Rafael Garcia-Suarez [Thu, 20 Apr 2006 20:24:46 +0000]
Add test boilerplate and use test.pl instead of Test::More
p4raw-id: //depot/perl@27923
Rafael Garcia-Suarez [Thu, 20 Apr 2006 20:22:23 +0000]
Add new tests for bug #32840 provided by David Landgren,
as a new file rxcode.t (they mostly test $^R for now)
p4raw-id: //depot/perl@27922
Nicholas Clark [Thu, 20 Apr 2006 18:51:02 +0000]
Tweak change 23256 to continue passing on 5.8.x
p4raw-id: //depot/perl@27919
Nicholas Clark [Thu, 20 Apr 2006 17:37:31 +0000]
Convert two ok()s to better constructions.
p4raw-id: //depot/perl@27918
Jerry D. Hedden [Wed, 19 Apr 2006 05:32:14 +0000]
[REPOST] [PATCH] Remove superfluous counter in threads
From: "Jerry D. Hedden" <jerry@hedden.us>
Message-ID: <
20060419053214.fb30e530d17747c2b054d625b8945d88.4606123135.wbe@email.secureserver.net>
p4raw-id: //depot/perl@27917
Nicholas Clark [Thu, 20 Apr 2006 13:54:45 +0000]
Fix typo in the description of change 21323.
p4raw-id: //depot/perl@27913
Rafael Garcia-Suarez [Thu, 20 Apr 2006 12:59:50 +0000]
New test for the "Assignment to both a list and a scalar" error,
suggested by Stephen McCamant
p4raw-id: //depot/perl@27912
Dave Mitchell [Thu, 20 Apr 2006 12:33:09 +0000]
regmatch(): replace magic value -1000 with symbolic CHRTEST_VOID
p4raw-id: //depot/perl@27911
Andy Lester [Wed, 19 Apr 2006 23:16:53 +0000]
op_type shrinking
Message-ID: <
20060420041653.GA28523@petdance.com>
p4raw-id: //depot/perl@27910
Andy Lester [Wed, 19 Apr 2006 12:35:43 +0000]
regexec consting & localizing
Message-ID: <
20060419173543.GA8268@petdance.com>
p4raw-id: //depot/perl@27909
Andy Lester [Wed, 19 Apr 2006 23:44:01 +0000]
Localizing vars
Message-ID: <
20060420044401.GA2792@petdance.com>
p4raw-id: //depot/perl@27908
Dave Mitchell [Thu, 20 Apr 2006 00:30:53 +0000]
regmatch: merge the greedy and non-greedy branches of CURLYM
p4raw-id: //depot/perl@27907
Linda Walsh [Mon, 10 Apr 2006 18:53:46 +0000]
[perl #38892] PATCH: remove spurious warning from Benchmark.pm
From: Linda Walsh (via RT) <perlbug-followup@perl.org>
Message-ID: <rt-3.0.11-38892-132143.0.
516271519365361@perl.org>
p4raw-id: //depot/perl@27906
Steve Peters [Wed, 19 Apr 2006 16:32:26 +0000]
Changes to perlfaq8 "How do I find out if I'm running interactively
or not?" suggested by Slaven Rezic in RT #38901: perlfaq8 correction
p4raw-id: //depot/perl@27905
Steve Peters [Wed, 19 Apr 2006 14:57:36 +0000]
Add a TODO test for RT #1085: "what should be output of perl -we
'print do { foreach (1, 2) { 1; } }'"
p4raw-id: //depot/perl@27904
Dave Mitchell [Wed, 19 Apr 2006 13:56:07 +0000]
regmatch(): make IFMATCH use PUSH_STACK rather than fake recursion
p4raw-id: //depot/perl@27903
Rafael Garcia-Suarez [Wed, 19 Apr 2006 11:38:11 +0000]
Upgrade to Test::Harness 2.57_05
p4raw-id: //depot/perl@27902
Nicholas Clark [Wed, 19 Apr 2006 11:20:08 +0000]
newCONSTSUB needs to be robust in case CopFILE is NULL.
p4raw-id: //depot/perl@27901
Nicholas Clark [Wed, 19 Apr 2006 09:58:19 +0000]
Oops. Need a macro to convert cv_ckproto() to cv_ckproto_len().
p4raw-id: //depot/perl@27900
Rafael Garcia-Suarez [Wed, 19 Apr 2006 09:34:48 +0000]
There's a macro to get ${^RE_DEBUG_FLAGS} when debugging,
so use it (and this fixes a valgrind memory access error)
p4raw-id: //depot/perl@27899
Nicholas Clark [Wed, 19 Apr 2006 09:22:03 +0000]
Avoid temporarily writing over the prototype when reporting an error.
(And beef up the relevant tests to really check that it all works).
p4raw-id: //depot/perl@27898
Rafael Garcia-Suarez [Wed, 19 Apr 2006 09:09:14 +0000]
Replace some Copy() by Move() calls, because valgrind reports
we can have overlapping memory areas here
p4raw-id: //depot/perl@27897
Nicholas Clark [Wed, 19 Apr 2006 08:42:18 +0000]
Calling cv_undef() on the CV created by newCONSTSUB() would leak like
a Jumblie's preferred maritime craft. To free CvFILE for this case,
take advantage of the 0 length prototype that will also be there,
and hang it from the prototype. To do this properly means changing
code to actually pay attention to SvCUR() on prototypes. It turns out
that we always know the length of the prototype string, so this may
be faster. Certainly, it's a memory saving (even ignoring the leak).
p4raw-id: //depot/perl@27896
Nicholas Clark [Wed, 19 Apr 2006 07:55:18 +0000]
Fix compilation of microperl, which doesn't have INT_MAX.
p4raw-id: //depot/perl@27895
Nicholas Clark [Tue, 18 Apr 2006 19:52:19 +0000]
Revert 27869 at Jarkko's request.
p4raw-id: //depot/perl@27894
Dave Mitchell [Tue, 18 Apr 2006 17:03:15 +0000]
regmatch: fix a || vs | typo in CACHEsayYES and add a test from Hugo.
Also defer altering the cache bit until CACHEsayYES/NO is called
p4raw-id: //depot/perl@27893
Steve Peters [Tue, 18 Apr 2006 16:47:58 +0000]
Add a TODO regarding merging the win32 and wince codebases.
p4raw-id: //depot/perl@27892
Rafael Garcia-Suarez [Tue, 18 Apr 2006 16:05:15 +0000]
When checking whether perl is built with -g in optimisation
flags, use a more tolerant regexp, because we can now have
things like '-O2 -g'.
p4raw-id: //depot/perl@27891
Nicholas Clark [Tue, 18 Apr 2006 16:03:10 +0000]
Expand "am I hot or not" into a more general "Profile Perl" section.
p4raw-id: //depot/perl@27890
Steve Peters [Tue, 18 Apr 2006 15:47:48 +0000]
A fix to change #27845 to get it to compile with -DPERL_MAD.
p4raw-link: @27845 on //depot/perl:
4199688e3d699f53e17448d3bad6e93e57d49dcc
p4raw-id: //depot/perl@27889
Jan Dubois [Mon, 17 Apr 2006 01:48:34 +0000]
win32/Makefile changes for Windows 64 bit
From: "Jan Dubois" <jand@ActiveState.com>
Message-ID: <
036401c661fb$bc7a1200$2217a8c0@candy>
and make similar changes to win32/makefile.mk
and delete a spurious extra line from that file
p4raw-id: //depot/perl@27888
Rafael Garcia-Suarez [Tue, 18 Apr 2006 14:02:23 +0000]
Addition to README.riscos suggested by Nicholas
p4raw-id: //depot/perl@27887
Nicholas Clark [Tue, 18 Apr 2006 13:18:27 +0000]
Note that Rafael has done the first cut of state variables.
Document that UTF-8 cleansing the tokeniser and the pads is TODO.
p4raw-id: //depot/perl@27886
Nicholas Clark [Tue, 18 Apr 2006 13:10:35 +0000]
Relocatable @INC and parallel builds pretty much work now.
p4raw-id: //depot/perl@27885
Alex Waugh [Mon, 17 Apr 2006 00:14:26 +0000]
Support compiling for RISC OS
Message-ID: <
2498b2184e.ajw498@caramel.cp15.org>
p4raw-id: //depot/perl@27884
Nicholas Clark [Tue, 18 Apr 2006 11:18:12 +0000]
Coverity still thinks that there is a route through do_vop that can
leak resources. I believe that it's spotted that you can skip all the
cases in the switch. Plug that hole.
p4raw-id: //depot/perl@27883
Nicholas Clark [Tue, 18 Apr 2006 10:54:15 +0000]
Coverity insists that if we deference a pointer without checking, and
subsequently check that same pointer before deferencing it, something
in our code or our logic is bogus. So assert() that it's safe the
first time and remove the second check.
p4raw-id: //depot/perl@27882
Nicholas Clark [Mon, 17 Apr 2006 20:50:34 +0000]
Fix an error, spotted by Tim Bunce.
p4raw-id: //depot/perl@27881
Nicholas Clark [Mon, 17 Apr 2006 20:41:00 +0000]
Coderefs in @INC are done and documented.
p4raw-id: //depot/perl@27880
Nicholas Clark [Mon, 17 Apr 2006 20:35:52 +0000]
Document all the return values that code references in @INC can return.
p4raw-id: //depot/perl@27879
Steve Peters [Mon, 17 Apr 2006 20:05:03 +0000]
Patch to fix RT #31227: perldoc -f, -q ignores extra arguments with no notice.
Never made it to the mailing list though, so, was never found. Patch
previously provided to the Pod-Perldoc CPAN RT queue.
p4raw-id: //depot/perl@27878
Nicholas Clark [Mon, 17 Apr 2006 19:12:23 +0000]
Coverity is upset about a signed int in ext/Filter/Util/Call/Call.xs,
but the real problem is down in Perl_filter_read() in toke.c.
p4raw-id: //depot/perl@27877
Nicholas Clark [Mon, 17 Apr 2006 18:30:59 +0000]
Coverity is a persistent beast. Hot on the tails of fixing one leak,
it notices a slightly earlier leak.
(Which also suggests that marking bugs as RESOLVED isn't a great plan,
as it's not clear whether that calls off the dogs of war. I'd much
prefer it to be forced to scan again, and give a positve "all clear"
(or more accurately "nowt wrong I can see with that"))
p4raw-id: //depot/perl@27876
Nicholas Clark [Mon, 17 Apr 2006 16:52:54 +0000]
Turn on UTF8 cache assertions with -Ca
p4raw-id: //depot/perl@27875
Nicholas Clark [Mon, 17 Apr 2006 16:52:07 +0000]
Oops. Change 27826 had an error in the definition of BGET_PADOFFSET().
p4raw-id: //depot/perl@27874
Steve Peters [Mon, 17 Apr 2006 16:49:51 +0000]
Upgrade to CGI.pm-3.19.
p4raw-id: //depot/perl@27873
Nicholas Clark [Mon, 17 Apr 2006 15:33:21 +0000]
Note that PVIOs probably aren't worth shrinking. Note 3 possible
subflags for -C if the two Unicode TODOs are done.
p4raw-id: //depot/perl@27872
SADAHIRO Tomoyuki [Sun, 16 Apr 2006 18:45:37 +0000]
Re: [PATCH] update to pp_complement() via Coverity
Message-Id: <
20060416184044.4ADB.BQW10602@nifty.com>
Date: Sun, 16 Apr 2006 18:45:37 +0900
Subject: Re: [PATCH] update to pp_complement() via Coverity
From: SADAHIRO Tomoyuki <bqw10602@nifty.com>
Message-Id: <
20060417230614.20A5.BQW10602@nifty.com>
Date: Mon, 17 Apr 2006 23:06:21 +0900
p4raw-id: //depot/perl@27871
Nicholas Clark [Mon, 17 Apr 2006 14:48:13 +0000]
In Change 27764 I forgot to at PL_hints_mutex to the list of symbols to
exclude when not compiling with ithreads. It would be good to automate
the processing of this.
p4raw-id: //depot/perl@27870
Jarkko Hietaniemi [Sun, 16 Apr 2006 12:26:48 +0000]
(tentative) regcomp.c: try convincing Coverity that data ptr in study_chunk is non-NULL when needed
Message-Id: <
20060416092648.812D36CF2D@aprikoosi.hut.fi>
Date: Sun, 16 Apr 2006 12:26:48 +0300 (EEST)
p4raw-id: //depot/perl@27869
Nicholas Clark [Mon, 17 Apr 2006 13:15:34 +0000]
Note in av_store() that S_regclass relies on being able to pass NULL.
p4raw-id: //depot/perl@27868
Nicholas Clark [Mon, 17 Apr 2006 13:14:55 +0000]
Add assertions to cover cases where the Coverity scanner thinks we
could are deferencing a NULL pointer, but we think it missed something
it should have been capable of figuring out.
p4raw-id: //depot/perl@27867
Nicholas Clark [Mon, 17 Apr 2006 12:48:43 +0000]
Add Dave's explainations of why certain pointers are always non-NULL,
and assert() that it is true.
p4raw-id: //depot/perl@27866
Yitzchak Scott-Thoennes [Sun, 16 Apr 2006 18:05:20 +0000]
fix cygwin build
Message-ID: <
20060417010520.GA2888@efn.org>
p4raw-id: //depot/perl@27865
H.Merijn Brand [Mon, 17 Apr 2006 12:18:07 +0000]
Support for -DDEBUGGING and its alias -DEBUGGING
p4raw-id: //depot/perl@27863
Nicholas Clark [Mon, 17 Apr 2006 12:07:23 +0000]
Revert change 27818, as change 27853 fixes the real bug.
p4raw-id: //depot/perl@27862
Paul Marquess [Mon, 17 Apr 2006 10:58:30 +0000]
Compress::Zlib
From: "Paul Marquess" <Paul.Marquess@ntlworld.com>
Message-ID: <
019b01c66205$7dc7ea50$6601a8c0@myopwv.com>
p4raw-id: //depot/perl@27861
Andy Lester [Sun, 16 Apr 2006 23:40:35 +0000]
pp_ctl.c caching
Message-ID: <
20060417044035.GA28818@petdance.com>
Date: Sun, 16 Apr 2006 23:40:35 -0500
p4raw-id: //depot/perl@27860
Jarkko Hietaniemi [Mon, 17 Apr 2006 10:19:37 +0000]
dooop.c: the strong asserts in Sv* macros could cause memory leakage -- move the macro calls earlier (Coverity CID 84)
Message-Id: <
20060417071937.C13346CF2D@aprikoosi.hut.fi>
Date: Mon, 17 Apr 2006 10:19:37 +0300 (EEST)
p4raw-id: //depot/perl@27859
Nicholas Clark [Mon, 17 Apr 2006 10:28:40 +0000]
Revert 27856.
p4raw-id: //depot/perl@27857
Jarkko Hietaniemi [Sun, 16 Apr 2006 11:19:25 +0000]
The danger of piping an mbox to patch is that it contains more than
one message. So both:
Subject: [PATCH] doop.c: one more code path where memory could leak (Coverity)
Message-Id: <
20060416081925.680336CF2D@aprikoosi.hut.fi>
Date: Sun, 16 Apr 2006 11:19:25 +0300 (EEST)
and
Subject: [PATCH] doop.c: one more code path where memory could leak (Coverity)
From: jhi@cc.hut.fi (Jarkko Hietaniemi)
Message-Id: <
20060416081925.680336CF2D@aprikoosi.hut.fi>
Date: Sun, 16 Apr 2006 11:19:25 +0300 (EEST)
p4raw-id: //depot/perl@27856
Dave Mitchell [Sun, 16 Apr 2006 23:30:36 +0000]
generalize the 'continue on success' mechanism of deferred REs
p4raw-id: //depot/perl@27855
Dave Mitchell [Sun, 16 Apr 2006 23:12:31 +0000]
abstract regmatch slab access with SLAB_FIRST and SLAB_LAST macros
p4raw-id: //depot/perl@27854
Dave Mitchell [Sun, 16 Apr 2006 23:07:13 +0000]
stop PL_op and PL_curpad referring to different pads at start of doeval
(this could make find_uninit_var do Bad Things)
p4raw-id: //depot/perl@27853
Nicholas Clark [Sun, 16 Apr 2006 21:44:59 +0000]
Fix bug introduced with change 27841. (len + 1, not just len).
p4raw-id: //depot/perl@27852
Nicholas Clark [Sun, 16 Apr 2006 19:06:19 +0000]
Coverity is flagging a potential problem because it sees a check for
NULL and assumes that this means that the variable host could be
NULL. It can't, and the check added in change 13291 was a little bit
more than the minimal solution needed for the bug report
[ID
20011126.148]
Hence remove the NULL check.
p4raw-id: //depot/perl@27851
Andy Lester [Fri, 14 Apr 2006 23:29:24 +0000]
op.c patch, 2nd version
Message-ID: <
20060415042924.GA1786@petdance.com>
Date: Fri, 14 Apr 2006 23:29:24 -0500
p4raw-id: //depot/perl@27850
Nicholas Clark [Sun, 16 Apr 2006 15:04:36 +0000]
A scalar reference returned from a coderef in @INC is treated as the
initial "content" of the file. When it is exhausted input is taken
from a real file handle, or a generator sub, if either exists.
p4raw-id: //depot/perl@27849
Jan Dubois [Sun, 16 Apr 2006 02:05:00 +0000]
remove some cruft from makedef.pl
From: "Jan Dubois" <jand@ActiveState.com>
Message-ID: <
022e01c66134$d7a8dec0$2217a8c0@candy>
p4raw-id: //depot/perl@27848
Jan Dubois [Sun, 16 Apr 2006 00:57:00 +0000]
testcase for change #27832
From: "Jan Dubois" <jand@ActiveState.com>
Message-ID: <
022a01c6612b$575cb7e0$2217a8c0@candy>
p4raw-id: //depot/perl@27847
Steve Peters [Sun, 16 Apr 2006 13:44:47 +0000]
Stray pod/perlapi.pod change found after a regen.
p4raw-id: //depot/perl@27846
Andy Lester [Sun, 16 Apr 2006 00:29:36 +0000]
dump.c patch redux
Message-ID: <
20060416052936.GA19143@petdance.com>
p4raw-id: //depot/perl@27845
Nicholas Clark [Sun, 16 Apr 2006 13:25:18 +0000]
Perl_do_vop can profitably use Perl_sv_usepvn_flags, as it has
allocated space for an initialised a trailing NUL.
p4raw-id: //depot/perl@27843
Nicholas Clark [Sun, 16 Apr 2006 12:06:57 +0000]
In Perl_sv_usepvn_flags, assert() that the caller has RTFM.
p4raw-id: //depot/perl@27842
Nicholas Clark [Sun, 16 Apr 2006 11:35:38 +0000]
Provide a new flag for sv_usepvn_flags, SV_HAS_TRAILING_NUL, which
signals that the buffer passed in is already well-formed for use as
SvPVX (i.e. it is followed by a trailing NUL), and hence the realloc()
to add a NUL can be skipped.
p4raw-id: //depot/perl@27841
Nicholas Clark [Sun, 16 Apr 2006 11:03:28 +0000]
Merge sv_usepvn and sv_usepvn_mg into sv_usepvn_flags. "Promote" the
other two to mathoms.c
p4raw-id: //depot/perl@27840
Nicholas Clark [Sun, 16 Apr 2006 10:37:46 +0000]
In Perl_sv_usepvn(), force a move under DEBUGGING to shake out caller
bugs.
p4raw-id: //depot/perl@27839
Nicholas Clark [Sun, 16 Apr 2006 10:37:04 +0000]
Fix htmllink.t to cater for the case where perladmin isn't an empty
string.
p4raw-id: //depot/perl@27838
Nicholas Clark [Sun, 16 Apr 2006 09:36:18 +0000]
Coverity notes that we might be dereferencing fgv before a check that
it's not NULL. In fact, the code ordering meant that one "if (fgv)"
would always be true. So fix this.
p4raw-id: //depot/perl@27837
Andy Lester [Sun, 16 Apr 2006 00:14:23 +0000]
update to pp_complement() via Coverity
Message-ID: <
20060416051423.GA17063@petdance.com>
p4raw-id: //depot/perl@27836
Andy Lester [Sat, 15 Apr 2006 23:18:27 +0000]
clean up hv_assert()
Message-ID: <
20060416041827.GA3031@petdance.com>
p4raw-id: //depot/perl@27835
Andy Lester [Sat, 15 Apr 2006 23:32:53 +0000]
Removed unused var in a macro
Message-ID: <
20060416043253.GA3438@petdance.com>
p4raw-id: //depot/perl@27834
Jan Dubois [Fri, 14 Apr 2006 13:13:20 +0000]
Change anchor generation in Pod::Html for "=item item 2"
From: "Jan Dubois" <jand@ActiveState.com>
Message-ID: <
017a01c65fff$dfd0f370$2217a8c0@candy>
p4raw-id: //depot/perl@27833
Jan Dubois [Mon, 10 Apr 2006 15:32:23 +0000]
Pod::Html::depod() doesn't work correctly for multi-line strings
From: "Jan Dubois" <jand@ActiveState.com>
Message-ID: <
017b01c65cee$a2f5fe00$6062a8c0@candy>
p4raw-id: //depot/perl@27832
Jan Dubois [Thu, 13 Apr 2006 18:26:29 +0000]
Missing part of change 25186
From: "Jan Dubois" <jand@ActiveState.com>
Message-ID: <
009401c65f62$74fded70$2217a8c0@candy>
p4raw-id: //depot/perl@27831
Jan Dubois [Thu, 13 Apr 2006 18:19:26 +0000]
Missing part of change 25676
From: "Jan Dubois" <jand@ActiveState.com>
Message-ID: <
009001c65f61$78866d60$2217a8c0@candy>
p4raw-id: //depot/perl@27830
bah@longitude.com [Sat, 15 Apr 2006 14:13:41 +0000]
Handle null io in leavewrite (not possible?) (Coverity)
Message-ID: <1661.38.96.138.153.
1145124821.squirrel@mail.longitude.com>
Date: Sat, 15 Apr 2006 14:13:41 -0400 (EDT)
p4raw-id: //depot/perl@27829
bah@longitude.com [Sat, 15 Apr 2006 13:54:10 +0000]
Potential (?) memory leak in Storable.xs (Coverity)
Message-ID: <1561.38.96.138.153.
1145123650.squirrel@mail.longitude.com>
Date: Sat, 15 Apr 2006 13:54:10 -0400 (EDT)
p4raw-id: //depot/perl@27828
Nicholas Clark [Sat, 15 Apr 2006 22:44:30 +0000]
Teach Perl_do_sv_dump how to cope with IOs that have bogus things in
GV slots. (Which source filters are prone to do to them).
p4raw-id: //depot/perl@27827
Nicholas Clark [Sat, 15 Apr 2006 21:50:12 +0000]
Coverity was getting upset about an assignment from a function
returning int to a U8 variable. Curiously it was only getting upset
about the first of many.
However, we are doing NO ERROR CHECKING on whether we read in the
requested number of bytes of bytecode. All except BGET_op_tr_array
will now croak on a short read.
p4raw-id: //depot/perl@27826
Nicholas Clark [Sat, 15 Apr 2006 21:43:13 +0000]
Coverity reports that Storable can potentially cause a NULL pointer
dereference while iterating a hash. This should never happen unless
the hash lies about how many keys it has, so croak if a fib is spotted.
p4raw-id: //depot/perl@27825
Nicholas Clark [Sat, 15 Apr 2006 21:11:44 +0000]
Coverity notes that resource are leaked in the unfinished and never
called function _utf8_to_bytes.
p4raw-id: //depot/perl@27824
Nicholas Clark [Sat, 15 Apr 2006 20:55:20 +0000]
Coverity correctly reports that gv might (just) be NULL. So don't
derefernece it if it is.
p4raw-id: //depot/perl@27823
Nicholas Clark [Sat, 15 Apr 2006 20:54:44 +0000]
Update the assertion to a croak "Out of range special SV number"
to give ByteLoader more meaningful failure diagnostics.
p4raw-id: //depot/perl@27822
Nicholas Clark [Sat, 15 Apr 2006 20:19:22 +0000]
I believe that it makes little sense for the simple user filters
provided by coderefs to be run in block mode. Hence always read from
the upstream filter in line mode, and always call the user's filter
with 1 line.
p4raw-id: //depot/perl@27821
Nicholas Clark [Sat, 15 Apr 2006 20:01:58 +0000]
Fix a bug - the introduction of the caching code means that we should
initialise the SV upstream after it.
assert that filter_sub is always non-null now.
p4raw-id: //depot/perl@27820
Nicholas Clark [Sat, 15 Apr 2006 19:37:32 +0000]
Augment the user filter caching code so that if the user filter returns
multiple lines, only one line at a time is returned.
Rename the variable len to status, as it is the status value.
p4raw-id: //depot/perl@27819
Nicholas Clark [Sat, 15 Apr 2006 19:17:51 +0000]
Deep in the bowels of creating new and exciting bugs, I managed to
segfault the uninitialised value reporting code. This protects against
the segfault.
p4raw-id: //depot/perl@27818
Nicholas Clark [Sat, 15 Apr 2006 18:08:04 +0000]
$foo .= $bar doesn't warn if $foo is undefined, so simplify code.
p4raw-id: //depot/perl@27817
Nicholas Clark [Sat, 15 Apr 2006 18:05:12 +0000]
If the downstream caller wants block mode, and we're in line mode,
then don't return more bytes than they asked for. Hold bytes over
until next time if necessary.
p4raw-id: //depot/perl@27816
projects / p5sagit/p5-mst-13.2.git / log