2.6.3 -> 2.7.2) assuming that's a VERY GOOD THING. > > Maybe it wasn't. I'll restore my 2.6.3 and just put in the nwe emxrts > > tonight. Or, maybe I live with the separate link step until a 2.7.2 version > > is out? > > I'm afraid these two alternatives are really the only solutions. Oh. (seeking for cleenex). > Let's hope that somebody will create GPC 2.7.x soon. Yes, yes. I really would do it if I just could (where are those cleenex, again?). However, I haven't even had success compiling gcc / gpc before, so I am less than optimistic. Anybody out there who'd really know what to do? > > > > I think all these problems will disappear as soon as we will have a 2.7.x > > > version of GPC. (But this is not my job. :-) > > Well, if there is no 2.7.x in 1997, I will do the job, because I also > do not like the situation as is. But I would prefer somebody else to > do this, so I can concentrate on the Borland stuff. 1997? (Cleenex alert again!) > > > > Another suggestion would be > > > to make GPC independend of the C runtime library, but include the required > > > stuff into gpclib.a. > > > > Is this how g77 does it with f2c.lib? > > I don't know but I could imagine. Does anybody know? > > > Oh -- by the way: What's the reasone for gpc/emx being >= 3 MB as opposed to > > 1.7 or so it was before? Is it the turbo stuff, or is it debug info that one > > could safely strip out? > > I think it's debug info. If so, I will strip it out in the next release of > binaries. So, could I just run strip against the binaries and they'd still work? > > > PS: A last one: I see mail to gpc(a)hut.fi bouncing all the time. You too? > > So do I. Does anybody know the address? > > So long, > > Peter ========================================================================== Stefan A. Deutscher, sad(a)utk.edu, (001)-423-[522-7845|974-7838|574-5897] home^ UTK^ ORNL^ ========================================================================== If there is software you'd like to have in a native version, visit the: OS/2 E-mail Campaign Page http://www.andrews.edu/~boyko/email.html --------------------------------------------------------------------------

1 0

Q: Is there a gpc/emx for emx09b already ?
by sad＠utkux.utcc.utk.edu 18 Mar '96

18 Mar '96

Hi all, hi Peter, I was wondering whether it is just me or a known (and maybe even fixed?) problem: I just finished installing emx09b on my machine to be able to run things like the latest gnuplot beta and all that stuff that gave me warnings about having to use emx09b. So, all seemed well. In fact, even g77 worked, it still does after I have installed the version that had been recompiled with emx09b. GPC doesn't like me anymore, though. It compiles, but linking doesn't seem to work, I get stuff like: rts-file.c:221 (e:/bin/emx/lib/gpc.a(rts-file.o)): Undefined symbol _write refer enced from text segment and lots of it. I then fetched a later version from ftp.uni-augsburg.de since I can't seem to get through to kampi.hut.fi tonight, and installed it: Still the same problem. I understand that emxgpc was compiled (or patched) for 2.6.3, but I was wondering whether there is any hope on the horizon for a transparent gpc under emx? Interesting enough, gpc --version reports %PROG2%, so this may serve as version indication: rwxa-- 161799 Dec 13 00:39 e:/bin/emx/bin/gpc.exe* but more interesting (and temporary relief!) is, that gpc -c test.pas gcc test.o -lgpc does produce a working executable. Any ideas, pointers, promises are most welcome! Cheers! Stefan

2 2

Q: How do you use UNITs in TurboGPC? Plus general comments ....
by lotu＠wam.umd.edu 18 Mar '96

18 Mar '96

I just downloaded the precompiled ELF binary version of "Turbo" GPC for Linux. I've been playing around with it for the past few minutes, but I can't seem to be able to figure out how to make and use Borland Pascal UNITs. To try it out, I made a simple program (test1.p) that called the procedure "Intro" which was in another unit (unit1.p). I first compiled the unit (thinking that gpc doesn't have a built-in make like BPC/TPC does) with the following command line: "gpc -c unit1.p". I then proceeded to compile and link test1.p -- but as I expected it didn't work. I get the message: "No exported interface matching 'Unit1'' "undeclared indentifier 'Intro' .....' I sort of expected this to happen 'cause how would gpc know about unit1.p? I'm obviously doing something wrong here. Can somebody please clue me in? Thanks! And on a seperate, but related note .... How goes the work on "Turbo" GPC? How about GPC in general? (as far as getting full Extended Pascal compliance is concerned). Copies of the ISO Pascal and Extended Pascal standards are available off the 'net courtesy of John Reagan, if anybody is interested. He also said that he was gonna put the new Object Pascal standard up for ftping as well ... so perhaps Object Pascal can be incorporated into GPC ... shouldn't be too hard now ... with "Borland Pascal with Objects" already in place. =============================================================================== Arcadio Alivio Sincero, Jr. Sophomore, Computer Science Major at the University of Maryland at College Park Amateur competitive bodybuilder Email: lotu(a)wam.umd.edu, WWW: <still working on it!> "D.A.R.E. .... to keep cops of donuts."

2 1

2nd attemtpt -- gpc and emx09b ?
by sad＠utkux.utcc.utk.edu 17 Mar '96

17 Mar '96

Sorry if this shows up twice but something bounced, so I try it again. Stefan Hi all, hi Peter, I was wondering whether it is just me or a known (and maybe even fixed?) problem: I just finished installing emx09b on my machine to be able to run things like the latest gnuplot beta and all that stuff that gave me warnings about having to use emx09b. So, all seemed well. In fact, even g77 worked, it still does after I have installed the version that had been recompiled with emx09b. GPC doesn't like me anymore, though. It compiles, but linking doesn't seem to work, I get stuff like: rts-file.c:221 (e:/bin/emx/lib/gpc.a(rts-file.o)): Undefined symbol _write refer enced from text segment and lots of it. I then fetched a later version from ftp.uni-augsburg.de since I can't seem to get through to kampi.hut.fi tonight, and installed it: Still the same problem. I understand that emxgpc was compiled (or patched) for 2.6.3, but I was wondering whether there is any hope on the horizon for a transparent gpc under emx? Interesting enough, gpc --version reports %PROG2%, so this may serve as version indication: rwxa-- 161799 Dec 13 00:39 e:/bin/emx/bin/gpc.exe* but more interesting (and temporary relief!) is, that gpc -c test.pas gcc test.o -lgpc does produce a working executable. Any ideas, pointers, promises are most welcome! Cheers! Stefan

1 0

None
by Marcus Pless 26 Feb '96

26 Feb '96

lists

1 0

suscribe gpc
by JanJaap van der Heijden 04 Feb '96

04 Feb '96

suscribe gpc -- Science is what I'm doing when I don't know what I'm doing Werner von Braun

1 0

request
by Harald Wiedemann 30 Jan '96

30 Jan '96

I would like to participate in the gpc list.

1 0

no
by Emmanuel Tychon 25 Jan '96

25 Jan '96

mailing list -- ||| | Emmanuel Tychon, manu(a)netline.be O-O | SemaDigit: 018/458701 (Belgium Only) (_) | NetLine sprl, <http://www.netline.be> oOO-----OOo | Linux Team , <http://www.linux.netline.be> | Linux | | \-------/ | NetLine IS your interNet partner.

1 0

[Help] gpc for dos
by b2506033 25 Jan '96

25 Jan '96

Hello, When compiled a small program, I got a lot of error message. I use gpc-263b got from kampi.hut.fi. Please help me. Thanks. <My program> program testgpc; var i : integer; begin i := 20; end. <environment variables> c_include_path=c:/emx/include; library_path=c:/emx/lib; emx=c:/emx/bin/emx.exe path=c:/dos;c:/emx/bin;......; <error message> rts-file.c: 221 c:/emx/lib/gpc.a(rts-file.o)): Undefined Symbol _write referenced from text segment ....... _read ....... _unlink ....... _fstat ....... _isatty ....... _access ....... _stat ....... _bcopy ....... ...... <file > kampi.hut.fi:/jtv/gnu-pascal/EMX 1005157 Sep 15 16:52 gpc-263b.zip --- b2506033(a)csie.ntu.edu.tw Herge

2 1

[Help] gpc for dos
by berend＠beard.nest.nl 24 Jan '96

24 Jan '96

b2506033 wrote in a message to Berend de Boer: b> When compiled a small program, I got a lot of error message. How do you call your compiler? Did you include all the right libs as -Xgpc? Groetjes, Berend (-: fido: 2:281/527.23 email: berend(a)beard.nest.nl

1 0

help
by CLink 27 Dec '95

27 Dec '95

hep

1 0

None
by CLink 27 Dec '95

27 Dec '95

help

1 0

compiling gpc with gcc-2.7.2
by Cesar Romani 22 Dec '95

22 Dec '95

Is it possible to compile gpc-1.1-2.6.3 with gcc-2.7.2 ? Thanks in advance Cesar Romani Hamburg, Germany

1 0

BO5
by Peter Gerwinski 19 Dec '95

19 Dec '95

Hello, world! This is an announcement of a new library for GNU Pascal: BO5 It is only a pre-alpha release (version 0.01), but it should give you - some Borland-like tools such as Assign, FillChar, ParamStr, GetEnv, Random, - direct keyboard access via KeyPressed and ReadKey function, - direct screen access with colors. You need turbo-1.1-gpc-2.6.3 (i.e. the first bug fix for the GPC with Borland extensions, released together with BO5 on 19. December 1995) to compile BO5. BO5 is intended to run on *all* platforms on which GPC runs. If not, or if you have some other comment/suggestion/bug report, please write to pege(a)mail.theo-phys.uni-essen.de If your comment is of general interest for all GPC users, you should better write to the gpc-list gpc(a)hut.fi I have uploaded BO5 to kampi.hut.fi and to ftp.uni-augsburg.de. In Augsburg, it can already be downloaded from /pub/gnu/gnu-pascal/turbo-alpha For more information, please read the files README.BO5 and README.001. Merry Christmas and a happy New Year! Yours, Peter

1 0

BO5
by Peter Gerwinski 19 Dec '95

19 Dec '95

-------------------------------------------------------------------------------- Dipl. Phys. Peter Gerwinski Fachbereich Physik Universitaet-GH Essen Phone: +49-201-183-2763 D-45117 Essen Fax: +49-201-183-2120 Germany e-mail: pege(a)mail.theo-phys.uni-essen.de --------------------------------------------------------------------------------

1 0

Borland extensions: new version
by Peter Gerwinski 19 Dec '95

19 Dec '95

Hello, GPC-list! I have uploaded the 1.1 version of the Borland Extensions for GPC to kampi.hut.fi and to ftp.uni-augsburg.de. In Augsburg, it can already be downloaded from the directory /pub/gnu/gnu-pascal/turbo-alpha. The file turbo-1.1-gpc-2.6.3.tar.gz contains the new patched source, binaries for EMX and Linux have been updated. There are no new features, but some bugs have been fixed. Objects (virtual methods) are more stable now, -g should work now, etc. That's not all ... second mail follows ... Peter -------------------------------------------------------------------------------- Dipl. Phys. Peter Gerwinski Fachbereich Physik Universitaet-GH Essen Phone: +49-201-183-2763 D-45117 Essen Fax: +49-201-183-2120 Germany e-mail: pege(a)mail.theo-phys.uni-essen.de --------------------------------------------------------------------------------

1 0

bug in records?
by Phil Nelson 12 Dec '95

12 Dec '95

gpc-2.6.3/sparc I believe the following program should generate an error. (But it doesn't.) program variant(output); type myrec = record case integer of 1 : (f1:integer); 2 : (f1:real) end; begin end. -- Phil Nelson e-mail: phil(a)cs.wwu.edu http://www.cs.wwu.edu/~phil

1 0

Dispose problem
by Peter Gerwinski 12 Dec '95

12 Dec '95

Hello, Peter, hello, gpc-list! > pcc version 2.6.3 > sparc sunos4.1.3 > > program test; > type > num_pointer = ^integer; > var > num : num_pointer; > begin > new(num); > num^ := 3; > dispose(num); > end. > > ==> gpc: Internal compiler error: program gpc1 got fatal signal 10 I can reproduce the error with EMX (not with Linux). I hope to be able to release a bug fix this week. Yours, Peter -------------------------------------------------------------------------------- Dipl. Phys. Peter Gerwinski Fachbereich Physik Universitaet-GH Essen Phone: +49-201-183-2763 D-45117 Essen Fax: +49-201-183-2120 Germany e-mail: pege(a)mail.theo-phys.uni-essen.de --------------------------------------------------------------------------------

1 0

Re: GPC problem..
by Janos Vegh 06 Dec '95

06 Dec '95

At 02:16 PM 12/5/95 +0100, you wrote: >Or you can wait a little: I am writing a system-independent >replacement for crt, graph, and the whole Turbo Vision (but in >graphics mode). (Some people say, there would be no need for it, >but I won't stop. :-) I hope to be ready for the first release >(of only the crt part) this year. > I am happy to hear about! Dr Janos Vegh, physicist Janos.Vegh(a)ATOMKI.HU Institut of Nuclear Research of Vegh_J(a)Tigris.KLTE.HU the Hungarian Academy of Science Janos.Vegh(a)helka.iif.hu H-4001 Debrecen, POB 51, Hungary Tel: +36(52)417266 ext 1215 (H-4028 Debrecen, Bem ter 18/C) Fax: +36(52)413945 or 416181

1 0

Re: GPC problem..
by Peter Gerwinski 05 Dec '95

05 Dec '95

According to pege: >From pege Tue Dec 5 14:15:44 1995 Subject: Re: GPC problem.. To: mamoman(a)xs4all.nl (Martin Moerman) Date: Tue, 5 Dec 1995 14:15:44 +0100 (MEZ) In-Reply-To: <Pine.SUN.3.90.951203013417.29850A(a)xs1.xs4all.nl> from "Martin Moerman" at Dec 3, 95 02:35:38 pm X-Mailer: ELM [version 2.4 PL13] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 8bit Content-Length: 1494 According to Martin Moerman: > > When i use gpc i can't use the command clrsrc (clear screen). > Normaly in pascal i have to do (uses dos;), you mean: uses crt; > how can i solve this problem > in gpc? You have to find some standard library (dependend of your operating system) which contains something equivalent to clrscr. On DOS and OS/2, look into the EMX documentation libref.doc (it is a C library, but never mind -- GPC.GUIDE explains how to use it with GPC). With Linux, write the terminal escape sequences to the screen to emulate functions of CRT -- see the etc/termcap file for details. Or you can wait a little: I am writing a system-independent replacement for crt, graph, and the whole Turbo Vision (but in graphics mode). (Some people say, there would be no need for it, but I won't stop. :-) I hope to be ready for the first release (of only the crt part) this year. > ps. Is there also a manual for it (not gpc.guide).. I am afraid, no. But read README.TURBO, there are some things documented, how GPC differs from Turbo. Bye, Peter -------------------------------------------------------------------------------- Dipl. Phys. Peter Gerwinski Fachbereich Physik Universitaet-GH Essen Phone: +49-201-183-2763 D-45117 Essen Fax: +49-201-183-2120 Germany e-mail: pege(a)mail.theo-phys.uni-essen.de -------------------------------------------------------------------------------- -------------------------------------------------------------------------------- Dipl. Phys. Peter Gerwinski Fachbereich Physik Universitaet-GH Essen Phone: +49-201-183-2763 D-45117 Essen Fax: +49-201-183-2120 Germany e-mail: pege(a)mail.theo-phys.uni-essen.de --------------------------------------------------------------------------------

1 0

GPC problem..
by Martin Moerman 03 Dec '95

03 Dec '95

I have a problem when using gpc.. I was a big fan of borland pascal (turbo pascal 3 to 7).. When i use gpc i can't use the command clrsrc (clear screen). Normaly in pascal i have to do (uses dos;), how can i solve this problem in gpc. ps. Is there also a manual for it (not gpc.guide).. Martin mamoman(a)xs4all.nl

1 0

GNU Pascal Binary
by root 01 Dec '95

01 Dec '95

I've just installed Slackware Linux 3.0 on my system and I'm trying to install the GNU pascal compiler. Unfortunately, the only version of GCC I have is 2.7.0. I don't know where to find GCC 2.6.3 to run with GPC. Do you know where I could find this, or know where I could find a GPC compiler compatable with 2.7.0? Any help would gladly be appreciated. Thanks, Greg Sibley n9345955(a)cain.lake.cs.wwu.edu <== mail to this address

1 0

Re: string(1) to char?
by Thomas Dunbar 30 Nov '95

30 Nov '95

Jukka Virtanen wrote: > > > > can one convert a string of length one into a char? > > > > > > ie the below wont work: > > > > > > program t(input,output); > > > var > > > ch:char; > > > s:string(10); > > > begin > > > s:='teststring'; > > > writeln(substr(s,2,1)); > > > ch:=substr(s,2,1); > > > writeln(ord(ch)); > > > end. > > > > Yes, one can. The correct assignment reads > > > > ch:= s [ 2 ]; > > > > I think, this is Standard Pascal, nothing special. At least, it is > > common between UCSD, Borland and GNU Pascal. > > In extended pascal it is also allowed to assign a string > to a char if the length of the string is 1, which is the case > here. you are welcome :) but note that i _can_ do the assignment via: readstr(substr(s,2,1),ch); > > I think you found a bug in gpc, thanks for the report :-) > > Juki > jtv(a)hut.fi -- Thomas Dunbar 540 231-3938 (fax 3714) http://gserver.grads.vt.edu/

1 0

emx libs
by tdunbar＠gserver.grads.vt.edu 30 Nov '95

30 Nov '95

when i compile: program t(input,output); type s1=string(10); var s:s1; begin readln(s); writeln(substr(s,2,3)); end. with gpc on solaris or linux, it compiles fine. when i compile with emxgpc, i get: ... ld -o t.exe /emx/lib/crt0.o -L/emx/lib/st -L/emx/lib C:\DOS\cce00002 -lgpc -lgcc -lc -lc_app -lc -lgcc -lemx -los2 -lemx2 rts-string.c:144 (/emx/lib/gpc.a(rts-string.o)): Undefined symbol ___gcc_bcmp referenced from text segment rts-string.c:188 (/emx/lib/gpc.a(rts-string.o)): Undefined symbol ___gcc_bcmp referenced from text segment rts-string.c:190 (/emx/lib/gpc.a(rts-string.o)): Undefined symbol ___gcc_bcmp referenced from text segment am i using the wrong emx package? (other things compile fine..just problems working with strings) where can i get libs compatible with the latest emxgpc? -- Thomas Dunbar 540 231-3938 (fax 3714) http://gserver.grads.vt.edu/

3 5

gpc - questions
by kraemer＠ervax1.decnet.bosch.de 30 Nov '95

30 Nov '95

Robert Bosch GmbH Erbach, 30.11.95 IA1 / ESP4 Peter Kraemer-Eis Postfach 1162 D-64701 Erbach Tel.: 06062/78-463 Germany Fax.: 06062/78-771 Dear Mr. Virtanen, in our company, we have developed the software for a robot controler. Up to now, we used the National NS32000 family as target-cpu. Except of the hardware-near parts, our present software is written in NS-cross-pascal. The whole software for the robot-controler needs 2 MByte code and is split into about 200 files (modules). Now we are forced to replace the target-cpu by an intel-cpu (i486 or i586) and we hope, that most of our present software is portable to your GNU-Pascal. For a test, we have installed 'Linux SuSE August 95', which includes your GPC version 1.1 (2.6.3). First results are quite promising, but there are some questions: (1) Is there any reference manual (in addition to the GPC.GUIDE, which is too short, to work with it)? If yes, how can I get it? (2) Is there any possibility, to avoid the case-sensitivity of the names of variables/procedures? Our present software is completely written in capital letters and therefore we have problems when debugging with GDB. (3) Is there any restriction to the number of import/export of variables or procedures? (4) How can I get the source-files of your GPC? I havn't found them on the CD-Roms of 'Linux SuSE August 95'. Best regards, yours Peter Kraemer-Eis

1 0

string(1) to char?
by berend＠beard.nest.nl 30 Nov '95

30 Nov '95

Thomas Dunbar wrote in a message to Berend de Boer: It works on my Prospero Extended Pascal compiler. The works maybe with gpc: TD> program t(input,output); TD> var TD> ch:char; TD> s:string(10); TD> begin TD> s:='teststring'; TD> writeln(substr(s,2,1)); ch:=substr[2]; TD> writeln(ord(ch)); TD> end. Groetjes, Berend (-: CIS: 100120,3121 email: berend(a)beard.nest.nl

1 0

None
by Erik Karas 30 Nov '95

30 Nov '95

Will there be a version of gpc for gcc 2.7.0? If so when? Thanks

1 0

Binaries for OS/2 and DOS
by A.A. Olowofoyeku 28 Nov '95

28 Nov '95

Hi Can you please point me to the location of the OS/2 and DOS binaries (with the Borland compatibility stuff)? I presume that there is a ready-to-run package somewhere (lost most of my saved mail recently, and so I can't trace earlier posts). Do I have to download GCC as well? Thanks. Warmest regards, The Chief --------- Dr. Abimbola A. Olowofoyeku (The African Chief) Keele University, England (and, The Great Elephant) Email: laa12(a)keele.ac.uk or, chief(a)mep.com

4 5

Re: Prospero compiler
by A.A. Olowofoyeku 28 Nov '95

28 Nov '95

Hello, Peter Gerwinski! You wrote: > > Dear Chief, > > > Prospero's compiler is commercial. There is one for OS/2 (which I reviewed > > in the May '95 edition of the Pascal Magazine) and there is one for DOS > > (which I have never seen). The OS/2 version is pretty good, and also > > supports ISO Pascal standards. > > That's nice, but I have no money to get anything commercial in the > moment :-( I have spent all my money on Borland compiler updates. I know the feeling - and I am not happy with Borland at all. > (But nothing for Windoze! Only for (Novell) DOS.) Lucky you. I have some shareware programs (a Windows installer and uninstaller, and a command interpreter for Windows), and my users are demanding Win32 versions. MS have moved the goal posts (-->> Win32) and I have no choice other than follow, or lose my users to competitors, some of whom already have Win32 versions, while I am still waiting for Borland. > > I think you can get Prospero by e-mail; > > tpm2info(a)prospero.demon.co.uk > > To buy or to look at or to steal? That's an important difference for me, > especially because the only thing I would do with it would be to extract > out the Extended Pascal standard in order to get a better feeling how to > improve GPC further. Hmmmm .... I am not sure you would get anything from Prospero that you couldn't get by just reading the EP specifications, so you don't really need Prospero at all. It is not cheap either. If you are prepared to review the compiler for some computer magazine, they might send you a free review copy. They sent me one, for my review in the Pascal Magazine. -- Warmest regards, The Chief --------- Dr. Abimbola A. Olowofoyeku (The African Chief) Keele University, England (and, The Great Elephant) Email: laa12(a)keele.ac.uk or, chief(a)mep.com

1 0

bug with strings in EMX port of gpc
by Thomas Dunbar 27 Nov '95

27 Nov '95

the EMX port of gpc that i've been using for my students seems to have problems with strings .. it's like the bcmp routine didnt get included in libs works fine on linux and solaris i dont even remember where i downloaded the emx version. where can i find it, or better, an update? -- Thomas Dunbar 540 231-3938 (fax 3714) http://gserver.grads.vt.edu/

1 0

Re: Files
by Pavel Petrovic 27 Nov '95

27 Nov '95

Hello, > Now I also have problems with the binding of external files to > Pascal file variables ... > > I tried to use Extended Pascal "binding" to assign a file variable > Type it in -- voila. I defined a semi-dummy > "Assign" procedure which prompts "please type in )-: myfile135.dat". > It could not be done with "myexecutable <filenames.in", so I really > must type in each filename by hand. > > What shall I do ??? I used -Grts -a file_variable:filename switch to my program, which used file file_variable (it was mentioned in program header). This worked like static assign. Try to look somewhere in docs (use grep), I found it there. I am not sure, whether some special switch should be given to the compiler. If you find any better solution, I am interested in it. Sincerely, Pavel Petrovic http://pascal.fmph.uniba.sk/~3petrovice

1 0

Files
by Peter Gerwinski 27 Nov '95

27 Nov '95

Hello! Now I also have problems with the binding of external files to Pascal file variables ... I tried to use Extended Pascal "binding" to assign a file variable to an external -- existing -- file. I just copied the "Assign" pro- cedure by Berend as published in borland2ep.doc. It does not always work: I often get a runtime error saying that the file did not exist. (But it exists -- I can edit it, etc..) The only solution I could figure out: Mention each file in the program's header, forget about an "Assign" procedure, just reset it. Then you are asked for the external file name. Type it in -- voila. I defined a semi-dummy "Assign" procedure which prompts "please type in )-: myfile135.dat". It could not be done with "myexecutable <filenames.in", so I really must type in each filename by hand. This error is not reproducable. I have written some programs which can read files using Berend's "Assign", but not write. And vice versa. Currently, both does not work. What shall I do ??? Peter -------------------------------------------------------------------------------- Dipl. Phys. Peter Gerwinski Fachbereich Physik Universitaet-GH Essen Phone: +49-201-183-2763 D-45117 Essen Fax: +49-201-183-2120 Germany e-mail: pege(a)mail.theo-phys.uni-essen.de --------------------------------------------------------------------------------

1 0

GCC 2.7.x and GPC 2.6.3
by Peter Gerwinski 27 Nov '95

27 Nov '95

Hello, everybody! Somehow good news: GCC 2.7.x and GPC 2.6.3 *can* peacefully coexist on the same system. You just have to make the linuxaout-libraries visible for GPC (for example with the SlackWare 3.0 directory tree, use "gpc -L /usr/i486-linuxaout/lib myfile.pas") and to replace temporarily the new library libgcc.a by the old one. Only this one -- not the elf versions, and not libc.a, etc.. The linker will complain about some entry point, but the executable will work. I suggest to use an alias or a shell script for all that. Yours, Peter -------------------------------------------------------------------------------- Dipl. Phys. Peter Gerwinski Fachbereich Physik Universitaet-GH Essen Phone: +49-201-183-2763 D-45117 Essen Fax: +49-201-183-2120 Germany e-mail: pege(a)mail.theo-phys.uni-essen.de --------------------------------------------------------------------------------

1 0

f2p, c2p
by Peter Gerwinski 24 Nov '95

24 Nov '95

Hello, folks! Only one small question/suggestion: Does there exist a f2p or c2p translator program which translates Fortran resp. C source to Pascal source? If not, I suggest that somebody writes it because of the following reasons: - Practical reason: It would be nice to import libraries written in C, C++ or Fortran automatically. For most purposes, it would be sufficient to have a program which ports C header files to Pascal Unit (Module) interfaces. - Psychological reason: The existence of a c2p would demonstrate that Pascal is at least as powerful as C. The existence of a p2c (which does in fact *not* support everything which is possible in Borland Pascal -- I am thinking of the Object extensions) is ac- cepted by many people as a "proof" that C would be superior to Pascal, so we should "proove" the opposite direction. :-) I am almost sure that at least the c2p program does not exist because Standard Pascal has indeed less abilities than C. However, you can do everything which is possible in C with GNU Pascal (including e.g. pointer arithmetics -- just do a type cast to Integer), so the main difficulty will be to *understand* the C source, not to translate it. C++ could be more complicated as long as GNU Pascal does not (yet) have multiple inheritance. For Fortran, I don't see any problems, since there are complex numbers in GNU (Extended) Pascal, but one should translate goto statements to structured statements wherever possible.. One idea how to write it: Just use the front-ends of the GNU compi- lers and replace the assembler back-end by a Pascal back-end. Once you have understood how the front-end passes information to the back-end, the job will be straightforward. (One could even easily generalize it to a general x2y translator ...! :-) I have not the time to do it myself :-( so I ask *you* if you are willing to do this important contribution. Yours, Peter -------------------------------------------------------------------------------- Dipl. Phys. Peter Gerwinski Fachbereich Physik Universitaet-GH Essen Phone: +49-201-183-2763 D-45117 Essen Fax: +49-201-183-2120 Germany e-mail: pege(a)mail.theo-phys.uni-essen.de --------------------------------------------------------------------------------

2 1

Re: borland extensions ("PS")
by Peter Gerwinski 23 Nov '95

23 Nov '95

According to pege: >From pege Thu Nov 23 19:21:21 1995 Subject: Re: borland extensions ("PS") To: Jukka.Virtanen(a)hut.fi Date: Thu, 23 Nov 1995 19:21:21 +0100 (MEZ) In-Reply-To: <Pine.BSD.3.91.951122221113.395X-100000(a)kampi.hut.fi> from "Jukka Virtanen" at Nov 22, 95 10:19:12 pm X-Mailer: ELM [version 2.4 PL13] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 8bit Content-Length: 1297 Hello again! > I find that the extended pascal is a good language in itself, > but I agree that there are far too many borland programs > around to ignore them. I think there is yet another reason: When you want to do lo-level programming in Pascal, you *need* the Borland extensions. (Or, you must write large parts in assembler, which is also non-ISO.) And there are more examples where things can easily be done with Borland Pascal but are very uncomfortable or even impossible with Extended Pascal. > (Also, the object pascal language > looks nice and would be fun to support. Any volunteers? :-) The Borland "nonstandard" supports objects, so the object pascal language already *is* implemented in some sense -- I expect that only a few names must be changed. If a volunteer wants to imple- ment Object Pascal: It won't be too much work! Yours, Peter -------------------------------------------------------------------------------- Dipl. Phys. Peter Gerwinski Fachbereich Physik Universitaet-GH Essen Phone: +49-201-183-2763 D-45117 Essen Fax: +49-201-183-2120 Germany e-mail: pege(a)mail.theo-phys.uni-essen.de -------------------------------------------------------------------------------- -------------------------------------------------------------------------------- Dipl. Phys. Peter Gerwinski Fachbereich Physik Universitaet-GH Essen Phone: +49-201-183-2763 D-45117 Essen Fax: +49-201-183-2120 Germany e-mail: pege(a)mail.theo-phys.uni-essen.de --------------------------------------------------------------------------------

1 0

GPC
by Dave Smith 23 Nov '95

23 Nov '95

Hi, I've downloaded the GPC binaries for Ms-dos, but I don't have an emx driver. Could you please point me in the direction to obtain these drivers for dos. I've searched the web, but can only find OS2 ones :(, so your help here would be much appreciated. Cheers Dave Smith

2 1

gcc 2.7.1
by Philippe.Defert＠cern.ch 23 Nov '95

23 Nov '95

What about gpc for gcc 2.7.1 ? Thanks for your help. -- ************************************************************************ * Philippe Defert: Computing and Networks Division * * CERN, European Laboratory for Particle Physics * ************************************************************************ * Earth mail: CERN / CN / CO | E*mail: defert(a)cern.ch * * CH-1211 Geneve 23. Switzerland. | defert(a)cernvm.bitnet * * Voice: + 41 22 7673923 | * ************************************************************************ * __ * * Un monde nouveau, tu comprends ////\ * * Rien ne sera plus jamais comme avant \\\// * * C'est la fin de l'histoire, le rouge apres le noir | | * * J.J. Goldman | | * ************************************************************************

6 6

GPC and OS/2
by alland＠whidbey.com 23 Nov '95

23 Nov '95

Hello, I installed the version of GPC patched for emx running in OS/2. Has anyone made an OS/2 header file that works for API calls? Can someone show me a snippet of code showing how to incorporate an assembler procedure into a program? WriteStr doesn't work - "unresolved external". Writeln is sometimes called multiple times. TIA Allan D. Derickson -- Derickson Forestry Services -- Freeland, WA USA ----------------------------------------------------------- alland(a)whidbey.com -----------------------------------------------------------

1 0

GPC for Win32
by berend＠beard.nest.nl 22 Nov '95

22 Nov '95

A.A. Olowofoyeku wrote in a message to Berend de Boer: AAO> I have been reading somewhere about GCC for Win32 - has AAO> anyone done a port of GPC for the Win32 platforms? I probably try it soon. But I think I need 2.7.x, but let's see if gpc2.6.3 works on top of gcc 2.7 Groetjes, Berend (-: CIS: 100120,3121 email: berend(a)beard.nest.nl

1 0

GPC for Win32
by A.A. Olowofoyeku 22 Nov '95

22 Nov '95

Hello, I have been reading somewhere about GCC for Win32 - has anyone done a port of GPC for the Win32 platforms? Thanks. Warmest regards, The Chief --------- Dr. Abimbola A. Olowofoyeku (The African Chief) Keele University, England (and, The Great Elephant) Email: laa12(a)keele.ac.uk or, chief(a)mep.com

1 0

NEED LINUX/ELF VERSION OF GPC!!
by Arcadio Alivio Sincero 14 Nov '95

14 Nov '95

Can somebody PLEASE make an ELF version of GPC for Linux that works with GCC v.2.7.0???? Or give me some pointers on how to do it myself???? I downloaded the Linux version of GPC that Peter Gerwinski made, but it doesn't work on my system. I'm thinking maybe because I have SlackWare 3.0/ELF installed. When I first installed GPC, the compilation process would break when it came to the linking phase. ld, the GNU Linker, would say that it's unable to find file "crt0.o". I remembered seeing that file way back in my SlackWare 2.2 days, so kicked out the old SlackWare 2.2 CD-ROM and copied a "crt0.o" into my /usr/lib directory. Now, when I try to compile and link, I just get a whole bunch of unresolved reference errors. However, GPC itself appears to compile alright (except it produces a.out object files instead of ELF). It's just when it comes time to link that it goes haywire. Please help somebody! I need this compiler soon for a school project ... Arcadio

3 3

Binaries and IDE for Turbo-GPC
by Peter Gerwinski 11 Nov '95

11 Nov '95

Hello! I have uploaded binaries for the Turbo-GPC for EMX (i.e. DOS and OS/2) and for Linux to kampi.hut.fi and also to ftp.uni-augsburg.de. In Augsburg, they already can be downloaded immediately from /pub/incoming/gnu-pascal/turbo-alpha, but I asked them to move the tree to /pub/gnu and to set up some symbolic links. By the way: there is a Borland-style integrated development system named XWPE which is designed to run gcc, but can handle gpc, too (of course). It runs fine with X as well as with Linux in text mode. It can be downloaded (for example) from cranach.rz.tu-ilmenau.de from the directory /pub/unix/linux/distributions/DEBIAN/debian-0.93/source/devel. Have fun, Peter -------------------------------------------------------------------------------- Dipl. Phys. Peter Gerwinski Fachbereich Physik Universitaet-GH Essen Phone: +49-201-183-2763 D-45117 Essen Fax: +49-201-183-2120 Germany e-mail: pege(a)mail.theo-phys.uni-essen.de --------------------------------------------------------------------------------

2 3

Announcement: Borland extensions for GPC
by Peter Gerwinski 08 Nov '95

08 Nov '95

Hello, folks! I have implemented most of the Borland extensions -- and some other stuff -- into GPC. My extensions include: - Units - Compiler directives - Bit manipulations: and, or, shl, shr - Increment, decrement, min, max - Absolute variables - Objects - User-definable operators The modified sources are available on kampi.hut.fi in the directory /gpc/turbo-alpha. I am looking forward to receive your comments. :-) Yours, Peter -------------------------------------------------------------------------------- Dipl. Phys. Peter Gerwinski Fachbereich Physik Universitaet-GH Essen Phone: +49-201-183-2763 D-45117 Essen Fax: +49-201-183-2120 Germany e-mail: pege(a)mail.theo-phys.uni-essen.de --------------------------------------------------------------------------------

4 5

gpc and gcc 2.7 on Linux....
by Michael Slavitch 02 Nov '95

02 Nov '95

Has anyone successfully used gpc with gcc 2.7.0? Michael

1 0

Re: HELP (files)
by Peter Gerwinski 02 Nov '95

02 Nov '95

> Hello, Hello as well! > I have a problem using files with gpc: > > In Turbo Pascal I usually use the following: > > var > f:text; > > begin > assign(f,'filename'); > reset(f); > ... > close(f); > > But this doesn't work with gpc !! Look into the file borland2ep.doc by Berend de Boer which comes with GPC. There is documented how to write `assign' by yourself using GPC's `bind' procedure. This has proven to work in some cases, but not in all ... perhaps it will work when we also take into account Arcadio's hint. Good luck, Peter -------------------------------------------------------------------------------- Dipl. Phys. Peter Gerwinski Fachbereich Physik Universitaet-GH Essen Phone: +49-201-183-2763 D-45117 Essen Fax: +49-201-183-2120 Germany e-mail: pege(a)mail.theo-phys.uni-essen.de --------------------------------------------------------------------------------

1 0

HELP !
by i085741＠spv10.ing.uniroma1.it 02 Nov '95

02 Nov '95

Hello, I have a problem using files with gpc: In Turbo Pascal I usually use the following: var f:text; begin assign(f,'filename'); reset(f); ... close(f); But this doesn't work with gpc !! I tryed: reset(f,'filename'); This time it also compiled OK, but gives a runtime error !! Please HELP me ! Best regards Mauro Ottaviani

2 1

converter bp->ep
by Christophe Ponsard 01 Nov '95

01 Nov '95

I'd like to convert some of my borland pascal units in extended pascal modules (or in gpc modules). Does such a converter exist (and where) ? Otherwise I'll write one using flex & bison on a subset of the extended pascal grammar. It would be nice if I could get it (the .lex and .yacc files are not in the gpc snapshot) Is there a solution to the following constraint : the interface part of all the modules used by a program have to be included at his beginning. The solution of deducing the name of the module file from the import name (as borland do) was excluded (not portable ?) My converter would implement the solution using #include "unit.ph" directives in the main program. Each borland pascal unit would be splitted in a 'unit.pas' file (implementation) and a 'unit.ph' file (interface). --------------------------------------- PONSARD Christophe Universiti catholique de Louvain Dipartement d'inginierie informatique Bbtiment Riaumur, place Sainte Barbe 1348 Louvain-la-Neuve til. : 010/47 31 30 e-mail: chp(a)info.ucl.ac.be

1 0

test
by root＠txcc.net 23 Oct '95

23 Oct '95

hello..

1 0

gpc
by peter＠idiap.ch 06 Oct '95

06 Oct '95

Hi I translated a big pc-program (former Turbo-Pascal) into a gpc-program. I found some problems and have some questions! -gcc version 2.6.3 -can not make comments with included comments program test; begin { this is ok! } (* this also *) { (* you get a error*) } { { also get a error } } end. How dos it work >>EXPORT, IMPORT<<? MODULE test INTERFACE; EXPORT Test = (testproc); PROCEDURE testproc; END. MODULE command_line IMPLEMENTATION; PROCEDURE testproc; BEGIN ... END; END. PROGRAM testp; IMPORT Test; BEGIN testproc; END. >gpc -c test.pas ==>test.o >gpc test.o testp.pas ==>No exported interface matching `Test' I found only one way to compile this: >gpc -include test.pas testp.pas How do you compile this? Have you implemented routines to handle the I/O Errors? (e.g. Turbo Pascal => IOResult ); ,~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~. | --------,\ | | Peter Weber / / / | | IDIAP Martigny Peter(a)IDIAP.CH / / _/ | | TU Ilmenau Peter.Weber(a)E-Technik.TU-ILMENAU.DE / / | | _/ _/ | |----------------------------------------------------------------------| | http://www.idiap.ch/html/idiap-networks.html | | http://www.rz.tu-ilmenau.de/~webmast/ | | http://www.rz.tu-ilmenau.de/~hsf/hsf.html | `~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'

1 0

None
by merkle＠inf.ufsc.br 29 Sep '95

29 Sep '95

subscribe gpc merkle(a)inf.ufsc.br

1 0

Re: GPC for EMX works! Announcments?
by sad＠utkux.utcc.utk.edu 21 Sep '95

21 Sep '95

> > According to sad(a)utkux.utcc.utk.edu: > > > > Hi Jukka and Peter, > > > > do you think (or mind) announcing this os/2 port of gpc in the > > appropriate news groups, like comp.os.os2.announce, > > comp.os.os2.programming, and comp.lang.pascal.ansi-iso? > > There are still rumors that there would be no os/2 port, you know, > > eventhough I have downloaded it already ... :-) > > > > Later. Stefan > > > > PS: I you give me the go-ahead I don't mind doing it. > > > I don't mind, I just hope it works :-) > Juki > > I don't mind either. I also only hope it works. :-> > > Peter Okay, so I take it one of you will take care of this, right? Or do you mean that neither of you does mind *me* doing it? Lemme know. Stefan :-] ---- now where is that .sig of mine?!

2 1

Re: GPC for EMX works!
by Jukka Virtanen 21 Sep '95

21 Sep '95

On Fri, 15 Sep 1995, Peter Gerwinski wrote: > Hello, folks! > > I have done the patch of GPC for EMX. It was not straightforward but not more > work than I had expected. I upload the binaries and the source patch to > kampi.hut.fi. > > Have fun with it! > > Peter Folks, The EMX binaries made by Peter are now available in kampi.hut.fi directory jtv/gnu-pascal/EMX There is a README file, please read it. Contents of the EMX directory: -r--r--r-- 1 jtv 35231 Sep 15 16:57 tar.exe -r--r--r-- 1 jtv 167476 Sep 15 16:57 tar-os2.exe -r--r--r-- 1 jtv 5468 Sep 15 16:55 README -r--r--r-- 1 jtv 188138 Sep 15 16:54 make-gpc.zip -r--r--r-- 1 jtv 39910 Sep 15 16:53 gzip.exe -r--r--r-- 1 jtv 1005157 Sep 15 16:52 gpc-263b.zip If I understood correctly, the make-gpc.zip contains a couple of modified source files (compared to gpc-2.6.3 sources you can find from jtv/gnu-pascal/gpc-1.1-2.6.3.tar.gz) The file gpc-263b.zip contains binaries for gpc driver program, gpc1 compiler, the gpc-cccp preprocessor and the gpc library. Juki ps. I don't have any systems around with emx in it, so I am unable to verify if these programs work...

3 3

Re^2: From Borland to GNU ...
by None 18 Sep '95

18 Sep '95

Hello, Juki! > Also I have no knowledge of turbo pascal, so your contribution > is very important!!! My problem is that the Borland "standard" is the only standard I know. I think I will be able to implement it (within some __long__ time ...) but I can not give a warranty that ISO standard will still work. Where did you get that standard? > I have the C++ version of gperf. As far as I rememer > Doug Schmidt wrote also a C version of it, but > I don't know where it currently is. If you need > it I could try to find it. I know the place in principle, but my access to there failed. In the moment, it is not this important. I will cry for help, if it becomes. > Where did you get the EMX changes from? Are they > already in the 2.7.0 gcc release? If they are, > it makes things a lot easier. I found version 2.6.3 at ftp.uni-paderborn.de. The patched version 2.7.0 is present at ftp.uni-stuttgart.de, directory /pub/systems/os2/emx-0.9a/gcc-2.7.0. My upload: I was knocked out of the system several times. Now I had success by changing the upload names to README.new and make-gpc.zip.new. Please change them back to README and make-gpc.zip since DOS will reject filenames longer than 8.3. Yours, Peter -------------------------------------------------------------------------------- Dipl. Phys. Peter Gerwinski Fachbereich Physik Universitaet-GH Essen Phone: +49-201-183-2763 D-45117 Essen Fax: +49-201-183-2120 Germany e-mail: pege(a)mail.theo-phys.uni-essen.de --------------------------------------------------------------------------------

1 0

Re: GPC for EMX works!
by Arcadio Alivio Sincero 16 Sep '95

16 Sep '95

On Fri, 15 Sep 1995, Peter Gerwinski wrote: > Hello, folks! > > I have done the patch of GPC for EMX. It was not straightforward but not more > work than I had expected. I upload the binaries and the source patch to > kampi.hut.fi. > Have fun with it! Where on kampi.hut.fi? Or you probably haven't done it yet when I read the message? Ok ... when you do you upload it, be sure to tell us where it is. kampi.hut.fi is in Europe and I'm in the U.S., so the connection is S-L-O-W. I'd hate having to hunt around on kampi.hut.fi ... thanks! Arcadio

1 0

Re: GPC for EMX works!
by sad＠utkux.utcc.utk.edu 15 Sep '95

15 Sep '95

Hi, Peter, this is very good news. Thanks for the effort! I will d/l it tonight and play with it. BTW, can it peacefully coexist with g77 (the gnu fortran compiler, which changes the gcc.exe)? If not, would you care to have a chat with the porter of g77? (He also happens to be German.) Cheers again! Stefan > > Hello, folks! > > I have done the patch of GPC for EMX. It was not straightforward but not more > work than I had expected. I upload the binaries and the source patch to > kampi.hut.fi. > > Have fun with it! > > Peter

1 0

GPC for EMX works!
by None 15 Sep '95

15 Sep '95

Hello, folks! I have done the patch of GPC for EMX. It was not straightforward but not more work than I had expected. I upload the binaries and the source patch to kampi.hut.fi. Have fun with it! Peter -------------------------------------------------------------------------------- Dipl. Phys. Peter Gerwinski Fachbereich Physik Universitaet-GH Essen Phone: +49-201-183-2763 D-45117 Essen Fax: +49-201-183-2120 Germany e-mail: pege(a)mail.theo-phys.uni-essen.de --------------------------------------------------------------------------------

1 0

GPC for EMX, objects, gperf
by berend＠beard.nest.nl 13 Sep '95

13 Sep '95

Jukka Virtanen wrote in a message to Berend de Boer: JV> gpc makefiles are quite standard, they should not require JV> any specific makefile features, except the VPATH JV> variable, which tells where the make program should look JV> for sources. Gnu make is fine. Most other new make JV> programs are also fine. The problem is dos' command.com is no bourne shell... That's the hardest part. Groetjes, Berend (-: CIS: 100120,3121 email: berend(a)beard.nest.nl

1 0

GPC for EMX, objects, gperf
by berend＠beard.nest.nl 13 Sep '95

13 Sep '95

Peter Gerwinski wrote PG> 1) I am thinking about porting GPC to the DOS / OS/2 EMX PG> environment. Does anybody already work on that task or PG> has even finished it? Yes. There is a djgpp recompile and an emx. Both should be easy, just a recompile if you have a bourne shell, else you need to create your own makefile. PG> 2) I would like to know more about when there will be PG> Object Pascal extensions in GPC, and how they will look PG> like. (E.g. "class" instead of Borland's "object"? PG> Does "override" mean what I know as "virtual"?) See the Ansi OOP draft, probably online soon (follow comp.lang.pascal.ansi-iso). Groetjes, Berend (-: CIS: 100120,3121 email: berend(a)beard.nest.nl

1 0

GPC for EMX, objects, gperf
by Peter Gerwinski 13 Sep '95

13 Sep '95

Hello everybody. I have three questions to the mailing list. 1) I am thinking about porting GPC to the DOS / OS/2 EMX environment. Does anybody already work on that task or has even finished it? 2) I would like to know more about when there will be Object Pascal extensions in GPC, and how they will look like. (E.g. "class" instead of Borland's "object"? Does "override" mean what I know as "virtual"?) 3) While trying to implement some Borland extensions into the GPC (i.e. shl/shr operators, OOP, ...) I stumbled about the utility program "gperf". I got the source - one as stand-alone C++ source, one as part of the C++ library - but could not compile it with gcc-2.6.3 running under Linux kernel 1.2.8. Does anybody know where to get Linux binaries for gperf? So far for the moment. Thank you, Peter -------------------------------------------------------------------------------- Dipl. Phys. Peter Gerwinski Fachbereich Physik Universitaet-GH Essen Phone: +49-201-183-2763 D-45117 Essen Fax: +49-201-183-2120 Germany e-mail: pege(a)mail.theo-phys.uni-essen.de --------------------------------------------------------------------------------

3 3

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

Gpc