Chip Salzenberg [Thu, 15 May 1997 22:15:00 +0000]
[differences between cumulative patch application and perl5.004]
Perl 5 Porters [Thu, 15 May 1997 22:15:00 +0000]
[inseparable changes from match from perl-5.003_99a to perl5.004]
BUILD PROCESS
Subject: Don't use 'unset' in Configure
From: Chip Salzenberg <chip@perl.com>
Files: Configure
Subject: Protect against having no such command as 'cc'
Date: Mon, 12 May 1997 16:35:34 -0400 (EDT)
From: Hans Mulder <hansm@icgned.nl>
Files: Configure
Msg-ID: 1997May12.163534.2006434@hmivax.humgen.upenn.edu
(applied based on p5p patch as commit
3bf198a5e20d135d4136d3233d58cf49a70772d9)
Subject: minor wording enhancement for Configure
Date: Sat, 10 May 1997 13:38:31 +0300 (EET DST)
From: Jarkko Hietaniemi <Jarkko.Hietaniemi@cc.hut.fi>
Files: Configure
private-msgid:
199705101038.NAA00471@alpha.hut.fi
CORE LANGUAGE CHANGES
Subject: Make C<m//g> reset pos on failure; make C<m//gc> not reset
From: Chip Salzenberg <chip@perl.com>
Files: dump.c op.c op.h pod/perldelta.pod pod/perlfaq6.pod pod/perlop.pod pod/perlre.pod pp_ctl.c pp_hot.c regcomp.c t/op/pat.t toke.c
Subject: SECURITY: Forbid exec() if $ENV{BASH_ENV} is tainted
From: Chip Salzenberg <chip@perl.com>
Files: pod/perldelta.pod pod/perlrun.pod pod/perlsec.pod t/op/taint.t taint.c
Subject: Allow exec() if $ENV{TERM} is tainted but innocuous
From: Chip Salzenberg <chip@perl.com>
Files: pod/perldelta.pod pod/perlrun.pod pod/perlsec.pod t/op/taint.t taint.c
Subject: Allow globbing when tainted under VMS (no external program)
From: Chip Salzenberg <chip@perl.com>
Files: pp_sys.c t/op/taint.t
CORE PORTABILITY
Subject: Win32 update (three patches)
From: Gurusamy Sarathy <gsar@engin.umich.edu>
Files: README.win32 perl.c win32/Makefile win32/config.H win32/config_h.PL win32/config_sh.PL win32/makedef.pl win32/win32.c win32/win32.h win32/win32io.c win32/win32io.h win32/win32iop.h
Subject: Don't require executable bit on perl -S if DOSISH
Date: Fri, 09 May 1997 12:33:18 -0400
From: Danny Sadinoff <sadinoff@olf.com>
Files: perl.c
Msg-ID:
337351CE.79B28DE3@olf.com
(applied based on p5p patch as commit
7596f71a28f72f9e3abd6d3962d29a7752cd9303)
DOCUMENTATION
Subject: Tweaks for perldelta
Date: Sun, 11 May 97 01:46:00 +0200
From: Unknown Contributor <hansm@euronet.nl>
Files: pod/perldelta.pod
Msg-ID:
199705102346.BAA17300@mail.euronet.nl
(applied based on p5p patch as commit
3e10809228cc961223b894e1639b44f8e2b64de0)
Subject: Mention perlfaq.pod and perlmodlib.pod in perldelta.pod
From: Chip Salzenberg <chip@perl.com>
Files: pod/perldelta.pod
Subject: Fix example of use of lexicals with formats
From: Chip Salzenberg <chip@perl.com>
Files: pod/perldelta.pod
OTHER CORE CHANGES
Subject: In C<eval &func>, always call &func in scalar context
From: Chip Salzenberg <chip@perl.com>
Files: op.c
Subject: Fix recursive substitution
From: Chip Salzenberg <chip@perl.com>
Files: cop.h global.sym pp_ctl.c proto.h scope.c
Subject: Fix core dump from get*() functions returning no alias array
From: Chip Salzenberg <chip@perl.com>
Files: pp_sys.c
Subject: Fix typo
Date: Sat, 10 May 1997 17:28:35 -0500
From: Mark K Trettin <mkt@lucent.com>
Files: pp_sys.c
private-msgid:
199705102228.RAA11163@gv18c.ih.lucent.com
Unknown Contributor [Sat, 10 May 1997 23:46:00 +0000]
Tweaks for perldelta
Here are a few more doc fixes.
The last one merely reformats a paragraph, so that a C<...> construct
is no longer spread across a line boudary. The perlpod page suggests
this is allowed, but does not actually say so. However, pod2html can't
handle it and all other instaces have been fixed recently (which would
suggest that it' not allowed).
p5p-msgid:
199705102346.BAA17300@mail.euronet.nl
Hans Mulder [Mon, 12 May 1997 20:35:34 +0000]
Protect against having no such command as 'cc'
ilya@math.ohio-state.edu (Ilya Zakharevich) wrote:
>> At some stage Configure asks "Which memory models are supported?"
>> In order to find the default to this question, Configure compiles
>> and runs a little test program. To compile this test program,
>> Configure uses straigt "cc", not "$cc", presumably because it
>> doesn't know $cc at that point (that's the next question).
>This is very suspicious, since I do not have cc at home (OS/2), but
>no failure of Configure happens. Or did you imply that Configure just
>ignores the failure, or runs it on some platforms only?
It ignores the error. On my unix box, I get a message
Configure: cc: not found
You could suppress that message with this patch:
p5p-msgid: 1997May12.163534.2006434@hmivax.humgen.upenn.edu
Tom Christiansen [Wed, 14 May 1997 14:56:30 +0000]
More detailed IO::Socket documentation
private-msgid:
199705141456.IAA19061@jhereg.perl.com
Danny Sadinoff [Fri, 9 May 1997 16:33:18 +0000]
Don't require executable bit on perl -S if DOSISH
update to previous bugreport regarding the -x filetest and running
perl -S filename.
I misstated the patch necessary for turning off the executability check
for the -S option. It should be:
p5p-msgid:
337351CE.79B28DE3@olf.com
Chip Salzenberg [Thu, 15 May 1997 22:15:00 +0000]
[dummy merge]
This merge exists so that the p5p version of the patch and the applied
version are both in the history
Chip Salzenberg [Thu, 15 May 1997 22:15:00 +0000]
Note that DATA filehandle is unavailable during BEGIN {}
(this is the same change as commit
1d79a27f835368158fa27370bebb63728faf77b2, but as applied)
Neil Bowers [Mon, 12 May 1997 12:27:43 +0000]
Note that DATA filehandle is unavailable during BEGIN {}
Subject: DOC PATCH: restriction on DATA filehandle?
This is a very minor patch to add in a clarification on the use of DATA
within a BEGIN block. Thanks to Chip for confirming this.
p5p-msgid:
199705121227.NAA29718@tardis.cre.canon.co.uk
Chip Salzenberg [Thu, 15 May 1997 22:15:00 +0000]
[dummy merge]
This merge exists so that the p5p version of the patch and the applied
version are both in the history
Chip Salzenberg [Thu, 15 May 1997 22:15:00 +0000]
Explain that destruction order is not defined
(this is the same change as commit
e195ec09596d2fe0e528026e48bc809e69e1165e, but as applied)
Gurusamy Sarathy [Thu, 15 May 1997 06:00:23 +0000]
Explain that destruction order is not defined
Subject: Re: perl debugger problem with object destructors?
On Wed, 14 May 1997 15:12:38 MDT, Tom Christiansen wrote:
>
>Actually, deterministic destruction/finalization is oft-deemed
>a feature. If you could check what perltoot has to say about destructors,
>gsar, and suggest an appropriate patch if necessary, I'd appreciate it.
I'd say something like this:
p5p-msgid:
199705150600.CAA13550@aatma.engin.umich.edu
Jeff Okamoto [Fri, 9 May 1997 23:48:20 +0000]
h2xs and @EXPORT_OK
Currently, h2xs emits the line:
use vars qw($VERSION @ISA @EXPORT);
Should @EXPORT_OK also be added to this list?
p5p-msgid:
199705092348.AA057881699@hpcc123.corp.hp.com
Chuck D. Phillips [Thu, 15 May 1997 17:35:41 +0000]
Fix sleep test: sleep(N) is defined to allow sleeping N-1
> But I found minor problem here. In some situation, BSD/OS sleep does
> not sleep two seconds (at least in my configuration) and it says
> op/sleep.t failure. If I ran the test suite again, it passed and show
> me 100% OK result.
I just read over t/op/sleep.t. Frankly, I'm suprised I haven't seen
it fail on HPUX also. On HPUX, SIGALRM is delivered only _on_ the
second. Thus, sleep(1) waits until the next even second, sleep(2)
waits until the next even second after that, etc.
The side effect of this is that sleep(1) causes a delay of 0-1 second,
sleep(2) causes a delay of 1-2 seconds, etc. This *should* cause
intermittant failure of the sleep test as currently written. I don't
have access to a BSD system, but something similar could be happening
there.
You could argue that this is a bug in the OS sleep(). However, it
means that a loop like...
while (foo()) {
bar();
sleep();
}
...will cycle once-per-second (as long as bar() executes in < 1
second) instead of cycling once-per-(second + bar()-delay). This
could be construed as a feature.
p5p-msgid:
199705151735.KAA01143@palrel1.hp.com
Tom Phoenix [Sun, 11 May 1997 02:57:30 +0000]
Make rand.t vanishingly unlikely to give false failure
On Wed, 7 May 1997, Larry Schwimmer wrote:
> Subject: FYI: perl5.00399/t/op/rand.t test 7
> I know this sounds perverse, but it did happen to fail test 7
> of op/rand.t the very first time I ran make test on a Solaris 2.5.1
> machine, and the test does say to mail the developers if that
> happened. (-:
And thank you for doing so. Ya done good. :-)
When I wrote that test, I had thought (erroneously, as it turns out) that
that test would never fail, or virtually never. Actually, on Solaris, it
can report a false positive about one time in two-to-the-15th tests.
That test attempts to ensure that srand's default seed isn't the same
twice in a row, which it shouldn't be. But was your test result falsely
positive, or was it a bug for real? We have no way to know.
> It worked fine the next 100 times I ran it and on the other seven
> builds,
Okay, if you had success the next 100 times, it's _probably_ a fluke.
There's no way to know for sure, though, short of finding a bug in the
srand code. :-(
I'm supplying a patch which makes the test more reliable without reducing
the sensitivity to bugs. This should effectively eliminate this problem,
except for unavoidable coincidences.
> but it might be nice to run the test file multiple times to
> reduce the likelihood of a false failure while still catching errant
> builds.
Actually, that wouldn't do the trick. If we ran it five times, and one of
those attempts gets the same srand seed twice, that's _still_
unacceptable. The program has to notify a human, since recompiling perl,
checking the source, and asking for advice are things that humans still do
better than machines. But this patch will make the machine a little better
at knowing when to cry "Wolf!" :-)
Thanks!
p5p-msgid: Pine.GSO.3.96.
970510190846.23340K-100000@kelly.teleport.com
Chip Salzenberg [Tue, 4 Feb 1997 16:37:26 +0000]
Fix core dump on IO::Seekable::setpos($fh, undef)
Chip Salzenberg [Wed, 9 Apr 1997 19:48:55 +0000]
In IO::File::open, prepend './' less often (for Win32 et al)
Lincoln Stein [Mon, 28 Apr 1997 17:58:26 +0000]
Refresh CGI.pm to 2.36
Chip Salzenberg [Thu, 15 May 1997 22:15:00 +0000]
[dummy merge]
This merge exists so that the p5p version of the patch and the applied
version are both in the history
Chip Salzenberg [Thu, 15 May 1997 22:15:00 +0000]
Make read with <> from a TTY notice EOF
(this is the same change as commit
451e086022facb7867a6764d6ffca2b71e86ff73, but as applied)
Jonathan I. Kamens [Mon, 12 May 1997 11:47:13 +0000]
Make read with <> from a TTY notice EOF
The patch I just sent in wasn't quite correct. I should have tested
it more before sending it :-). Here's a correct patch.
p5p-msgid:
199705121147.HAA03845@jik.saturn.net
Tim Bunce [Tue, 18 Oct 1994 16:46:27 +0000]
Fix recursive substitution [test]
Jonathan Hudson [Thu, 1 May 1997 02:20:23 +0000]
VMS update, including socket support (four patches)
Signed-off-by: Peter Prymmer <pvhp@forte.com>
Signed-off-by: Dan Sugalski <sugalsd@lbcc.cc.or.us>
Jeff Okamoto [Tue, 13 May 1997 22:28:04 +0000]
Fix default HP-UX installation path
private-msgid:
199705132228.AA227042483@hpcc123.corp.hp.com
Chip Salzenberg [Thu, 8 May 1997 15:03:58 +0000]
Make Irix hints adapt when n32 libm.so is missing
Chip Salzenberg [Wed, 7 May 1997 12:00:00 +0000]
[differences between cumulative patch application and perl-5.003_99a]
Perl 5 Porters [Wed, 7 May 1997 12:00:00 +0000]
[inseparable changes from match from perl-5.003_99 to perl-5.003_99a]
BUILD PROCESS
Subject: AFS patches
From: Chip Salzenberg <chip@perl.com>
Files: Configure installperl
CORE LANGUAGE CHANGES
Subject: SECURITY: Forbid glob() when tainting (-T or setuid)
From: Chip Salzenberg <chip@perl.com>
Files: pod/perlrun.pod pod/perlsec.pod pp_sys.c
Subject: SECURITY: Forbid exec() if $ENV{TERM} or $ENV{ENV} is tainted
From: Chip Salzenberg <chip@perl.com>
Files: pod/perlrun.pod pod/perlsec.pod t/op/taint.t taint.c
CORE PORTABILITY
Subject: (NeXT|Open)Step update
Date: Wed, 7 May 97 17:47:02 -0500
From: Gerd Knops <gerti@BITart.com>
Files: Configure MANIFEST config_h.SH hints/next_3.sh hints/next_4.sh
private-msgid:
9705072247.AA18882@BITart.com
Subject: Win32 update (consolidated patch plus three followups)
From: Gurusamy Sarathy <gsar@engin.umich.edu>
Files: EXTERN.h README.win32 lib/Sys/Hostname.pm pod/perldelta.pod win32/config.H win32/config.w32 win32/config_sh.PL win32/perllib.c win32/win32.c win32/win32.h win32/include/sys/socket.h
DOCUMENTATION
Subject: Updates to perldelta
From: Chip Salzenberg <chip@perl.com>
Files: pod/perldelta.pod
Subject: Document 'Possible attempt to separate words with commas'
Date: 06 May 1997 23:27:55 +0200
From: Gisle Aas <gisle@aas.no>
Files: pod/perlop.pod
Msg-ID: hyb9snvdw.fsf@bergen.sn.no
(applied based on p5p patch as commit
18270fd3b8aafde2f9ea21ea13adde95ef24b149)
Subject: Document that C<m?x?> is just like C<?x?>
From: Chip Salzenberg <chip@perl.com>
Files: pod/perlop.pod
OTHER CORE CHANGES
Subject: Fix for redefined sort subs nastiness
Date: Thu, 08 May 1997 20:04:18 -0400
From: Gurusamy Sarathy <gsar@engin.umich.edu>
Files: op.c pod/perldelta.pod pod/perldiag.pod sv.c t/op/sort.t
Msg-ID:
199705090004.UAA15032@aatma.engin.umich.edu
(applied based on p5p patch as commit
e9e069932a0db06904b29e2b09a435afd40ed35c)
Gisle Aas [Tue, 6 May 1997 21:27:55 +0000]
Document 'Possible attempt to separate words with commas'
Perhaps we should just document it:
p5p-msgid: hyb9snvdw.fsf@bergen.sn.no
Ilya Zakharevich [Thu, 8 May 1997 00:46:45 +0000]
OS/2 doc patch for _99
It adds some info people judged hard-to-find...
Enjoy,
p5p-msgid:
199705080046.UAA23466@monk.mps.ohio-state.edu
Gurusamy Sarathy [Fri, 9 May 1997 00:04:18 +0000]
Fix for redefined sort subs nastiness
>sub sortfunc { &once }
>sub once {
> eval q{
> print "Eval from ", join(':', caller(0)), "\n";
> sub rest {
> print "REST: $a <=> $b\n";
> $a <=> $b
> }
> };
> *sortfunc = *rest;
> &sortfunc;
>}
>@x = sort sortfunc 10, 5, 2.5, 1.25;
That misbehaves due to the redefinition of the sort sub
while the sort is active. That's a big no-no, because
the sortcop was pointing to the CvSTART of the original,
and it will now point to freed memory (if the sub really
got undefined).
Here's a tested patch that does nothing but die under
that circumstance.
p5p-msgid:
199705090004.UAA15032@aatma.engin.umich.edu
Chip Salzenberg [Wed, 7 May 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
Chip Salzenberg [Wed, 7 May 1997 12:00:00 +0000]
Mention the Regular Expressions book
(this is the same change as commit
7dd178533182fc85a734d7633d0a938b5d1c853a, but as applied)
Stephen Potter [Wed, 7 May 1997 17:37:37 +0000]
Mention the Regular Expressions book
| Could somebody please add Jfrield's Mastering Regeular Expressions to
| perlbook.pod, and also to a SEE ALSO in perlre.pod?
p5p-msgid:
199705071548.KAA05924@psa.pencom.com
private-msgid:
199705071737.MAA18799@psa.pencom.com
Chip Salzenberg [Tue, 29 Apr 1997 04:33:47 +0000]
Refresh description of sprintf()
Ilya Zakharevich [Thu, 8 May 1997 01:07:14 +0000]
Debugger docs patch
New bit 16 of the `frame' option and the new `ornaments' option is
documented.
Enjoy,
p5p-msgid:
199705080107.VAA24317@monk.mps.ohio-state.edu
John D Groenveld [Tue, 6 May 1997 17:41:12 +0000]
perlfaq9, hostname
Simple patch...
p5p-msgid:
199705061741.NAA22777@cse.psu.edu
Andy Dougherty [Thu, 8 May 1997 21:14:54 +0000]
More explicit Solaris instructions
Here are two patches to try to make the gcc -B/usr/ccs/bin/ trick
more apparent. They were suggested by a user who didn't find the current
hints and INSTALL file sufficiently specific.
As a side effect, you get whatever else I had cooked up to go into the
next INSTALL update.
p5p-msgid: Pine.SOL.3.95q.
970508171206.438A-100000@fractal.lafayette.edu
Ilya Zakharevich [Thu, 8 May 1997 00:32:46 +0000]
Newer CPerl mode
Some major flaws became appparent in older CPerls, and newer ones
prove themselves reasonably good, so here it is (for inclusion into
5.004):
Description of changes is one page down,
Enjoy,
p5p-msgid:
199705080032.UAA22532@monk.mps.ohio-state.edu
Ralf S. Engelschall [Mon, 5 May 1997 14:47:03 +0000]
Trivial bugfix for pod of xsubpp
private-msgid:
199705051447.QAA09995@en1.engelschall.com
Unknown Contributor [Mon, 5 May 1997 22:28:06 +0000]
Three bugs in pod2html
I had three problems with the pod2html distributed with perl 5.00399:
1. The HTML code generated begins with a series of hyperlinks to all
sections of the file. Pod directives occurring in the section
titles (=head directives) were not translated.
2. C<foo()> sets translates to <CODE>foo</CODE> -- the () disappear.
3. Perlpod promises that example code gets typeset with tab stops
every 8 columns. HTML makes no such promise, hence pod2html must
expand tabs.
The first three hunks of the patch below solve #1, hunk 4 fixes a
call to &process_text with the wrong number of args, hunk 5 solves #3
and hunk 6 solves #2.x
p5p-msgid:
199705052228.AAA25351@mail.euronet.nl
Russ Allbery [Thu, 8 May 1997 18:36:12 +0000]
Another blank line patch to Pod::Text
If an =over/=back block ends just before a =head1 tag, there is an
extraneous blank line before the heading. This patch fixes that and
doesn't appear to have any negative effects. The blank line after the
last =item should be provided by the =item's paragraph break.
p5p-msgid: m3afm5g6ar.fsf@windlord.Stanford.EDU
Chip Salzenberg [Wed, 30 Apr 1997 17:05:21 +0000]
Fix wantarray() in sort subs [fixes metaconfig]
Gary Clark [Mon, 5 May 1997 17:49:03 +0000]
Win32 boot_DynaLoader problem in 99
In makedef.pl, boot_DynaLoader is defined in $skip_sym. It should be
included in the __DATA__ section to force it into the dll exports.
p5p-msgid: 1997May05.105000.1708.84476@mail.jeld-wen.com
Chuck D. Phillips [Mon, 5 May 1997 05:48:39 +0000]
HPUX: patch for ext/DynaLoader/dl_hpux.xs
By default, doing a "make test" on a dynamically loaded module with
unresolved symbols will result with a useless error message. Adding
the BIND_VERBOSE flag will add the names of undefined symbols to the
error message already printed to stderr. This does *not* affect
default operation; only when PERL_DL_NONLAZY is set does it make a
p5p-msgid:
199705050548.WAA21260@palrel1.hp.com
Scott Henry [Tue, 6 May 1997 18:09:56 +0000]
Irix hint update
Hopefully, this is the final set of patches for the IRIX hints file
hints/irix_6.sh. This is rather a multi-person effort, Jarkko and
John Stoffel wrote the major changes (and changed the default to
-n32), I made the last tweaks so that -o32 still works without
errors on the various compiler releases. I've tested on IRIX-6.2,
IRIX64-6.2, IRIX64-6.4 and IRIX-6.5alpha.
p5p-msgid: yd83es0fp57.fsf@hoshi.engr.sgi.com
Giles Lean [Mon, 5 May 1997 13:46:37 +0000]
NetBSD hint update
There is a missing character in hints/netbsd.sh.
Line 51
case "$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
Chip Salzenberg [Wed, 30 Apr 1997 12:00:00 +0000]
[differences between cumulative patch application and 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)
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
Chip Salzenberg [Wed, 30 Apr 1997 13:57:56 +0000]
Fix typo in perldelta
Gisle Aas [Thu, 1 May 1997 10:42:46 +0000]
Fix description of av_undef() in perlguts
private-msgid:
199705011042.MAA09897@bergen.sn.no
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
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)
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
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: v
03102801af8e160d3879@[194.51.248.68]
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
Chip Salzenberg [Thu, 6 Mar 1997 20:53:34 +0000]
Add new globals to perl.exp
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
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
Chip Salzenberg [Fri, 25 Apr 1997 01:55:09 +0000]
Handle tainted values in lists returned from subs, evals
Chip Salzenberg [Thu, 24 Apr 1997 14:34:21 +0000]
Forget prototype of subroutine after C<undef &subr>
Chip Salzenberg [Wed, 30 Apr 1997 05:27:58 +0000]
Fix NUL-termination bug in delimcpy()
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
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>
Chip Salzenberg [Sat, 26 Apr 1997 00:39:00 +0000]
Describe Math::Trig in perlmodlib
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
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
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)
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
Chip Salzenberg [Fri, 18 Apr 1997 14:42:33 +0000]
Fix C< hex('80') * 0x1000000 >
Chip Salzenberg [Sat, 26 Apr 1997 12:00:00 +0000]
[differences between cumulative patch application and 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
Chip Salzenberg [Thu, 24 Apr 1997 16:55:38 +0000]
Doc fix for close of pipe handle
Chip Salzenberg [Thu, 24 Apr 1997 16:16:44 +0000]
Describe __PACKAGE__ in perldelta
Tom Christiansen [Sat, 26 Apr 1997 00:50:09 +0000]
Split perlmod
private-msgid:
199704260050.RAA02468@toy.perl.com
Jesse Glick [Thu, 2 Jan 1997 23:18:01 +0000]
SelfLoader: fix prototype pattern, rename intrusive lexical
Signed-off-by: Jesse Glick <jesse@ginger>
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
Lincoln Stein [Thu, 17 Apr 1997 18:22:51 +0000]
Refresh CGI.pm to 2.35
Chip Salzenberg [Sat, 1 Mar 1997 06:14:33 +0000]
Fix possible buffer overflow under VMS
Chip Salzenberg [Wed, 23 Apr 1997 18:29:12 +0000]
Fix typo in NeXT dynaloader
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>
Chip Salzenberg [Wed, 23 Apr 1997 14:27:36 +0000]
Fix lingering '%S' in XS_VERSION_BOOTCHECK
Gurusamy Sarathy [Thu, 24 Apr 1997 17:13:30 +0000]
Win32 update (three patches)
Signed-off-by: Gurusamy Sarathy <gsar@engin.umich.edu>
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
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
Chip Salzenberg [Thu, 24 Apr 1997 12:00:00 +0000]
[differences between cumulative patch application and 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
Chip Salzenberg [Wed, 23 Apr 1997 23:48:51 +0000]
Document new {,s}printf() behavior
Chip Salzenberg [Thu, 10 Apr 1997 19:49:15 +0000]
Fix buffer overflow in a2p
Gurusamy Sarathy [Thu, 24 Apr 1997 22:21:44 +0000]
Math::{Complex,Trig} update
private-msgid:
199704242221.BAA30363@alpha.hut.fi
Chip Salzenberg [Wed, 16 Apr 1997 17:47:04 +0000]
Handle symlinks, high permission bits in File::Path
Chip Salzenberg [Thu, 19 Dec 1996 22:11:20 +0000]
Don't core dump if my_inet_aton() get a NULL
Chip Salzenberg [Fri, 18 Apr 1997 16:37:00 +0000]
Always NUL-terminate opsets
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
Chip Salzenberg [Sun, 13 Apr 1997 09:06:24 +0000]
Undo private patch
Doug MacEachern [Thu, 24 Apr 1997 17:06:21 +0000]
Fix scalar leak in pp_prtf()
private-msgid:
199704241706.NAA19140@postman.osf.org
Chip Salzenberg [Tue, 22 Apr 1997 12:00:00 +0000]
[differences between cumulative patch application and 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
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
Jarkko Hietaniemi [Sun, 20 Apr 1997 19:38:13 +0000]
Two doublewords less
private-msgid:
199704201938.WAA07722@alpha.hut.fi