http://ASPN.ActiveState.com/ASPN/Mail/Browse/Threaded/perl6-Porters For the maintenance, development, and porting of the programming language Perl. en-us Copyright 2005, ActiveState aspn-feedback@activestate.com aspn-feedback@activestate.com http://ASPN.ActiveState.com/ASPN/img/logo_78x25.gif http://ASPN.ActiveState.com/ASPN/Mail/Browse/Threaded/perl6-Porters http://ASPN.ActiveState.com/ASPN/Mail/Message/perl6-Porters/481346 <PRE>Edward Peschko wrote: > > > I personally think that this is something Larry is going to have to > > decide. However, I would like to note that leaving these off by default > > lowers the transition curve to Perl 6 immensely for those people that > > use Perl as a glue/scripting/sysadmin language. > > Right, but what I don't understand is that its two extra characters at the end > of a command line... whats the big deal about typing '-q' on one line in > scripts? Its easy enough to advertise '-q' and put it in big lights... Yes and no, I think. As Alan Burlinson pointed out earlier, this is *not* a purely clean-slate we have here to work with. There are lots and lots and lots of users who know how Perl works. Many of these users don't even know that you can do this: my $x = 5; Let alone that this: my $x, $y, $z; Doesn't DWIM, again according to what most people think. > > Key: Not everyone becomes a Perl expert. Many people never leave > > novice/intermediate level. This doesn't mean that we should design the > > language for these people, but it also doesn't mean we should thumb our > > noses at them. > > So - why is it a religious issue then? I respect the fact that you want to > write scripts without 'use strict'. I do this all the time. But I really think > that its a small price to pay to put '-q' on the #!/usr/bin/perl command line > for the vast benefits that it would give us all as far as CPAN goes. Let's separate the CPAN thing out for a moment. "use strict" and "use warnings" are not the universal solutions to CPAN quality issues, at least as far as I'm aware. As for the -q thing, I think it is far *less* of a burden to add "use strict" and "use warnings" when you're writing a big piece of code. When you're writing 5 lines, every extra character counts. When you're writing 500 or 5000 lines, 2 lines of "use" statements are nothing. > So - in the place of a '-q', would you support a mechanism for making > sure that CPAN is '-w' clean? If so, how would you enforce it? Absolutely, 100%. But this is a separate issue. For example, a big problem with CPAN is also packaging. There's lots of modules that don't use the h2xs/Makefile.PL approach. Also, many modules on CPAN have been in beta for years, and not just 0.99 beta but 0.02 beta, broken and uninstallable. There's also a lot of modules that don't run under -T. But "strict" won't fix these issues. If we're interested in increased CPAN quality, there's a bunch of stuff we can do. We can have a standard test suite that's run against every module uploaded to check if it's installable and compiles basically. We can check for -w, -T, use strict, and tons of other stuff. We can check for the packaging to make sure that "perl -MCPAN -e shell" will install the stuff cleanly. If there's simple problems (misnamed tar files, a $VERSION split onto multiple lines), we can even auto-fix this. Then send all the results to the user who uploaded the module. Heck, I'd even volunteer to head up a project to do this. But I think it's a separate issue from "use strict" and "use warnings", let's not confuse the two. -Nate </PRE> http://ASPN.ActiveState.com/ASPN/Mail/Message/perl6-Porters/481270 <PRE>> I personally think that this is something Larry is going to have to > decide. However, I would like to note that leaving these off by default > lowers the transition curve to Perl 6 immensely for those people that > use Perl as a glue/scripting/sysadmin language. Right, but what I don't understand is that its two extra characters at the end of a command line... whats the big deal about typing '-q' on one line in scripts? Its easy enough to advertise '-q' and put it in big lights... My point is that '-w' is pretty much useless because people contribute modules to CPAN that don't have '-w' turned on, and you get lots of junk output from your scripts because other people don't adhere to the '-w'. Hell, *I* don't use '-w' because of this. > Key: Not everyone becomes a Perl expert. Many people never leave > novice/intermediate level. This doesn't mean that we should design the > language for these people, but it also doesn't mean we should thumb our > noses at them. So - why is it a religious issue then? I respect the fact that you want to write scripts without 'use strict'. I do this all the time. But I really think that its a small price to pay to put '-q' on the #!/usr/bin/perl command line for the vast benefits that it would give us all as far as CPAN goes. So - in the place of a '-q', would you support a mechanism for making sure that CPAN is '-w' clean? If so, how would you enforce it? Ed ( ps - if you want to take this offline, its fine by me. If this issue has been covered before, you can tell me the arguments for/against it. But right now I do consider '-w' broken because of this; and it doesn't give me much comfort that lots of the modules on CPAN could have hidden bugs in them because -w is too difficult to use. ) </PRE> http://ASPN.ActiveState.com/ASPN/Mail/Message/perl6-Porters/478511 <PRE>Hiya, I wasnt expecting this but here is my $.02, Eric Raymond, is interesting but his contribtution is going to cause a lot more grief than contracting revenue and salary bonuses. Perl is frozen at 5.5.3 for me, which is not a problem. I would like a virtual machine built into the kernel, but more than anything, my industry (unix) needs a rock solid kernel to get control of the operations. Edward Peschko and Simon Cozens : > > Perl usage has grown respectably, but the language itself has been > stagnant > > > for two years or more. > > > > Bah. Looks like my Perl5-Porters summaries have been completely in vain. :) > > yeah, he's full of BS here. ===== John van Vlaanderen ############################################# # CXN, Inc. Contact: john@... # # # Proud Sponsor of Perl/Unix of NY # # http://puny.vm.com # ############################################# __________________________________________________ Do You Yahoo!? Get personalized email addresses from Yahoo! Mail - only $35 a year! http://personal.mail.yahoo.com/ </PRE> http://ASPN.ActiveState.com/ASPN/Mail/Message/perl6-Porters/477641 <PRE>On Fri, Feb 09, 2001 at 07:06:46PM +0000, Simon Cozens wrote: > Eric Raymond's book-in-development ``The Art of Unix Programming'' says > this about the future of Perl: > > > Perl usage has grown respectably, but the language itself has been stagnant > > for two years or more. > > Bah. Looks like my Perl5-Porters summaries have been completely in vain. :) yeah, he's full of BS here. > > If that other is Perl 6, I don't think we're stalled, are we? Language design > is waiting on Larry to produce the spec, and internals design is going on > quietly but steadily. We're in the design stage. That'll probably last a while > because scripting languages and interpreters aren't easy things to design, and > are even harder to get right. Well, sometimes I think things *are* stalled. I'm thinking of updating my book to perl 6, and maybe writing a new one, except that it is very difficult to start, let alone convince an editor to put the time and effort into such a project if you can't even point to a spec. And - correct me if I'm wrong - but there isn't even a status available on how far the spec process is coming. I'm not saying that every day or even every week - but once a month might be nice. Last time I heard, he was going through 'topic by topic' and was going to release verdicts bit-by-bit. But that was on Dec. 20th. Ed </PRE> http://ASPN.ActiveState.com/ASPN/Mail/Message/perl6-Porters/420918 <PRE>This will be the last posting on this list before I remove it. mail bootstrap-subscribe@... to join the new incarnation. - ask -- ask bjoern hansen - <http://www.netcetera.dk/~ask/> more than 70M impressions per day, <http://valueclick.com> </PRE> http://ASPN.ActiveState.com/ASPN/Mail/Message/perl6-Porters/420917 <PRE>On Sat, 24 Oct 1998, "Felix S. Gallo" <fsg@... wrote: > Chip writes: > >According to Felix S. Gallo: > >> 1. Describe the goal. I don't believe Perl can be usefully rewritten in > >> C++ [*] before the next Perl conference in any case [...] > > > >Yes, that's the target. > >> [*] as if > >We shall see. (Were I a betting man I'd clean up. :-)) > > Never bet against a British pedant -- I can always win by redefining > usefully. For my current claim, I think 'usefully' means that it's written > in C++ style; Perl is right now written in C++ if you make a few > minor changes. Below are those minor changes (with gratuitous Linux dependencies), for anyone wishing to play around. I'm interested in Joel Spolsky's article, <URL:http://joel.editthispage.com/stories/storyReader$47>, on not rewriting big programs from scratch. Apologies for straying from the path. :-) -John This is not a proper patch and requires care and feeding. Here is how I apply it, more or less: cd perl-5.6.0 chmod u+w hints/linux.sh echo timetype=time_t >> hints/linux.sh ./Configure -ds !patch -p0 < patchfile make all test Then do stuff like this until it's Perl 6: inline NV& sv::NVX () { return ((XPVNV*)sv_any)->xnv_nv; } #define SvNVX(sv) ((sv)->NVX()) --- gv.h~ Sun Feb 6 14:32:59 2000 +++ gv.h Sun Jun 25 11:17:29 2000 @... -19,7 +19,7 @... U32 gp_cvgen; /* generational validity of cached gv_cv */ U32 gp_flags; /* XXX unused */ line_t gp_line; /* line first declared at (for -w) */ - char * gp_file; /* file first declared in (for -w) */ + const char *gp_file; /* file first declared in (for -w) */ }; #if defined(CRIPPLED_CC) && (defined(iAPX286) || defined(M_I286) || defined(I80286)) --- gv.c~ Tue Mar 21 00:28:10 2000 +++ gv.c Sun Jun 25 01:22:20 2000 @... -982,7 +982,7 @... gv_check(hv); /* nested package */ } else if (isALPHA(*HeKEY(entry))) { - char *file; + const char *file; gv = (GV*)HeVAL(entry); if (SvTYPE(gv) != SVt_PVGV || GvMULTI(gv)) continue; --- mg.c~ Fri Mar 17 20:24:04 2000 +++ mg.c Sun Jun 25 11:54:19 2000 @... -21,6 +21,10 @... # include <unistd.h> #endif +#ifdef I_GRP +# include <grp.h> +#endif + #if defined(HAS_GETGROUPS) || defined(HAS_SETGROUPS) # ifndef NGROUPS # define NGROUPS 32 @... -688,9 +692,7 @... break; case '~': s = IoFMT_NAME(GvIOp(PL_defoutgv)); - if (!s) - s = GvENAME(PL_defoutgv); - sv_setpv(sv,s); + sv_setpv(sv,s ? s : GvENAME(PL_defoutgv)); break; #ifndef lint case '=': --- pp_hot.c~ Fri Mar 17 22:11:42 2000 +++ pp_hot.c Sun Jun 25 10:10:29 2000 @... -836,60 +836,52 @... } if (PL_delaymagic & ~DM_DELAY) { if (PL_delaymagic & DM_UID) { -#ifdef HAS_SETRESUID - (void)setresuid(PL_uid,PL_euid,(Uid_t)-1); -#else -# ifdef HAS_SETREUID +#ifdef HAS_SETREUID (void)setreuid(PL_uid,PL_euid); -# else -# ifdef HAS_SETRUID +#else +# ifdef HAS_SETRUID if ((PL_delaymagic & DM_UID) == DM_RUID) { (void)setruid(PL_uid); PL_delaymagic &= ~DM_RUID; } -# endif /* HAS_SETRUID */ -# ifdef HAS_SETEUID +# endif /* HAS_SETRUID */ +# ifdef HAS_SETEUID if ((PL_delaymagic & DM_UID) == DM_EUID) { (void)seteuid(PL_uid); PL_delaymagic &= ~DM_EUID; } -# endif /* HAS_SETEUID */ +# endif /* HAS_SETEUID */ if (PL_delaymagic & DM_UID) { if (PL_uid != PL_euid) DIE(aTHX_ "No setreuid available"); (void)PerlProc_setuid(PL_uid); } -# endif /* HAS_SETREUID */ -#endif /* HAS_SETRESUID */ +#endif /* HAS_SETREUID */ PL_uid = PerlProc_getuid(); PL_euid = PerlProc_geteuid(); } if (PL_delaymagic & DM_GID) { -#ifdef HAS_SETRESGID - (void)setresgid(PL_gid,PL_egid,(Gid_t)-1); -#else -# ifdef HAS_SETREGID +#ifdef HAS_SETREGID (void)setregid(PL_gid,PL_egid); -# else -# ifdef HAS_SETRGID +#else +# ifdef HAS_SETRGID if ((PL_delaymagic & DM_GID) == DM_RGID) { (void)setrgid(PL_gid); PL_delaymagic &= ~DM_RGID; } -# endif /* HAS_SETRGID */ -# ifdef HAS_SETEGID +# endif /* HAS_SETRGID */ +# ifdef HAS_SETEGID if ((PL_delaymagic & DM_GID) == DM_EGID) { (void)setegid(PL_gid); PL_delaymagic &= ~DM_EGID; } -# endif /* HAS_SETEGID */ +# endif /* HAS_SETEGID */ if (PL_delaymagic & DM_GID) { if (PL_gid != PL_egid) DIE(aTHX_ "No setregid available"); (void)PerlProc_setgid(PL_gid); } -# endif /* HAS_SETREGID */ -#endif /* HAS_SETRESGID */ +#endif /* HAS_SETREGID */ PL_gid = PerlProc_getgid(); PL_egid = PerlProc_getegid(); } --- sv.c~ Wed Mar 22 21:44:37 2000 +++ sv.c Sun Jun 25 11:55:29 2000 @... -15,6 +15,11 @... #define PERL_IN_SV_C #include "perl.h" +/* XXX If this causes problems, set i_unistd=undef in the hint file. */ +#ifdef I_UNISTD +#include <unistd.h> +#endif + #define FCALL *f #define SV_CHECK_THINKFIRST(sv) if (SvTHINKFIRST(sv)) sv_force_normal(sv) @... -2540,7 +2545,7 @... if (dtype <= SVt_PVGV) { glob_assign: if (dtype != SVt_PVGV) { - char *name = GvNAME(sstr); + const char *name = GvNAME(sstr); STRLEN len = GvNAMELEN(sstr); sv_upgrade(dstr, SVt_PVGV); sv_magic(dstr, dstr, '*', name, len); --- pp_sys.c~ Sun Mar 19 02:18:10 2000 +++ pp_sys.c Sun Jun 25 12:53:08 2000 @... -36,10 +36,16 @... #endif #ifdef HAS_SYSCALL -#ifdef __cplusplus +#ifdef __cplusplus +#ifndef __GLIBC__ +/* XXX Who put this here? Is someone compiling Perl with a C++ compiler + on a system that has syscall and doesn't declare it in a header file? + Really?? This breaks g++ on Linux because syscall returns long. + Hence my GNU Libc exclusion. -jtobey */ extern "C" int syscall(unsigned long,...); #endif #endif +#endif #ifdef I_SYS_WAIT # include <sys/wait.h> @... -3971,7 +3977,13 @... #ifdef HAS_GETPRIORITY who = POPi; which = TOPi; +#if defined(__GLIBC__) && defined(__cplusplus) + /* XXX GNU Libc documents `which' as int but declares it as enum + __priority_which in <sys/resource.h>, so let's not take chances. */ + SETi( ((int (*)(int, int))getpriority)(which, who) ); +#else SETi( getpriority(which, who) ); +#endif RETURN; #else DIE(aTHX_ PL_no_func, "getpriority()"); @... -3989,7 +4001,13 @... who = POPi; which = TOPi; TAINT_PROPER("setpriority"); +#if defined(__GLIBC__) && defined(__cplusplus) + /* XXX GNU Libc documents `which' as int but declares it as enum + __priority_which in <sys/resource.h>, so let's not take chances. */ + SETi( ((int (*)(int, int, int))setpriority)(which, who, niceval) >= 0 ); +#else SETi( setpriority(which, who, niceval) >= 0 ); +#endif RETURN; #else DIE(aTHX_ PL_no_func, "setpriority()"); --- sv.h~ Thu Mar 9 12:40:40 2000 +++ sv.h Sun Jun 25 11:49:08 2000 @... -290,7 +290,7 @... HV* xmg_stash; /* class package */ GP* xgv_gp; - char* xgv_name; + const char* xgv_name; STRLEN xgv_namelen; HV* xgv_stash; U8 xgv_flags; --- op.c~ Tue Mar 21 00:06:34 2000 +++ op.c Sun Jun 25 11:51:36 2000 @... -5498,7 +5498,7 @... /* is this op a FH constructor? */ if (is_handle_constructor(o,numargs)) { - char *name = Nullch; + const char *name = Nullch; STRLEN len; flags = 0; --- ext/DB_File/DB_File.xs~ Tue Feb 15 00:42:40 2000 +++ ext/DB_File/DB_File.xs Sun Jun 25 12:38:45 2000 @... -140,6 +140,10 @... #include <fcntl.h> +#ifdef __cplusplus +extern "C" void __getBerkeleyDBInfo(); +#endif + /* #define TRACE */ #define DBM_FILTERING @... -380,7 +384,7 @... #endif /* DBM_FILTERING */ -#define my_sv_setpvn(sv, d, s) sv_setpvn(sv, (s ? d : (void*)""), s) +#define my_sv_setpvn(sv, d, s) sv_setpvn(sv, (s ? (const char*)d : ""), s) #define OutputValue(arg, name) { if (RETVAL == 0) { @... -507,8 +511,8 @... PUSHMARK(SP) ; EXTEND(SP,2) ; - PUSHs(sv_2mortal(newSVpvn(data1,key1->size))); - PUSHs(sv_2mortal(newSVpvn(data2,key2->size))); + PUSHs(sv_2mortal(newSVpvn((const char *)data1,key1->size))); + PUSHs(sv_2mortal(newSVpvn((const char *)data2,key2->size))); PUTBACK ; count = perl_call_sv(CurrentDB->compare, G_SCALAR); @... -563,8 +567,8 @... PUSHMARK(SP) ; EXTEND(SP,2) ; - PUSHs(sv_2mortal(newSVpvn(data1,key1->size))); - PUSHs(sv_2mortal(newSVpvn(data2,key2->size))); + PUSHs(sv_2mortal(newSVpvn((const char *)data1,key1->size))); + PUSHs(sv_2mortal(newSVpvn((const char *)data2,key2->size))); PUTBACK ; count = perl_call_sv(CurrentDB->prefix, G_SCALAR); @... -768,7 +772,6 @... SV ** svp; HV * action ; DB_File RETVAL = (DB_File)safemalloc(sizeof(DB_File_type)) ; - void * openinfo = NULL ; INFO * info = &RETVAL->info ; STRLEN n_a; @... -808,7 +811,6 @... croak("DB_File can only tie an associative array to a DB_HASH database") ; RETVAL->type = DB_HASH ; - openinfo = (void*)info ; svp = hv_fetch(action, "hash", 4, FALSE); @... -843,7 +845,6 @... croak("DB_File can only tie an associative array to a DB_BTREE database"); RETVAL->type = DB_BTREE ; - openinfo = (void*)info ; svp = hv_fetch(action, "compare", 7, FALSE); if (svp && SvOK(*svp)) @... -892,7 +893,6 @... croak("DB_File can only tie an array to a DB_RECNO database"); RETVAL->type = DB_RECNO ; - openinfo = (void *)info ; info->db_RE_flags = 0 ; @... -1011,7 +1011,7 @... Flags |= DB_TRUNCATE ; #endif - status = db_open(name, RETVAL->type, Flags, mode, NULL, openinfo, &RETVAL->dbp) ; + status = db_open(name, RETVAL->type, Flags, mode, NULL, info, &RETVAL->dbp) ; if (status == 0) #if DB_VERSION_MAJOR == 2 && DB_VERSION_MINOR < 6 status = (RETVAL->dbp->cursor)(RETVAL->dbp, NULL, &RETVAL->cursor) ; @... -1027,9 +1027,9 @... #else #if defined(DB_LIBRARY_COMPATIBILITY_API) && DB_VERSION_MAJOR > 2 - RETVAL->dbp = __db185_open(name, flags, mode, RETVAL->type, openinfo) ; + RETVAL->dbp = __db185_open(name, flags, mode, RETVAL->type, info) ; #else - RETVAL->dbp = dbopen(name, flags, mode, RETVAL->type, openinfo) ; + RETVAL->dbp = dbopen(name, flags, mode, RETVAL->type, info) ; #endif /* DB_LIBRARY_COMPATIBILITY_API */ #endif --- ext/DB_File/version.c~ Sun Jan 23 08:15:45 2000 +++ ext/DB_File/version.c Sun Jun 25 12:34:01 2000 @... -25,7 +25,7 @... #include <db.h> -void +EXTERN_C void __getBerkeleyDBInfo() { SV * version_sv = perl_get_sv("DB_File::db_version", GV_ADD|GV_ADDMULTI) ; --- ext/B/B.xs~ Thu Feb 24 20:49:18 2000 +++ ext/B/B.xs Sun Jun 25 11:23:20 2000 @... -1054,7 +1054,7 @... GvLINE(gv) B::GV gv -char * +const char * GvFILE(gv) B::GV gv --- ext/B/typemap~ Thu Oct 28 17:35:07 1999 +++ ext/B/typemap Sun Jun 25 11:18:38 2000 @... -31,6 +31,7 @... SSize_t T_IV STRLEN T_IV PADOFFSET T_UV +const char * T_PV INPUT T_OP_OBJ --- config.sh.orig Sun Jun 25 13:43:19 2000 +++ config.sh Sun Jun 25 13:49:12 2000 @... -52,7 +52,7 @... c='' castflags='0' cat='cat' -cc='cc' +cc='g++' cccdlflags='-fpic' ccdlflags='-rdynamic' ccflags='-fno-strict-aliasing -I/usr/local/include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64' @... -76,8 +76,8 @... cppflags='-fno-strict-aliasing -I/usr/local/include' cpplast='-' cppminus='-' -cpprun='cc -E' -cppstdin='cc -E' +cpprun='g++ -E' +cppstdin='g++ -E' cppsymbols='_FILE_OFFSET_BITS=64 __GNUC_MINOR__=95 _LARGEFILE_SOURCE=1 _POSIX_C_SOURCE=199506 _POSIX_SOURCE=1 __STDC__=1 __i386=1 __i386__=1 __linux=1 __linux__=1 __unix=1 __unix__=1' crosscompile='undef' cryptlib='' @... -383,7 +383,7 @... dlsrc='dl_dlopen.xs' doublesize='8' drand01='drand48()' -dynamic_ext='B ByteLoader DB_File Data/Dumper Devel/DProf Devel/Peek Fcntl File/Glob GDBM_File IO IPC/SysV ODBM_File Opcode POSIX SDBM_File Socket Sys/Hostname Sys/Syslog attrs re' +dynamic_ext='B ByteLoader DB_File Data/Dumper Devel/DProf Devel/Peek Fcntl File/Glob GDBM_File IO IPC/SysV Opcode POSIX SDBM_File Socket Sys/Hostname Sys/Syslog attrs re' eagain='EAGAIN' ebcdic='undef' echo='echo' @... -392,7 +392,7 @... eunicefix=':' exe_ext='' expr='expr' -extensions='B ByteLoader DB_File Data/Dumper Devel/DProf Devel/Peek Fcntl File/Glob GDBM_File IO IPC/SysV ODBM_File Opcode POSIX SDBM_File Socket Sys/Hostname Sys/Syslog attrs re Errno' +extensions='B ByteLoader DB_File Data/Dumper Devel/DProf Devel/Peek Fcntl File/Glob GDBM_File IO IPC/SysV Opcode POSIX SDBM_File Socket Sys/Hostname Sys/Syslog attrs re Errno' fflushNULL='define' fflushall='undef' find='' @... -532,7 +532,7 @... known_extensions='B ByteLoader DB_File Data/Dumper Devel/DProf Devel/Peek Fcntl File/Glob GDBM_File IO IPC/SysV NDBM_File ODBM_File Opcode POSIX SDBM_File Socket Sys/Hostname Sys/Syslog Thread attrs re' ksh='' large='' -ld='cc' +ld='g++' lddlflags='-shared -L/usr/local/lib' ldflags=' -L/usr/local/lib' ldlibpthname='LD_LIBRARY_PATH' --- ext/Devel/DProf/DProf.xs~ Fri Feb 4 11:43:02 2000 +++ ext/Devel/DProf/DProf.xs Sun Jun 25 11:59:02 2000 @... -3,6 +3,11 @... #include "perl.h" #include "XSUB.h" +/* XXX If this causes problems, set i_unistd=undef in the hint file. */ +#ifdef I_UNISTD +#include <unistd.h> +#endif + /* For older Perls */ #ifndef dTHR # define dTHR int dummy_thr @... -60,7 +65,7 @... clock_t tms_utime; /* cpu time spent in user space */ clock_t tms_stime; /* cpu time spent in system */ clock_t realtime; /* elapsed real time, in ticks */ - char *name; + const char *name; U32 id; opcode ptype; }; @... -210,7 +215,7 @... } static void -prof_dumps(pTHX_ U32 id, char *pname, char *gname) +prof_dumps(pTHX_ U32 id, const char *pname, const char *gname) { PerlIO_printf(g_fp,"& %"UVxf" %s %s\n", (UV)id, pname, gname); } @... -241,8 +246,8 @... } else if (ptype == OP_GV) { U32 id = g_profstack[base++].id; - char *pname = g_profstack[base++].name; - char *gname = g_profstack[base++].name; + const char *pname = g_profstack[base++].name; + const char *gname = g_profstack[base++].name; prof_dumps(aTHX_ id, pname, gname); } @... -318,7 +323,8 @... { SV **svp; - char *gname, *pname; + const char *gname; + const char *pname; CV *cv; cv = INT2PTR(CV*,SvIVX(Sub)); --- ext/File/Glob/bsd_glob.c~ Thu Mar 2 12:53:17 2000 +++ ext/File/Glob/bsd_glob.c Sun Jun 25 12:08:02 2000 @... -63,6 +63,11 @... #include <perl.h> #include <XSUB.h> +/* XXX If this causes problems, set i_unistd=undef in the hint file. */ +#ifdef I_UNISTD +#include <unistd.h> +#endif + #include "bsd_glob.h" #ifdef I_PWD # include <pwd.h> @... -637,6 +642,12 @... /* NOTREACHED */ } +#ifdef __cplusplus +typedef Direntry_t *(*readdirfunc_t)(DIR*); +#else +typedef Direntry_t *(*readdirfunc_t)(); +#endif + static int glob3(Char *pathbuf, Char *pathend, Char *pattern, Char *restpattern, glob_t *pglob) @... -646,14 +657,7 @... int err; int nocase; char buf[MAXPATHLEN]; - - /* - * The readdirfunc declaration can't be prototyped, because it is - * assigned, below, to two functions which are prototyped in glob.h - * and dirent.h as taking pointers to differently typed opaque - * structures. - */ - Direntry_t *(*readdirfunc)(); + readdirfunc_t readdirfunc; *pathend = BG_EOS; errno = 0; @... -689,9 +693,9 @... /* Search directory for matching names. */ if (pglob->gl_flags & GLOB_ALTDIRFUNC) - readdirfunc = pglob->gl_readdir; + readdirfunc = (readdirfunc_t)pglob->gl_readdir; else - readdirfunc = my_readdir; + readdirfunc = (readdirfunc_t)my_readdir; while ((dp = (*readdirfunc)(dirp))) { register U8 *sc; register Char *dc; @... -859,7 +863,7 @... g_Ctoc(str, buf); if (pglob->gl_flags & GLOB_ALTDIRFUNC) - return((*pglob->gl_opendir)(buf)); + return((DIR*)(*pglob->gl_opendir)(buf)); else return(PerlDir_open(buf)); } -- John Tobey, late nite hacker <jtobey@... \\\ /// ]]] With enough bugs, all eyes are shallow. [[[ /// \\ </PRE> http://ASPN.ActiveState.com/ASPN/Mail/Message/perl6-Porters/420916 <PRE>On a counter-note, I was kinda anti-ANSI, too. But a late night discussion on #perl has me swayed a little bit, Abigail and mjd scored some good points for standardization: Schwern: Hmmm, Mr. Rosler misses the point of standardization. yrlnry: schwern: I somehow doubt that. Schwern: Well, why "officially" standardize something that's not fractured? pudge_: Schwern, so businesses can rely on it. Schwern: Yay. They already can rely on it. pudge_: no, not on perl pudge_: on a standard Schwern: Perl people are absolutely crazy about backwards compatibility. pudge_: i am not saying i agree pudge_: you asked why standardize, and i told you :) pudge_: actually Schwern: ...but that's his argument. Yeah. yrlnry: crazy in the sense of ignoring it in an irrational and self-destructive way, perhaps. pudge_: one could make an argument that perl IS fractured Schwern: I'm just afraid it'll get frozen in time like C. pudge_: you havce 5.004 over there, 5.005 here, and 5.6 out there Schwern: Pudge: Standardization won't help that anymore than it helped C++ pudge_: Schwern, my hope is that Perl WILL get frozen in time. Schwern: There's still tons of non-ANSI C++ compilers out there. pudge_: when it is done. fimmtiu: C++'s problems are not a result of standardization. fimmtiu: C++'s problems are because it's a poorly designed language. Schwern: fimmtiu: Not what I'm saying. pudge_: C++ has only one problem. pudge_: it's not perl! **later** Abigail: What do people think of the Larry Rosler interview? Schwern: Abigail: His arguments for turning Perl into hundreds of pages of unreadable documents are less than convincing. PerlJam: Abigail: I'm composing email to Larry Rosler about it right now. I think he's clued but misguided. Schwern: Abigail: THen again... we already have man pages... PerlJam: What Schwern just said. Abigail: I didn't see him talking about "unreadable" documentation. Schwern: Abigail: Ever read an ANSI spec? Abigail: Schwern: yes. PerlJam: Abigail: What do *you* think about it? Schwern: Abigail: Not the clearest things in the universe. Schwern: And has a language as complicated as Perl ever been spec'd? PerlJam: Schwern: C++ yrlnry: schwern: Have you read the ANSI C standard? I don't think it's unreadable. Schwern: C++ has nothing on Perl. Schwern: mjd: My copies in storage, I've seen a few gems. quidity: If we had a standards committee, would we have to adopt strict definitions of all the Larry Wall quotes as well? Abigail: I think that it would benefit Perl in the long run to standardize it. However, it should be a standard that allows things that are "deprecated" to be removed in future versions. uri: mjd: ever read the pl/1 specs? total gibberish. yrlnry: uri: No, I haven't had that pleasure. uri: mjd: the pleasure^Wpain was all mine Abigail: One of the problems now is that certain things aren't well defined, and if behaviour chances from version to version, a bugreport can be dismissed as "well, it was never well defined, so it isn't a bug". Schwern: I'm more afraid of calcification of Perl than anything else. uri: mjd: i wrote massive runtime library stuff from that spec. i still have the scars dngor: That's what Ilya means when he says Perl is nondeterministic, right? yrlnry: Another problem is that the current development team doesn't seem to particularly care about backward compatibility anyway. Abigail: For instance, what should the following print: sub AUTOLOAD {print $AUTOLOAD} *foo = *bar; foo (); ? (It prints different things in 5.005 and 5.6 for instance). Schwern: mjd: Odd, I've always found p5p to be absolutely rabid about that. yrlnry: schwern: You haven't been paying very much attention then. Schwern: Not since 5.6.0 came out. yrlnry: Not since about three yeasr ago, it appears. yrlnry: Try reading perldelta sometime. clintp: The theme is "Well, we can break backward compatability, as long as it's not something important that anyone in p5p uses." yrlnry: Or if Larry decides that the new way is more useful. clintp: For varying degrees of "important" Abigail: "Important" often meaning "I don't like the new way of doing it, so the old way is important". Schwern: Here's the question. Would an ANSI spec chisel Perl in stone? Would there be no more new features because of it (or effectively because they weren't in the standard)? yrlnry: Well, that didn't happen to C. Why would it happen to Perl? PerlJam: Schwern: in as much as C is "chiselled in stone" Schwern: C relies more on its libraries than its core language. uri: schwern: i think the issue is less about features than clarifying what perl is supposed to do in odd places. it is too empirical sometimes. PerlJam: Perl needs to be redesigned from the ground up (again) *** mandrake- (mandrake@... has joined channel #perl Abigail: Having a Perl standard means that other people could write compatible Perl compilers/interpreters. mandrake-: huh Schwern: I'd consider C to be (arguably) "complete" yrlnry: Schwern: Regardless of what you think, the new standard is standardizing features that have been introduced after 1989. Schwern: mjd: What new standard? PerlJam: uri: ah yrlnry: schwern: There is a new ANSI C standard in development. Abigail: C9x Schwern: There's a new ANSI C standard? I had no idea. Schwern: of course, this will take another decade to be widely adopted... yrlnry: That makes like five things you have cirticized and argued about in the last twelve hours that you had no idea about. Schwern: mjd: That's why I have you here, dear. PerlJam: Schwern: Have you seen GNU C lately? Abigail: Schwern: there are a hell of a lot more C compilers and run time environments out there then there are Perl compilers. Abigail: (Perl compiler being perl, not perlcc). -- Michael G Schwern http://www.pobox.com/~schwern/ schwern@... <GuRuThuG> make a channel called Perl, and infest it with joking and fun....it doesnt make alot of sense. </PRE> http://ASPN.ActiveState.com/ASPN/Mail/Message/perl6-Porters/420915 <PRE>Perl is my only language, its a monoculture like a family, dysfunctional at times, but show me a family isn't (at times). In fact the whole public domain seems a little goofy at times but I stick with it because I know the end results are going to be fantastic. Perl is designed in a true democracy, chaotic at times. Even the group decision to keep the Y2K "bug" intact is endearing. It seems Larry Rosler wants to change all that. I am cut+pasting from http://www.perl.com/pub/2000/06/rosler.html?wwwrrr_20000606.txt In his own words: LR: I believe that [ Perl needs ANSI standardization ], in order to increase its acceptability... ME: Perl is already everywhere, it is the single most used language on the web, Java lacks performance an always seems to be in court. A few stuffed shirts seem to be in denial about perl but their days are numbered anyway. LR: I was experimenting with CGI programming using shell ... Soon I discovered that Perl ... was much more expressive and much ME: Duh !! Shell is nearly useless. LR: ... 'official' Perl semantics has [ err, have ] never been adequately characterized independent of the implementation, so is subject to arbitrary change. ME: Arbitrary ?? I can honestly say that I have never seen a single perl feature that didn't make me proud of the perl way of making decisions. Its a model for the whole human race :) LR: Building on quicksand is acceptable for 'scripts' of limited longevity ME: Hugh ?? Knowing systems operations, I am guessing most scripts will out live Bell Labs, even UN*X LR: Ilya Zakharevitch is most outspoken in his view that Perl is not (yet?) a 'programming' language! ME: Applications building language !! Perl does not seek to replace C, but work w/ it (and the others.) Its a philosophy _and_ a desert topping. LR: people should devote their attention to firming up the semantics and making sure that the implementation conforms to those semantics, rather than the other way around ME: No more swiss-army-chainsaw, :`( Joe Johnston: Or to Perl's charmingly permissive OO implementation. ME: Gush !! LR: I am less concerned about individual programmers [ and ] more concerned about major corporations or government agencies ME: Perl is a collage of individuals who wont be denied, even in the workplace. It seems LR wants to hand it over to the suits. The new economy smart enough to reject corporate welfare and bold enough to take on the lazy big-caps. Perl is the language to do it. LR: Samuel Johnson: "When a man knows he is to be hanged in a fortnight, it concentrates his mind wonderfully." ME: Better start thinking a little smarter because the small-caps are the clear winners, the old economy is hanging by a thread and doesn't even know it. LR: Python is [in]sufficient to give it a critical mass and I don't think significant inroads are being made. ME: My sources tell me that Python is _the_ language to learn from. It will never achieve chainsaw status, and thats probably a good thing. Small is beautiful. Thanks for denigrating it, now I'm going to take the plunge. LR: I have never needed to write an object in any program. And, as I said, to me OO is a big yawner. ME: Hey, news flash: CPAN = OO modules !! Joe Johnston: How much longer do you see the "Internet Goldrush" continuing? LR: Judging by the recent performance of the NASDAQ, it may be over already. ME: Well since then the gold rush has half returned. This statement is typical of the cynicism and ignorance that caused the tumble in the first place. I support the small-caps but dont want work for them for the simple reason that they will make me work. Fact is that you can actually fall asleep in your big-cap cubicle for years... I dont sleep, obviously, but I spend all my time fathoming how perl is going to save our werld. The economy today is global. The NASDAQ collapse was thought of as an internet failure, but that was the media talking out their ass. A quick lesson: There are several indexes which show the whole picture. While the connection between the stocks and investment is dubious at best, its the only game in the global village. We, the privileged, vote for our economy by buy stocks in the sectors we support, usually through funds. DOW: The fat cats, receivers of corporate welfare S&P 500: Generally midcaps, it spans the distance to the next layer, the Russell 2000: Small, aggressive successful companies who pray each and every day for lower interest rates NASDAQ: Similar to the Russell 2000, in fact the charts are parallel. The trend since y2k is clear. Smaller companies, despite high interest rates, started eating the bigcaps for lunch. Technology is a key booster, but other stuff is going on. Leading up to March 10th peak the whole world was elated. To quote some old cowboy, "you spit in the desert and a flower blooms." In a way, these broken investments are the best because recovery is pure growth, and growth is profit. For example, the application of translation software to an array of B2B applications simply drops state of the art business services on any broken nation instantly fixing it making it a viable trading partner. From y2k until the end of March, the naz and the dow were oscillating against each other. Growth possibilities brought down the dow, interest rate fears boosted the dow and killed the naz. Now both are growing, but, face it, the big caps are doomed, though we may have to topple the berlin wall here in US. Clearly, the big-caps are not interested in seeing their coporate welfare evaoprate, nor do they want to be gobbled alive by the up-and-comings. Greenspan exists to keep their machine running. His contradictory statements would have made Nixon blush. His claims of controlling inflation are pure lie, he only seeks to lower one number, OUR salaries. That and only that number will determine interest rate hikes. The problem is that we are worth the money, and always have been. And perlers are worth even ore. LR: Each of us is working on the Perl Golf tournament ME: There ya go !! ===== John van Vlaanderen ############################################# # CXN, Inc. Contact: john@... # # # Proud Sponsor of Perl/Unix of NY # # http://puny.vm.com # ############################################# __________________________________________________ Do You Yahoo!? Yahoo! Photos -- now, 100 FREE prints! http://photos.yahoo.com </PRE> http://ASPN.ActiveState.com/ASPN/Mail/Message/perl6-Porters/420914 <PRE>From: Chip Salzenberg [mailto:chip@... > > The standard C++ vector<> container is almost perfect for use in Perl > arrays and the data stack. But it has a major weakness... It's _so_ > general that it doesn't use memcpy() [or moral equivalent] when it has > to grow. Instead, it uses copy constructors in a loop. That could be > a major performance issue with vector<Counted_p<T>>. > > Now, I could write a vector<> -- it's not that hard, as C++ goes -- > but I'm wondering if anyone has a lead on such a thing already, and/or > library(ies) of such handy classes. I don't know if/how such details are or should be handled, but you might want to take a look at boost (www.boost.org) or ask on the boost mailing list (boost@... The people there have done a lot of work on extensions to the STL classes, and may have what you want (I recall something about a fast allocator, which may help performance of the standard containers - if not, I'm sure there are people on the list who would be interested in thrashing out performance issues like this). On a related note, boost already has a counted pointer class, and they are working hard on efficiency issues for it, so that may be worth a look, too. Paul. </PRE> http://ASPN.ActiveState.com/ASPN/Mail/Message/perl6-Porters/420913 <PRE>I like it. Its 4K+ lines are clearly enough written to break them up into a number of smaller modules specific to their packages and methods. This would allow hooks for upstream applications to implement aspects of the module. A lot of the methods only use packages as a place to keep them, they are independant of the class structures. These could go into a utility modules like the ones I am creating from the perl power tools for NT. Documentation at the end implies that its success depends on the evolution of modules in the rest of perl. Nice !!! ===== John van Vlaanderen ############################################# # CXN, Inc. Contact: john@... # # # Proud Sponsor of Perl/Unix of NY # # http://www.thinman.com/puny # ############################################# __________________________________________________ Do You Yahoo!? Send instant messages & get email alerts with Yahoo! Messenger. http://im.yahoo.com/ </PRE>