M.J.T. Guy [Fri, 21 Mar 1997 09:58:17 +0000 (09:58 +0000)]
Re: NUL in die and other messages
There has been no response to this on p5p, Chip hasn't included it
in his "More seeds of _94" and I can't find it on the archive.
So I suppose it got lost.
I note that the patch needs to be merged with a more recently posted
one handling references.
p5p-msgid: E0w815V-0005xs-00@ursa.cus.cam.ac.uk
Roderick Schertler [Thu, 13 Mar 1997 23:40:51 +0000 (18:40 -0500)]
fix for Exporter's $SIG{__WARN__} handler
Exporter's $SIG{__WARN__} handler is too zealous. It fails here:
$ cat t
require Carp;
package Foo;
@ISA = qw(Exporter);
@EXPORT_OK = @EXPORT_FAIL = qw(foo);
sub export_fail {
my $self = shift;
Carp::carp("carp from export_fail");
return ();
}
package main;
import Foo 'foo';
$ perl5.00393 t
carp from export_fail at t line 16
at t line 16
This is because the carp() inside export_fail triggers Exporter's
__WARN__ handler which calls carp() a second time which adds the second
" at t line 16".
I was surprised to learn that carp() adds its string to an error message
which ends with a newline, I thought it followed warn()'s lead on that.
I checked 5.003 and it was the same there, too, though.
Also, shouldn't Exporter's __WARN__ handler increment $Carp::CarpLevel
rather than setting it to 1? I didn't include that in the patch because
I'm not sure, but it seems the right thing to do to me.
p5p-msgid: 2282.
858296451@eeyore.ibcinc.com
Ilya Zakharevich [Fri, 21 Mar 1997 04:13:31 +0000 (23:13 -0500)]
Problems with SKIP in makemaker
The example in perlxstut (which I created with a lot of pain) works by
coincidence only:
It has
WriteMakefile(
NAME => 'Mytest2::mylib',
SKIP => [qw(all static static_lib dynamic dynamic_lib)],
clean => {'FILES' => 'libmylib$(LIB_EXT)'},
);
sub MY::top_targets {
'
all :: static
static :: libmylib$(LIB_EXT)
libmylib$(LIB_EXT): $(O_FILES)
$(AR) cr libmylib$(LIB_EXT) $(O_FILES)
$(RANLIB) libmylib$(LIB_EXT)
';
}
and work only because $self->MM::top_targets is not called. The reason is
that 'all' chunk is written anyway not depending on the value of SKIPHASH.
In my eText package I do $self->MM::top_targets inside
MY::top_targets, and this bombs.
Patch follows (only for 'all', I did not check any other target):
(The alternative is to patch perlxstut, but this may break some other
programs, not only mine...)
Enjoy,
p5p-msgid:
199703210413.XAA21601@monk.mps.ohio-state.edu
Paul Marquess [Wed, 12 Mar 1997 15:51:14 +0000 (15:51 +0000)]
Refresh DB_File to 1.12
> Title: "Regularize headings in DB_File documentation"
> From: Chip Salzenberg
> Files: ext/DB_File/DB_File.pm
Could you merge this patch please. It documents the fact that DB_File
will not yet work with Berkeley DB version 2.
p5p-msgid:
9703121551.AA07435@claudius.bfsec.bt.co.uk
Ilya Zakharevich [Fri, 21 Mar 1997 00:34:30 +0000 (19:34 -0500)]
New module: autouse.pm
Subject: Newer autouse.pm
I included support for prototypes (and some warnings in POD) into the
package, but cannot fully check them, since they trigger a lot of bugs
in support of prototypes in Perl. I hope that when the bugs are
corrected, this will work better.
Enjoy,
p5p-msgid:
199703210034.TAA13469@monk.mps.ohio-state.edu
Spider Boardman [Sun, 16 Mar 1997 07:21:35 +0000 (02:21 -0500)]
3_93 doesn't install pods
My patch (which is in _93) broke installing .pod files. At least,
I think it's my fault. Anyway, here's the fix.
p5p-msgid:
199703160721.
CAA08339@Orb.Nashua.NH.US
Doug MacEachern [Wed, 19 Mar 1997 23:45:53 +0000 (18:45 -0500)]
Warn about missing -DMULTIPLICITY if likely a problem
Subject: Re: multiple embedded interpreter example in perlembed.pod fails
Ken Fox <kfox@ford.com> wrote:
> I wrote:
>
> > Multiple embedded interpreters don't work. Here's a transcript:
> ...
> > ccflags =''
>
> Ugh. Stupid user error. I ran Configure with -DMULTIPLICITY instead
> of -Dccflags='-DMULTIPLICITY'. Sorry for the confusion.
you can also 'make LIBPERL=libperlm.a', libperld, libperldm, etc., or
edit cflags by hand after Configure.
> Could the
> runtime error message be improved to let people know when they've made
> this mistake?
good idea, something like the patch below? I'm not sure of other
cases where e_fp would be bogus.
p5p-msgid:
199703192345.SAA15070@postman.osf.org
Roderick Schertler [Wed, 19 Mar 1997 17:42:50 +0000 (12:42 -0500)]
printf format corrections for -DDEBUGGING
Here are some more printf format fixes. These are all in -DDEBUGGING
code.
p5p-msgid: 26592.
858793370@eeyore.ibcinc.com
Charles Bailey [Wed, 12 Mar 1997 03:00:55 +0000 (22:00 -0500)]
VMS update
Subject: Re: _93 is a keeper
In article <
199703112146.QAA12455@nielsenmedia.com>, Chip Salzenberg <salzench@nielsenmedia.com> writes:
> Based on feedback so far, I think we can release _93 as the first beta.
> ("And there was much rejoicing." "yeah") However, the announcement will
> refer to the Win32 support as "not yet ready"; that will wait for _94.
Here's my $0.02, just in from the field :-):
5.003_93 unchanged:
VMS hmivax.humgen.upenn.edu 6.0 MicroVAX 3400 Series
All tests successful.
u=1649.55 s=0 cu=0 cs=0 files=120 tests=3337
5.003_93 with minor patches:
VMS ssrl.slac.stanford.edu 6.2 AlphaServer 2100 4/275:
All tests successful.
u=352.57 s=0 cu=0 cs=0 files=118 tests=3328
I've appended a few patches below, some necessary for the AXP build,
and others just cleaning up the last remnants of old VMS subprocess
status code and some other nits.
If you're just going to rename _93 on CPAN, don't worry about this stuff.
OTOH, if you're planning a new kit, it'd be nice if you could sneak in the
vms/descrip.mms patch (necessary for build on some systems) and the
utils/perlbug.PL patch (necessary for perlbug to work on all systems).
Thanks again for all the work!
Off to the docs . . .
p5p-msgid: 1997Mar11.220056.1873182@hmivax.humgen.upenn.edu
David Dyck [Mon, 10 Mar 1997 23:55:44 +0000 (15:55 -0800)]
allow C<substr 'hello', -10>
Subject: patch substr to fetch rightmost n characters
This is not a 'porting' issue, but I've
cc'd the list since many of the experts are there.
to extract the the first 3 (or less characters in a
string one can say
$ perl -le '$x="abcdefg"; print substr($x , 0, 3)'
abc
$ perl -le '$x="ab"; print substr($x , 0, 3)'
ab
but to print the last 3 characters (or less)
the analogy doesn't work.
$ perl -le '$x="abcdefg"; print substr($x , -3)'
efg
$ perl -le '$x="ab"; print substr($x , -3)'
I was trying to let a string grow, but keep
it shorter than some maximum length.
Of course a work around is to check the length of
the string, eg
$ perl -le '$x="ab"; print length($x)<3?$x:substr($x , 0, 3)'
ab
but this doesn't seem reasonable.
Can anyone think of a reason against changing substr
to allow this feature. I had expected it to work this way.
(as it does with the following patch)
$ ./perl -wle '$x="ab"; print substr($x , -3)'
ab
p5p-msgid: 97Mar10.155517pst.35716-2@gateway.fluke.com
Tom Phoenix [Mon, 10 Mar 1997 22:33:20 +0000 (14:33 -0800)]
Re: Inline PI function
private-msgid: Pine.GSO.3.95q.
970310143125.22489V-100000@kelly.teleport.com
Larry W. Virden [Tue, 11 Mar 1997 15:53:49 +0000 (10:53 -0500)]
Update site list
private-msgid:
9703111053.AA20051@cas.org
Signed-off-by: x2487 <lvirden@cas.org>
Chip Salzenberg [Sat, 8 Mar 1997 23:57:19 +0000 (11:57 +1200)]
[dummy merge]
This merge exists so that the p5p version of the patch and the applied
version are both in the history
Chip Salzenberg [Sat, 8 Mar 1997 23:57:19 +0000 (11:57 +1200)]
INSTALL: How to enable debugging
(this is the same change as commit
bbf3737a130f2476a9fbd9397c661175a0027d33, but as applied)
Andy Dougherty [Fri, 21 Mar 1997 16:25:32 +0000 (11:25 -0500)]
INSTALL: How to enable debugging
On Fri, 21 Mar 1997, Tom Christiansen wrote:
> BTW, why isn't there a makefile option to do this? I looked in
> INSTALL, but really saw no obvious way to 'make me a perl with cc -g
> style debugging'. There should be, or if there is, there should be
> clearer directions.
It's all spelled out for you when you run Configure interactively, of
course, but you already knew that :-)
p5p-msgid: Pine.SOL.3.95q.
970321112326.1414A-100000@fractal.lafayette.edu
private-msgid: Pine.SOL.3.95q.
970321112326.1414A-100000@fractal.lafayette.e
Roderick Schertler [Wed, 12 Mar 1997 00:09:31 +0000 (19:09 -0500)]
Re: bug in pod2man (5.00326): section=3 for .pm modules
On Tue, 11 Mar 1997 21:25:10 +0100 (MET), Slaven Rezic <eserte@cs.tu-berlin.de> said:
>
> pod2man does not recognize the proper section for .pm modules. You
> have to delete the "$section = $opt_section || $DEF_SECTION;" in line
> 378 in .../pod/pod2man.PL, since it overrides the proper setting in
> line 362.
Good catch. Here's that fix plus code to use the man3 extension
specified to Configure (I use 3pm myself).
p5p-msgid: pzn2sat1hg.fsf@eeyore.ibcinc.com
M.J.T. Guy [Tue, 18 Mar 1997 20:50:16 +0000 (20:50 +0000)]
Fix typos in IO::Socket documentation
At line 156, IO::Socket::configure reads
croak "IO::Socket: Cannot configure socket in domain '$domain'"
unless ref($fh) eq "IO::Socket";
Doesn't this mean that IO::Socket fails the 'empty subclass' test?
Should the condition be replaced with 'isa' or something?
And the following patch mends a few minor typos in the pod.
p5p-msgid: E0w75po-0003yh-00@taurus.cus.cam.ac.uk
Chip Salzenberg [Thu, 2 Jan 1997 23:18:00 +0000 (11:18 +1200)]
Add entry for prototype() in Pod::Functions
Chip Salzenberg [Sat, 8 Mar 1997 23:57:19 +0000 (11:57 +1200)]
[dummy merge]
This merge exists so that the p5p version of the patch and the applied
version are both in the history
Chip Salzenberg [Sat, 8 Mar 1997 23:57:19 +0000 (11:57 +1200)]
Fix typo in -l*perl* pattern
(this is the same change as commit
ddab96b77391e288333a995edd0cbe8bc02d2481, but as applied)
Doug MacEachern [Tue, 11 Mar 1997 03:58:38 +0000 (22:58 -0500)]
Fix typo in -l*perl* pattern
one tiny Embed.pm patch for libperlm and friends
p5p-msgid:
199703110400.XAA20694@sinistar.idle.com
private-msgid:
199703110414.XAA12884@berlin.atlantic.net
Chip Salzenberg [Tue, 4 Mar 1997 22:15:47 +0000 (10:15 +1200)]
In debugger, don't reference %{$f{$g}} if $f{$g} doesn't exist
Chip Salzenberg [Sat, 8 Mar 1997 23:57:19 +0000 (11:57 +1200)]
[dummy merge]
This merge exists so that the p5p version of the patch and the applied
version are both in the history
Chip Salzenberg [Sat, 8 Mar 1997 23:57:19 +0000 (11:57 +1200)]
New subroutine Symbol::qualify_to_ref()
(this is the same change as commit
578b087ab63db59886b8f9a1e8bebc8524523fa5, but as applied)
Roderick Schertler [Wed, 12 Mar 1997 00:39:36 +0000 (19:39 -0500)]
New subroutine Symbol::qualify_to_ref()
Subject: Re: select under use strict
On Tue, 11 Mar 1997 16:36:25 -0500 (EST), Trevor Blackwell <tlb@viaweb.com> said:
>
> use strict;
> syswrite(select,"",0) || warn("$!");
[...]
> yields:
>
> Can't use string ("main::STDOUT") as a symbol ref [...]
It seems a reasonable candidate for a new function in Symbol.pm, a
function just like qualify() which returns a globref. Here's an
implementation. I'll send this in again after 5.004. In the meanwhile
you could just put something like it in your own code.
Usage would be like
syswrite ref_qualify(select), "foo\n", 4 or die $!;
p5p-msgid: pzlo7ut03b.fsf@eeyore.ibcinc.com
Jarkko Hietaniemi [Thu, 2 Jan 1997 23:17:59 +0000 (11:17 +1200)]
Math::Complex update
Chip Salzenberg [Wed, 15 Jan 1997 21:36:47 +0000 (09:36 +1200)]
Don't use $(LIBS) when creating shared libperl
Tom Phoenix [Sun, 16 Mar 1997 21:40:35 +0000 (13:40 -0800)]
More MachTen hints
private-msgid: Pine.GSO.3.95q.
970316133852.27997A-100000@kelly.teleport.com
Chip Salzenberg [Sat, 8 Mar 1997 23:57:19 +0000 (11:57 +1200)]
[dummy merge]
This merge exists so that the p5p version of the patch and the applied
version are both in the history
Chip Salzenberg [Sat, 8 Mar 1997 23:57:19 +0000 (11:57 +1200)]
BSD/OS 3.0 hints
(this is the same change as commit
077b59947527f35fc97006c59d4f42e5fbe0b977, but as applied)
Christopher Davis [Fri, 14 Mar 1997 21:20:46 +0000 (16:20 -0500)]
BSD/OS 3.0 hints
These are based on the 2.1 hints and the config.sh that BSDI used to build
their "perl5" executable. It compiles and passes "make test" but has not
been tested extensively.
p5p-msgid: w47mjakw5t.fsf@loiosh.kei.com
Chip Salzenberg [Thu, 6 Mar 1997 15:46:38 +0000 (03:46 +1200)]
Ignore whitespace before +*? in //x
Chip Salzenberg [Thu, 6 Mar 1997 16:01:12 +0000 (04:01 +1200)]
[differences between cumulative patch application and perl-5.003_93]
Perl 5 Porters [Thu, 6 Mar 1997 16:01:12 +0000 (04:01 +1200)]
[inseparable changes from match from perl-5.003_92 to perl-5.003_93]
BUILD PROCESS
Subject: Fix for Unisys UNIX and libperl.so
Date: Thu, 6 Mar 97 16:28 GMT0
From: Alan Burlison <aburlison@cix.compulink.co.uk>
Files: Configure
Msg-ID: memo.147328@cix.compulink.co.uk
(applied based on p5p patch as commit
2f525a329506abe067c4ec9d4c7ae13d8b223081)
Subject: Allow './Configure -Uoptimize'
Date: Thu, 06 Mar 1997 11:15:47 -0500 (EST)
From: Andy Dougherty <doughera@fractal.phys.lafayette.edu>
Files: Configure
Msg-ID: Pine.SOL.3.95q.
970306110532.11070A-100000@fractal.lafayette.
(applied based on p5p patch as commit
59ae47780c168857b216412f8354d10ae6dd0b61)
Subject: Use 'test -f', not 'test -x'
Date: Fri, 7 Mar 1997 19:53:00 -0500
From: Spider Boardman <spider@web.zk3.dec.com>
Files: Configure
Msg-ID:
199703080053.TAA13943@web.zk3.dec.com
(applied based on p5p patch as commit
64b4562546c4e74f87c21f65a76ec05e96a1b74f)
CORE LANGUAGE CHANGES
Subject: Don't autovivify array and hash elements in sub parameters
Date: Thu, 06 Mar 1997 14:12:09 -0500
From: Gurusamy Sarathy <gsar@engin.umich.edu>
Files: op.c pod/perldelta.pod pod/perlsub.pod pod/perltrap.pod
Msg-ID:
199703061912.OAA20606@aatma.engin.umich.edu
(applied based on p5p patch as commit
0ffc6623ceb5acc7b954e8cdbaeb8ba319474d7b)
Subject: Support READ and GETC for tied handles
Date: Sat, 08 Mar 1997 19:19:38 -0500
From: Doug MacEachern <dougm@opengroup.org>
Files: pod/perldelta.pod pod/perltie.pod pp_sys.c t/op/misc.t
Msg-ID:
199703090019.TAA32591@postman.osf.org
(applied based on p5p patch as commit
b06b64f805517c26cbd7c4d2f74efd5f36b4692c)
Subject: Warn on C<@x =~ /a/> and C<%x =~ s/a/b/>
From: Chip Salzenberg <chip@perl.com>
Files: op.c pod/perldiag.pod
CORE PORTABILITY
Subject: VMS update
Date: Fri, 07 Mar 1997 22:49:46 -0500 (EST)
From: Charles Bailey <bailey@HMIVAX.HUMGEN.UPENN.EDU>
Files: lib/ExtUtils/MM_VMS.pm vms/descrip.mms vms/gen_shrfls.pl vms/sockadapt.h
private-msgid: 01IG8KN5R28M00661G@hmivax.humgen.upenn.edu
DOCUMENTATION
Subject: Consolidated INSTALL updates since _92
Date: Sat, 08 Mar 1997 13:21:22 -0500 (EST)
From: Andy Dougherty <doughera@fractal.phys.lafayette.edu>
Files:
Msg-ID: Pine.SOL.3.95q.
970308131806.23766F-100000@fractal.lafayette.
(applied based on p5p patch as commit
e6834c60cd85866fc530e8b4bb831af6186bad4d)
Subject: PODs corrections
Date: Fri, 7 Mar 1997 21:53:04 -0500 (EST)
From: Ilya Zakharevich <ilya@math.ohio-state.edu>
Files: ext/DB_File/DB_File.pm ext/Socket/Socket.pm lib/Class/Template.pm lib/ExtUtils/Embed.pm lib/ExtUtils/MM_VMS.pm lib/ExtUtils/Mksymlists.pm lib/File/Basename.pm lib/File/stat.pm lib/Time/gmtime.pm lib/Time/localtime.pm lib/Time/tm.pm lib/User/grent.pm lib/User/pwent.pm pod/perlcall.pod pod/perldebug.pod pod/perlfunc.pod pod/perlguts.pod pod/perllocale.pod pod/perlop.pod pod/perlsub.pod
Msg-ID:
199703080253.VAA24975@monk.mps.ohio-state.edu
(applied based on p5p patch as commit
72451f4af0d31f24ef5b12bc5d034e3e8b35d43d)
OTHER CORE CHANGES
Subject: Fix imbalanced ENTER/LEAVE from C<BEGIN{die}>
From: Chip Salzenberg <chip@perl.com>
Files: op.c perl.c proto.h
Subject: perl -P path patch
Date: Sat, 08 Mar 1997 12:45:08 -0500 (EST)
From: Andy Dougherty <doughera@fractal.phys.lafayette.edu>
Files: config_H config_h.SH perl.c plan9/config.plan9 t/comp/cpp.t vms/config.vms win32/config.H
Msg-ID: Pine.SOL.3.95q.
970308120242.23766D-100000@fractal.lafayette.
(applied based on p5p patch as commit
bba014945c609b5474f61f5e82ed2ff3e83a6e47)
Ilya Zakharevich [Sat, 8 Mar 1997 02:53:04 +0000 (21:53 -0500)]
PODs corrections
Below I fix misfeatures of PODs found by pod2ipf. Just for fun note that it
auto-recognizes hyperlinks to WWW now.
p5p-msgid:
199703080253.VAA24975@monk.mps.ohio-state.edu
Chip Salzenberg [Thu, 6 Mar 1997 16:01:12 +0000 (04:01 +1200)]
[dummy merge]
This merge exists so that the p5p version of the patch and the applied
version are both in the history
Andy Dougherty [Sat, 8 Mar 1997 18:21:22 +0000 (13:21 -0500)]
Consolidated INSTALL updates since _92
I've incorporated the various suggestions for INSTALL. Here's
the current version. (I took the opportunity to regularize some
of the malloc-related sections as well as add the appropriate
info.)
Note: There's one error here: I refer users to L<perlvar> for
the description of $^M, but I don't see it there!
p5p-msgid: Pine.SOL.3.95q.
970308131806.23766F-100000@fractal.lafayette.edu
private-msgid: Pine.SOL.3.95q.
970308131806.23766F-100000@fractal.lafayette.
Chip Salzenberg [Thu, 6 Mar 1997 16:01:12 +0000 (04:01 +1200)]
OS/2 doc update
(this is the same change as commit
de0c99f683f89c0419c77decc2ae7cee1432e5d4, but as applied)
Ilya Zakharevich [Sat, 8 Mar 1997 05:37:30 +0000 (00:37 -0500)]
OS/2 doc update
The patch below fixes some other goofs in README.os2 and INSTALL
discovered by pod2ipf (again, broken links + minor updates to OS/2).
p5p-msgid:
199703080537.
AAA25157@monk.mps.ohio-state.edu
Spider Boardman [Sat, 8 Mar 1997 00:53:00 +0000 (19:53 -0500)]
Use 'test -f', not 'test -x'
Subject: 3_92 get "test: argument expected" during Configure
This is a bug report for perl from spider@web.zk3.dec.com,
generated with the help of perlbug 1.16 running under perl 5.00392.
p5p-msgid:
199703080053.TAA13943@web.zk3.dec.com
Chip Salzenberg [Tue, 31 Dec 1996 06:32:43 +0000 (18:32 +1200)]
Warn about '.' terminating E-Mail
Andy Dougherty [Thu, 6 Mar 1997 16:15:47 +0000 (11:15 -0500)]
Allow './Configure -Uoptimize'
Subject: Re: miniperl core dumps during build of 5.003_91 on Unisys SVR4
On Thu, 6 Mar 1997, Alan Burlison wrote:
> In-Reply-To: <memo.119650@cix.compulink.co.uk>
> > miniperl dumps core during the build on Unisys SVR4 version 1.4.
> Apparently there is a bug in the C compiler optimiser in this version
> of the OS. I'll attribute this problem to that, & just turn off
Our record in this regard is mixed, at best, I think. (Some
problems that appeared to be optimizer problems have been made to go
away by fixing possibly-related bugs in the perl source.) Still, if it
works . . . .
> debugging. I am using Configure -des -Doptimize=' ' to do that. Is
> this the best way? -Uoptimize causes Configure to die.
Oops. That's not nice. The way I do this is
sh Configure -Doptimize=none
though your way is quite fine too.
In general, there's no particular reason to expect optimize='undef' to
do what you want, but it's not hard to handle it gracefully either, so
we could apply this low-urgency but low-risk patch:
p5p-msgid: Pine.SOL.3.95q.
970306110532.11070A-100000@fractal.lafayette.edu
private-msgid: Pine.SOL.3.95q.
970306110532.11070A-100000@fractal.lafayette.
Chip Salzenberg [Thu, 6 Mar 1997 15:26:22 +0000 (03:26 +1200)]
Fix more E-Mail addresses in pods
Alan Burlison [Thu, 6 Mar 1997 03:28:00 +0000 (16:28 +1300)]
Fix for Unisys UNIX and libperl.so
Subject: Re: Why libperl.a instead of libperl.so in 5.003_91? - A PATCH
In-Reply-To: <memo.141901@cix.compulink.co.uk>
> On the hunt of the elusive cause of the dynamic linker errors I am
> getting in 5.003_91, I have noticed that unlike every other vesion of
> perl5 prior to 5.002 that I have built on this platform, Configure
> now builds a libperl.a instead of a libperl.so. Why? I also note
> that the -Kpic flag has disappeared from the cc command lines. Why?
The following patch to Configure solves the problem. Do I win the prize for
the smallest ever patch to Perl?
p5p-msgid: memo.147328@cix.compulink.co.uk
Chip Salzenberg [Tue, 4 Mar 1997 21:55:21 +0000 (09:55 +1200)]
Quote pathname before using as pattern
Andy Dougherty [Sat, 8 Mar 1997 17:45:08 +0000 (12:45 -0500)]
perl -P path patch
On Fri, 7 Mar 1997, Robin Barker wrote:
> This is a bug report for perl from rmb1@npl.co.uk,
> generated with the help of perlbug 1.16 running under perl 5.00392.
>
>
> -----------------------------------------------------------------
> [Please enter your report here]
>
> I can't get perl -P to work.
>
> In particular
> % perl -P t/comp/cpp.aux
> sh: /home/rmb1/local/perl/bin/cppstdin: not found
>
> the make process constructs cppstdin in the src directory
> and would copy that to $installbin
> but both t/comp/cpp.t and the perl binary look for a cppstdin
> executable in $scriptdir.
Thanks for the report. This is a bug. cppstdin belongs in $installbin,
not $scriptdir.
Why? Because cppstdin is architecture-dependent. It depends, in detail,
on the particular compiler required for that architecture. True, we could
rewrite it in portable perl to use all the appropriate Config.pm
variables, but really, why bother? It's a tiny script that is rarely
used, and I don't want to further encourage its use.
Here's a patch. Hmm. It's a bit longer than I first imagined because I
need to use BIN_EXP instead of BIN, and since perl currently doesn't use
BIN_EXP, metaconfig hadn't stuck it in config.h.
Alternatively, we could have put cppstdin in $installarch or something
like that, but we'd better keep it in $bin just in case some-one's been
relying on that.
Ok. Here it is. (Mind you, I haven't yet *tested* this on a machine that
needs cppstdin :-).
p5p-msgid: Pine.SOL.3.95q.
970308120242.23766D-100000@fractal.lafayette.edu
private-msgid: Pine.SOL.3.95q.
970308120242.23766D-100000@fractal.lafayette.
Chip Salzenberg [Wed, 15 Jan 1997 03:42:51 +0000 (15:42 +1200)]
Ignore backup files in strict.t and warning.t
Doug MacEachern [Sun, 9 Mar 1997 00:19:38 +0000 (19:19 -0500)]
Support READ and GETC for tied handles
Subject: Re: Seeds of _93: planting TIEHANDLE/READ
Ilya Zakharevich <ilya@math.ohio-state.edu> wrote:
> Doug MacEachern writes:
> >
> > Chip Salzenberg <salzench@nielsenmedia.com> wrote:
> >
> > > Here's what I have for _93. Speak now, etc.
> >
> > sorry to plant this seed so late, but I think it's important for the
> > growth of young TIEHANDLE. We have PRINT and READLINE, why no READ?
>
> If you added READ, where is getc? Or is it covered by the patch?
sorry, I was being selfish, I don't care about getc :-) new patch
against fresh _92 below. I know this all came up in the past, there
are more missing, why where they not covered before?
p5p-msgid:
199703090019.TAA32591@postman.osf.org
Hugo van der Sanden [Thu, 6 Mar 1997 19:03:35 +0000 (19:03 +0000)]
Fix counts in output of TEST
Solaris 2.5.1: clean make, test and install, except for a slight bug in
t/TEST itself. Before the patch below it reports:
All tests successful.
u=0.45 s=0.44 cu=27.36 cs=13.53 scripts=1 tests=17
and after, it reports:
All tests successful.
u=0.44 s=0.35 cu=28.39 cs=13.93 scripts=145 tests=3618
p5p-msgid:
331F1507.4BE8@iii.co.uk
Gurusamy Sarathy [Thu, 6 Mar 1997 19:12:09 +0000 (14:12 -0500)]
Don't autovivify array and hash elements in sub parameters
On Thu, 06 Mar 1997 10:02:33 PST, Larry Wall wrote:
>: =head2 Why does passing a subroutine an undefined element in a hash create it?
>:
>: If you say something like:
>:
>: somefunc($hash{"nonesuch key here"});
>:
>: Then that element "autovivifies"; that is, it springs into existence
>: whether you store something there or not. That's because functions
>: get scalars passed in by reference. If somefunc() modifies C<$_[0]>,
>: it has to be ready to write it back into the caller's version.
>
>I still think this behavior is bogus. By default, it should simply be
>an error to pass an undefined value to a routine that wants to modify
>it. (We can maybe modify that in the future with prototypes.)
>
>: =for p5p is this still true?
>
>Yes, unfortunately. Is my editorial opinion leaking through here? :-)
I hope you considered my one-line patch to revert this behavior to that of
perl5.001 (and perl4). I still think we'd be doing a service to include
this patch in 5.004 (and remove that section from the FAQ and the docs).
One may still modify real scalars via $_[0] with the patch, so it has
little potential to cause breakage.
p5p-msgid:
199703061912.OAA20606@aatma.engin.umich.edu
Roderick Schertler [Fri, 7 Mar 1997 00:19:07 +0000 (19:19 -0500)]
Open[23] autoflush docs
p5p-msgid: 7939.
857693947@eeyore.ibcinc.com
Roderick Schertler [Sat, 8 Mar 1997 17:22:04 +0000 (12:22 -0500)]
@EXPORT_FAIL fix for Exporter.pm
The default export_fail method in Exporter.pm needs to drop the first
argument (the class name).
p5p-msgid: 24884.
857841724@eeyore.ibcinc.com
M.J.T. Guy [Sat, 8 Mar 1997 20:12:17 +0000 (20:12 +0000)]
Carp with multiple arguments
Subject: Carp with multiple arguments (and lack of tests)
carp/croak/confess are documented as similar to warn/die, but whereas
warn/die print multiple arguments, carp/croak/confess print only the
first. Assuming that this is a bug rather than a feature, the following
patch corrects it. (And if it _is_ a feature, the pod needs updating
instead, to explain it away.)
Trying to be polite, I thought I'd update the Carp tests to cover this
case. But I couldn't find any Carp tests ... I'll have a go
at rectifying this lacuna. But not till tomorrow.
p5p-msgid: E0w3STZ-0007RW-00@taurus.cus.cam.ac.uk
Chip Salzenberg [Thu, 6 Mar 1997 16:01:12 +0000 (04:01 +1200)]
[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, 6 Mar 1997 16:01:12 +0000 (04:01 +1200)]
Don't count on 'trap 0' inside () in shell script
(this is the same change as commit
2626e112ed2a75058c6a5e98cce21f1af08a8ace, but as applied)
Alan Burlison [Thu, 6 Mar 1997 03:28:00 +0000 (16:28 +1300)]
Don't count on 'trap 0' inside () in shell script
When building 5.003_91, I noticed that Configure leaves a load of files lying
around in /tmp, for example exp21743c, etc. These appear to contain lists of
symbols. A bit of digging reveals that the offender is the perl_exp.SH
script. This contains a bit of code of the form
(
trap "rm -f ...." 0
...
)
I guess that when the sub-shell invoked by the ( ... ) construct exits, the
trap 0 should be taken, and the files removed. Unfortunately, I have noticed
that on at least 2 systems (UnixWare & Unisys SVR4) this doesn't happen. The
following patch fixes it.
p5p-msgid: memo.147326@cix.compulink.co.uk
Chip Salzenberg [Thu, 6 Mar 1997 15:46:34 +0000 (03:46 +1200)]
Make conversion of @_ to real array work right after C<shift>
Norbert Pueschel [Sat, 8 Mar 1997 11:50:15 +0000 (12:50 +0100)]
AmigaOS hint patch
92 compiles and tests 100% under AmigaOS and NetBSD 1.2.
The following patch is needed to prevent Configure from
blowing up pdksh. Don't ask me why, I haven't the foggiest idea. :-(
p5p-msgid:
77724767@Armageddon.meb.uni-bonn.de
Chip Salzenberg [Thu, 6 Mar 1997 15:46:37 +0000 (03:46 +1200)]
Warn on %{+undef} and @{+undef}
Chip Salzenberg [Sat, 1 Mar 1997 06:40:49 +0000 (18:40 +1200)]
[differences between cumulative patch application and perl-5.003_92]
Perl 5 Porters [Sat, 1 Mar 1997 06:40:49 +0000 (18:40 +1200)]
[inseparable changes from match from perl-5.003_91 to perl-5.003_92]
CORE LANGUAGE CHANGES
Subject: Strictly follow lexical context of C<eval ''> and nested subs
From: Chip Salzenberg <chip@perl.com>
Files: op.c
Subject: Make ::SUPER and UNIVERSAL work together
From: Chip Salzenberg <chip@perl.com>
Files: gv.c pod/perlguts.pod
CORE PORTABILITY
Subject: OS/2 patches
Date: Wed, 5 Mar 1997 22:08:43 -0500 (EST)
From: Ilya Zakharevich <ilya@math.ohio-state.edu>
Files: hints/os2.sh lib/ExtUtils/MakeMaker.pm t/op/taint.t
Msg-ID:
199703060308.WAA22211@monk.mps.ohio-state.edu
(applied based on p5p patch as commit
eda4d5189d403b15f244b4696a710fb91d15053e)
Subject: VMS patches
Date: Wed, 05 Mar 1997 23:10:24 -0500 (EST)
From: Charles Bailey <bailey@HMIVAX.HUMGEN.UPENN.EDU>
Files: lib/ExtUtils/MM_VMS.pm lib/ExtUtils/Manifest.pm perlsdio.h t/op/runlevel.t t/op/taint.t vms/descrip.mms vms/perly_c.vms vms/sockadapt.c vms/sockadapt.h vms/vms_yfix.pl
private-msgid: 01IG5SQE4A6U00661G@hmivax.humgen.upenn.edu
DOCUMENTATION
Subject: Add taint checks and srand to perldelta
Date: Sun, 2 Mar 1997 11:56:08 -0800 (PST)
From: Tom Phoenix <rootbeer@teleport.com>
Files: pod/perldelta.pod
Msg-ID: Pine.GSO.3.95q.
970302115355.23058D-100000@kelly.teleport.com
(applied based on p5p patch as commit
b28e0bc0aa3232e18d1bacb3efcbfb755ad100e0)
Subject: Don't call FileHandle 'deprecated'
From: Chip Salzenberg <chip@perl.com>
Files: pod/perldelta.pod
Subject: Improve sample module header
Date: Sat, 01 Mar 1997 10:32:31 -0700
From: Tom Christiansen <tchrist@jhereg.perl.com>
Files: pod/perlmod.pod
Msg-ID:
199703011732.KAA14693@jhereg.perl.com
(applied based on p5p patch as commit
3e1e15658152387f41e00ded4796cede4e1e10d3)
Subject: Update list of CPAN sites
Date: Sun, 2 Mar 1997 16:54:22 +0200 (EET)
From: Jarkko Hietaniemi <jhi@iki.fi>
Files: pod/perlmod.pod
Msg-ID:
199703021454.QAA07446@alpha.hut.fi
(applied based on p5p patch as commit
9423903e60e6c92c1893f5f4cab2476f403f8a4b)
Subject: Enhance description of 'server error'
Date: Tue, 4 Feb 1997 21:03:23 +0200 (EET)
From: Jarkko Hietaniemi <jhi@cc.hut.fi>
Files: pod/perldiag.pod
private-msgid:
199702041903.VAA16070@alpha.hut.fi
Subject: Regularize format of E-Mail addresses in *.pod
From: Chip Salzenberg <chip@perl.com>
Files: pod/*.pod
LIBRARY AND EXTENSIONS
Subject: Use IV instead of double for tms structure members
From: Chip Salzenberg <chip@perl.com>
Files: ext/POSIX/POSIX.xs
OTHER CORE CHANGES
Subject: Make sure $^X is tainted when ARG_ZERO_IS_SCRIPT
From: Chip Salzenberg <chip@perl.com>
Files: toke.c
Subject: Clarify '-T too late' error
From: Chip Salzenberg <chip@perl.com>
Files: perl.c pod/perldiag.pod
Subject: Warn when redefining or undefining a constant sub
From: Chip Salzenberg <chip@perl.com>
Files: pod/perldiag.pod pp.c sv.c
Subject: Don't generate spurious 'not imported' warning
From: Chip Salzenberg <chip@perl.com>
Files: gv.c t/pragma/strict-vars pod/perldiag.pod
Subject: Clarify message re: @host in string
From: Chip Salzenberg <chip@perl.com>
Files: pod/perldiag.pod pod/perltrap.pod toke.c
Subject: Disconnect refs that are targets of pp_readline
From: Chip Salzenberg <chip@perl.com>
Files: pp_hot.c
Subject: Fix typo in test of HvFILL()
From: Chip Salzenberg <chip@perl.com>
Files: op.c
Subject: Allow for pad name array to be shorter than pad array
From: Chip Salzenberg <chip@perl.com>
Files: op.c
Subject: Eliminate format-string type warnings
Date: Mon, 3 Mar 1997 10:15:11 +0100 (MET)
From: Hallvard B Furuseth <h.b.furuseth@usit.uio.no>
Files: doio.c ext/POSIX/POSIX.xs gv.c hints/dec_osf.sh pp.c pp_ctl.c pp_hot.c run.c sv.c x2p/a2py.c
private-msgid:
199703030915.KAA11634@bombur2.uio.no
Subject: Update copyright dates
From: Chip Salzenberg <chip@perl.com>
Files: *.[hc] x2p/*.[hc] win32/EXTERN.h vms/vmsish.h vms/vms.c
TESTS
Subject: Smarter t/op/taint.t
Date: Mon, 3 Mar 1997 10:31:54 -0800 (PST)
From: Tom Phoenix <rootbeer@teleport.com>
Files: t/op/taint.t
private-msgid: Pine.GSO.3.95q.
970303103047.24000A-100000@kelly.teleport.com
Subject: Fix taint test for systems without csh
From: Chip Salzenberg <chip@perl.com>
Files: t/op/taint.t
Jarkko Hietaniemi [Sun, 2 Mar 1997 14:54:22 +0000 (16:54 +0200)]
Update list of CPAN sites
p5p-msgid:
199703021454.QAA07446@alpha.hut.fi
Chip Salzenberg [Sat, 1 Mar 1997 06:40:49 +0000 (18:40 +1200)]
[dummy merge]
This merge exists so that the p5p version of the patch and the applied
version are both in the history
Tom Christiansen [Sat, 1 Mar 1997 17:32:31 +0000 (10:32 -0700)]
Improve sample module header
Subject: Re: addition to perlmod and perlfaq?
Here's the "final" version. Larry and I have to
head off to a book signing now. Chip, please have
this replace the old stuff I have in L<perlmod/"Perl Modules">
that begins with ``package Fred;''.
p5p-msgid:
199703011732.KAA14693@jhereg.perl.com
Signed-off-by: Graham Barr <gbarr@ti.com>
Chip Salzenberg [Sat, 1 Mar 1997 06:40:49 +0000 (18:40 +1200)]
Clarify C<crypt> documentation
(this is the same change as commit
f827b0a6c8a4efd53c50a16b098f59ac04fd670d, but as applied)
Tom Phoenix [Fri, 28 Feb 1997 21:18:25 +0000 (13:18 -0800)]
Clarify C<crypt> documentation
From a suggestion I got in private e-mail. Most readers of this list will
quickly recognize this as the part of perlfunc which explains crypt. This
should apply over my previous patch, but if you've already fixed my typo,
you'll have to fix something by hand.
p5p-msgid: Pine.GSO.3.95q.
970228131112.12357D-100000@kelly.teleport.com
Tom Phoenix [Sun, 2 Mar 1997 19:56:08 +0000 (11:56 -0800)]
Add taint checks and srand to perldelta
p5p-msgid: Pine.GSO.3.95q.
970302115355.23058D-100000@kelly.teleport.com
(name lost) [Sat, 25 Jan 1997 02:01:05 +0000 (14:01 +1200)]
Support 'long long' in h2ph
Ilya Zakharevich [Thu, 6 Mar 1997 03:08:43 +0000 (22:08 -0500)]
OS/2 patches
Chip Salzenberg writes:
>
> Below is a raw list of the changes I've incorporated into my working
> source tree. I've not broken them down by category yet, or edited
> their titles; they're in chronological order of application.
Here is what I have:
a) Minor fix for hints/os2;
b) 1 taint test minorly broken (does not unset IFS), and should be
switched off on DOSISH systems anyway :-(;
c) One additional word made legal in MakeMaker (It was supported
before, but a warning was issued).
Enjoy,
p5p-msgid:
199703060308.WAA22211@monk.mps.ohio-state.edu
Chip Salzenberg [Sat, 1 Mar 1997 06:40:49 +0000 (18:40 +1200)]
[dummy merge]
This merge exists so that the p5p version of the patch and the applied
version are both in the history
Chip Salzenberg [Sat, 1 Mar 1997 06:40:49 +0000 (18:40 +1200)]
Update pod2html
(this is the same change as commit
ffcd4c5fd14c54dc48f279f8fb1699214b296ed6, but as applied)
William Middleton [Mon, 3 Mar 1997 00:25:03 +0000 (16:25 -0800)]
Update pod2html
> [...]
> # Lost interest? It's so bad I've lost apathy.
You too, eh?
However, here's a patch which brings the _91 version up to date.
YES, it still slurps chunks. ( No comment )
YES, it still uses XMP. (Nothing else works right)
YES, it still is hard to follow. (But oh, the joy of figuring it out!)
YES, it works on all platforms with configuration, including Macs...
I had pretty much given up on this, just telling people to grab
the latest from my CPAN dir, but I get alot of mail with fixes
against 1.15, and this is version is vastly better... Please
try it, and apply to 5.004, if it's not too late.
I will (respectfully) ignore any whining about the inappropriateness
of some of the things I had to do in here. Also any comments
about other renditions, I've tried them all, and this one still
excels; besides working on all platforms. I'm in the middle of a
rather difficult project, and I took the day off today to get this
together.
p5p-msgid:
199703030025.QAA08106@ducks
Chip Salzenberg [Tue, 18 Feb 1997 16:22:49 +0000 (04:22 +1200)]
Don't test locales if there is no setlocale()
Daniel S. Lewart [Sat, 1 Mar 1997 18:04:09 +0000 (12:04 -0600)]
Make op/TEST silent under -w
Below is a unidiff to make t/TEST "-w" clean.
p5p-msgid:
199703011821.NAA13037@sinistar.idle.com
Chip Salzenberg [Mon, 20 Jan 1997 16:11:30 +0000 (04:11 +1200)]
Document IO::File::new_tmpfile
Hugo van der Sanden [Thu, 6 Mar 1997 07:55:25 +0000 (07:55 +0000)]
Fix quotewords
Subject: Re: Fix for ::SUPER with UNIVERSAL
In <
199703052319.SAA32568@rio.atlantic.net>, Chip Salzenberg writes:
:Cute. It turns out that FileHandle and IO::Handle were okay, but the
:handling of ::SUPER (which is used by IO::Handle::new) was broken in
:the face of UNIVERSAL.
:
:Here's a patch.
That fixes it.
Here's one back - I think Text::ParseWords->quotewords is more broken than
this, but it's a start.
p5p-msgid:
199703060755.HAA15060@crypt.compulink.co.uk
Hugo van der Sanden [Tue, 4 Mar 1997 11:32:11 +0000 (11:32 +0000)]
Benchmark: using code refs
It often irritates me that I can't supply a coderef to the Benchmark::
routines: here's a small patch to allow them.
p5p-msgid:
199703041132.LAA07613@tyree.iii.co.uk
Johan Vromans [Fri, 28 Feb 1997 18:11:34 +0000 (06:11 +1200)]
Refresh Getopt::Long to 2.9
Chip Salzenberg [Sat, 1 Mar 1997 06:40:49 +0000 (18:40 +1200)]
[dummy merge]
This merge exists so that the p5p version of the patch and the applied
version are both in the history
Chip Salzenberg [Sat, 1 Mar 1997 06:40:49 +0000 (18:40 +1200)]
Newer ReadLine
(this is the same change as commit
bcbbe6e503cc1899ede8fc1ac0c1c14e432c4f60, but as applied)
Ilya Zakharevich [Tue, 4 Mar 1997 06:34:28 +0000 (01:34 -0500)]
Newer ReadLine
Finally I could compile GNU ReadLine2.1, so I got some experience with
T::R::Gnu. Unfortunately, debugger could not use any advanced features
of T::R::Gnu since it was accessing the features of T::R::Perl via
backdoors.
I reworked the interface to use object-oriented methods so that it
should not know anything about the particular ReadLine package it uses
(as far as it conforms to some simple interface).
Below is that part of the patch which should go into standard
distribution
Consider it as mildly emergent bug fix, it is -w-safe,
Enjoy,
p5p-msgid:
199703040634.
BAA19919@monk.mps.ohio-state.edu
Chip Salzenberg [Fri, 28 Feb 1997 17:28:12 +0000 (05:28 +1200)]
Change 'continuing anyway' to 'probably harmless'
Chip Salzenberg [Sat, 1 Mar 1997 06:40:49 +0000 (18:40 +1200)]
[dummy merge]
This merge exists so that the p5p version of the patch and the applied
version are both in the history
Chip Salzenberg [Sat, 1 Mar 1997 06:40:49 +0000 (18:40 +1200)]
near-harmless bug in _91's Configure
(this is the same change as commit
79893acbe8efa42713540e8af91f667f182baa34, but as applied)
Roderick Schertler [Sun, 2 Mar 1997 02:26:49 +0000 (21:26 -0500)]
near-harmless bug in _91's Configure
At line 8118 of _91's Configure is this:
echo"(I can't seem to compile the test program...)"
note missing space after echo.
p5p-msgid: pzg1yfuiza.fsf@eeyore.ibcinc.com
Tom Phoenix [Wed, 5 Mar 1997 17:47:22 +0000 (09:47 -0800)]
Re: The continuing MachTen saga
private-msgid: Pine.GSO.3.95q.
970305091611.3572E-100000@kelly.teleport.com
Raphael Manfredi [Thu, 6 Mar 1997 13:10:38 +0000 (14:10 +0100)]
HP-UX hint update
Quoting Chip Salzenberg:
:If you have a second, please look them over and see if anything
:important is out or anything broken is in.
I don't see my hint/hpux.sh patch. Here it is again (in unified format
this time):
p5p-msgid: 1479.
857653838@lyon.grenoble.hp.com
Ilya Zakharevich [Thu, 6 Mar 1997 03:08:43 +0000 (22:08 -0500)]
OS/2 patches
Chip Salzenberg writes:
>
> Below is a raw list of the changes I've incorporated into my working
> source tree. I've not broken them down by category yet, or edited
> their titles; they're in chronological order of application.
Here is what I have:
a) Minor fix for hints/os2;
b) 1 taint test minorly broken (does not unset IFS), and should be
switched off on DOSISH systems anyway :-(;
c) One additional word made legal in MakeMaker (It was supported
before, but a warning was issued).
Enjoy,
p5p-msgid:
199703060308.WAA22211@monk.mps.ohio-state.edu
Chip Salzenberg [Sat, 1 Mar 1997 06:40:49 +0000 (18:40 +1200)]
[dummy merge]
This merge exists so that the p5p version of the patch and the applied
version are both in the history
Chip Salzenberg [Sat, 1 Mar 1997 06:40:49 +0000 (18:40 +1200)]
Clarify C<crypt> documentation
(this is the same change as commit
033135fa782fb902eccfa5fabb72b7be2050ba12, but as applied)
Tom Phoenix [Fri, 28 Feb 1997 21:18:25 +0000 (13:18 -0800)]
Clarify C<crypt> documentation
p5p-msgid: <Pine.GSO.3.95q.
970228131112.12357D-100000@kelly.teleport.com>
(name lost) [Sat, 25 Jan 1997 02:01:05 +0000 (14:01 +1200)]
Support 'long long' in h2ph
Chip Salzenberg [Sat, 1 Mar 1997 06:40:49 +0000 (18:40 +1200)]
[dummy merge]
This merge exists so that the p5p version of the patch and the applied
version are both in the history
Chip Salzenberg [Sat, 1 Mar 1997 06:40:49 +0000 (18:40 +1200)]
Update pod2html
(this is the same change as commit
90841c69f02b801a9d408ee4b2ed3da4664a144d, but as applied)
William Middleton [Mon, 3 Mar 1997 00:25:03 +0000 (16:25 -0800)]
Update pod2html
> [...]
> # Lost interest? It's so bad I've lost apathy.
You too, eh?
However, here's a patch which brings the _91 version up to date.
YES, it still slurps chunks. ( No comment )
YES, it still uses XMP. (Nothing else works right)
YES, it still is hard to follow. (But oh, the joy of figuring it out!)
YES, it works on all platforms with configuration, including Macs...
I had pretty much given up on this, just telling people to grab
the latest from my CPAN dir, but I get alot of mail with fixes
against 1.15, and this is version is vastly better... Please
try it, and apply to 5.004, if it's not too late.
I will (respectfully) ignore any whining about the inappropriateness
of some of the things I had to do in here. Also any comments
about other renditions, I've tried them all, and this one still
excels; besides working on all platforms. I'm in the middle of a
rather difficult project, and I took the day off today to get this
together.
p5p-msgid:
199703030025.QAA08106@ducks
Chip Salzenberg [Tue, 18 Feb 1997 16:22:49 +0000 (04:22 +1200)]
Don't test locales if there is no setlocale()
Daniel S. Lewart [Sat, 1 Mar 1997 18:04:09 +0000 (12:04 -0600)]
Make op/TEST silent under -w
Below is a unidiff to make t/TEST "-w" clean.
p5p-msgid:
199703011821.NAA13037@sinistar.idle.com
Chip Salzenberg [Mon, 20 Jan 1997 16:11:30 +0000 (04:11 +1200)]
Document IO::File::new_tmpfile
Hugo van der Sanden [Thu, 6 Mar 1997 07:55:25 +0000 (07:55 +0000)]
Fix quotewords
Subject: Re: Fix for ::SUPER with UNIVERSAL
In <
199703052319.SAA32568@rio.atlantic.net>, Chip Salzenberg writes:
:Cute. It turns out that FileHandle and IO::Handle were okay, but the
:handling of ::SUPER (which is used by IO::Handle::new) was broken in
:the face of UNIVERSAL.
:
:Here's a patch.
That fixes it.
Here's one back - I think Text::ParseWords->quotewords is more broken than
this, but it's a start.
p5p-msgid:
199703060755.HAA15060@crypt.compulink.co.uk
Hugo van der Sanden [Tue, 4 Mar 1997 11:32:11 +0000 (11:32 +0000)]
Benchmark: using code refs
It often irritates me that I can't supply a coderef to the Benchmark::
routines: here's a small patch to allow them.
p5p-msgid:
199703041132.LAA07613@tyree.iii.co.uk