Larry Wall [Mon, 13 Mar 1995 23:55:18 +0000]
cleaner version of the official unofficial patch
: There were some warnings (AIX, xlc compiler):
Here's a cleaner version of the official unofficial patch, based on 5.001.
Larry Wall [Tue, 14 Mar 1995 23:25:31 +0000]
Okay, here's your official unofficial closure leak patch
Larry Wall [Mon, 13 Mar 1995 06:32:14 +0000]
Perl 5.001
[See the Changes file for a list of changes]
Andy Dougherty [Fri, 10 Mar 1995 23:34:12 +0000]
perl5.000 patch.0o: [address] a few more Configure and build nits.
This patch addresses a few more Configure and build nits. Full
details are given below, but the main hightligths are (slightly)
better support for nested extensions and DLD and AIX MakeMaker fixes.
Configure
Detect MachTen. Thanks to Mark Pease <peasem@primenet.com>.
Delete some tabs that caused a MachTen /bin/sh core dump!
Detect extensions nested 1 level deep, e.g. Devel/DProf/DProf.xs
MANIFEST
MANIFEST.new
Include new hints/machten.sh hint file.
Makefile.SH
Document why we use ./makedir instead of mkdir.
U/Extensions.U
Detect extensions nested 1 level deep, e.g. Devel/DProf/DProf.xs
U/dist3_051.pat
Include MachTen patches.
configpm
Convert nested extension names from filesytem-dependent Devel/DProf
to perl5's internal naming scheme Devel::DProf.
doio.c
A dup-related buglet fix from Hallvard B. Furuseth
<h.b.furuseth@usit.uio.no>.
ext/DB_File/DB_File.pm
ext/DynaLoader/DynaLoader.pm
ext/Fcntl/Fcntl.pm
ext/GDBM_File/GDBM_File.pm
ext/POSIX/POSIX.pm
ext/Socket/Socket.pm
Throw a qw() around @ISA elements to show "good style".
hints/machten.sh
new file.
lib/ExtUtils/MakeMaker.pm
Bump version number to 4.086.
Fix AIX buglet -- needed to specify NAME.
Linux/DLD/gcc-2.6.2: We no longer load .sa libraries (except
libm.sa, which is apparently still o.k.
util.c
Another dup-related buglet fix.
Andy Dougherty [Sun, 5 Mar 1995 22:11:42 +0000]
perl5.000 patch.0n: [address Configure and build issues]
This is my patch patch.0n for perl5.000.
This _very_ small patch
1. updates the linux and dec_osf hints files,
2. adds sv_isobject to global.sym,
3. updates Configure to deal with recent Linux nm output, and
4. fixes the names in File::Path.
This patch addresses only Configure and build issues for which I have
tested solutions. It does nothing else. Maybe some of the other patches
floating around should be included. Maybe not. I'm afraid I just
don't have time to think about them now.
Unless something's broken, I hope not to issue any more patches :-)
(Yes, I've said that before, but this one's _really_ small, and linux
support was broken.:-) Thanks to Kenneth Albanowski for researching,
implementing, and testing the Linux patch.
Andy Dougherty [Mon, 27 Feb 1995 22:35:59 +0000]
perl5.000 patch.0m: [various fixes, hint file updates and documentation]
This is my patch patch.0m for perl5.000.
This patch fixes all remaining problems that I am aware of, and for
which I have a solution. It also updates some hint files and
documentation.
Here's what's new:
Configure
Protect against spaces in uname -m output (unicos).
Look in <stdlib.h> for malloctype and freetype.
Check if user has void free() or int free().
Look in linux/signal.h for signal names.
MANIFEST
MANIFEST.new
Two new hint files: cxux.sh and PowerUNIX.sh.
Sorted.
README
Indicate what gets installed and where it usually goes.
Thanks to Hallvard B. Furuseth <h.b.furuseth@usit.uio.no>
for suggesting this.
U/Myinit.U
Update extliblist comment.
U/dist3_051.pat
This file contains patches to dist 3 (PL 51) that I used to generate
Configure for perl.
U/mallocsrc.U
Look in <stdlib.h> for malloctype and freetype.
Check if user has void free() or int free().
config_h.SH
config.H
Add Free_t to handle void free() vs. int free().
ext/DynaLoader/README
Updated comment.
ext/POSIX/POSIX.pm
creat() has 2 arguments, not 3 (thanks, Paul).
ext/POSIX/POSIX.xs
Fix return type of lseek.
ext/SDBM_File/sdbm/sdbm.h
Add I_STDLIB guard on #include <stdlib.h>
ext/util/extliblist
Add note indicating this is obsolete. Don't remove it because
people might be using it for their own private extensions.
hints/PowerUNIX.sh
hints/cxux.sh
New files. Written by Tom.Horsley@mail.hcsc.com
hints/linux.sh
Simplified.
lib/ExtUtils/MakeMaker.pm
Typo fixed, only affected aix?
malloc.c
Allow for possible int free().
perl.h
pp_sys.c
util.c
If the user is not using vfork, move the #define vfork fork
util after various #include files. Since vfork() and fork() might
have different prototypes, the #define could cause a conflict in
system header files. (Reported for 386bsd.)
Makefile.SH
make realclean will remove h2xs and makeaperl (but leave behind
the .SH versions, of course).
Andy Dougherty [Tue, 21 Feb 1995 01:14:24 +0000]
perl5.000 patch.0l: MakeMaker 4.085 and upgrade Configure to dist3 PL 51.
Here's what's new:
Configure
Generated by metaconfig PL 51.
Correctly set ./mips on a MIPS system.
Improved (we hope) handling of $archname.
MANIFEST
MANIFEST.new
Made .SH files out of h2xs and makeaperl so that they get the
correct path-to-perl at the top.
Makefile.SH
Propagate $(perllib) to extensions.
U/dist3_051.patches
Two patches to apply on top of metaconfig PL 51. I've sent
them off for inclusion in the next metaconfig update.
config_h.SH
config.H
Regenerated. Only the order of elements has changed.
ext/DB_File/Makefile.PL
ext/GDBM_File/Makefile.PL
ext/NDBM_File/Makefile.PL
Add -L/usr/local/lib to LIBS variable.
ext/POSIX/POSIX.xs
Place #ifdef around FD_CLOEXEC (needed for Apollo).
ext/SDBM_File/sdbm/Makefile.PL
Simplified, thanks to MakeMaker enhancements.
ext/util/make_ext
Pass through $(perllib) argument for cflags.
h2xs.SH
Changed from h2xs to h2xs.SH. Now finds correct path to perl.
hints/next_3_2.sh
Updated for hppa.
hints/solaris_2.sh
Remove potentially problematic -lmalloc from $libswanted.
hints/unicos.sh
Look in /usr/include/rpcsvc for dbm.h.
installperl
Install h2xs.
lib/Cwd.pm
Use 'my' variable to avoid clobbering $_.
lib/ExtUtils/MakeMaker.pm
Upgraded from 4.06 to 4.085.
Lots of documentation improvements.
EXE_FILES to refer to an array of executable files to install.
Reduce chatter during build process.
Don't count a useful -L/path option as a successful search for
a particular library.
Cleanup handling of aix external symbols.
Create/update perllocal.pod to indicate what we've done.
makeaperl.SH
Changed from makeaperl to makeaperl.SH. Now finds correct path
to perl.
x2p/util.c
Delete unused setenv() and envix() functions.
x2p/util.h
Delete unused setenv() and envix() prototypes.
vms/config.vms
Define I_SYS_STAT and I_SYS_TYPES.
Andy Dougherty [Sat, 11 Feb 1995 01:17:38 +0000]
perl5.000 patch.0k: MakeMaker 4.06 and to fix minor portability and build problems reported even after patches 0a through 0j
MakeMaker 4.06 allows you to build extensions away from the source
tree with either static or dynamic loading.
In a rare act of prescience, I've also fixed some un-reported bugs.
Specifically, there were several places where Configure said you could
specify things using ~name notation, but, in fact, you couldn't.
In detail, here's what's included:
Configure
Check I_SYS_TYPES for x2p/a2p.h
Improve and generalize $osvers detection for DEC Alpha
(now will work even for osvers > 3.)
No longer override hint-file setting of $archname.
Don't tell users ~name is ok for Dynamic loading file. It's not.
MANIFEST
MANIFEST.new
Updated.
Makefile.SH
Some trailing ' ' removed from lines.
New target lib/ExtUtils/Miniperl.pm built. This stashes away
miniperlmain.c in the library so new static extensions can be
built away from the source tree.
Minor cleanup.
U/Oldconfig.pat.2
This is a patch to be applied against dist-PL 50 to upgrade
the DEC OSF/1 version detection.
U/archlib.U
Preserve previous value for $archname. Otherwise this is
identical to the unit in dist-PL 50.
U/dlsrc.U
Users may not use ~name notation to find the dynamic loading
module. (Back in early alpha days they could, but that hasn't
worked since the DynaLoader module was introduced.
config.H
Updated.
config_h.SH
Updated.
hints/dec_osf.sh
Updated. Simplified. Don't use ld -no_archive (at least as
the default). It only worked because some versions *ignored* it.
hints/mpeix.sh
Add a few comments. I should have added more.
hints/next_3_0.sh
New hint file from Kevin White <klwhite@magnus.acs.ohio-state.edu>
hints/ultrix_4.sh
Separate out flags not appropriate for gcc.
installperl
Install sperl.o.
lib/ExtUtils/MakeMaker.pm
Upgraded from 4.03 to 4.06. Many improvements. Now possible
to build and install new extensions outside the source tree,
for both static and dynamic loading.
lib/File/Path.pm
New. Creates or removes a series of directories
makeaperl
New utility to create a new perl binary from static extensions
minimod.PL
New. minimod.PL writes the contents of miniperlmain.c into the
module ExtUtils::Miniperl for later perusal (when the perl
source is deleted)
perl.c
ARCHLIB and PRIVLIB changed to ARCHLIB_EXP and PRIVLIB_EXP,
since perl is not prepared to deal with ~name expansion. The
_EXP variables are pre-expanded by Configure.
proto.h
NeXt 3.0 couldn't handle the #ifdef __attribute line.
It said 'illegal #ifdef'.
vms/config.vms
s/ARCHLIB/ARCHLIB_EXP/;
s/PRIVLIB/PRIVLIB_EXP/;
Add in I_SYS_STAT and I_SYS_TYPES, since the source now looks
for them.
vms/ext/MM_VMS.pm
New file.
x2p/a2p.h
Include <sys/types.h>
Andy Dougherty [Tue, 7 Feb 1995 01:51:12 +0000]
perl5.000 patch.0j: fix minor portability and build problems remaining even after patches 0a through 0i
Specifically, here's what's included:
Configure
Regenerated with metaconfig patchlevel 50. This changed
a variety of things, mostly related to selecting and changing
the installation prefix.
Handle csh, sed, and byacc no matter what the setting of
d_portable. (This was causing glob problems in patch.0i).
Set d_portable to default to 'y'. It doesn't matter anyway,
but gives people a warm fuzzy feeling.
Remove useless d_group and d_passwd tests.
Add check for <sys/stat.h>.
Improve & generalize AIX version detection.
Consider /opt/man/man1 as a possible place to install man pages.
Be a little more robust about OS version changes when deciding
if the output of uname -a has really changed.
MANIFEST
MANIFEST.new
Added hints/mpeix.sh.
README
Tell users the Configure defaults are probably right.
Makefile.SH
Better detection of whether user has byacc.
Use $(MAKE) instead of make.
U/Loc_sed.U
Works again with d_portable='define'.
U/Myinit.U
Set d_portable=define as default.
U/d_byacc.U
Detect whether user has byacc even if d_portable=define.
U/d_csh.U
Works again with d_portable='define'.
U/d_group.U
Empty file to avoid useless metaconfig test.
U/d_passwd.U
Empty file to avoid useless metaconfig test.
U/dist.patch
This file contains two minor updates to dist3 PL50 that were used
to generage Configure.
U/i_sysstat.U
New test. See if sys/stat.h is available.
config.H
Updated.
config_h.SH
Updated to metaconfig patchlevel 50.
ext/NDBM_File/Makefile.PL
ext/ODBM_File/Makefile.PL
Add -lucb for SVR4 systems.
handy.h
Protect agains g++-2.6.3, which predefines bool. g++ can be
used to compile an extension, but not perl itself. Still, the
extension will #include "perl.h", which eventually gets
"handy.h", which #define's bool. If this happens to you, add
-DHAS_BOOL to your ccflags in your extension, or else ensure that
_G_config.h is #include'd before perl.h. (_G_config.h will define
_G_HAVE_BOOL, if indeed your version of g++ has bool.)
hints/aix.sh
Updated. Handles AIX 3.2.x and 4.1. Comments included!
hints/hpux_9.sh
Updated.
hints/irix_4.sh
Updated. Includes comments for IRIX 4.0.4
hints/linux.sh
Updated. Beginnings of ELF support added, but completely
untested.
hints/mpeix.sh
New hint file.
hints/solaris_2.sh
Useless ccflags="$ccflags" line removed.
hints/svr4.sh
Updated.
installperl
Doesn't use Config anymore (it already reads config.sh
directly. That's probably backwards, but, oh well.
Install perl.exp for AIX.
lib/ExtUtils/MakeMaker.pm
Upgraded from 4.01 to 4.03.
makedepend.SH
Use $MAKE instead of plain make.
Index: op.c
Remove overlapping strcpy().
perl.h
Add test for <sys/stat.h>.
Delete unused VOIDSIG stuff.
Delete unused typedef struct lstring Lstring;
perl_exp.SH
Add safexxxx calls.
pp_sys.c
Delete wayward break in HAS_ALARM section.
proto.h
Change true and false (!) in function prototypes to please
g++-2.6.3, which has true and false built in. (See notes for
handy.h.)
Index: unixish.h
Long-overdue housekeeping.
HAS_GROUP and HAS_PASSWD are always defined.
util.c
Yet another (char*) cast for bcmp.
vms/config.vms
Changed comments to match unixish.h.
writemain.SH
Now correctly handles nested static extensions. Recent
MakeMakers have moved where they get built.
x2p/a2p.h
More definitions that will doubtless cause trouble somewhere
else.
x2p/a2py.c
x2p/walk.c
Remove unprotected char *strchr();
Andy Dougherty [Thu, 26 Jan 1995 00:40:50 +0000]
perl5.000 patch.0i: fix glaring mistakes in patches a-h
This patch does the following things:
1. Fix various bonehead errors I introduced in patches a-g.
2. Incorporate MakeMaker changes to bring it up to version 4.01 (mostly).
3. Stick in things I forgot in patches a-g (e.g. AIX).
4. Some minor additional cleanup in x2p/ for even pickier compilers.
5. More hints updates (hpux and next).
6. Include newest dl_hpux.xs.
I didn't have time to
1. Fix the overlapping strcpy() in op.c
2. Restore h2xs to Larry's original design to process <>.
3. take out unnecessary "use Config" in installperl.
4. Add in vms patches.
I forgot to
[If I remembered what i forgot, I wouldn't have forgotten it. :]
I deliberately decided *not* to
1. Touch pod/*
2. deal with overloading
Specifically, here's what's included:
Configure
Regenerated to be sure it's up-to-date.
Makefile.SH
Build extension libraries right into lib/auto/whatever.
Don't set CCCDLFLAGS since we don't use it anyway.
Take care to avoid modifying lib/Config.pm without reason
Visit DynaLoader for `make clean'. (Previously only did
so for `make realclean'.)
@echo "Note that make realclean does not delete config.sh"
Include config.h dependency.
U/i_db.U
config_h.SH
config.H
Remove unwanted quotes around db_hashtype and db_prefixtype.
configpm
Allow specification of alternate name for lib/Config.pm,
so the makefile mv-if-diff trick saves needless re-making.
ext/DynaLoader/DynaLoader.pm
Updated warning messages and comments.
ext/DynaLoader/dl_hpux.xs
Updated to version 2.1. Now uses bootstrap files.
ext/util/make_ext
Explicitly use #!/bin/sh to start it up. This is useful
for testing make_ext.
Improve & simplify Nested::Extension::Processing.
More robust handling of `make clean'.
hints/hpux_9.sh
Support both the bundled and unbundled compilers.
hints/next_3_2.sh
Back to using -posix rather than POSIX_SOURCE. And that only
for ext/POSIX/POSIX.xs.
installperl
Special ranlib treatment for NeXT, which gets confused about
timestamps in libraries, even when you just copy the library.
Supply missing '$' in samepath() function.
lib/AutoSplit.pm
New parameters.
lib/ExtUtils/MakeMaker.pm
Upgraded from 3.8 to 4.01.
lib/ExtUtils/xsubpp
Documentation changed from man to pod.
lib/Getopt/Long.pm
Avoid typo warning.
Drop unused $optx.
lib/Text/Tabs.pm
Fix package name.
makedepend.SH
Explicitly start with $startsh.
Catch cpp that prints # <stdin> instead of line numbers.
perl.h
Fix bonehead mistake that ended up calling my_fmod even if
not needed.
perl_exp.SH
also add symbols from interp.sym
proto.h
Delete 2 redundant prototypes (newBINOP and newUNOP).
util.c
Add (char *) casts to unsigned char args to bcmp.
x2p/a2p.h
Rearrange order of <string?.h> and bcopy & bzero stuff.
Change a few function prototypes to void, to reflect actual
usage.
x2p/a2py.c
Change a few function types to void, to reflect actual usage.
x2p/handy.h
Make *alloc declarations match those in x2p/util.c.
x2p/util.c
Make *alloc declarations match those in x2p/handy.h.
x2p/walk.c
Add a (Size_t) cast for comparison of 1 to the result of strlen().
Thanks to all who's work is included here. Little of it is mine.
Larry Wall [Wed, 11 Jan 1995 22:06:42 +0000]
[fix crash in regexec.c]
In article <3ekgoo@giga.bga.com> jamshid@ses.com (Jamshid Afshar) writes:
: I'm getting some unexpected behavior for a small perl script on SunOS
: 4.1.3 perl 4.0p36. The same script crashes under perl 5.0 on HP-UX.
: That script is at the end of this article -- it's as small as I could
: make it and still reproduce the crash.
Here's an unofficial patch for the problem in Perl 5.
Larry
Larry Wall [Fri, 13 Jan 1995 18:28:56 +0000]
"unofficial" patches for some of the more spectacular [memory leaks]
To: Simon Parsons <S.Parsons@fulcrum.co.uk>
: I am on a Sun sparc running Solaris 5.3 / 5.4
:
: Are there any patches available for perl5.000, or a list of know bugs?
: I am having problems with a script running out of memory, which may be
: causes by memory leaks. The process size grows steadily up to approx
: 8Meg (over a couple of minutes) an then grows to approx 22Meg (in 2 or
: 3 seconds) before running out of memory. purify indicates memory
: leaks, but I am not sure whether this is as a result of perl
: abandoning memory as it exits.
5.001 will contain fixes for a number of memory leaks. Here are some
unofficial patches for some of the more spectacular ones. The one
for sv.c is the likeliest one to be affecting you, unless you're doing
a lot of evals.
Larry Wall [Sat, 19 Nov 1994 01:01:34 +0000]
[return values correctly with G_EVAL]
You need this patch for G_EVAL to return values correctly.
Larry Wall [Wed, 11 Jan 1995 19:01:09 +0000]
duplicate DESTROY
In order to fix the duplicate DESTROY bug, I need to remove [the
modified] lines from sv_setsv.
Basically, copying an object shouldn't produce another object without an
explicit blessing. I'm not sure if this will break anything. If Ilya
and anyone else so inclined would apply this patch and see if it breaks
anything related to overloading (or anything else object-oriented), I'd
be much obliged.
By the way, here's a test script for the duplicate DESTROY. You'll note
that it prints DESTROYED twice, once for , and once for . I don't
think an object should be considered an object unless viewed through
a reference. When accessed directly it should behave as a builtin type.
#!./perl
= new main;
= '';
sub new {
my ;
local /tmp/ssh-vaEzm16429/agent.16429 = bless $a;
local = ; # Bogusly makes an object.
/tmp/ssh-vaEzm16429/agent.16429;
}
sub DESTROY {
print "DESTROYED\n";
}
Larry
Andy Dougherty [Wed, 18 Jan 1995 03:44:04 +0000]
perl5.000 patch.0g: [various portability fixes, and use latest metaconfig for Configure]
This patch incorporates various portability fixes and uses the latest
metaconfig to generate Configure (and config_h.SH).
It would take a long time to summarize all that I've changed. I
haven't included many code changes because I'm trying *not* to
duplicate bug fixes Larry may already have applied.
Here's an older description I prepared that's still mostly accurate:
I've also included a few portability fixes in the main source, but
these are certainly not a complete set of everything that's been
reported.
Don't be put off by the size of the patch. Mostly, it's just
rearrangement of the parts in Configure and some cosmetic changes.
Since gcc often supports long long, I had started to add quad support
to Configure. Since SunOS 4.1.3 defines a conflicting "quad"
structure, I changed the name from 'quad' to Quad_t, consistent with
other Configure "types." I also changed "QUAD" to "HAS_QUAD".
However, it turns out it's pretty hard to actually *use* Quad_t.
Neither system I have access to can sprintf() a "long long", nor can
they carry one around in an IV, unless I make IV "long long", which I
didn't want to force generally. Thus I wonder whether any but a
precious few could actually use Quad_t, and dropped the tests from
Configure. I left in the s/quad/Quad_t/ and s/QUAD/HAS_QUAD/ stuff in
case someone else wants to pick it up, and also because I was too lazy
to take it back out :-).
Some highlights:
Configure
Several new options. Use Configure -h to learn more. Also,
read the directions Configure prints. :-)
Spaces now allowed in -D command line options.
New -O option that overrides config.sh.
You can start interactively and then change that to accepting
all the defaults by specifying &-d at any Configure prompt. This
is useful if you have to re-run Configure to only change a few
settings.
Signal type set correctly for the cast{i32,neg} tests.
archname detection improved a bit
guard against ksh users who have set -u
Oldconfig.U cleaned up and regularized a bit more.
Guard against hint files using (and over-writing) $tmp.
Command line options now are processed after metaconfig INIT
lines. Thus things like Configure -Uuseposix should work now.
Various miscellaneous clean-ups.
better use/detection of tr.
i_db.U now checks for hash and prefix type (I think!) I can't
test it here.
i_?db*.U now all check for an associated function before deciding
to include or not the header.
MANIFEST
MANIFEST.new
Sorted & updated.
Makefile.SH
Some shells/makes bombed out on test -d lib/auto || mkdir lib/auto
Use makedir instead.
README
Some additional notes that people won't read :-).
cflags.SH
Now calls $startsh. Weird things were happening on Intergraph,
and this might be related.
config.H
Updated.
config_h.SH
Regenerated.
deb.c
Varargs dependencies on STANDARD_C replaced by I_STDARG.
doop.c
quad stuff.
ext/DB_File/DB_File.xs
Use the new DB_Hash_t and DB_Prefix_t symbols.
ext/SDBM_File/sdbm/sdbm.h
Fix #defines to be more robust.
mg.c
Replace VOIDSIG by metaconfig's Signal_t.
opcode.h
opcode.pl
semop only takes 2 arguments, not 3.
perl.c
Better guard on getenv() prototype. A hint file can use this, if
necessary. Me, I think some compilers are just too picky.
perl.h
The (very) beginnings of some Quad support. See above.
Remove the very troublesome sprintf() prototype. Since we don't
_use_ the return value anyway (since it's not portable) this
should be o.k. The problem was that some systems CAN_PROTOTYPE
but include char *sprintf(); in <stdio.h>. That's incompatible
with the version we used to have in perl.h. Most people have
a prototype for sprintf() in <stdio.h>. Those that don't probably
can get by without it anyway.
Protect the timesbuf by the specific HAS_TIMES test. Some older
gcc-2.something/Solaris 2.something installations apparently don't
have times.
pp.c
More quad stuff.
pp_ctl.c
s/STANDARD_C/I_STDARG/ for varargs stuff.
pp_sys.c
use Signal_t.
proto.h
Update to match new metaconfig names.
util.c
s/STANDARD_C/I_STDARG/ for varargs stuff.
comment out <unistd.h>. A pause prototype was causing problems on
some systems.
vms/config.vms
Changed to use Signal_t.
Andy Dougherty [Wed, 18 Jan 1995 02:44:39 +0000]
perl5.000 patch.0f: [enable metaconfig (PL48) users to regenerate Configure]
This patch enables metaconfig (PL48) users to regenerate Configure. A few
of the units are considerably enhanced from those distributed with
perl5.000. (This patch assumes that you do *not* have the units from
perl5beta3i installed. Sorry if that's an inconvenience.)
In particular, *dbm testing also checks for an appropriate
library function, not just the right header files.
The dynamic loading unit U/dlsrc.U has also been updated to support
SVR4 dynamic loading.
The U/i_db.U unit now checks for the types for the various structure
members that have been causing problems.
Note that at a minimum, you need to be sure you have
the cext='.xs' line from this .package file in your own. That tells
metaconfig that it should look in .xs files (as well as the usual
.[chy] files) for C symbols.
Andy Dougherty [Wed, 18 Jan 1995 02:37:01 +0000]
perl5.000 patch.0e: fix various non-broken things in the x2p/ directory
This patch fixes various non-broken things in the x2p/ directory.
Mostly, I've supplied function prototypes to satisfy particularly
picky compilers.
I've also updated Makefile.SH to know that the byacc-generated a2p.c
is now included with the distribution so that we no longer need to go
looking for yacc/bison/byacc and deal with various library issues or
command line options to support those various compiler compilers.
I've included a2p.c generated by byacc-1.9. Larry, feel free to
use your own from byacc-1.8 instead.
Andy Dougherty [Wed, 18 Jan 1995 02:27:49 +0000]
perl 5.000d : [hint file updates]
This patch consolidates most of the hint file updates I've been
collecting since perl5.000's release. Some of the updates don't make
any sense until you also apply the Configure patches in patch.0g.
(A few late arrivals are also in patch.0g.)
Tim Bunce [Tue, 17 Jan 1995 04:09:00 +0000]
MakeMaker 3.8
Tim Bunce [Thu, 29 Dec 1994 19:47:28 +0000]
MakeMaker 3.7
This patch patches the following:
- lib/ExtUtils/MakeMaker.pm
Most, if not all, the MakeMaker support for no perl source
is now included. Recent ld and mkbootstrap patches applied.
-lX11_s suffix fix applied.
- Makefile.SH
Fix nested module problem which affected make_ext
- ext/DynaLoader/DynaLoader.pm
Change error message to
"Can't load module $module, dynamic loading not available in this perl"
- ext/util/make_ext
A very minor tweak to allow for Deeply::Nested::Modules
- h2xs
Major reorganisation. Mainly aimed at simplifying for later
enhancements. The constant() and AUTOLOAD functions can no
longer be individually enabled or disabled - it never made
any sense - they need each other. Header file parsing code
has been simplified (may allow prototypes to be parsed later).
The .pm file always inherits from AutoLoader.
I hope not to issue another MakeMaker patch till after Perl5.001!
If you want to play with the (as yet untested) no-perl-source mechanism
you'll need to start by doing something like this:
cp ext/xsubpp ext/typemap $(PERL_LIB)/ExtUtils
cp *.h $(PERL_ARCHLIB)/CORE
And then try executing Makefile.PL away from (not under) the perl
source code. You should get a 'Unable to locate perl source' warning
and the PERL_SRC macro will be undefined. Let me know how it goes but
be aware that any problems/fixes are unlikely to turn up in an official
MakeMaker patch till after Perl5.001.
*Please* test this patch and report your findings back to the list so
Larry knows that all is well (or not :-).
Best wishes for a Happy New Year to you all.
Tim Bunce.
Tim Bunce [Mon, 19 Dec 1994 22:27:00 +0000]
This is my patch patch.0a for perl5.000.
[Actually, that's a lie. This is just MakeMaker 3.6. I've just
usurped the letter 'a' to fit it into my patch sequence.]
Andy Dougherty doughera@lafcol.lafayette.edu
Dept. of Physics
Lafayette College, Easton PA
this patch includes:
- My recently posted 'Very small patches to AutoSplit.pm and Cwd.pm'
(with no changes).
- A previous small patch to DynaLoader .bs handling with one addition:
! if (-f $bs) {
! if (-s $bs) { # only read file if it's not empty
- A recently posted patch to hints/aix.sh (with cosmetic changes).
Hopefully no further changes to MakeMaker will be needed before perl5.001.
If any changes are required I intend that they will be release as patches
to be applied over this one. This is the last MakeMaker jumbo patch for
perl5.000.
Patch and enjoy.
Regards,
Tim Bunce.
p.s. I'll be around until about 4pm GMT tomorrow (Tuesday), after that
I'm off for Christmas. This has been a great year for me. I have very
much enjoyed working with the perl5-porters and I wish you all a
wonderful and merry Christmas and a very happy New Year.
Larry Wall [Mon, 17 Oct 1994 23:00:00 +0000]
perl 5.000
[editor's note: this commit combines approximate 4 months of furious
releases of Andy Dougherty and Larry Wall - see pod/perlhist.pod for
details. Andy notes that;
Alas neither my "Irwin AccuTrack" nor my DC 600A quarter-inch cartridge
backup tapes from that era seem to be readable anymore. I guess 13 years
exceeds the shelf life for that backup technology :-(.
]
Larry Wall [Wed, 4 May 1994 23:00:00 +0000]
perl 5.0 alpha 9
[editor's note: the sparc executables have not been included,
and emacs backup files have been removed]
Andy Dougherty [Mon, 4 Apr 1994 00:00:00 +0000]
perl 5.0 alpha 8
[the last one taken from the September '94 InfoMagic CD; a similar
style of cleanup as the previous commits was performed]
Larry Wall [Fri, 18 Mar 1994 00:00:00 +0000]
perl 5.0 alpha 6
[editor's note: cleaned up from the September '94 InfoMagic CD, just
like the last commit]
Larry Wall [Fri, 10 Dec 1993 00:00:00 +0000]
perl 5.0 alpha 5
[editor's note: the sparc executables have not been included,
and emacs backup files and other cruft such as patch backup files have
been removed. This was reconstructed from a tarball found on the
September 1994 InfoMagic CD]
Larry Wall [Fri, 10 Dec 1993 00:00:00 +0000]
perl5a5:pat/inherit.pat
[random patch found in the perl-5alpha5 tarball applied separately]
Larry Wall [Fri, 10 Dec 1993 00:00:00 +0000]
perl5a5:pat/env.pat
[random patch found in the perl-5alpha5 tarball applied separately]
Larry Wall [Wed, 10 Nov 1993 00:00:00 +0000]
perl 5.0 alpha 4
[editor's note: the sparc executables have not been included, and
emacs backup files have been removed. This was reconstructed from a
tarball found on the September 1994 InfoMagic CD; the date of this is
approximate]
Larry Wall [Sun, 10 Oct 1993 00:00:00 +0000]
perl 5.0 alpha 3
[editor's note: the sparc executables have not been included,
and emacs backup files have been removed]
Larry Wall [Thu, 7 Oct 1993 23:00:00 +0000]
perl 5.0 alpha 2
[editor's note: from history.perl.org. The sparc executables
originally included in the distribution are not in this commit.]
Larry Wall [Thu, 4 Feb 1993 22:50:33 +0000]
perl 4.0 patch 36: (combined patch)
Since Ed Barton sent me a patch for the malignent form of "Malformed
cmd links", I finally broke down and made a patch for the various
other little things that have been accumulating on version 4.
Larry Wall [Mon, 22 Jun 1992 23:34:36 +0000]
perl 4.0 patch 35: (combined patch)
Subject: bad interaction between backslash and hyphen in tr///
Among other things, tr/\040-\126/ / was not doing a character range,
due to a earlier botched fix to make \- work right.
Subject: Configure test for presence of nroff was wrong
If Loc doesn't find nroff, it sets $nroff to 'nroff'. The man
page test was tesing against the null string.
Subject: installperl error message printed file mode in decimal, not octal
A real, honest-to-goodnes nit.
Subject: fixed up some filenames in MANIFEST
Erroneously contained "pstruct", omitted hints/isc_3_2_3.sh.
Larry Wall [Thu, 11 Jun 1992 08:21:11 +0000]
perl 4.0 patch 34: (combined patch)
Here's the typical cleanup patch that follows any large
set of patches. My testing organization is either too large
or too small, depending on how you look at it, sigh...
Larry Wall [Mon, 8 Jun 1992 04:52:17 +0000]
perl 4.0 patch 33: patch #20, continued
See patch #20.
Larry Wall [Mon, 8 Jun 1992 04:53:03 +0000]
perl 4.0 patch 32: patch #20, continued
See patch #20.
Larry Wall [Mon, 8 Jun 1992 04:52:59 +0000]
perl 4.0 patch 31: patch #20, continued
See patch #20.
Larry Wall [Mon, 8 Jun 1992 04:52:11 +0000]
perl 4.0 patch 30: patch #20, continued
See patch #20.
Larry Wall [Mon, 8 Jun 1992 04:51:08 +0000]
perl 4.0 patch 29: patch #20, continued
See patch #20.
Larry Wall [Mon, 8 Jun 1992 04:50:30 +0000]
perl 4.0 patch 28: patch #20, continued
See patch #20.
Larry Wall [Mon, 8 Jun 1992 04:49:57 +0000]
perl 4.0 patch 27: patch #20, continued
See patch #20.
Larry Wall [Mon, 8 Jun 1992 04:52:08 +0000]
perl 4.0 patch 26: patch #20, continued
See patch #20.
Larry Wall [Mon, 8 Jun 1992 04:52:56 +0000]
perl 4.0 patch 25: patch #20, continued
See patch #20.
Larry Wall [Mon, 8 Jun 1992 04:49:46 +0000]
perl 4.0 patch 24: patch #20, continued
See patch #20.
Larry Wall [Mon, 8 Jun 1992 04:52:29 +0000]
perl 4.0 patch 23: patch #20, continued
See patch #20.
Larry Wall [Mon, 8 Jun 1992 04:52:53 +0000]
perl 4.0 patch 22: patch #20, continued
See patch #20.
Larry Wall [Mon, 8 Jun 1992 04:52:51 +0000]
perl 4.0 patch 21: patch #20, continued
See patch #20.
Larry Wall [Mon, 8 Jun 1992 04:49:43 +0000]
perl 4.0 patch 20: (combined patch)
ENHANCEMENTS
Subject: relaxed requirement for semicolon at the end of a block
Subject: scalar keys %array now counts keys for you
Subject: added ... as variant on ..
Subject: get*by* routines now return something useful in a scalar context
Subject: form feed for formats is now specifiable via $^L
Subject: PERLLIB now supports multiple directories
Subject: paragraph mode now skips extra newlines automatically
MANPAGE
Subject: documented that numbers may contain underline
Subject: clarified that DATA may only be read from main script
Subject: documented need for 1; at the end of a required file
Subject: extended bracket-style quotes to two-arg operators: s()() and tr()()
Subject: documented PERLLIB and PERLDB
Subject: documented limit on size of regexp
CONFIGURATION
Subject: bcopy() and memcpy() now tested for overlap safety
Subject: isascii() may now be supplied by a library routine
Subject: Configure now allows optional continuation with files missing
Subject: many more hints files added
Subject: many more hints added
Subject: hints now auto selected on uname -s as well as uname -m
Subject: OSF/1 support added
Subject: Configure growing-library-list bug fixed
Subject: seekdir(), telldir() and rewinddir() now checked for independently
Subject: cray didn't give enough memory to /bin/sh
Subject: perl -P now uses location of sed determined by Configure
Subject: SH files didn't work well with symbolic links
Subject: makefiles now display new shift/reduce expectations
Subject: support added to installperl for cross-compilation
Subject: a2p was installed unexecutable
Subject: installperl didn't warn on failed manpage installation
Subject: disabled cpp test if cppstdin not yet installed
PORTABILITY
Subject: O_PIPE conflicted with Atari
Subject: config.H updated to reflect more recent config.h
Subject: removed implicit int declarations on functions
Subject: added Atari ST portability
Subject: some machines don't define ENOTSOCK in errno.h
Subject: added explicit time_t support
Subject: alternate config.h files upgraded
Subject: new OS/2 support
COMPILER
Subject: various error messages have been clarified
Subject: the switch optimizer didn't do anything in subroutines
Subject: clarified debugging output for literals and double-quoted strings
Subject: new warning for use of x with non-numeric right operand
Subject: illegal lvalue message could be followed by core dump
Subject: new warning for ambiguous use of unary operators
Subject: eval "1 #comment" didn't work
Subject: semantic compilation errors didn't abort execution
Subject: an expression may now start with a bareword
Subject: if {block} {block} didn't work any more
Subject: "$var{$foo'bar}" didn't scan subscript correctly
Subject: an EXPR may now start with a bareword
Subject: print $fh EXPR can now expect term rather than operator in EXPR
Subject: new warning on spurious backslash
Subject: new warning on missing $ for foreach variable
Subject: "foo"x1024 now legal without space after x
Subject: new warning on print accidentally used as function
Subject: 2. now eats the dot
Subject: <@ARGV> now notices @ARGV
Subject: tr/// now lets you say \-
RUNTIME
Subject: an eval block containing a null block or statement could dump core
Subject: modulus with highest bit in left operand set didn't always work
Subject: join() now pre-extends target string to avoid excessive copying
Subject: subroutines didn't localize $`, $&, $', $1 et al correctly
Subject: usersub routines didn't reclaim temp values soon enough
Subject: ($<,$>) = ... didn't work on some architectures
Subject: fixed memory leak on system() for vfork() machines
Subject: @ in unpack failed too often
Subject: slice on null list in scalar context returned random value
Subject: splice with negative offset didn't work with $[ = 1
Subject: fixed some memory leaks in splice
Subject: dbmclose(%array) didn't work
Subject: delete could cause %array to give too low a count of buckets filled
Subject: hash tables now split only if the memory is available to do so
Subject: realloc(0, size) now does malloc in case library routines call it
Subject: running taintperl explicitly now does checks even if $< == $>
Subject: fixed memory leak in doube-quote interpretation
Subject: a splice on non-existent array elements could dump core
Subject: tr/stuff// wasn't working right
I/O
Subject: new warnings for failed use of stat operators on filenames with \n
Subject: wait failed when STDOUT or STDERR reopened to a pipe
Subject: end of file latch not reset on reopen of STDIN
Subject: seek(HANDLE, 0, 1) went to eof because of ancient Ultrix workaround
Subject: h_errno now accessible via $?
REGEXP
Subject: pattern modifiers i and o didn't interact right
Subject: g pattern modifer sometimes returned extra values
Subject: m/$pattern/g didn't work
Subject: /^stuff/ wrongly assumed an implicit $* == 1
Subject: /x{0}/ was wrongly interpreted as /x{0,}/
Subject: added \W, \S and \D inside /[...]/
Subject: pattern modifiers i and g didn't interact right
Subject: in some cases $` and $' didn't get set by match
Subject: made /\$$foo/ look for literal '$foo'
LIBRARIES
Subject: big*.pl library files upgraded
Subject: better support in chat2 for multiple children
Subject: &ctime didn't handle $[ != 0
Subject: find.pl got confused by unreadable directories
Subject: new version of newgetopt.pl
Subject: Tom's famous double-ended pipe opener, open2(), is now included
Subject: support added to pwd.pl to strip automounter crud
Subject: &shellwords looped on bad input, and used inefficient regular exprs
Subject: termcap.pl didn't parse termcap terminal names right
Subject: timelocal could loop on bad input
Subject: timelocal now calculates DST itself
Subject: &getcap eventually dumped core in bsdcurses
DEBUGGER
Subject: support for MSDOS folded into perldb.pl
Subject: perldb couldn't debug file containing '-', such as STDIN designator
Subject: the debugger now warns you on lines that can't set a breakpoint
Subject: the debugger made perl forget the last pattern used by //
Subject: fixed double debug break in foreach with implicit array assignment
Subject: debugger sometimes displayed wrong source line
INTERSTICES
Subject: Perl now distinguishes overlapped copies from non-overlapped
Subject: fixed confusion between a *var's real name and its effective name
Subject: deleted some minor memory leaks
Subject: couldn't require . files
Subject: -e 'cmd' no longer fails silently if /tmp runs out of space
Subject: function key support added to curses.mus
TRANSLATORS
Subject: find2perl assumed . in PATH
Subject: find2perl didn't output portable startup code
Subject: find2perl didn't always stat at the right time
Subject: s2p didn't output portable startup code
Subject: s2p didn't translate s/pat/\&/ or s/pat/\$/ or s/pat/\\1/ right
Subject: in a2p, getline should allow variable to be array element
Subject: in a2p, now warns about spurious backslashes
Subject: in a2p, now allows [ to be backslashed in pattern
Subject: in a2p, now allows numbers of the form 2.
Subject: in a2p, simplified the filehandle model
Subject: in a2p, made RS="" translate to $/ = "\n\n"
Subject: in a2p, do {...} while ... was missing some reconstruction code
Larry Wall [Mon, 11 Nov 1991 03:50:16 +0000]
perl 4.0 patch 19: (combined patch)
Ok, here's the cleanup patch I suggested you wait for. Have at it...
Subject: added little-endian pack/unpack options
This is the only enhancement in this patch, but it seemed unlikely
to bust anything else, and added functionality that it was very
difficult to do any other way. Compliments of David W. Sanderson.
Subject: op/regexp.t failed from missing arg to bcmp()
Subject: study was busted by 4.018
Subject: sort $subname was busted by changes in 4.018
Subject: default arg for shift was wrong after first subroutine definition
Things that broke in 4.018. Shame on me.
Subject: do {$foo ne "bar";} returned wrong value
A bug of long standing. How come nobody saw this one? Or if you
did, why didn't you report it before now? Or if you did, why did
I ignore you? :-)
Subject: some machines need -lsocket before -lnsl
Subject: some earlier patches weren't propagated to alternate 286 code
Subject: compile in the x2p directory couldn't find cppstdin
Subject: more hints for aix, isc, hp, sco, uts
Subject: installperl no longer updates unchanged library files
Subject: uts wrongly defines S_ISDIR() et al
Subject: too many preprocessors can't expand a macro right in #if
The usual pastiche of portability kludges.
Subject: deleted some unused functions from usersub.c
And fixed the spelling of John Macdonald's name, and included his
suggested workaround for a certain vendor's stdio bug...
Subject: added readdir test
Subject: made op/groups.t more reliable
Subject: added test for sort $subname to op/sort.t
Subject: added some hacks to op/stat.t for weird filesystem architectures
Improvements (hopefully) to the regression tests.
Larry Wall [Tue, 5 Nov 1991 09:55:53 +0000]
perl 4.0 patch 18: patch #11, continued
See patch #11.
Larry Wall [Tue, 5 Nov 1991 06:28:31 +0000]
perl 4.0 patch 17: patch #11, continued
See patch #11.
Larry Wall [Tue, 5 Nov 1991 06:28:06 +0000]
perl 4.0 patch 16: patch #11, continued
See patch #11.
Larry Wall [Tue, 5 Nov 1991 06:28:23 +0000]
perl 4.0 patch 15: patch #11, continued
See patch #11.
Larry Wall [Tue, 5 Nov 1991 06:28:36 +0000]
perl 4.0 patch 14: patch #11, continued
See patch #11.
Larry Wall [Tue, 5 Nov 1991 06:28:20 +0000]
perl 4.0 patch 13: patch #11, continued
See patch #11.
Larry Wall [Tue, 5 Nov 1991 06:26:52 +0000]
perl 4.0 patch 12: patch #11, continued
See patch #11.
Larry Wall [Tue, 5 Nov 1991 10:12:55 +0000]
perl 4.0 patch 11: (combined patch)
Subject: added eval {}
Subject: eval 'stuff' now optimized to eval {stuff}
This set of patches doesn't have many enhancements but this is
one of them. The eval operator has two distinct semantic functions.
First, it runs the parser on some random string and executes it.
Second, it traps exceptions and returns them in $@. There are times
when you'd like to get the second function without the first. In
order to do that, you can now eval a block of code, which is parsed
like ordinary code at compile time, but which traps any run-time
errors and returns them in the $@ variable. For instance, to
trap divide by zero errors:
eval {
$answer = $foo / $bar;
};
warn $@ if $@;
Since single-quoted strings don't ever change, they are optimized
to the eval {} form the first time they are encountered at run-time.
This doesn't happen too often, though some of you have written things
like eval '&try_this;'. However, the righthand side of s///e is
evaluated as a single-quoted string, so this construct should run
somewhat faster now.
Subject: added sort {} LIST
Another enhancement that some of you have been hankering for.
You can now inline the sort subroutine as a block where the
subroutine name used to go:
@articles = sort {$a <=> $b;} readdir(DIR);
Subject: added some support for 64-bit integers
For Convexen and Crayen, which have 64-bit integers, there's
now pack, unpack and sprintf support for 64-bit integers.
Subject: sprintf() now supports any length of s field
You can now use formats like %2048s and %-8192.8192s. Perl will
totally bypass your system's sprintf() function on these. No,
you still probably can't say %2048d. No, I'm not going to
change that any time soon.
Subject: substr() and vec() weren't allowed in an lvalue list
Subject: extra comma at end of list is now allowed in more places (Hi, Felix!)
Subject: underscore is now allowed within literal octal and hex numbers
Various syntactic relaxations. You can now get away with
(substr($foo,0,3), substr($bar,0,3)) = ('abc', 'def');
(1,2,3,)[$x];
$addr = 0x1a20_ff0b;
Subject: safe malloc code now integrated into Perl's malloc when possible
To save a bunch of subroutine calls. If you use your system's
malloc it still has to use wrappers.
Subject: added support for dbz
By saying "make dbzperl" you can make a copy of Perl that can
access C news's dbz files. You still have to follow the dbz rules,
though, if you're going to try to write a dbz file.
Subject: there are now subroutines for calling back from C into Perl
Subject: usub/curses.mus now supports SysV curses
More C linkage support. I still haven't got Perl embeddable, but
we're getting there. That's too big an enhancement for this
update, in which I've been trying to stick to bug fixes, with some
success.
Subject: prepared for ctype implementations that don't define isascii()
A larger percentage of this update consists of code to do
consistent ctype processing whether or not <ctype.h> is 8-bit
clean.
Subject: /$foo/o optimizer could access deallocated data
Subject: certain optimizations of //g in array context returned too many values
Subject: regexp with no parens in array context returned wacky $`, $& and $'
Subject: $' not set right on some //g
Subject: grep of a split lost its values
Subject: # fields could write outside allocated memory
Subject: length($x) was sometimes wrong for numeric $x
Recently added or modified stuff that you kind of expect to be
a bit flaky still. Well, I do...
Subject: passing non-existend array elements to subrouting caused core dump
Subject: "foo" x -1 dumped core
Subject: truncate on a closed filehandle could dump
Subject: a last statement outside any block caused occasional core dumps
Subject: missing arguments caused core dump in -D8 code
Subject: cacheout.pl could dump core from invalid comparison operator
Subject: *foo = undef coredumped
Subject: warn '-' x 10000 dumped core
Subject: index("little", "longer string") could visit faraway places
A bunch of natty little bugs that you wouldn't generally run into
unless you're trying to be coy.
Subject: hex() didn't understand leading 0x
It wasn't documented that it should work, but oct() understands 0x,
so why not hex()? I dunno...
Subject: "foo\0" eq "foo" was sometimes optimized to true
Subject: eval confused by string containing null
Yet more holdovers from the time before Perl was 8-bit clean.
Subject: foreach on null list could spring memory leak
Subject: local(*FILEHANDLE) had a memory leak
Kind of slow leaks, as leaks go. Still...
Subject: minimum match length calculation in regexp is now cumulative
More substitutions can be done in place now because Perl knows
that patterns like in s/foo\s+bar/1234567/ have to match a
certain number of characters total. It used to be on that
particular pattern that it only knew that it had to match at
least 3 characters. Now it know it has to match at least 7.
Subject: multiple reallocations now avoided in 1 .. 100000
You still don't want to say 1 .. 1000000, but at least it will
refrain from allocating intermediate sized blocks while it's
constructing the value, and won't do the extra copies implied
by realloc.
Subject: indirect subroutine calls through magic vars (e.g. &$1) didn't work
Subject: defined(&$foo) and undef(&$foo) didn't work
Subject: certain perl errors should set EBADF so that $! looks better
Subject: stats of _ forgot whether prior stat was actually lstat
Subject: -T returned true on NFS directory
Subject: sysread() in socket was substituting recv()
Subject: formats didn't fill their fields as well as they could
Subject: ^ fields chopped hyphens on line break
Subject: -P didn't allow use of #elif or #undef
Subject: $0 was being truncated at times
Subject: forked exec on non-existent program now issues a warning
Various things you'd expect to work the way you expect, but
didn't when you did, or I did, or something...
Subject: perl mistook some streams for sockets because they return mode 0 too
Subject: reopening STDIN, STDOUT and STDERR failed on some machines
Problems opening files portably. So what's new?
Subject: cppstdin now installed outside of source directory
Subject: installperl now overrides installer's umask
People who used cppstdin for the cpp filter or who had their
umask set to 700 will now be happier. (And Configure will now
prefer /lib/cpp over cppstdin like it used to. If this gives
your machine heartburn because /lib/cpp doesn't set the symbols
it should, write a hints file to poke them into ccflags.)
Subject: initial .* in pattern had dependency on value of $*
An initial .* was optimized to have a ^ on the front to avoid retrying
when we know it won't match. Unfortunately this implicit ^ was
paying attention to $*, which it shouldn't have been.
Subject: certain patterns made use of garbage pointers from uncleared memory
Many of you saw this as a failure in t/op/pat.t.
Subject: perl now issues warning if $SIG{'ALARM'} is referenced
Since the book mentions "SIGALARM", I thought we needed this.
Subject: solitary subroutine references no longer trigger typo warnings
You can now use -w (more) profitably on programs that require
other files. I figured if you mistype a subroutine name you'll
get a fatal error anyway, unlike a variable, which just defaults
to being undefined.
Subject: $foo .= <BAR> could overrun malloced memory
Good old-fashioned bug.
Subject: \$ didn't always make it through double-quoter to regexp routines
Subject: \x and \c were subject to double interpretation in regexps
Subject: nested list operators could miscount parens
Subject: sort eval "whatever" didn't work
Syntactic misfeatures of various sorts.
Subject: find2perl produced incorrect code for -group
Subject: find2perl could be confused by names containing whitespace
Subject: in a2p, split on whitespace produced extra null field
Translator stuff.
Subject: new complete.pl from Wayne Thompson
Subject: assert.pl and exceptions.pl from Tom Christiansen
Subject: added Tom's c2ph stuff
Subject: getcwd.pl from Brandon S. Allbery
Subject: fastcwd.pl from John Basik
Subject: chat2.pl from Randal L. Schwartz
New contributed stuff. Thanks!
(Not that a lot of the other stuff isn't contributed too...)
Subject: debugger got confused over nested subroutine definitions
Subject: once-thru blocks didn't display right in the debugger
Subject: perldb.pl modified to run within emacs in perldb-mode
Debugger stuff. The first two were caused by not saving line
numbers at exactly the right moment.
Subject: documented meaning of scalar(%foo)
I also updated the Errata section of the man page.
Subject: various portability fixes
Subject: random cleanup
Subject: saberized perl
Type casts, saber warning message suppression, hints files and various
metaconfig fiddlehoods.
Larry Wall [Sun, 9 Jun 1991 12:36:21 +0000]
perl 4.0 patch 10: (combined patch)
Subject: pack(hh,1) dumped core
Subject: read didn't work from character special files open for writing
Subject: close-on-exec wrongly set on system file descriptors
Subject: //g only worked first time through
Subject: perl -v printed incorrect copyright notice
Subject: certain pattern optimizations were botched
Subject: documented some newer features in addenda
Subject: $) and $| incorrectly handled in run-time patterns
Subject: added tests for case-insensitive regular expressions
Subject: m'$foo' now treats string as single quoted
Larry Wall [Thu, 6 Jun 1991 23:28:30 +0000]
perl 4.0 patch 9: patch #4, continued
See patch #4.
Larry Wall [Thu, 6 Jun 1991 23:28:14 +0000]
perl 4.0 patch 8: patch #4, continued
See patch #4.
Larry Wall [Thu, 6 Jun 1991 23:28:07 +0000]
perl 4.0 patch 7: patch #4, continued
See patch #4.
Larry Wall [Thu, 6 Jun 1991 23:28:02 +0000]
perl 4.0 patch 6: patch #4, continued
See patch #4.
Larry Wall [Thu, 6 Jun 1991 23:27:54 +0000]
perl 4.0 patch 5: patch #4, continued
See patch #4.
Larry Wall [Thu, 6 Jun 1991 23:27:37 +0000]
perl 4.0 patch 4: (combined patch)
Random patches, mostly bugs and portability stuff. //g is the
only major new feature. Additionally, there is now an alternate
license you can distribute Perl under.
Larry Wall [Thu, 11 Apr 1991 20:29:16 +0000]
perl 4.0 patch 3: Patch 1 continued
Larry Wall [Thu, 11 Apr 1991 20:32:32 +0000]
perl 4.0 patch 2: Patch 1 continued
Larry Wall [Thu, 11 Apr 1991 20:32:40 +0000]
perl 4.0 patch 1: (combined patch)
Subject: Configure now handles defaults much better
Subject: Configure now knows if config.sh was built on this machine
Subject: Configure now checks file existence more efficiently
Subject: Configure now handles stupid SCO csh
Configure has been heavily revised. Many of the tests that used
to simply force a decision now check that decision against the
previous value of the variable, and offer to let you change it.
The default now is to keep the old value, so that you don't lose
information from your previous run.
Because of this, it's now more important to know whether, in fact,
config.sh was produced on this machine and on this version of
the operating system. config.sh now contains a lastuname variable
which contains the output of uname -a. If this matches the current
output of uname -a, Configure defaults to including the old config.sh.
Otherwise not. If there is no valid config.sh, then Configure looks
defaults for the current architecture in the hints/ subdirectory
instead. The guesswork I've done in this section of code is
phenomenal, so you'll have to instruct me where I've misparsed
the output of uname (a problem in portability all of its own).
Subject: Configure now differentiates getgroups() type from getgid() type
Subject: Configure now figures out malloc ptr type
Subject: Configure now does better on sprintf()
Configure was assuming that the array of values returned from
getgroups was the same type as the gids returned by other system
calls. Unfortunately, reality set in. Likewise for malloc() and
sprintf(), which there is only one portable way to find out the return
value of: try it one way or the other, and see if it blows up.
Subject: C flags are now settable on a per-file basis
Subject: reduced maximum branch distance in eval.c
Certain compilers and/or optimizers get bozoed out by large
compilation units, or by large structures within those units.
Previously, you either had to change the compiler flags for all
the files, or do hairy editing in Makefile.SH and remake the Makefile,
necessitating a make depend. Now there is a script called cflags.SH
whose duty it is to return the proper CFLAGS for any given C file.
You can change the flags in just one spot now and they will be
immediately reflected in the next make (or even in the current
make, if one is running). Eventually I expect that any of the hints
files could modify cflags.SH, but I haven't done that yet.
The particular problem of long jump offsets in eval.c has been at
least partially alleviated by locating some of the labels in the
middle of the function instead of at the end. This still doesn't
help the poor Vax when you compile with -g, since it puts a jump
to the end of the function to allocate the stack frame and then
jumps back to the beginning of the function to execute it. For
now Vaxen will have to stick with -O or hand assemble eval.c and
teval.c with a -J switch.
Subject: fixed "Bad free" error
Subject: fixed debugger coredump on subroutines
Subject: regexec only allocated space for 9 subexpresssions
These are problems that were reported on the net and had unofficial
patches. Now they have official patches. Be sure to patch a
copy of your files without the unofficial patches, or the patch
program will get confused.
Subject: you may now use "die" and "caller" in a signal handler
Someone pointed out that using die to raise an exception out
of a signal handler trashed the expression value stack if the
exception was caught by eval. While fixing that, I also fixed
the longstanding problem that signal handlers didn't have a normal
call frame, which prevented the caller function from working.
Subject: fixed undefined environ problem
Subject: hopefully straightened out some of the Xenix mess
Subject: random cleanup in cpp namespace
Just keeping up with the current progress in non-standardization.
Subject: fixed failed fork to return undef as documented
The open function returns undef on failed implicit forks. The Book
assumed that the same was true of an explicit fork. I've made the
function behave like the Book says. It's a pity there's no way
to have an undefined value that returns -1 in a numeric context
but false in a boolean context...
Subject: generalized the yaccpar fixer some
Thanks to Andy Dougherty, perly.fixer now knows how to fix SVR3 2.2's
yaccpar code to do dynamic parse stack allocation. He also made it
easy for other people to insert their code there. Hooray!
Subject: find2perl sometimes needs to stat on the 2nd leg of a -o
Subject: find2perl didn't correctly handle switches with an argument of 0
In attempting to delay the lstat to the last moment, in case a filename
could be rejected on the basis of its name, find2perl neglected to
take into account the fact that control might pass to the 2nd half
of a -o without executing all of the 1st half, in particular without
executing the lstat.
find2perl was wisely removing leading zeroes from numbers that would
mistakenly be interpreted as octal numbers by Perl. Unfortunately,
this caused it to delete the number 0 entirely.
Subject: fixed dumpvar not to dump internal debugging info
Subject: substr($ENV{"PATH"},0,0) = "/foo:" didn't modify environment
Subject: $foo .= <BAR> could cause core dump for certain lengths of $foo
Subject: perl -de "print" wouldn't stop at the first statement
Random glitchy little things.
Subject: I'm at NetLabs now
I'm now working for NetLabs, Inc., and I hadn't changed my
address everywhere.
Larry Wall [Thu, 21 Mar 1991 00:00:00 +0000]
perl 4.0.00: (no release announcement available)
So far, 4.0 is still a beta test version. For the last production
version, look in pub/perl.3.0/kits@44.
Larry Wall [Fri, 11 Jan 1991 08:58:45 +0000]
perl 3.0 patch #44 patch #42, continued
See patch #42.
Larry Wall [Fri, 11 Jan 1991 05:46:37 +0000]
perl 3.0 patch #43 patch #42, continued
See patch #42.
Larry Wall [Fri, 11 Jan 1991 05:47:59 +0000]
perl 3.0 patch #42 (combined patch)
Most of these patches are pretty self-explanatory. Much of this
is random cleanup in preparation for version 4.0, so I won't talk
about it here. A couple of things should be noted, however.
First, there's a new -0 option that allows you to specify (in octal)
the initial value of $/, the record separator. It's primarily
intended for use with versions of find that support -print0 to
delimit filenames with nulls, but it's more general than that:
null
^A
default
CR
paragraph mode
file slurp mode
This feature is so new that it didn't even make it into the book.
The other major item is that different patchlevels of perl can
now coexist in your bin directory. The names "perl" and "taintperl"
are just links to "perl3.044" and "tperl3.044". This has several
benefits. The perl3.044 invokes the corresponding tperl3.044 rather
than taintperl, so it always runs the correct version. Second, you can
"freeze" a script by putting a #! line referring to a version that
it is known to work with. Third, you can put a new version out
there to try out before making it the default perl. Lastly, it
sells more disk drives. :-)
Barring catastrophe, this will likely be the last patch before
version 4.0 comes out.
Larry Wall [Tue, 13 Nov 1990 02:28:59 +0000]
perl 3.0 patch #41 (combined patch)
Here's the requisite dinky patch to fix the problems of the
preceding large set of patches. In this case, a couple of
malloc/free problems--one of which involved overrunning the end
of an allocated string, and the other of which involved freeing
with invalid pointers. (There was also a bug in there involving
variable magicalness propagating incorrectly, which resulting in
a dbm anomoly.)
I updated README to mention that dnix needs to avoid -O.
I added the hp malloc union overhead strut that Jan Dr{rv posted.
(Eventually this should be determined by Configure, but laziness
has its advantages.)
Larry Wall [Fri, 9 Nov 1990 13:38:50 +0000]
perl 3.0 patch #40 patch #38, continued
See patch #38.
Larry Wall [Fri, 9 Nov 1990 13:37:16 +0000]
perl 3.0 patch #39 patch #38, continued
See patch #38.
Larry Wall [Fri, 9 Nov 1990 13:39:17 +0000]
perl 3.0 patch #38 (combined patch)
Forget the description, it's too late at night...
Larry Wall [Fri, 19 Oct 1990 13:31:07 +0000]
perl 3.0 patch #37 (combined patch)
I tried to take the strlen of an integer on systems without wait4()
or waitpid(). For some reason this didn't work too well...
In hash.c there was a call to dbm_nextkey() which needed to be
ifdefed on old dbm systems.
A pattern such as /foo.*bar$/ was wrongly optimized to do
tail matching on "foo". This was a longstanding bug that
was unmasked by patch 36.
Some systems have some SYS V IPC but not all of it. Configure
now figures this out.
Patch 36 put the user's PATH in front of Configures, but to make
it work right I needed to change all calls of loc to ./loc in
Configure.
$cryptlib needed to be mentioned in the Makefile.
Apollo 10.3 and Sun 3.5 have some compilation problems, so I
mentioned them in README.
Cray has weird restrictions on setjmp locations--you can't say
if (result = setjmp(...))
Random typos and cleanup.
Larry Wall [Mon, 15 Oct 1990 23:07:21 +0000]
perl 3.0 patch #36 patch #29, continued
See patch #29.
Larry Wall [Mon, 15 Oct 1990 23:05:15 +0000]
perl 3.0 patch #35 patch #29, continued
See patch #29.
Larry Wall [Mon, 15 Oct 1990 23:06:41 +0000]
perl 3.0 patch #34 patch #29, continued
See patch #29.
Larry Wall [Mon, 15 Oct 1990 23:03:11 +0000]
perl 3.0 patch #33 patch #29, continued
See patch #29.
Larry Wall [Tue, 16 Oct 1990 02:28:17 +0000]
perl 3.0 patch #32 patch #29, continued
See patch #29.
Larry Wall [Tue, 16 Oct 1990 02:30:59 +0000]
perl 3.0 patch #31 patch #29, continued
See patch #29.
Larry Wall [Mon, 15 Oct 1990 23:06:25 +0000]
perl 3.0 patch #30 patch #29, continued
See patch #29.
Larry Wall [Mon, 15 Oct 1990 23:06:10 +0000]
perl 3.0 patch #29 (combined patch)
This set of patches pretty much brings you up to the functionality
that version 4.0 will have. The Perl Book documents version 4.0.
Perhaps these should be called release notes... :-)
Enhancements:
Many of the changes relate to making the debugger work better.
It now runs your scripts at nearly full speed because it no longer
calls a subroutine on every statement. The debugger now doesn't
get confused about packages, evals and other filenames. More
variables (though still not all) are available within the debugger.
Related to this is the fact that every statement now knows which
package and filename it was compiled in, so package semantics are
now much more straightforward. Every variable also knows which
package it was compiled in. So many places that used to print
out just the variable name now prefix the variable name with the
package name. Notably, if you print *foo it now gives *package'foo.
Along with these, there is now a "caller" function which returns
the context of the current subroutine call. See the man page for
more details.
Chip Salzenberg sent the patches for System V IPC (msg, sem and shm)
so I dropped them in.
There was no way to wait for a specific pid, which was silly, since
Perl was already keeping track of the information. So I added
the waitpid() call, which uses Unix's wait4() or waitpid() if
available, and otherwise emulates them (at least as far as letting
you wait for a particular pid--it doesn't emulate non-blocking wait).
For use in sorting routines, there are now two new operators,
cmp and <=>. These do string and numeric comparison, returning
-1, 0 or 1 when the first argument is less than, equal to or
greater than the second argument.
Occasionally one finds that one wants to evaluate an operator in
a scalar context, even though it's part of a LIST. For this purpose,
there is now a scalar() operator. For instance, the approved
fix for the novice error of using <> in assigning to a local is now:
local($var) = scalar(<STDIN>);
Perl's ordinary I/O is done using standard I/O routines. Every
now and then this gets in your way. You may now access the system
calls read() and write() via the Perl functions sysread() and
syswrite(). They should not be intermixed with ordinary I/O calls
unless you know what you're doing.
Along with this, both the sysread() and read() functions allow you
an optional 4th argument giving an offset into the string you're
reading into, so for instance you can easily finish up partial reads.
As a bit of syntactic sugar, you can now use the file tests -M, -A
and -C to determine the age of a file in (possibly fractional) days
as of the time the script started running. This makes it much
easier to write midnight cleanup scripts with precision.
The index() and rindex() functions now have an optional 3rd argument
which tells it where to start looking, so you can now iterate through
a string using these functions.
The substr() function's 3rd argument is now optional, and if omitted,
the function returns everything to the end of the string.
The tr/// translation function now understands c, d and s options, just
like the tr program. (Well, almost just like. The d option only
deletes characters that aren't in the replacement string.) The
c complementes the character class to match and the s option squishes
out multiple occurrences of any replacement class characters.
The reverse function, used in a scalar context, now reverses its
scalar argument as a string.
Dale Worley posted a patch to add @###.## type fields to formats.
I said, "Neat!" and dropped it in, lock, stock and sinker.
Kai Uwe Rommel sent a bunch of MSDOS and OS/2 updates, which I (mostly)
incorporated. I can't vouch for them, but they look okay.
Any data stored after the __END__ marker can be accesses now via
the DATA filehandle, which is automatically opened onto the script
at that point. (Well, actually, it's just kept open, since it
was already open to read the script.)
The taintperl program now checks for world writable PATH components,
and complains if any are found (if PATH is used).
Bug fixes:
It used to be that you could get core dumps by such means as
@$foo=();
@foo[42];
(1,2,3)[42];
$#foo = 50;
foreach $elem (@foo) {
$elem = 1;
}
This is no longer so. (For those who are up on Perl internals, the
stack policy no longer allows Nullstr--all undefined values must
be passed as &str_undef.)
If you say something like
local($foo,$bar);
or
local($initialized,$foo,$bar) = ('one value');
$foo and $bar are now initialized to the undefined value, rather
than the defined null string.
Array assignment to special arrays is now better supported. For
instance, @ENV = () clears the environment, and %foo = () will
now clear any dbm file bound to %foo.
On the subject of dbm files, the highly visible bugs at patchlevel
28 have been fixed. You can now open dbm files readonly, and you
don't have to do a dummy assignment to make the cache allocate itself.
The modulus operator wasn't working right on negative values because
of a misplaced cast. For instance, -5 % 5 was returning
the value 5, which is clearly wrong.
Certain operations coredumped if you didn't supply a value:
close;
eof;
Previously, if the subroutine supplied for a sort operation didn't
exist, it failed quietly. Now it produces a fatal error.
The bitwise complement operator ~ didn't work on vec() strings longer
than one byte because of failure to increment a loop variable.
The oct and hex functions returned a negative result if the highest
bit was set. They now return an unsigned result, which seems a
little less confusing. Likewise, the token 0x
80000000 also produces
an unsigned value now.
Some machines didn't like to see 0x
87654321 in an #ifdef because
they think of the symbols as signed. The tests have been changed
to just look at the lower 4 nybbles of the value, which is sufficient
to determine endianness, at least as far as the #ifdefs are concerned.
The unshift operator did not return the documented value, which
was the number of elements in the new array. Instead it returned
the last unshifted argument, more or less by accident.
-w sometimes printed spurious warnings about ARGV and ENV when
referencing the arrays indirectly through shift or exec. This
was because the typo test was misplaced before the code that
exempts special variables from the typo test.
If you said 'require "./foo.pl"', it would look in someplace like
/usr/local/lib/perl/./foo.pl instead of the current directory. This
works more like people expect now. The require error messages also
referred to wrong file, if they worked at all.
The h2ph program didn't translate includes right--it should have
changed .h to .ph.
Patterns with multiple short literal strings sometimes failed.
This was a problem with the code that looks for a maximal literal
string to feed to the Boyer-Moore searching routine. The code
was gluing together literal strings that weren't continuous.
The $* variable controls multi-line pattern matching. When it's
0, patterns are supposed to match as if the string contained a
single line. Unfortunately, /^pat/ occasionally matched in middle
of string under certain conditions.
Recently the regular expression routines were upgraded to do
{n,m} more efficiently. In doing this, however, I manufactured
a couple of bugs: /.{n,m}$/ could match with fewer than n characters
remaining on the line, and patterns like /\d{9}/ could match more
than 9 characters.
The undefined value has an actual physical location in Perl, and
pointers to it are passed around. By certain circuitous routes
it was possible to clobber the undefined value so that it
was no longer undefined--kind of like making /dev/null into
a real file. Hopefully this can't happen any more.
op.stat could fail if /bin/0 existed, because of a while (<*>) {...
This has been changed to a while (defined($_ = <*>)) {...
The length of a search pattern was limited by the length of
tokenbuf internally. This restriction has been removed.
The null character gave the tokener indigestion when used as
a delimiter for m// or s///.
There was a bunch of other cleanupish things that are too trivial
to mention here.
Larry Wall [Mon, 13 Aug 1990 09:45:26 +0000]
perl 3.0 patch #28 (combined patch)
Certain systems, notable Ultrix, set the close-on-exec flag
by default on dup'ed file descriptors. This is anti-social
when you're creating a new STDOUT. The flag is now forced
off for STDIN, STDOUT and STDERR.
Some yaccs report 29 shift/reduce conflicts and 59 reduce/reduce
conflicts, while other yaccs and bison report 27 and 61. The
Makefile now says to expect either thing. I'm not sure if there's
a bug lurking there somewhere.
The defined(@array) and defined(%array) ended up defining
the arrays they were trying to determine the status of. Oops.
Using the status of NSIG to determine whether <signal.h> had
been included didn't work right on Xenix. A fix seems to be
beyond Configure at the moment, so we've got some OS dependent
#ifdefs in there.
There were some syntax errors in the new code to determine whether
it is safe to emulate rename() with unlink/link/unlink. Obviously
heavily tested code... :-)
Patch 27 introduced the possibility of using identifiers as
unquoted strings, but the code to warn against the use of
totally lowercase identifiers looped infinitely.
I documented that you can't interpolate $) or $| in pattern.
It was actually implied under s///, but it should have been
more explicit.
Patterns with {m} rather than {m,n} didn't work right.
Tests io.fs and op.stat had difficulties under AFS. They now
ignore the tests in question if they think they're running under
/afs.
The shift/reduce expectation message was off for a2p's Makefile.
Larry Wall [Wed, 8 Aug 1990 17:07:27 +0000]
perl 3.0 patch #27 patch #19, continued
See patch #19.
Larry Wall [Wed, 8 Aug 1990 17:06:25 +0000]
perl 3.0 patch #26 patch #19, continued
See patch #19.
Larry Wall [Wed, 8 Aug 1990 17:07:07 +0000]
perl 3.0 patch #25 patch #19, continued
See patch #19.
Larry Wall [Wed, 8 Aug 1990 17:04:39 +0000]
perl 3.0 patch #24 patch #19, continued
See patch #19.
Larry Wall [Wed, 8 Aug 1990 17:06:03 +0000]
perl 3.0 patch #23 patch #19, continued
See patch #19.
Larry Wall [Wed, 8 Aug 1990 17:01:53 +0000]
perl 3.0 patch #22 patch #19, continued
See patch #19.
Larry Wall [Wed, 8 Aug 1990 17:07:00 +0000]
perl 3.0 patch #21 patch #19, continued
See patch #19.
Larry Wall [Wed, 8 Aug 1990 17:07:00 +0000]
perl 3.0 patch #20 patch #19, continued
See patch #19.
Larry Wall [Wed, 8 Aug 1990 17:02:14 +0000]
perl 3.0 patch #19 (combined patch)
You now have the capability of linking C subroutines into a
special version of perl. See the files in usub/ for an example.
There is now an operator to include library modules with duplicate
suppression and error checking, called "require". (makelib has been
renamed to h2ph, and Tom Christiansen's h2pl stuff has been included
too. Perl .h files are now called .ph files to avoid confusion.)
It's now possible to truncate files if your machines supports any
of ftruncate(fd, size), chsize(fd, size) or fcntl(fd, F_FREESP, size).
Added -c switch to do compilation only, that is, to suppress
execution. Useful in combination with -D1024.
There's now a -x switch to extract a script from the input stream
so you can pipe articles containing Perl scripts directly into perl.
Previously, the only places you could use bare words in Perl were as
filehandles or labels. You can now put bare words (identifiers)
anywhere. If they have no interpretation as filehandles or labels,
they will be treated as if they had single quotes around them.
This works together nicely with the fact that you can use a
symbol name indirectly as a filehandle or to assign to *name.
It basically means you can write subroutines and pass filehandles
without quoting or *-ing them. (It also means the grammar is even
more ambiguous now--59 reduce/reduce conflicts!!! But it seems
to do the Right Thing.)
Added __LINE__ and __FILE__ tokens to let you interpolate the
current line number or filename, such as in a call to an error
routine, or to help you translate eval linenumbers to real
linenumbers.
Added __END__ token to let you mark the end of the program in
the input stream. (^D and ^Z are allowed synonyms.) Program text
and data can now both come from STDIN.
`command` in array context now returns array of lines. Previously
it would return a single element array holding all the lines.
An empty %array now returns 0 in scalar context so that you can
use it profitably in a conditional: &blurfl if %seen;
The include search path (@INC) now includes . explicity at the
end, so you can change it if you wish. Library routines now
have precedence by default.
Several pattern matching optimizations: I sped up /x+y/ patterns
greatly by not retrying on every x, and disabled backoff on
patterns anchored to the end like /\s+$/. This made /\s+$/ run
100 times faster on a string containing 70 spaces followed by an X.
Actual improvements will generally be less than that. I also
sped up {m,n} on simple items by making it a variant of *.
And /.*whatever/ is now optimizaed to /^.*whatever/ to avoid
retrying at every position in the event of failure. I fixed
character classes to allow backslashing hyphen, by popular
request.
In the past, $ in a pattern would sometimes match in the middle
of the string and sometimes not, if $* == 0. Now it will never
match except at the end of the string, or just before a terminating
newline. When $* == 1 behavior is as before.
In the README file, I've expanded on just how I think the GNU
General Public License applies to Perl and to things you might
want to do with Perl.
The interpreter used to set the global variable "line" to be
the current line number. Instead, it now sets a global pointer
to the current Perl statement, which is no more overhead, but
now we will have access to the file name and package name associated
with that statement, so that the debugger soon be upgraded to
allow debugging of evals and packages.
In the past, a conditional construct in an array context passed
the array context on to the conditional expression, causing
general consternation and confusion. Conditionals now always
supply a scalar context to the expression, and if that expression
turns out to be the one whose value is returned, the value is
coerced to an array value of one element.
The switch optimizer was confused by negative fractional values,
and truncating them the wrong direction.
Configure now checks for chsize, select and truncate functions, and
now asks if you want to put scripts into some separate directory
from your binaries. More and more people are establishing a common
directory across architectures for scripts, so this is getting
important.
It used to be that a numeric literal ended up being stored both
as a string and as a double. This could make for lots of wasted
storage if you said things like "$seen{$key} = 1;". So now
numeric literals are now stored only in floating point format,
which saves space, and generates at most one extra conversion per
literal over the life of the script.
The % operator had an off-by-one error if the left argument was
negative.
The pack and unpack functions have been upgraded. You
can now convert native float and double fields using f and d.
You can specify negative relative positions with X<n>, and absolute
positions in the record with @<n>. You can have a length of *
on the final field to indicate that it is to gobble all the rest
of the available fields. In unpack, if you precede a field
spec with %<n>, it does an n-bit checksum on it instead of the
value itself. (Thus "%16C*" will checksum just like the Sys V sum
program.) One totally wacked out evening I hacked a u format
in to pack and unpack uudecode-style strings.
A couple bugs were fixed in unpack--it couldn't unpack an A or a
format field in a scalar context, which is just supposed to
return the first field. The c and C formats were also calling
bcopy to copy each character. Yuck.
Machines without the setreuid() system call couldn't manipulate
$< and $> easily. Now, if you've got setuid(), you can say $< = $>
or $> = $< or even ($<, $>) = ($uid, $uid), as long as it's
something that can be done with setuid(). Similarly for setgid().
I've included various MSDOS and OS/2 patches that people have sent.
There's still more in the hopper...
An open on a pipe for output such as 'open(STDOUT,"|command")' left
STDOUT attached to the wrong file descriptor. This didn't matter
within Perl, but it made subprocesses expecting stdout to be on fd 1
rather irate.
The print command could fail to detect errors such as running out
room on the disk. Now it checks a little better.
Saying "print @foo" might only print out some of the elements
if there undefined elements in the middle of the array, due to
a reversed bit of logic in the print routine.
On machines with vfork the child process would allocate memory
in the parent without the parent knowing about it, or having any way
to free the memory so allocated. The parent now calls a cleanup
routine that knows whether that's what happened.
If the getsockname or getpeername functions returned a normal
Unix error, perl -w would report that you tried I/O on an
unopened socket, even though it was open.
MACH doesn't have seekdir or telldir. Who ever uses them anyway?
Under certain circumstances, an optimized pattern match could
pass a hint into the standard pattern matching routine which
the standard routine would then ignore. The next pattern match
after that would then get a "panic: hint in do_match" because the
hint didn't point into the current string of interest.
The $' variable returned a short string if it contained an
embedded null.
Two common split cases are now special-cased to avoid the regular
expression code. One is /\s+/ (and its cousin ' ', which also
trims leading whitespace). The other is /^/, which is very useful
for splitting a "here-is" quote into lines:
@lines = split(/^/, <<END);
Element 0
Element 1
Element 2
END
You couldn't split on a single case-insensitive letter because
the single character split optimization ignore the case folding
flag.
Sort now handles undefined strings right, and sorts lists
a little more efficiently because it weeds them out before
sorting so it doesn't have to check for them on every comparison.
The each() and keys() functions were returning garbage on null
keys in DBM files because the DBM iterator merely returns a pointer
into the buffer to a string that's not necessarily null terminated.
Internally, Perl keeps a null at the end of every string (though
allowing embedded nulls) and some routines make use of this
to avoid checking for the end of buffer on every comparison. So
this just needed to be treated as a special case.
The &, | and ^ operators will do bitwise operations on two strings,
but for some reason I hadn't implemented ~ to do a complement.
Using an associative array name with a % in dbmopen(%name...)
didn't work right, not because it didn't parse, but because the
dbm opening routine internally did the wrong thing with it.
You can now say dbmopen(name, 'filename', undef) to prevent it
from opening the dbm file if it doesn't exist.
The die operator simply exited if you didn't give an argument,
because that made sense before eval existed. But now it will be
equivalent to "die 'Died';".
Using the return function outside a subroutine returned a cryptic
message about not being able to pop a magical label off the stack.
It's now more informative.
On systems without the rename() system call, it's emulated with
unlink()/link()/unlink(), which could clobber a file if it
happened to unlink it before it linked it. Perl now checks to
make sure the source and destination filenames aren't in fact
the same directory entry.
The -s file test now returns size of file. Why not?
If you tried to write a general subroutine to open files, passing
in the filehandle as *filehandle, it didn't work because nobody
took responsibility to allocate the filehandle structure internally.
Now, passing *name to subroutine forces filehandle and array
creation on that symbol if they're already not created.
Reading input via <HANDLE> is now a little more efficient--it
does one less string copy.
The dumpvar.pl routine now fixes weird chars to be printable, and
allows you to specify a list of varables to display. The debugger
takes advantage of this. The debugger also now allows \ continuation
lines, and has an = command to let you make aliases easily. Line
numbers should now be correct even after lines containing only
a semicolon.
The action code for parsing split; with no arguments didn't
pass correct a corrent value of bufend to the scanpat it was
using to establish the /\s+/ pattern.
The $] variable returned the rcsid string and patchlevel. It still
returns that in a string context, but in a numeric context it
returns the version number (as in 4.0) + patchlevel / 1000.
So these patches are being applied to 3.018.
The variables $0, %ENV, @ARGV were retaining incorrect information
from the previous incarnation in dumped/undumped scripts.
The %ENV array is suppose to be global even inside packages, but
and off-by-one error kept it from being so.
The $| variable couldn't be set on a filehandle before the file
was opened. Now you can.
If errno == 0, the $! variable returned "Error 0" in a string
context, which is, unfortunately, a true string. It now returns ""
in string context if errno == 0, so you can use it reasonable in
a conditional without comparing it to 0: &cleanup if $!;
On some machines, conversion of a number to a string caused
a malloc string to be overrun by 1 character. More memory is
now allocated for such a string.
The tainting mechanism didn't work right on scripts that were setgid
but not setuid.
If you had reference to an array such as @name in a program, but
didn't invoke any of the usual array operations, the array never
got initialized.
The FPS compiler doesn't do default in a switch very well if the
value can be interpreted as a signed character. There's now a
#ifdef BADSWITCH for such machines.
Certain combinations of backslashed backslashes weren't correctly
parsed inside double-quoted strings.
"Here" strings caused warnings about uninitialized variables because
the string used internally to accumulate the lines wasn't initialized
according to the standards of the -w switch.
The a2p translator couldn't parse {foo = (bar == 123)} due to
a hangover from the old awk syntax. It also needed to put a
chop into a program if the program referenced NF so that the
field count would come out right when the split was done.
There was a missing semicolon when local($_) was emitted.
I also didn't realize that an explicity awk split on ' ' trims
leading whitespace just like the implicit split at the beginning
of the loop. The awk for..in loop has to be translated in one
of two ways in a2p, depending on whether the array was produced
by a split or by subscripting. If the array was a normal array,
a2p put out code that iterated over the array values rather than
the numeric indexes, which was wrong.
The s2p didn't translate \n correctly, stripping the backslash.
Larry Wall [Tue, 27 Mar 1990 04:46:23 +0000]
perl 3.0 patch #18 patch #16, continued
See patch #16.
Larry Wall [Tue, 27 Mar 1990 04:26:14 +0000]
perl 3.0 patch #17 patch #16, continued
See patch #16.
Larry Wall [Tue, 27 Mar 1990 20:20:03 +0000]
perl 3.0 patch #16 (combined patch)
There is now support for compiling perl under the Microsoft C
compiler on MSDOS. Special thanks go to Diomidis Spinellis
<dds@cc.ic.ac.uk> for this. To compile under MSDOS, look at the
readme file in the msdos subdirectory.
As a part of this, six files will be renamed when you run
Configure. These are config.h.SH, perl.man.[1-4] and t/op.subst.
Suns (and perhaps other machines) can't cast negative floating
point numbers to unsigned ints reasonably. Configure now detects
this and takes appropriate action.
Configure looked for optional libraries but then didn't ever use
them, even if there was no config.sh value to override.
System V Release 4 provides us with yet another nm format for
Configure to parse. No doubt it's "better". Sigh.
MIPS CPUs running under Ultrix were getting configured for volatile
support, but they don't like volatile when applied to a type generated
by a typedef. Configure now tests for this.
I've added two new perl library routines: ctime.pl from
Waldemar Kebsch and Marion Hakanson, and syslog.pl from Tom
Christiansen and me.
In subroutines, non-terminal blocks should never have arrays
requested of them, even if the subroutine call's context is
looking for an array.
Formats didn't work inside eval. Now they do.
Any $foo++ that doesn't return a value is now optimized to ++$foo
since the latter doesn't require generation of a temporary to hold
the old value.
A self-referential printf pattern such as sprintf($s,...,$s,...)
would end up with a null as the first character of the next field.
On machines that don't support executing scripts in the kernel,
perl has to emulate that when an exec fails. In this case,
the do_exec() routine can lose arguments passed to the script.
A memory leakage in pattern matching triggered by use of $`, $& or $'
has been fixed.
A splice that pulls up the front of an array such as splice(@array,0,$n)
can cause a duplicate free error.
The grep operator blew up on undefined array values. It now handles
them reasonably, setting $_ to undef.
The .. operator in an array context is used to generate number
ranges. This has been generalized to allow any string ranges that
can be generated with the magical increment code of ++. So
you can say 'a' .. 'f', '000'..'999', etc.
The ioctl function didn't return non-zero values correctly.
Associative array slices from dbm files like @dbmvalues{'foo','bar'}
could use the same cache entry for multiple values, causing loss of
some of the values of the slice. Cache values are now not flushed
until the end of a statement.
The do FILE operator blew up when used inside an eval, due to trying
to free the eval code it was still executing.
If you did s/^prefix// on a string, and subsequently assigned a
value that didn't contain a string value to the string, you could
get a bad free error.
One of the taint checks blew up on undefined array elements, which
showed up only when taintperl was run.
The final semicolon in program is supposed to be optional now.
Unfortunately this wasn't true when -p or -n added extra code
around your code. Now it's true all the time.
A tail anchored pattern such as /foo$/ could cause grief if you
searched a string that was shorter than that.
Larry Wall [Tue, 13 Mar 1990 23:33:04 +0000]
perl 3.0 patch #15 (combined patch)
In patch 13, there was a fix to make the VAR=value construct
in a command force interpretation by the shell. This was botched,
causing an argv list to be occasionally allocated with too small
a size. This problem is hidden on some machines because of
BSD malloc's semantics.
The lib/dumpvar.pl file was missing final 1; which made it
difficult to tell if it loaded right.
The lib/termcap.pl Tgetent subroutine didn't interpret ^x right
due to a missing ord().
In the section of the man page that gives hints for C programmers,
it falsely declared that you can't subscript array values. As of
patch 13, this statement is "inoperative".
The t/op.sleep test assumed that a sleep of 2 seconds would always
return a value of 2 seconds slept. Depending on the load and
the whimsey of the scheduler, it could actually sleep longer than
2 seconds upon occasion. It now allows sleeps of up to 10 seconds.
Larry Wall [Mon, 12 Mar 1990 04:13:22 +0000]
perl 3.0 patch #14 patch #13, continued
See patch #13.
Larry Wall [Mon, 12 Mar 1990 04:09:28 +0000]
perl 3.0 patch #13 (combined patch)
I added the list slice operator: (LIST)[LIST]
$hexdigit = (0..9,'a','b','c','d','e','f')[$fourbits]
There was no way to cut stuff out of the middle of an array
or to insert stuff without copying the head and tail of the array,
which is gross. I added the splice operator to do this:
@oldelems = splice(@array,$offset,$len,LIST)
Equivalencies:
splice(@array,0,1)
splice(@array,0,0,$x,$y)
splice(@array,-1,1)
splice(@array,$#array+1,0,$x,$y)
splice(@array,$x,1,$y)
Having -lPW as one of the libraries that Configure looks for
was causing lots of people grief. It was only there for
people using bison who otherwise don't have alloca(), so I
zapped it.
Some of the questions that supported the ~name syntax didn't
say so, and some that should have supported it didn't. Now they do.
If you selected the manp directory for your man pages, the manext
variable was left set to 'n'.
When Configure sees that the optional libraries have previously
been determined in config.sh, it now believes it rather than using
the list it generates.
In the test for byteorder, some compilers get indigestion on the
constant 0x
0807060504030201. It's now split into two parts.
Some compilers don't like it if you put CCFLAGS after the .c file
on the command line. Some of the Configure tests did this.
On some systems, the test for vprintf() needs to have stdio.h
included in order to give valid results.
Some machines don't support the volatile declaration as applied
to a pointer. The Configure test now checks for this.
Also, cmd.c had some VOLATILE declarations on pointed-to items
rather than the pointers themselves, causing MIPS heartburn.
In Makefile.SH, some of the t*.c files needed to have dependencies
on perly.h. Additionally, some parallel makes can't handle a
dependency line with two targets, so the perly.h and perl.c lines
have been separated. Also, when perly.h is generated, it will
now have a declaration added to it for yylval--bison wasn't supplying
this.
The construct "while (s/x//) {}" was partially fixed in patch 9, but
there were still some weirdnesses about it. Hopefully these are
ironed out now.
If you did a switch structure based on numeric value, and there
was some action attached to when the variable is greater than
the maximum specified value, that action would not happen. Instead,
any action for values under the minimum value happened.
The debugger had some difficulties after patch 9, due to changes
in the meaning of @array in a scalar context, and because of
an pointer error in patch 9.
Because of the fix in patch 9 to let return () work right, the
construct "return (@array)" did counter-intuitive things. It
now returns an array value. "return @array" and "return (@array)"
now mean the same thing.
A pack of ascii strings could call str_ncat() with negative length
when the length of the string was greater than the length specified
for the field.
Patch 9 fixed *name values so that the wouldn't collide with ordinary
string values, but there were two places I missed, one in perldb,
and one in the sprintf code.
Perl looks at commands it is going to execute to see if it can
bypass /bin/sh and execute them directly. Ordinarily = is not
a shell metacharacter, but in a command like "system 'FOO=bar command'"i
it indicates that /bin/sh should be used, since it's setting an
environment variable. It now does that (other than that construct,
the = character is still not a shell metacharacter).
If a runtime pattern to split happens to be null, it was being
interpreted as if it were a space, that is, as the awk-emulating
split. It now splits all characters apart, since that's more in
line with what people expect, and the other behavior wasn't documented.
Patch 9 added the reserved word "pipe". The scripts eg/g/gsh and
/eg/scan/scanner used pipe as filehandle since they were written
before the recommendation of upper-case filehandles was devised.
They now use PIPE.
The undef $/ command was supposed to let you slurp in an entire
binary file with one <>, but it didn't work as advertised.
Xenix systems have been having problems with Configure setting
up ndir right. Hopefully this will work better now, but it's
possible the changes will blow someone else up. Such is life...
The construct (LIST,) is now legal, so that you can say
@foo = (
1,
2,
3,
);
Various changes were made to the documentation.
In double quoted strings, you could say \0 to mean the null
character. In pattern matches, only \000 was allowed since
\0 was taken to be a \<digit> backreference. Since it doesn't
make sense to refer to the whole matched string before it's done,
there's no reason \0 can't mean null in a pattern too. So now
it does.
You could modify a numeric variable by using substr as an lvalue,
and if you then reference the variable numerically, you'd get
the old number out rather than one derived from the new string.
Now the old number is invalidated on lvalued substr.
The test t/op.mkdir should create directories 0777 rather than 0666.
As Randal requested, the last semicolon of a program is now optional.
Actually, he just asked for -e 'prog' to have that behaviour, but
it seemed reasonable to generalize it slightly. It's been that
way with eval for some time.