| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
| |
This is my patch _01 to perl 5.002.
The bulk of this patch consists of VMS and OS/2 updates, and the
conversion to use EMBED as the default when building Perl. It
also includes miscellaneous bugfixes, corrections to the pods,
and minor additions, notably the new $^E and $^O variables.
[editor's note: the changes between this and the perl-5.002_01 tag are
the individual changes in the patch set]
|
|
|
|
|
| |
[editor's note: changes seem to be mostly module updates,
documentation changes and some perl API macro additions]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
>From roehrich@cray.comTue Feb 6 10:10:30 1996
>Date: Mon, 5 Feb 1996 20:30:34 -0600
>From: Dean Roehrich <roehrich@cray.com>
>To: perl5-porters@africa.nicoh.com
>Subject: h2xs 1.15
>
>This is h2xs version 1.15. Apply over beta3.
>
>From roehrich@cray.comWed Feb 7 09:17:51 1996
>Date: Tue, 6 Feb 1996 18:31:53 -0600
>From: Dean Roehrich <roehrich@cray.com>
>To: perl5-porters@africa.nicoh.com
>Subject: h2xs 1.16
>
>Following a suggestion from PaulM, I have updated the AUTOLOAD() that h2xs
>creates. It uses Carp::croak now.
|
| |
|
|
|
|
| |
Move perl_init_i18nl14n here from miniperlmain.c.
|
|
|
|
|
| |
This patch fixes a file leak triggered by "use" or "require" errors.
I think it's of fairly high priority...
|
|
|
|
| |
This patch ensures that hex() and oct() always return positive values.
|
| |
|
|
|
|
|
| |
This does not change actual tests, but improves support code such that
an early bail out does not happen if environment is not Unixish.
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
>From salzench@dun.nielsen.comSat Feb 3 15:18:05 1996
>Date: Sat, 3 Feb 1996 12:45:32 -0500 (EST)
>From: Chip Salzenberg <salzench@dun.nielsen.com>
>Reply to: chip@atlantic.net
>To: Perl 5 Porters <perl5-porters@africa.nicoh.com>
>Subject: Beta3: Combine code: provide_refs
>From lwall@sems.comWed Feb 7 09:10:55 1996
>Date: Tue, 06 Feb 96 14:52:41 -0800
>From: Larry Wall <lwall@sems.com>
>To: perl5-porters@africa.nicoh.com
>Subject: study still busted
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
>From Richard_Ohnemus@dallas.csd.sterling.comThu Feb 8 10:09:31 1996
>Date: Wed, 7 Feb 1996 20:47:25 -0600 (CST)
>From: Richard Ohnemus <Richard_Ohnemus@dallas.csd.sterling.com>
>To: perl5-porters@africa.nicoh.com
>Subject: small bug in pod2man
pod2man will generate incorrect man pages if the pod file name has more than
1 dot ('.') in it.
>From kjahds@kjahds.comFri Feb 9 12:56:50 1996
>Date: Fri, 9 Feb 1996 12:06:23 -0500 (EST)
>From: Kenneth Albanowski <kjahds@kjahds.com>
>To: Peter Wolfe <wolfe@teloseng.com>
>Cc: tchrist@mox.perl.com, roderick@gate.net, doughera@lafcol.lafayette.edu,
> perl5-porters@africa.nicoh.com
>Subject: Re: pod2man patch
[Fixes .bd command ]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
>From okamoto@hpcc123.corp.hp.comSun Feb 11 21:28:55 1996
>Date: Fri, 9 Feb 1996 17:32:31 -0800
>From: Jeff Okamoto <okamoto@hpcc123.corp.hp.com>
>To: perl5-porters@africa.nicoh.com
>Subject: Perlxstut, 2/9 version
>
>Here's the latest version. It should address all the comments that were made
>in the last release.
>From Tim.Bunce@ig.co.ukSun Feb 11 21:29:11 1996
>Date: Sat, 10 Feb 1996 08:34:48 +0000
>From: Tim Bunce <Tim.Bunce@ig.co.uk>
>To: perl5-porters@africa.nicoh.com, okamoto@hpcc123.corp.hp.com
>Cc: doughera@lafcol.lafayette.edu
>Subject: Re: Perlxstut, 2/9 version
[Use Mytest instead of mytest -- reserve lower case for pragmas.]
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
Chip Salzenberg writes:
>
> According to Paul Marquess:
> > use vars qw($VERSION @ISA @EXPORT);
>
> It would be safe to add @EXPORT_OK, @EXPORT_FAIL, %EXPORT_TAGS, and
> %OVERLOAD.
You remind me ;-)
|
|
|
|
|
|
|
| |
Here's some documentation for tied(), there doesn't seem to be any in
2b3. Also included is a small comment in the FIRSTKEY() hash example of
perltie.pod, I had to look twice when I first saw the code and I thought
others might benefit from the explanation.
|
| |
|
| |
|
|
|
|
|
|
| |
Change message to 'gamma'.
Here's a "gamma safe" patch.
|
| |
|
|
|
|
|
|
|
| |
[next note is from Andy]
[ This didn't apply cleanly to my source. If I delete the (int)
casts in the first part of the patch, it does apply cleanly. I worry
about that a little, but I put in the patch anyway. --AD ]
|
| |
|
|
|
|
|
|
| |
Moved function to util.c It was C++ unfriendly and probably
didn't need to be here anyway. (When it was just a small 2 or 3
liner, it made sense . . . ).
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
>From salzench@dun.nielsen.comSat Feb 3 15:17:03 1996
>Date: Sat, 3 Feb 1996 12:43:39 -0500 (EST)
>From: Chip Salzenberg <salzench@dun.nielsen.com>
>Reply to: chip@atlantic.net
>To: Perl 5 Porters <perl5-porters@africa.nicoh.com>
>Subject: Beta3: Possible typo in sv_unmagic() [edited]
>From salzench@dun.nielsen.comTue Feb 6 09:45:25 1996
>Date: Mon, 5 Feb 1996 00:09:09 -0500 (EST)
>From: Chip Salzenberg <salzench@dun.nielsen.com>
>Reply to: chip@atlantic.net
>To: Perl 5 Porters <perl5-porters@africa.nicoh.com>
>Subject: Beta3: Fix for 2nd magic SEGV
>From lwall@sems.comWed Feb 7 09:10:55 1996
>Date: Tue, 06 Feb 96 14:52:41 -0800
>From: Larry Wall <lwall@sems.com>
>To: perl5-porters@africa.nicoh.com
>Subject: study still busted
>The study itself work fine, but if you modify the string, it still thinks
>it is studied. In some ways this is worse than study not working at all,
>as it did (or rather didn't) before. Here's the test case:
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
[This finishes a job started by Randy Ray with perl5.002b1f.]
This patch updates the files lib/TieHash.pm and lib/SubstrHash.pm. Those
two files are moved to a new directory called lib/Tie. A new tie-related
class for scalars, lib/Tie/Scalar.pm, is added. Each of these has pod
documentation that passes through pod2man without complaint. All package
names are updated to reflect this structure, and the "standard" sub-classes
for hashes and scalars are named Tie::StdHash and Tie::StdScalar, in
keeping with the group consensus.
With this, the documentation for Tie::Hash should be fine, Tie::Scalar
exists and is documented, and Tie::SubstrHash (formerly just SubstrHash)
is also documented, albeit weakly. I have no examples of usage to show, and
I suspect a possible bug or two, but without examples it's hard to say.
As it happens, this patch also impacts the DBM-style extensions, as they
(DB_File, NDBM_File, GBM_File, ODBM_File, SDBM_File) all inherited from
the old TieHash.pm as a base class.
|
| |
|
| |
|
| |
|
|
|
|
|
|
| |
This moved from pod/PerlDoc/Functions.pm. It's used by the buildtoc
script, and might be used by a future html2pod script. Or not. It
needed a home somewhere, and Pod::Functions.pm seemed sensible.
|
|
|
|
|
| |
I have replaced the version in 5.002 beta 3 with GetoptLong-2.1.
This has some nice new features and has been extensively tested.
|
|
|
|
| |
The following patch fixes a warning from File::Path. Apply over beta3.
|
| |
|
| |
|
|
|
|
|
|
|
| |
Updated for MakeMaker 5.21. This version handles local customization
of installation directories in a much easier-to-manipulate manner.
This is an important improvement and made it worthwhile to upgrade
from the version in 5.002 beta 3.
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
| |
Here is a correction for autoloader. Note that one cannot
use vars '$package::var';
, alas.
|
|
|
|
|
|
| |
Fix 2 installman bugs:
* If the old manpage existed, it was removed even if $pod2man failed.
* If installman aborted, an incomplete manpage was left in $MANPATH.
|
|
|
|
|
|
| |
Add comment at tzname.
Add -DSTRUCT_TM_HASZONE flag for POSIX.
|