[prev in list] [next in list] [prev in thread] [next in thread]
List: perl5-changes
Subject: [perl.git] annotated tag v5.23.8, created. v5.23.8
From: Sawyer X <xsawyerx () gmail ! com>
Date: 2016-02-20 22:17:54
Message-ID: E1aXFqc-0007Mp-TB () camel-001 ! ams6 ! corp ! booking ! com
[Download RAW message or body]
In perl.git, the annotated tag v5.23.8 has been created
<http://perl5.git.perl.org/perl.git/commitdiff/d8e3b2382b32ad92abea50a30ace2879999d783d?hp=0000000000000000000000000000000000000000>
at d8e3b2382b32ad92abea50a30ace2879999d783d (tag)
tagging 0d316f7c6263594f0d3ceabf61cf4089beadc4f5 (commit)
replaces v5.23.7
tagged by Sawyer X
on Sat Feb 20 21:47:35 2016 +0100
- Log -----------------------------------------------------------------
Ninth release of the v5.23 series
Aaron Crane (1):
regen/mk_invlists.pl: add braces round subobject initialisers
Chris 'BinGOs' Williams (2):
Fix Module-CoreList version in Changes file
Putting the core into corelist
Craig A. Berry (14):
TODO long-failing argv.t tests on VMS.
Limit index arg to logicals in vmstrnenv().
Use vmstrnenv() to look up PERL5LIB/PERLLIB on VMS.
Integrate podlators 4.05.
Update utils commands in configure.com.
VMS does have the siginfo_si_* fields.
Tweaks to podlators integration.
Do environ key case consistently on VMS.
Integrate podlators 4.06.
perldelta for podlators update.
perldelta: recent %ENV changes on VMS.
Make File::Spec::VMS->abs2rel handle Unix-format input.
DynaLoader shouldn't use mod2fname when finding .bs files.
Cast PL_dump_re_max_len to avoid type mismatch warning.
Dagfinn Ilmari Mannsåker (5):
[perl #121351] Remove non-doio.c uses of PL_statbuf
Move email alias from AUTHORS to checkAUTHORS.pl
Add SA_SIGINFO 'code' constants to POSIX
Probe for and expose more fields for SA_SIGINFO
Skip SA_SIGINFO uid/pid test on POSIX-deficient OS X versions
Daniel Dragan (12):
backport EUMM commits
add parallelness to win32/GNUmakefile
add MSVC support to win32/GNUmakefile
win32/GNUmakefile collapse shell echos into one liners
XS DLLs shouldn't have comctl32.dll manifest in them, they dont link to it
fix link failure of APItest.dll on VC 6
fix a race condition in parallel builds with Visual C
fix op/infnan.t test fails with NAN conversion on VC 6
add shortcut around syscalls when file not found in win32_stat
fix win32 gmake with win64 VC with 32 bit GCC in PATH build failure
minor comment improvements in hv.h and scope.h
[perl #127556] update installperl to new location of W32 libperl link lib
David Mitchell (212):
eliminate an SAVEFREESV(cv) from PUSHSUB
SvREFCNT_inc(cv) recursive subs
pp_goto: use cx->blk_oldscopesp
pp_goto(): reorder LEAVE_SCOPE for consistency
pp_goto: SvREFCNT_dec(oldcv) *after* undef test
pp_entersub: remove extraneous SAVETMPS
save old PL_comppad in CXt_SUB/FORMAT block
document unrolled PUSHSUB/POPSUB
eliminate cx->blk_sub.oldcomppad
make "for my $lex {}" faster under ITHREADS
undef *_; goto &f: update cx.argarray
pp_goto: skip restoring PL_comppad
Unwind save stack in sync with POPEVAL
assert that it's safe to remove CX.argarray field
eliminate the argarray field from the CX struct
improve @_ commentary in pp_goto
pp_entersub: skip resetting @_
t/perf/benchmarks: add a few sub and goto tests
make POP_SAVEARRAY() safe
pp_goto: a couple of cosmetic changes
eliminate a goto in pp_goto (!)
pp_goto: do the DIEing before the POPing
goto.t: add freeing CV test
avoid leaking @_ in goto
add Perl_clear_defarray()
pp_goto(): use clear_defarray()
clear_defarray(): clear @_ if possible
pp_leavesub: simplify recursion test
pp_entersub(): mortal-copy args earlier
pp_entersub(): remove an unnecessary condition
pp_entersub(): move SAVETMPS next to PUSHSUB
pp_goto: do SAVETMPS etc in XS and non-XS branches
pp_sort: move SAVETMPS later
pp_dbstate: do SAVETMPS etc in both branches
PUSH_MULTICALL: move SAVETMPS later
add old_tmpsfloor field to CXt_SUB context frame
pp_entersub(): eliminate a label
pp_entersub(): simplify autoload logic
pp_hot.c: skip unnecessary test
Eliminate ENTER/LEAVE from sub calls
tweak POPSUB()
revamp pp_entersub()'s CV locating code
pp_entersub(): don't prematurely calc hasargs
pp_entersub(): reduce scope of gimme
pp_entersub: unroll some CvFLAGS(cv) tests
pp_entersub(): reduce life of padlist var
CXt_FORMAT: save ss_ix and tmps_floor in CX struct
call_sv(), fold_const(): different CX pop test
move SAVETMPS next to PUSHEVAL
CXt_EVAL: save savestack_ix and tmps_floor in CX
only call leave_common in non-void context
pp_last: use debugging LEAVE variant
for my $x (...): $x is always stale
S_leave_common(): simplify SVs_PADTMP handling
pp_leaveloop: use SVs_PADTMP|SVs_TEMP
pp_return(): handle dounwind() freeing args
pp_return(): reindent following previous commit
POPLOOP: call LEAVE_SCOPE()
Add itersave field to LOOP context struct
POPLOOP(): no need to mortalise current item
for loops: don't refcount bump orig var
pp_enteriter: tidy itervar-setting code
add CXp_FOR_PAD, CXp_FOR_GV flags
pp_enteriter: don't create new SV for GvSV slot
pp_enteriter, POPLOOP: simplify some code
context: move couple of fields into block struct
pp_given: avoid using savestack for old var
Consistently call leave_common() before POPFOO
Make remaining context types avoid ENTER/LEAVE
add loop benchmark tests
rename S_dopoptogiven() to S_dopoptogivenfor()
pp_leavewhen: simply for() handling
pp_leavewhen(): skip POPWHEN()
add PUSH/POPBASICBLK macros.
move POPBLOCK after arg stack munging
POPBLOCK: don't set newsp and gimme
call LEAVE_SCOPE() before POPEVAL()
add CX_LEAVE_SCOPE(cx) macro
move CX_LEAVE_SCOPE into POPEVAL
make LEAVESUB() always immediately follow POPSUB()
eliminate LEAVESUB() macro
eliminate weird gimme calc in pp_leave()
simplify S_leave_common() and callers
make PL_tmps_floor restore consistent
reverse the order of POPBLOCK; POPFOO
remove newpm param from POPBLOCK() macro.
pp_sort: add missing CX_LEAVE_SCOPE()
sort(!) out CXt_NULL and CXp_MULTICALL
do PL_tmps_floor restore in POPBLOCK
do PL_tmps_floor save in PUSHBLOCK
move CX_LEAVE_SCOPE outside the POPFOO's
move sbu_oldsave into base of CX struct
move blku_old_savestack_ix to base of cxt struct
dounwind: CX_LEAVE_SCOPE for CXt_SUBST too
dowinwind(): move common CX_LEAVE_SCOPE() outside
make POPSUBST not cxstack_ix--
split some common POPSUB code into a macro
pp_coreargs: rationalise @_ code
factor common code into POPSUB_ARGS()
remove redundant croak code in pp_leavesublv
simplify CHANGE_MULTICALL_FLAGS
add CX_POP(cx) macro: glorified cxstack_ix--
POPEVAL: don't set optype
add S_undo_inc_then_croak()
Perl_die_unwind(): remove unneeded assert
pp_return: avoid potential CX stack realloc prob
remove obsolete panic from die_unwind()
S_doeval(): tidy up comments
rename S_doeval() to S_doeval_compile()
pp_leaveeval: reset stack in VOID context
stop S_undo_inc_then_croak() doing CX_POP
benchmarks: add some 'for' array iterating
split CXt_LOOP_FOR into CXt_LOOP_LIST,CXt_LOOP_ARY
pp_iter(): optimise stack handling
Eliminate cx->blk_loop.resetsp
PUSHLOOP_PLAIN: don't set unused fields
simplify two conditions in pp_iter:
move and rename cx_old_savestack_ix
reorder 'struct block' fields.
make POPSUB and POPFORMAT re-entrant safe
make POPEVAL safe against re-entrancy
make POPSUB re-entrant safe
make POPGIVEN re-entrant safe
make POPSUBST re-entrant safe
add CX_CUR() macro
rename DEBUG_CX() to CX_DEBUG()
make TOPBLOCK(cx) not set cx
op/sub.t: fix ticket number in comment
Always copy return values when exiting scope
test that pp_leavesub copies returned PADTMPs.
extend magic copy test to all scope exit types
restore PL_tmps_floor on exit
Perl_free_tmps(): don't test for PL_sv_undef
pp_leavesub: reset SP in void context
intrvar.h: document PL_tmps_max
optimise sv_setsv_flags()
Perl_runops_debug(): do FREETMPS
move SET_SVANY_FOR_BODYLESS_IV() from sv.c to sv.h
pp_leavesub(): call FREETMPS and optimise
pp_leavesublv(): croak on *all* PADTMPs
pp_leavesublv(): document PL_sv_undef exception
make pp_leavesublv use S_leavesub_adjust_stacks()
make pp_return() use leave_adjust_stacks()
replace leave_common() with leave_adjust_stacks()
pp_ctl.c: s/newsp/oldsp/g
POPSUB_ARGS: move a code comment to the right line
TOPBLOCK: make comment clear its used by goto too
provide some basic documentation for Perl_dounwind
pp_enteriter: use efficient SvREFCNT_inc variant
fix *_ = "" for 0 .. 1;
pp_iter: optimise integer setting in for (1..10) {}
fixup cx_dup()
APItest.xs: fixup clone_with_stack()
clarify code comment in pp_goto(()
give POP_SAVEARRAY() macro a cx arg
only set CXp_FOR_DEF with CXp_FOR_GV
fix CxFOREACH
tweak POPLOOP and CXt_LOOP_* order
pp_break(): don't use TOPBLOCK
dounwind(): do a POPBLOCK for final cx frame.
factor out common actions in TOPBLOCK and POPBLOCK
S_unwind_loop(): return pointer rather than index
S_unwind_loop(): remove opname param
make S_unwind_loop static again
optimise bare 'next'
pp_redo()): reorder some stuff
rename POPFOO() to CX_POPFOO()
CX_POPFOO(): assert cx is of the right type
Document CxLVAL()
MULTICALL should clear scope after each call
sort compare subs: don't do unnecessary scope work
eliminate PERL_STACK_OVERFLOW_CHECK
PUSH_MULTICALL: use SAVEOP()
move PL_savestack_ix saving into PUSHBLOCK
PUSHBLOCK: don't use implicit args
PUSHSUB: don't use implicit args
PUSHFORMAT: don't use implicit args
PUSHEVAL: make retop a parameter
PUSHSUB: make retop a parameter
PUSHEVAL: make n param an SV rather than a string
consolidate common code in PUSHLOOP_FOR,_PLAIN
pp_enteriter: add comment about setting cxt type
eliminate PUSH/POPBASICBLK macros
rename PUSHBLOCK,PUSHSUB etc to CX_PUSHBLOCK etc
add SAVEt_TMPSFLOOR save type and Perl_savetmps()
offset PL_savestack_max by SS_MAXPUSH
add a few grep and map benchmarks
convert CX_PUSH/POP/TOPBLOCK to inline fns
convert CX_PUSHSUB/POPSUB to inline fns
convert CX_PUSHFORMAT/POPFORMAT to inline fns
convert CX_PUSHEVAL/POPEVAL to inline fns
convert CX_PUSHLOOP*/POPLOOP to inline fns
convert CX_{PUSH|POP}{WHEN|GIVEN} to inline fns
dMULTICALL: remove unused vars
add blk_old_tmpsfloor shortcut
MULTICALL *shouldn't* clear savestack
fix cx_dup for CXt_LOOP_PLAIN
perlguts: add section on context stack
perlfunc: say what block types 'return' recognises
fix -DPERL_GLOBAL_STRUCT_PRIVATE
make gimme consistently U8
leave_adjust_stacks(): avoid accessing random tmps
leave_adjust_stacks() fix some code comments
remove dSP from a couple of pp_enter* fns
[MERGE] revamp context system
add perldelta entry for context stack work
Add support for bison 3.0
pp_enter: calculate gimme earlier in XS branch
regen_perly.pl: print command with -v
regen_perly.pl: improve action extracting
run regen_perly.pl
doop.c: fix typo in header comment
Ed J (2):
Cache is_make_type
Optimise is_make_type RE
H.Merijn Brand (1):
Updated outdated link to smoke reports for HP-UX
Herbert Breunung (1):
perlretut: typo correction
James E Keenan (2):
Update guidance on naming of modules.
perldelta for f94c6c536844091ca6a005e3e0398db8e1cc212e
Jarkko Hietaniemi (46):
[perl #127183] Non-canonical hexadecimal floats are parsed prematurely
Do not export no text symbols starting with S_.
Add STATICs to S_ functions.
Additional hexfp %a tests inspired by c95ea682.
Oddly placed unused decls for fma() and the gamma funcs.
We're against contractions.
POSIX version bump.
OpenBSD does not do si_uid with sigaction().
Cast away Solaris Studio 12.3 warning.
cmpVERSION STDERR messages for test failures.
assert() that itersvp is non-NULL.
Lexical scoping in case statement is tricky.
Whitespace only: zap empty lines.
If not using smallbuf and len > sizeof(d_name), Move() is illegal.
Check for invlist_search() returning negative array indices.
assert(cv) before doing CvROOT(cv)
Check against negative uid/gid for fchown().
Do not try to fchown() to uid -1 and gid -1.
Assert no bad array access.
assert(PL_parser)
Add missing break in switch.
Add missing break in switch.
ODBM_File: Avoid TOCTOU and using negative returns.
XS-APItest: Length cannot be negative.
Storable: Own ASSERT or no, we want to assert(prev).
POSIX: Check signal numbers against negatives.
POSIX: Check fds against negatives.
POSIX: strcmp NE strEQ().
POSIX version bump.
Storable version bump.
Sort the Time::HiRes constants, one per line
Add Linux-specific clock_gettime() constants.
Add FreeBSD specific clock_gettime() constants.
Add the new Time::HiRes constants to @EXPORT_OK.
OS X clock_gettime() and clock_getres() emulation
OS X clock_nanosleep() emulation
Mention the OS X get_clock...() emulations.
Add caveat on the clock_getres() resolution
Add the Time-HiRes Changes file from CPAN.
Whitespace only: zap EOL spaces
Time::HiRes version bump.
Solaris /usr/bin/sed cannot handle labels of length eight.
Skip the length sanity check if d_name is pointer or less.
Allow TIME_HIRES_DONT_RUN_PROBES=1 to aid cross-compiling
Time::HiRes version bump.
Upgrade to IPC-SysV 2.05.
Karl Williamson (63):
utf8.c: Add missing 'STATIC' to declaration
ext/POSIX/lib/POSIX.pod: Make verbatim line fit in 79 cols
utf8.c: Add cast to suppress a warning message
regexec.c: Macro needs param to be dereferenced
Escape t/test.pl got vs expected strings
Use table lookup for qr/\b{wb}/
regexec.c: Fix comment, white-space
regexec.c: Refactor \b{sb} handling
re/uniprops: Fix EBCDIC issue
perldelta: Make verbatim line fit in 79 columns
perlguts: Make verbatim lines fit in 79 cols
podcheck.t: regen db for overlong verbatim lines
perlapi: Clarify that a literal string must end in a NUL
perldelta, perlguts: Fix typos
locale.c: Improve -DL debug info
regcomp.c: Add code to compute edit distance (Damerau–Levenshtein)
Add Nick Logan to AUTHORS
autodoc.pl: Fix misspelled /[[:alpha:]]/
regcomp.c: White-space, variable name-change only
regcomp.c: Fix recursive parsing bug
PATCH: [perl #8904] Revamp [:posix:] parsing
podcheck.t: Need to translate E<lt> and E<gt>
porting/diag.t: Handle some E<> pod escapes
regcomp.c: Extract duped code into one fcn
regcomp.c: Fix some parsing glitches
regcomp.c, regexec.c: Comments, white-space only
utf8.h: Guard some macros against improper calls
regcomp.c: Add some grouping parens
regcomp.c: Avoid a function call in a common case
regcomp.c: Replace invalid assertion
regexec.c: Skip duplicate work
sv.c: Handle radix being multi-byte and not UTF-8
t/re/regex_sets.t: Add some tests
regcomp.c: Clarify error message
Remove POSIX isfoo() as scheduled
t/re/reg_mesg.t: Add a couple of tests
perlrecharclass: Add some missing info
Don't allow /\N{}/ under 're strict'
regcomp.c: Use macro to hide complexity
regcomp.c: Fix -Dr bug
regcomp.c: -Dr \xZZ instead of \x{ZZ}
regcomp.c: Clean up logic in function
regcomp.c: Simplify a few lines of code
regcomp.h: Not all ANYOF flags are in use.
perlapi: Hide the swash functions
Cast correctly to U8, not char
regcomp.c, toke.c: swap functions being inline static
PATCH: [perl 127537] /\W/ regression with UTF-8
regcomp.c: Avoid a segfault
regcomp.c: optimization for qr/[...]/il
regcomp.c: Move static declaration to file level
regcomp.c: Change structure element size and loc
Change private function to static
Add a parameter to a static function
regcomp.c: Comments, white-space, add grouping () for clarity
regcomp.c: Clarify -Dr output under /l
regcomp.c: Save a branch test
Add environment variable for -Dr: PERL_DUMP_RE_MAX_LEN
Revamp -Dr handling of /[...]/
regcomp.c: Backlslash {} in -Dr output
regcomp.c: Use colors for -Dr metanotation
regcomp.c: Can't do optimization if inverting
regcomp.c: White-space only
Lukas Mai (10):
perlhacktips: fix / properly break example of bad pointer access
mg.c: move declaration of i closer to its use
util.c: fix/simplify unused arg logic in my_vsnprintf
perlop: fix broken example by deleting it [perl #119667]
ensure isASCII argument is an integer
PerlIO::encoding: explicitly cast char * to STDCHAR *
README.android: make the POD a bit nicer
op.c: update comment about compiler warnings
tweak NOT_REACHED in DEBUGGING builds
Revert "tweak NOT_REACHED in DEBUGGING builds"
Niko Tyni (1):
Fix umask for mkstemp(3) calls
Pip Cet (1):
[perl #127474] fix operator precedence when (castflags & 2)
Ricardo Signes (3):
move Time-HiRes from cpan to dist
update release schedule for beginnings of 5.25
release schedule: September 2016 is scheduled
Sawyer X (10):
Remove outdated task in release:
Update perldelta.pod:
Document eba98284740b799def26803ebac340aaa8930437
Document e57270be442bfaa9dc23eebd67485e5a806b44e3:
Document 38e3b24
cleanup perldelta
update Module::CoreList (Porting/corelist.pl)
Update perldelta module versions (Porting/corelist-perldelta.pl)
Typos, POD errors, etc.
add new release to perlhist
Sisyphus (1):
Win32 gmake needs SHELL to be specified
Stevan Little (5):
updating the epigraph and release schedule
new perldelta
bumping version
manual tweak
more version bumping
Steve Hay (4):
Upgrade Encode from version 2.78 to 2.79
Porting/Miantainers.pl update following changes in 273df2b189...538ad527dc
perlipc.pod: Fix typo
Upgrade Encode from version 2.79 to 2.80
Todd Rinaldo (1):
Document broken SvSTASH for %version:: in B's test suite
Tom Hukins (2):
Time::HiRes moved from "cpan" to "dist" in 91ba54
Remove an unused variable
Tony Cook (34):
add 5.23.8 placeholders to Module::CoreList
[perl #126991] treat cop_line as unsigned in caller() (since it is)
perldelta for e9e9e546c676
[perl #127122] warn on unless (assignment) when syntax warnings are on
perldelta for 08b3e84fbb1c
bump $XS::APItest::VERSION
Add parallel build and MSVC support for the gmake makefile
perldelta for 273df2b1892a ... 273df2b1892
[perl #126632] improve diagnostics for the comctl32 test
[perl #126632] more diagnostics
Revert "[perl #126632] more diagnostics"
[perl #126045] part revert e9b19ab7 for vc2003 and earlier
perldelta for 6f6d1bab334
[perl #127426] fixes for 126045 patch, restrict to MSVC, add casts
[perl #127351] TODO tests for arrayref assigned to *Foo::ISA issues
[perl #127351] add isaelem magic on *Foo::ISA = arrayref
perldelta for 8452c1a03e174
Clarify sprintf() handling of + and space flags
[perl #125540] handle already being at EOF while not finding a heredoc \
terminator perldelta for 23c4e91245a4
[perl #126544] correct the first example in the fcntl documentation
perldelta the fix for [perl #126621]
add Pip Cet to AUTHORS
perldelta for 071db91b12fc
[perl #126410] keep the DESTROY cache in mro_meta
[perl #124387] TODO test for AUTOLOAD on DESTROY
[perl #124387] call AUTOLOAD when DESTROY isn't defined
perldelta for 27895dda8085, ac3b837b9e1b, f05081b8ef22, 000814da4770
[perl #127494] TODO test for $AUTOLOAD being set for DESTROY
[perl #127494] don't cache AUTOLOAD as DESTROY
perldelta for 7db8c4f1f19e and 958cdeac409
[perl #127334] S_incline: avoid overrunning the end of the parse buffer
perldelta for 1bb1a3d6d35
perldelta: move the two Win32 gmake improvements to where they belong
bulk88 (1):
t/echo.t needs SHELL env for Win32 gmake
-----------------------------------------------------------------------
--
Perl5 Master Repository
[prev in list] [next in list] [prev in thread] [next in thread]
Configure |
About |
News |
Add a list |
Sponsored by KoreLogic