27 years agoWin32 boot_DynaLoader problem in 99
Gary Clark [Mon, 5 May 1997 17:49:03 +0000]
Win32 boot_DynaLoader problem in 99

In makedef.pl, boot_DynaLoader is defined in $skip_sym.  It should be
included in the __DATA__ section to force it into the dll exports.

p5p-msgid: 1997May05.105000.1708.84476@mail.jeld-wen.com

27 years agoHPUX: patch for ext/DynaLoader/dl_hpux.xs
Chuck D. Phillips [Mon, 5 May 1997 05:48:39 +0000]
HPUX: patch for ext/DynaLoader/dl_hpux.xs

By default, doing a "make test" on a dynamically loaded module with
unresolved symbols will result with a useless error message.  Adding
the BIND_VERBOSE flag will add the names of undefined symbols to the
error message already printed to stderr.  This does *not* affect
default operation; only when PERL_DL_NONLAZY is set does it make a

p5p-msgid: 199705050548.WAA21260@palrel1.hp.com

27 years agoIrix hint update
Scott Henry [Tue, 6 May 1997 18:09:56 +0000]
Irix hint update

Hopefully, this is the final set of patches for the IRIX hints file
hints/irix_6.sh.  This is rather a multi-person effort, Jarkko and
John Stoffel wrote the major changes (and changed the default to
-n32), I made the last tweaks so that -o32 still works without
errors on the various compiler releases.  I've tested on IRIX-6.2,
IRIX64-6.2, IRIX64-6.4 and IRIX-6.5alpha.

p5p-msgid: yd83es0fp57.fsf@hoshi.engr.sgi.com

27 years agoNetBSD hint update
Giles Lean [Mon, 5 May 1997 13:46:37 +0000]
NetBSD hint update

There is a missing character in hints/netbsd.sh.

Line 51
case "$osver" in

should be
case "$osvers" in

Missing this case leads to the functions

setregid()
setreuid()
setrgid()
setruid()

being used, leading to a security problem on all released versions
of NetBSD.

p5p-msgid: 199705051346.XAA13584@topaz.nemeton.com.au

27 years agoDon't require CvDEPTH == 0 when bombing out of subs.
Malcolm Beattie [Fri, 2 May 1997 19:03:49 +0000]
Don't require CvDEPTH == 0 when bombing out of subs.

p4raw-id: //depot/thrperl@9

27 years ago[differences between cumulative patch application and perl-5.003_99]
Chip Salzenberg [Wed, 30 Apr 1997 12:00:00 +0000]
[differences between cumulative patch application and perl-5.003_99]

27 years ago[inseparable changes from match from perl-5.003_98 to perl-5.003_99]
Perl 5 Porters [Wed, 30 Apr 1997 12:00:00 +0000]
[inseparable changes from match from perl-5.003_98 to perl-5.003_99]

 CORE PORTABILITY

Subject: NeXT hints update
Date: Thu, 1 May 97 00:28:41 +0200
From: Unknown Contributor <hansm@euronet.nl>
Files: Configure Makefile.SH hints/next_4.sh

    private-msgid: 199704302229.AAA02690@mail.euronet.nl

Subject: Support shared libperl on AIX
Date: Sat, 26 Apr 1997 20:31:37 -0500
From: Eric Bartley <bartley@icd.cc.purdue.edu>
Files: Configure Makefile.SH hints/aix.sh
Msg-ID: 199704270131.UAA51426@icd.cc.purdue.edu

    (applied based on p5p patch as commit 019c9bb5516c92cceeeb4b050fa8075a1359c43c)

27 years agoSupport shared libperl on AIX
Eric Bartley [Sun, 27 Apr 1997 01:31:37 +0000]
Support shared libperl on AIX

The current distribution will not build a shared libperl on AIX.

Currently the Makefile generated uses the same ld options for building
a shared libperl and dynamicly loaded extentions.  AIX's compiler
requires that you tell it explicitly what symbols to export/import. So,
it needs a separate set of LD options for each.

On top of all this, AIX doesn't seem to have an equivilent to
LD_LIBRARY_PATH. So, the directories to search during runtime must be
explicitly compiled in. In the following patch I compiled in both the
compilation directory (for miniperl and testing) and the installation
directory. I wish someone could suggest something cleaner, but it works.

p5p-msgid: 199704270131.UAA51426@icd.cc.purdue.edu

27 years agoFix typo in perldelta
Chip Salzenberg [Wed, 30 Apr 1997 13:57:56 +0000]
Fix typo in perldelta

27 years agoFix description of av_undef() in perlguts
Gisle Aas [Thu, 1 May 1997 10:42:46 +0000]
Fix description of av_undef() in perlguts

private-msgid: 199705011042.MAA09897@bergen.sn.no

27 years ago[dummy merge]
Chip Salzenberg [Wed, 30 Apr 1997 12:00:00 +0000]
[dummy merge]

This merge exists so that the p5p version of the patch and the applied
version are both in the history

27 years agotypo fixes to installhtml
Chip Salzenberg [Wed, 30 Apr 1997 12:00:00 +0000]
typo fixes to installhtml

(this is the same change as commit de0d1a7847c16bea152243fdffd9a1bc0e6d8f8a, but as applied)

27 years agotypo fixes to installhtml
Larry W. Virden [Thu, 1 May 1997 11:14:31 +0000]
typo fixes to installhtml

I noticed a few typos in this file so here's a fix.  It's from a strictly
vanilla install.

p5p-msgid: 199705011114.HAA26968@cas.org

27 years agoDisable op/pipe.t test under Machten
Dominic Dunlop [Thu, 1 May 1997 10:48:26 +0000]
Disable op/pipe.t test under Machten

At 09:47 +0200 1997-05-01, I wrote:
>Everything else is as expected, but I'm looking into this:
>
>dialup02 domo$ ./TEST io/pipe.t
>io/pipe...........print failed: Broken pipe at ./io/pipe.t line 86.
>FAILED on test 9

Here's a patch.  It seems MachTen doesn't behave like native UNIX here, so
punt.

Given that VMS doesn't behave like native UNIX either (and there's been
nothing on the list yet about NT and other "aliens"), test 9 looks like a
good emulation exposer.  Is this what we want?  As far as I can make out
from a little mucking about and a reading of ISO 9945-1:1990, MachTen is
behaving in a way that POSIX allows: stdio elects not to buffer writes on
pipes (a defensible decision); a write on a broken pipe fails with EPIPE
and raises SIGPIPE (correct by POSIX); close of a broken pipe succeeds
(correct by POSIX,and OK by perl: nothing's buffered, because the write
failed) if the pipe reader exited with zero status.  Slightly
disconcertingly, I discover that, if I substitute the command 'false' for
'true' as the putative pipe reader, the (perl) close fails.  close() is not
documented as failing if a pipe reader has terminated with a non-zero exit
code.  Is this a documentation shortcoming, a general perl bug, or a
MachTen perl bug?  Whether close() suceeds or fails, $? is set
appropriately.

Whatever, so as to get 5.004 out of the door, either apply this patch or do
away with test 9 entirely, pending deeper reflection.

p5p-msgid: v03102801af8e160d3879@[194.51.248.68]

27 years agoRefresh DB_File to 1.14
Paul Marquess [Wed, 30 Apr 1997 20:45:09 +0000]
Refresh DB_File to 1.14

Something for _99.

Made it illegal to tie an associative array to a RECNO database and an
ordinary array to a HASH or BTREE database.

p5p-msgid: 9704302045.AA05484@claudius.bfsec.bt.co.uk

27 years agoAdd new globals to perl.exp
Chip Salzenberg [Thu, 6 Mar 1997 20:53:34 +0000]
Add new globals to perl.exp

27 years agoFix OS/2-specific buffer overflow
Ilya Zakharevich [Wed, 30 Apr 1997 19:20:01 +0000]
Fix OS/2-specific buffer overflow

private-msgid: 199704301920.PAA09681@monk.mps.ohio-state.edu

27 years agoFix sysread() on tied handle
Spider Boardman [Thu, 1 May 1997 06:01:20 +0000]
Fix sysread() on tied handle

Found in both perldelta.pod and perltie.pod:

: =item READ this LIST
:
: This method will be called when the handle is read from via the C<read>
: or C<sysread> functions.

This isn't true without the following patch:

p5p-msgid: 199705010601.CAA04926@Orb.Nashua.NH.US

27 years agoHandle tainted values in lists returned from subs, evals
Chip Salzenberg [Fri, 25 Apr 1997 01:55:09 +0000]
Handle tainted values in lists returned from subs, evals

27 years agoForget prototype of subroutine after C<undef &subr>
Chip Salzenberg [Thu, 24 Apr 1997 14:34:21 +0000]
Forget prototype of subroutine after C<undef &subr>

27 years agoFix NUL-termination bug in delimcpy()
Chip Salzenberg [Wed, 30 Apr 1997 05:27:58 +0000]
Fix NUL-termination bug in delimcpy()

27 years ago[inseparable changes from match from perl-5.003_97j to perl-5.003_98]
Perl 5 Porters [Tue, 29 Apr 1997 12:00:00 +0000]
[inseparable changes from match from perl-5.003_97j to perl-5.003_98]

 DOCUMENTATION

Subject: Add new diagnostics to perldelta
From: Chip Salzenberg <chip@perl.com>
Files: pod/perldelta.pod pod/perldiag.pod

27 years agoSupport C< $coderef->($x,$y) >
Chip Salzenberg [Tue, 29 Apr 1997 12:00:00 +0000]
Support C< $coderef->($x,$y) >

Randal Schwartz said:
> Some time in October, 1994, Larry Wall said:
> > : All other references you can follow with ->, what about code refs?
> > Aw, look, I'd really like that feature in, but I think it's a bit
> > too close to the release for that.
> Hey Chip,
> Look, Larry said he'd really like that feature

Larry said that?  That change is like two lines, it's in!

Tricked-into-doing-by: Randal Schwartz <merlyn@stonehenge.com>

27 years agoDescribe Math::Trig in perlmodlib
Chip Salzenberg [Sat, 26 Apr 1997 00:39:00 +0000]
Describe Math::Trig in perlmodlib

27 years agoFix warning wrt return value of PerlIO_getname()
Spider Boardman [Wed, 30 Apr 1997 04:48:13 +0000]
Fix warning wrt return value of PerlIO_getname()

Problem with -Duseperlio:

cc: Warning: perlio.c, line 302: Non-void function "PerlIO_getname" does not contain a return statement.
char *
^

p5p-msgid: 199704300448.AAA24174@Orb.Nashua.NH.US

27 years ago[dummy merge]
Chip Salzenberg [Tue, 29 Apr 1997 12:00:00 +0000]
[dummy merge]

This merge exists so that the p5p version of the patch and the applied
version are both in the history

27 years agoReset errno after failed piped close
Chip Salzenberg [Tue, 29 Apr 1997 12:00:00 +0000]
Reset errno after failed piped close

(this is the same change as commit 00db273fa22ecba6466df777a772c6017c403a96, but as applied)

27 years agoReset errno after failed piped close
Roderick Schertler [Mon, 28 Apr 1997 22:02:20 +0000]
Reset errno after failed piped close

Chip had asked me to redo the pclose patches against 97j.  I'm sending
them here to encourage others to try them out since _98 looms so near.
Also included are the patches I'd sent on Friday for the bugs Gisle
reported with Time::{localtime,gmtime}, with these this message contains
all my outstanding patches.

p5p-msgid: 28152.862264940@eeyore.ibcinc.com

27 years agoFix C< hex('80') * 0x1000000 >
Chip Salzenberg [Fri, 18 Apr 1997 14:42:33 +0000]
Fix C< hex('80') * 0x1000000 >

27 years ago[differences between cumulative patch application and perl-5.003_97j]
Chip Salzenberg [Sat, 26 Apr 1997 12:00:00 +0000]
[differences between cumulative patch application and perl-5.003_97j]

27 years ago[inseparable changes from match from perl-5.003_97i to perl-5.003_97j]
Perl 5 Porters [Sat, 26 Apr 1997 12:00:00 +0000]
[inseparable changes from match from perl-5.003_97i to perl-5.003_97j]

 CORE PORTABILITY

Subject: Return to favoring memset(,0,) over bzero()
From: Chip Salzenberg <chip@perl.com>
Files: perl.h

 OTHER CORE CHANGES

Subject: Update sprintf: '%hn'; '%s',NULL; panic on frexp() failure
From: Chip Salzenberg <chip@perl.com>
Files: perl.h pod/perldiag.pod sv.c

27 years agoDoc fix for close of pipe handle
Chip Salzenberg [Thu, 24 Apr 1997 16:55:38 +0000]
Doc fix for close of pipe handle

27 years agoDescribe __PACKAGE__ in perldelta
Chip Salzenberg [Thu, 24 Apr 1997 16:16:44 +0000]
Describe __PACKAGE__ in perldelta

27 years agoSplit perlmod
Tom Christiansen [Sat, 26 Apr 1997 00:50:09 +0000]
Split perlmod

private-msgid: 199704260050.RAA02468@toy.perl.com

27 years agoSelfLoader: fix prototype pattern, rename intrusive lexical
Jesse Glick [Thu, 2 Jan 1997 23:18:01 +0000]
SelfLoader: fix prototype pattern, rename intrusive lexical
Signed-off-by: Jesse Glick <jesse@ginger>

27 years agoRefresh DB_File to 1.13
Paul Marquess [Sun, 27 Apr 1997 14:12:59 +0000]
Refresh DB_File to 1.13

Subject: DB_File 1.13 patch

Here is the patch to (hopefully) fix the casting warnings.

p5p-msgid: 9704271413.AA08876@claudius.bfsec.bt.co.uk

27 years agoRefresh CGI.pm to 2.35
Lincoln Stein [Thu, 17 Apr 1997 18:22:51 +0000]
Refresh CGI.pm to 2.35

27 years agoFix possible buffer overflow under VMS
Chip Salzenberg [Sat, 1 Mar 1997 06:14:33 +0000]
Fix possible buffer overflow under VMS

27 years agoFix typo in NeXT dynaloader
Chip Salzenberg [Wed, 23 Apr 1997 18:29:12 +0000]
Fix typo in NeXT dynaloader

27 years agoEliminate Alpha warnings
Hallvard B Furuseth [Thu, 24 Apr 1997 21:45:10 +0000]
Eliminate Alpha warnings
Signed-off-by: Hallvard B Furuseth <h.b.furuseth@usit.uio.no>

27 years agoFix lingering '%S' in XS_VERSION_BOOTCHECK
Chip Salzenberg [Wed, 23 Apr 1997 14:27:36 +0000]
Fix lingering '%S' in XS_VERSION_BOOTCHECK

27 years agoWin32 update (three patches)
Gurusamy Sarathy [Thu, 24 Apr 1997 17:13:30 +0000]
Win32 update (three patches)
Signed-off-by: Gurusamy Sarathy <gsar@engin.umich.edu>

27 years agoHP-UX hint update
Chuck D. Phillips [Mon, 28 Apr 1997 05:35:07 +0000]
HP-UX hint update

Subject: suggested changes to hpux hints file

NOTE: It *might* not work under 11.x, but at least it's not guaranteed
to fail.  :-)

p5p-msgid: 199704280535.WAA22441@palrel1.hp.com

27 years agoNetBSD hint update
matthew green [Fri, 25 Apr 1997 10:18:02 +0000]
NetBSD hint update

hi folks... been a while since i posted here...

someone told me that 5.004 was happening RFSN... so i thought it was
time to update the hints/netbsd.sh file, as it needs it.

p5p-msgid: 199704251021.EAA22570@jhereg.perl.com

27 years ago[differences between cumulative patch application and perl-5.003_97i]
Chip Salzenberg [Thu, 24 Apr 1997 12:00:00 +0000]
[differences between cumulative patch application and perl-5.003_97i]

27 years ago[inseparable changes from match from perl-5.003_97h to perl-5.003_97i]
Perl 5 Porters [Thu, 24 Apr 1997 12:00:00 +0000]
[inseparable changes from match from perl-5.003_97h to perl-5.003_97i]

 CORE PORTABILITY

Subject: Provide memset() if it's missing
From: Chip Salzenberg <chip@perl.com>
Files: global.sym perl.h proto.h util.c

Subject: Don't tell GCC that warn(), croak(), and die() are printf-lik
From: Chip Salzenberg <chip@perl.com>
Files: proto.h

 DOCUMENTATION

Subject: FAQ udpate (24-apr-97)
Date: Thu, 24 Apr 1997 16:47:23 -0600 (MDT)
From: Nathan Torkington <gnat@prometheus.frii.com>
Files: pod/perlfaq*.pod

    private-msgid: 199704242247.QAA07010@prometheus.frii.com

 OTHER CORE CHANGES

Subject: Misc. sv_vcatpvfn() fixes
From: Hugo van der Sanden <hv@crypt.compulink.co.uk>
Files: gv.c mg.c op.c perl.c pp.c pp_ctl.c sv.c toke.c util.c

Subject: Enforce order of sprintf() elements
From: Chip Salzenberg <chip@perl.com>
Files: sv.c

Subject: Guard against long numbers, <<LONG_DELIM, and <long glob>
From: Chip Salzenberg <chip@perl.com>
Files: global.sym mg.c perl.c pod/perldiag.pod proto.h toke.c util.c

Subject: Guard against C<goto> to deeply nested label
From: Chip Salzenberg <chip@perl.com>
Files: pod/perldiag.pod pp_ctl.c

Subject: Guard against overflow in dup2() emulation
From: Chip Salzenberg <chip@perl.com>
Files: util.c

Subject: Win32: Guard against long function names
From: Chip Salzenberg <chip@perl.com>
Files: win32/win32sck.c

Subject: Make mess() always work, by using a non-arena SV
From: Chip Salzenberg <chip@perl.com>
Files: perl.c util.c

Subject: When copying a format line, take only its string value
From: Chip Salzenberg <chip@perl.com>
Files: sv.c

Subject: Fix LEAKTEST numbers
From: Chip Salzenberg <chip@perl.com>
Files: ext/DynaLoader/dl_vms.xs handy.h os2/os2.c util.c vms/vms.c win32/win32.c win32/win32sck.c

27 years agoDocument new {,s}printf() behavior
Chip Salzenberg [Wed, 23 Apr 1997 23:48:51 +0000]
Document new {,s}printf() behavior

27 years agoFix buffer overflow in a2p
Chip Salzenberg [Thu, 10 Apr 1997 19:49:15 +0000]
Fix buffer overflow in a2p

27 years agoMath::{Complex,Trig} update
Gurusamy Sarathy [Thu, 24 Apr 1997 22:21:44 +0000]
Math::{Complex,Trig} update

private-msgid: 199704242221.BAA30363@alpha.hut.fi

27 years agoHandle symlinks, high permission bits in File::Path
Chip Salzenberg [Wed, 16 Apr 1997 17:47:04 +0000]
Handle symlinks, high permission bits in File::Path

27 years agoDon't core dump if my_inet_aton() get a NULL
Chip Salzenberg [Thu, 19 Dec 1996 22:11:20 +0000]
Don't core dump if my_inet_aton() get a NULL

27 years agoAlways NUL-terminate opsets
Chip Salzenberg [Fri, 18 Apr 1997 16:37:00 +0000]
Always NUL-terminate opsets

27 years agoCope with a <db.h> that isn't related to DB
Jarkko Hietaniemi [Thu, 24 Apr 1997 17:28:39 +0000]
Cope with a <db.h> that isn't related to DB

private-msgid: 199704241728.UAA09951@alpha.hut.fi

27 years agoUndo private patch
Chip Salzenberg [Sun, 13 Apr 1997 09:06:24 +0000]
Undo private patch

27 years agoFix scalar leak in pp_prtf()
Doug MacEachern [Thu, 24 Apr 1997 17:06:21 +0000]
Fix scalar leak in pp_prtf()

private-msgid: 199704241706.NAA19140@postman.osf.org

27 years agoAdded programmer-level condition variables via "condpair" magic.
Malcolm Beattie [Wed, 23 Apr 1997 19:06:45 +0000]
Added programmer-level condition variables via "condpair" magic.
Added support for detached threads and tweaked a few things.

p4raw-id: //depot/thrperl@8

27 years ago[differences between cumulative patch application and perl-5.003_97h]
Chip Salzenberg [Tue, 22 Apr 1997 12:00:00 +0000]
[differences between cumulative patch application and perl-5.003_97h]

27 years ago[inseparable changes from match from perl-5.003_97g to perl-5.003_97h]
Perl 5 Porters [Tue, 22 Apr 1997 12:00:00 +0000]
[inseparable changes from match from perl-5.003_97g to perl-5.003_97h]

 BUILD PROCESS

Subject: Fix up Linux hints for tcsh, and Configure patch
Date: Tue, 22 Apr 1997 11:02:27 -0400 (EDT)
From: Andy Dougherty <doughera@lafcol.lafayette.edu>
Files: Configure hints/linux.sh
Msg-ID: Pine.SOL.3.95q.970422101051.2506C-100000@fractal.lafayette.e

    (applied based on p5p patch as commit 1eb1b1cb9647b817d039bb17afa3e74940b5ef92)

Subject: There is no standard answer to 'Use suidperl?'
From: Chip Salzenberg <chip@perl.com>
Files: hints/bsdos.sh hints/freebsd.sh hints/linux.sh hints/machten_2.sh

 CORE LANGUAGE CHANGES

Subject: Support PRINTF for tied handles
Date: Sun, 20 Apr 1997 18:26:13 -0400
From: Doug MacEachern <dougm@opengroup.org>
Files: pod/perldelta.pod pod/perltie.pod pp_sys.c t/op/misc.t
Msg-ID: 199704202226.SAA08032@postman.osf.org

    (applied based on p5p patch as commit e7c5525577c16ee25e3521e86aca2b5105dba394)

 CORE PORTABILITY

Subject: Fix bitwise shifts and pack('w') on Crays
From: Chip Salzenberg <chip@perl.com>
Files: pp.c

 DOCUMENTATION

Subject: FAQ udpate (23-apr-97)
Date: Wed, 23 Apr 1997 12:22:55 -0600 (MDT)
From: Nathan Torkington <gnat@prometheus.frii.com>
Files: pod/perlfaq*.pod

    private-msgid: 199704231822.MAA05074@prometheus.frii.com

 OTHER CORE CHANGES

Subject: Mondo Cool patch for buffer safety and convenience
From: Chip Salzenberg <chip@perl.com>
Files: XSUB.h doop.c dump.c ext/DynaLoader/dl_dlopen.xs ext/DynaLoader/dl_hpux.xs ext/DynaLoader/dl_next.xs ext/DynaLoader/dlutils.c ext/ODBM_File/ODBM_File.xs global.sym gv.c interp.sym mg.c op.c perl.c perl.h pod/perlguts.pod pp.c pp_ctl.c pp_hot.c pp_sys.c proto.h regcomp.c regexec.c sv.c toke.c util.c

Subject: Problems with glob
Date: Sun, 20 Apr 1997 02:44:32 -0400 (EDT)
From: Ilya Zakharevich <ilya@math.ohio-state.edu>
Files: op.c
Msg-ID: 1997Apr20.024432.1941365@hmivax.humgen.upenn.edu

    (applied based on p5p patch as commit a1230b335277820e65b8a9454ab751341204cf4f)

Subject: Fix scalar leak in closures
From: Chip Salzenberg <chip@perl.com>
Files: op.c scope.c

Subject: Refine error messages re: anon subs' prototypes
From: Chip Salzenberg <chip@perl.com>
Files: op.c

Subject: Outermost scope is void, not scalar
From: Chip Salzenberg <chip@perl.com>
Files: pp_ctl.c

27 years agoFix up Linux hints for tcsh, and Configure patch
Andy Dougherty [Tue, 22 Apr 1997 15:02:27 +0000]
Fix up Linux hints for tcsh, and Configure patch

perl5.003_97[mumble,mumble] contained some hints to work around a csh bug
in some versions of Linux.

Unfortunately, that hint doesn't work for a variety of reasons.  If
anyone's really interested in the anatomy of a failed hint, I can oblige.
Or you can just trust me :-)

p5p-msgid: Pine.SOL.3.95q.970422101051.2506C-100000@fractal.lafayette.edu
private-msgid: Pine.SOL.3.95q.970422101051.2506C-100000@fractal.lafayette.e

27 years agoTwo doublewords less
Jarkko Hietaniemi [Sun, 20 Apr 1997 19:38:13 +0000]
Two doublewords less

private-msgid: 199704201938.WAA07722@alpha.hut.fi

27 years agoProblems with glob
Ilya Zakharevich [Sun, 20 Apr 1997 06:44:32 +0000]
Problems with glob

This patch seems to make it on _93 (I recall that on _93 overwritten
`glob' just does not make it for <blah>):

p5p-msgid: 1997Apr20.024432.1941365@hmivax.humgen.upenn.edu

27 years agoDon't put leading newline on numeric strings
Andreas Koenig [Wed, 23 Apr 1997 08:47:00 +0000]
Don't put leading newline on numeric strings

Subject: NeXT fails one test due to atoi

pragme/constants has a test that does the equivalent of:

    print " \t\n1234" == 1234 ? "ok" : "not ok";

This test prints "not ok" on the NeXT. Here's some background how the
NeXT behaves with several leading whitespaces.

    koenig@franz:/M/extra/sources/perl/perl5.003_97% perl -le '
    print " \t\n1234" == 1234 ? "ok" : "not ok";
    print " 1234" == 1234 ? "ok" : "not ok";
    print "\t1234" == 1234 ? "ok" : "not ok";
    print "\n1234" == 1234 ? "ok" : "not ok";
    '
    not ok
    ok
    ok
    not ok

I'd recommend the following patch:

p5p-msgid: 199704230847.KAA22752@anna.in-berlin.de

27 years agoSupport PRINTF for tied handles
Doug MacEachern [Sun, 20 Apr 1997 22:26:13 +0000]
Support PRINTF for tied handles

A mod_perl user just asked why "print ..." is sent to the browser but
"printf ..." goes to the term window.  Sorry this is coming in late,
this question has been asked a few times in the past, but I forgot
about it :-(

p5p-msgid: 199704202226.SAA08032@postman.osf.org

27 years agoSilence warnings on simple C<use ops>
Roderick Schertler [Sat, 19 Apr 1997 14:22:43 +0000]
Silence warnings on simple C<use ops>

private-msgid: pzybafum6k.fsf@eeyore.ibcinc.com

27 years agoCroak on C<use autouse> without module name
Chip Salzenberg [Tue, 1 Apr 1997 18:17:04 +0000]
Croak on C<use autouse> without module name

27 years agoMath::Complex update
Jarkko Hietaniemi [Wed, 23 Apr 1997 00:00:00 +0000]
Math::Complex update

27 years agoWin32 update (two patches)
Gurusamy Sarathy [Wed, 23 Apr 1997 00:00:00 +0000]
Win32 update (two patches)

27 years ago[differences between cumulative patch application and perl-5.003_97g]
Chip Salzenberg [Fri, 18 Apr 1997 00:00:00 +0000]
[differences between cumulative patch application and perl-5.003_97g]

27 years ago[inseparable changes from match from perl-5.003_97f to perl-5.003_97g]
Perl 5 Porters [Fri, 18 Apr 1997 00:00:00 +0000]
[inseparable changes from match from perl-5.003_97f to perl-5.003_97g]

 CORE LANGUAGE CHANGES

Subject: Improve sysseek(), remove systell(), fix Opcode
From: Chip Salzenberg <chip@perl.com>
Files: doio.c ext/Opcode/Makefile.PL ext/Opcode/Opcode.pm ext/Opcode/Opcode.xs global.sym keywords.pl opcode.pl pod/perldelta.pod pod/perldiag.pod pod/perlfunc.pod pp_sys.c proto.h t/op/sysio.t toke.c

 DOCUMENTATION

Subject: Document {,un}pack changes
Date: Fri, 18 Apr 97 13:49:39 BST
From: Paul Marquess <pmarquess@bfsec.bt.co.uk>
Files: pod/perldelta.pod pod/perldiag.pod
Msg-ID: 9704181249.AA11733@claudius.bfsec.bt.co.uk

    (applied based on p5p patch as commit 7f505e5d2748dc2677688b22967678392a186b16)

 OTHER CORE CHANGES

Subject: SECURITY FIX: 'Identifier too long'
From: Chip Salzenberg <chip@perl.com>
Files: pod/perldelta.pod pod/perldiag.pod toke.c

27 years agoDocument {,un}pack changes
Paul Marquess [Fri, 18 Apr 1997 12:49:39 +0000]
Document {,un}pack changes

Here is a documentation patch for the changes to pack & unpack.

p5p-msgid: 9704181249.AA11733@claudius.bfsec.bt.co.uk

27 years agoMake h2ph generate constant subs
Roderick Schertler [Fri, 18 Apr 1997 18:23:46 +0000]
Make h2ph generate constant subs

>> Shouldn't these:
>>     eval 'sub RUN_LVL {1;}';
>> be:
>>     eval 'sub RUN_LVL() {1;}';
>
> I'd say so.  Who can tell us whether h2ph is easily changeable?

It is, here's a patch.  It could still use changes so that it emits new
style sub invocations so that more constant folding happens, but that's
trickier.  Additional changes in the patch are removal of a "chdir
/usr/include" (which prevented you from converting files in other
directories and wasn't needed for the standard invocation) and addition
of a -d switch to specify output directory (so I could test my changes
more easily).

Somewhere around here I've got the start of regression suite for h2ph,
remind me to dust it off for _50.

Another bug in h2ph it would be nice to fix:

    $ echo "
     #define FOO (BAR || BAZ)
#if FOO
#endif" | h2ph | perl
    Undefined subroutine &main::BAR called at - line 1.

I can't regenerate my vendor's current headers due to this one.

p5p-msgid: pz2088w5ot.fsf@eeyore.ibcinc.com

27 years agoFix NAME in SDBM_File build
Chip Salzenberg [Fri, 18 Apr 1997 00:00:00 +0000]
Fix NAME in SDBM_File build

27 years agoPatch to Getopt::Long
Johan Vromans [Fri, 18 Apr 1997 21:24:00 +0000]
Patch to Getopt::Long

private-msgid: m0wIKCO-00081IC@phoenix.squirrel.nl

27 years agoMinor, optional patch to Makefile.SH
Randy J. Ray [Fri, 18 Apr 1997 01:32:17 +0000]
Minor, optional patch to Makefile.SH

This is the (minor) change I made to add a "purecovperl" target to the
Makefile. Just a copy of the pureperl and quantperl targets:

p5p-msgid: rjray-9703180132.AA00374040@snakepit.ecte.uswc.uswest.com

27 years agoFix tcsh hack in Configure
Chip Salzenberg [Fri, 18 Apr 1997 00:00:00 +0000]
Fix tcsh hack in Configure

27 years agoFix unnecessary re-linking
Chip Salzenberg [Fri, 18 Apr 1997 00:00:00 +0000]
Fix unnecessary re-linking

27 years agoRemove pp_method() inefficiency from last patch
Chip Salzenberg [Fri, 18 Apr 1997 00:00:00 +0000]
Remove pp_method() inefficiency from last patch

27 years agoSECURITY FIX: Buffer overflow in gv_fetchfile()
Chip Salzenberg [Fri, 18 Apr 1997 00:00:00 +0000]
SECURITY FIX: Buffer overflow in gv_fetchfile()

27 years agoExtUtils/Miniperl.pm not built on Win32
Nick Ing-Simmons [Fri, 18 Apr 1997 17:42:32 +0000]
ExtUtils/Miniperl.pm not built on Win32

private-msgid: 199704181742.SAA08407@ni-s.u-net.com

27 years agoIrix update
Scott Henry [Fri, 18 Apr 1997 19:37:24 +0000]
Irix update

>>>>> "A" == Andy Dougherty <doughera@fractal.phys.lafayette.edu> writes:

A> On 17 Apr 1997, Scott Henry wrote:
>> My recommendation is that irix_6_3.sh, irix_6_4.sh should be
>> duplicates of (the recently changed) irix_6_2.sh.  As best as I can
>> determine this far ahead of release, irix_6_5.sh will be the same,
>> too.

A> In which case, they should all just be irix_6.sh.  If 6.0 and 6.1 need
A> separate hints files, then we can have irix_6_0.sh and irix_6_1.sh.

A> If we use the generic irix_6.sh, then
A> 1.  We don't have pointless, error-prone duplication of hint files.
A> 2.  We magically get covered even when 6.5 and 6.6 are released and we're
A> still stuck at 5.004. :-)

EXCELLENT idea!  I'll make up my patch that way.

[editor's note: the patch never hit the mailing list, but this message
seems relevant]

p5p-msgid: yd8afmwnu80.fsf@hoshi.engr.sgi.com
private-msgid: yd8d8rsi0ln.fsf@hoshi.engr.sgi.com

27 years agoFix (and test) spaces in {,un}pack()
Chip Salzenberg [Fri, 18 Apr 1997 00:00:00 +0000]
Fix (and test) spaces in {,un}pack()

27 years ago[differences between cumulative patch application and perl-5.003_97f]
Chip Salzenberg [Thu, 17 Apr 1997 00:00:00 +0000]
[differences between cumulative patch application and perl-5.003_97f]

27 years ago[inseparable changes from match from perl-5.003_97e to perl-5.003_97f]
Perl 5 Porters [Thu, 17 Apr 1997 00:00:00 +0000]
[inseparable changes from match from perl-5.003_97e to perl-5.003_97f]

 CORE LANGUAGE CHANGES

Subject: New operator systell()
From: Chip Salzenberg <chip@perl.com>
Files: doio.c ext/Opcode/Opcode.pm keywords.pl opcode.pl pod/perldelta.pod pod/perldiag.pod pod/perlfunc.pod pp_sys.c t/op/sysio.t toke.c

Subject: Allow constant sub to be optimized when called with parens
From: Chip Salzenberg <chip@perl.com>
Files: toke.c

Subject: Make {,un}pack fail on invalid pack types
From: Chip Salzenberg <chip@perl.com>
Files: pod/perldiag.pod pp.c

 CORE PORTABILITY

Subject: Fix bitwise ops and {,un}pack() on Cray CPUs
From: Chip Salzenberg <chip@perl.com>
Files: pp.c

Subject: VMS update
From: Charles Bailey <bailey@hmivax.humgen.upenn.edu>
Files: lib/Cwd.pm lib/File/Path.pm lib/FindBin.pm vms/perly_c.vms vms/vms.c vms/writemain.pl

Subject: Win32 update (three patches)
From: Gurusamy Sarathy <gsar@engin.umich.edu>
Files: lib/Cwd.pm lib/ExtUtils/MM_Unix.pm lib/ExtUtils/MM_Win32.pm lib/File/Basename.pm win32/Makefile win32/makedef.pl win32/perllib.c win32/win32.c win32/win32iop.h

 DOCUMENTATION

Subject: Document size restrictions for packed integers
From: Jarkko Hietaniemi <Jarkko.Hietaniemi@cc.hut.fi>
Files: pod/perlfunc.pod

 LIBRARY AND EXTENSIONS

Subject: Fix bug in Opcode when (maxo & 15) > 8
From: Chip Salzenberg <chip@perl.com>
Files: ext/Opcode/Makefile.PL ext/Opcode/Opcode.pm ext/Opcode/Opcode.xs

27 years agoDoc updates: INSTALL-1.13, pumpkin.pod-1.9
Andy Dougherty [Thu, 17 Apr 1997 00:00:00 +0000]
Doc updates: INSTALL-1.13, pumpkin.pod-1.9

27 years agoRevise quotewords()
Shishir Gundavaram [Tue, 1 Apr 1997 21:53:42 +0000]
Revise quotewords()

27 years agoCGI.pm broke again
Andreas Koenig [Thu, 17 Apr 1997 11:36:28 +0000]
CGI.pm broke again

Somehow this line was lost during recent patching. It was in 2.32 but
not in 2.33 up.

p5p-msgid: 199704171136.NAA24859@anna.in-berlin.de

27 years agoLinux hints: Allow build w/o suidperl, prefer tcsh to csh
Michael De La Rue [Thu, 17 Apr 1997 00:00:00 +0000]
Linux hints: Allow build w/o suidperl, prefer tcsh to csh

27 years agoFix use of var before init in util.c
Gurusamy Sarathy [Wed, 16 Apr 1997 23:42:41 +0000]
Fix use of var before init in util.c

I thought I sent this a few sub-subversions ago, but it
seems to be missing in 3_97e.

p5p-msgid: 199704162342.TAA20773@aatma.engin.umich.edu

27 years agoFix error messages on method lookup failure
Chip Salzenberg [Thu, 17 Apr 1997 00:00:00 +0000]
Fix error messages on method lookup failure

27 years ago[inseparable changes from patch from perl-5.003_97d to perl-5.003_97e]
Perl 5 Porters [Mon, 14 Apr 1997 12:00:00 +0000]
[inseparable changes from patch from perl-5.003_97d to perl-5.003_97e]

 CORE LANGUAGE CHANGES

Subject: New operator: sysseek()
From: Chip Salzenberg <chip@perl.com>
Files: doio.c ext/Opcode/Makefile.PL ext/Opcode/Opcode.pm global.sym keywords.pl opcode.pl pod/perldelta.pod pod/perlfunc.pod pp_sys.c t/op/sysio.t toke.c

Subject: Allow recursive substitution again
From: Chip Salzenberg <chip@perl.com>
Files: pod/perldelta.pod pod/perldiag.pod pp_hot.c

 CORE PORTABILITY

Subject: Use size_t for socket size parameters of GNU libc
From: Chip Salzenberg <chip@perl.com>
Files: doio.c pp_sys.c

Subject: Win32 update (four patches)
From: Gurusamy Sarathy <gsar@engin.umich.edu>
Files: MANIFEST README.win32 dosish.h ext/SDBM_File/Makefile.PL ext/SDBM_File/sdbm/Makefile.PL ext/SDBM_File/sdbm/sdbm.c ext/SDBM_File/sdbm/sdbm.h lib/ExtUtils/MM_Unix.pm perl.c utils/perlbug.PL utils/perldoc.PL win32/Makefile win32/TEST win32/config.H win32/config.w32 win32/config_h.PL win32/config_sh.PL win32/perllib.c win32/runperl.c win32/win32.c win32/win32io.c win32/win32sck.c

 DOCUMENTATION

Subject: Add CGI to perldelta.pod and improve its description in MANIFEST
From: Chip Salzenberg <chip@perl.com>
Files: MANIFEST pod/perldelta.pod

Subject: Describe probs with majordomo 1.94.1
From: Chip Salzenberg <chip@perl.com>
Files: pod/perldelta.pod

Subject: Fix description of /\G/g
From: Chip Salzenberg <chip@perl.com>
Files: pod/perlop.pod

Subject: Mention '...' operator in precedence table
Date: Sun, 13 Apr 1997 11:24:16 -0600
From: Tom Christiansen <tchrist@perl.com>
Files: pod/perlop.pod

    private-msgid: 199704131724.LAA23120@jhereg.perl.com

 OTHER CORE CHANGES

Subject: New API function: perl_eval_pv()
Date: Mon, 14 Apr 1997 17:13:41 -0400
From: Doug MacEachern <dougm@opengroup.org>
Files: perl.c pod/perlcall.pod pod/perldelta.pod pod/perlembed.pod pod/perlguts.pod proto.h

    private-msgid: 199704142113.RAA06823@postman.osf.org

Subject: Fix C< s//whatever/ >, which reuses old pattern
From: Chip Salzenberg <chip@perl.com>
Files: pp_hot.c regexec.c

27 years agoxsubpp incorrectly handles 'class::newthing()'
John Q. Linux [Sat, 12 Apr 1997 22:01:33 +0000]
xsubpp incorrectly handles 'class::newthing()'

private-msgid: 199704122201.PAA01780@jql.accessone.com

27 years agoClean up format of dlopen() debug info
Hallvard B Furuseth [Fri, 24 Jan 1997 15:29:00 +0000]
Clean up format of dlopen() debug info

27 years agoSys::Syslog: hyphens in hostnames
Jarkko Hietaniemi [Tue, 15 Apr 1997 14:21:53 +0000]
Sys::Syslog: hyphens in hostnames

Jerome Abela writes:
 >
 > In sub connect of Sys::Syslog, someone wrote:
 >         ($host) = $host_uniq =~ /(\w+)/;
 >
 > But, if the name contains a '-', it gets truncated. The RE should
 > rather be someting like [a-zA-Z0-9-]+, shouldn't it ?

p5p-msgid: 199704151421.RAA19693@alpha.hut.fi

27 years agoUser::pwent.pm: g{,e}cos
Tom Christiansen [Sun, 13 Apr 1997 01:35:54 +0000]
User::pwent.pm: g{,e}cos

private-msgid: 199704130135.TAA23274@jhereg.perl.com

27 years ago[dummy merge]
Chip Salzenberg [Mon, 14 Apr 1997 12:00:00 +0000]
[dummy merge]

This merge exists so that the p5p version of the patch and the applied
version are both in the history

27 years agoUpdate File::Path
Chip Salzenberg [Mon, 14 Apr 1997 12:00:00 +0000]
Update File::Path

(this is the same change as commit 8a37cc0e712cfd90de4a433c1b9be391b72daf86, but as applied)

27 years agoUpdate File::Path
Andreas Koenig [Tue, 15 Apr 1997 14:01:07 +0000]
Update File::Path

>>>>> Chip Salzenberg writes:

 > According to Andreas Koenig:
>> I'd prefer to have consistent semantics on all platforms. My patch
>> treats the third parameter ($safe) in a way that matches the current
>> description in the pods better. This means, on all systems a chmod +rw
>> is tried before removing a file or directory unless the $safe
>> parameter is specified.

 > No, that's not useful.  UNIX systems pay *no* attention to the
 > permissions of "x" when unlinking "/y/x"; all that matter are the
 > permissions of "/y".

Ouch. Too-Quick-oh. <Blush>

Sure thing is, I want to chmod 777 always for _directories_ before I
remove them, because they may contain subdirectories that need to be
removed recursively. So I want them both read- and writeable. Stupid
thing was that I applied the same idea to files.

>> In addition File::Path becomes strict clean.

 > Well, that would be useful.  Could you separate that part of the patch
 > from the always-chmod part?

I redid it. This time with a test that fails with current perl and
works with my patch in place.

p5p-msgid: 199704151401.QAA02556@anna.in-berlin.de

27 years agodiagnostics: $/ gotcha
Andreas Koenig [Tue, 15 Apr 1997 18:14:01 +0000]
diagnostics: $/ gotcha

diagnostics does not localize $\ when it comes to printing.

p5p-msgid: 199704151814.UAA03404@anna.in-berlin.de

27 years agoDebugger update
Ilya Zakharevich [Mon, 14 Apr 1997 21:15:27 +0000]
Debugger update

private-msgid: 199704142115.RAA09923@monk.mps.ohio-state.edu

27 years agoRefresh CGI to 2.34
Chip Salzenberg [Tue, 15 Apr 1997 00:00:00 +0000]
Refresh CGI to 2.34