Commit Graph
73 Commits
Author SHA1 Message Date
Sudheer Kumar Peddireddy 427ecfea83 Increase the initial sizes of hash tables that are guaranteed to grow. b=432633 r=dbaron 2008-06-03 15:25:31 -07:00
7b34ae7602 Back out bug 412320 and various followups because the perf impact was minimal and the code is more complicated 2008-02-21 12:39:20 -08:00
db8d572868 Bug 412320 - Try code-generation for QI impls, r=roc+luser 2008-01-25 09:14:48 -08:00
a113c5d648 Bug 387023: minor string cleanup in nsAtomTable.cpp. r=bsmedberg, a=bzbarsky. 2007-09-05 18:13:42 -07:00
8a277235d9 Bug 348748 - Cleanup from the handful of patches which have landed since the initial landing that have readded cast macros; I intend to remove the rest of the instances Very Soon, all but certainly before the M8 freeze, so the macro definitions can be removed, again all but certainly before the M8 freeze, if people are okay with them being removed. Still r=bsmedberg, a=no-functionality-change 2007-09-04 23:41:35 -07:00
126c483d51 Bug 387445: Remove thread safety from atoms. r+sr=dbaron, a=bzbarsky 2007-08-07 12:16:52 -07:00
8be89d4076 Fixing bug 377360. Fix atom table crash due to invalid UTF data in atoms. r=jonas@sicking.cc, sr=peterv@propagandism.org 2007-07-11 13:46:43 -07:00
98de634631 Bug 348748 - Replace all instances of NS_STATIC_CAST and friends with C++ casts (and simultaneously bitrot nearly every patch in existence). r=bsmedberg on the script that did this. Tune in next time for Macro Wars: Episode II: Attack on the LL_* Macros. 2007-07-08 00:08:04 -07:00
0f08b75950 Remove unused getKey callback from PLDHashTableOps/JSDHashTableOps. b=374906 r=bsmedberg 2007-03-27 08:33:38 -07:00
mats.palmgren@bredband.net 7f3bcc12a7 Don't crash on OOM. b=353997 r+sr=darin 2006-09-30 05:15:57 +00:00
dbaron@dbaron.org 3ad2463c7a Fix threadsafety regression so we're back to the old broken state. b=335734 r=sicking rs=brendan 2006-05-06 17:49:21 +00:00
dbaron@dbaron.org 7452284d8f Make IsPermanent non-virtual so it can be called in the destructor. b=334605 r=mrbkap rs=brendan 2006-04-19 21:46:47 +00:00
cbiesinger@web.de aae44b57d9 bug 183156 remove *UCS2* functions, replacing them with *UTF16* ones
r+sr=darin
2006-02-03 14:18:39 +00:00
bzbarsky@mit.edu b47c9f2b3e Implement a in-place (no copy) CompareUTF8toUTF16, and use it to make the atom
hashtable lookups zero-copy.  Patch by jst, bug 314465 (with lots of the
discussion in bug 277479), r=bsmedberg,dbaron,brendan (on the PLDHashTable
keyhash value assumptions), sr=bzbarsky, moa=shaver.
2005-11-04 19:52:18 +00:00
gerv@gerv.net c0cd46fc51 Bug 236613: change to MPL/LGPL/GPL tri-license. 2004-04-18 14:21:17 +00:00
jst@mozilla.jstenback.com 0c5e542fe7 Fixing bug 233307. deCOMtaminating nsIScript* and related interfaces. r+sr=bryner@brianryner.com. 2004-02-09 22:48:53 +00:00
bryner@brianryner.com 73c3254bfa Continuing removal of unneeded virtual destructors (bug 229875). r=dougt, sr=dbaron. 2004-01-22 01:01:19 +00:00
timeless@mozdev.org 09e16f2fe3 Bug 211275 GetAtomHashEntry doesn't check the return value of PL_DHashTableInit
r=dbaron sr=bz
2003-07-03 14:45:01 +00:00
alecf@flett.org b2eac91a89 fix bug 209699 r=dougt, sr=jst - move some consumers over to CopyUTF8toUTF16 2003-06-18 02:22:46 +00:00
alecf@netscape.com 80b5831587 fix for bug 195298 - pldhash ops tables should be const.. r=dougt, sr=sfraser 2003-04-04 15:10:37 +00:00
alecf@netscape.com 30294ad661 remove #ifdef NS_DEBUG as per bug 200241
code was already reviewed, just moved out of #ifdef per bryner
2003-04-02 15:47:16 +00:00
alecf@netscape.com 2fbddd4893 landing attempt #2 for 195262 - support for static atoms, and converting atoms over to UTF8 to reduce bloat - now without performance regressions!
sr=darin, r=dbaron
2003-03-25 18:55:59 +00:00
alecf@netscape.com 1f1c5bf595 backout my static atom patch for now until I figure out the performance loss! :(
bug 195262
2003-03-21 20:13:08 +00:00
alecf@netscape.com 5a84e994cb oops, don't free the arena pool if we never allocated it!
duh.
fixes a shutdown crash
2003-03-21 08:38:56 +00:00
alecf@netscape.com 34b709508c fix for bug 195262:
- convert atoms to UTF8 because 99% of the time, that is what they are used for
- add support for static atoms, where the strings are declared in a large const array and not copied into the heap
r=dbaron, sr=darin

this is an experimental landing - if there are performance issues, I will back it out.
2003-03-21 06:26:32 +00:00
dbaron@dbaron.org 834096485c Remove nsISizeOfHandler and associated SizeOf methods. b=106792 r=bzbarsky sr=jst 2003-02-22 15:34:38 +00:00
dbaron@dbaron.org 8c37890522 Bug 178643: Remove uses of NS_INIT_ISUPPORTS, since it is no longer needed. r=timeless sr=jag 2003-01-08 23:19:20 +00:00
seawood@netscape.com 0ae28f14fb Add CPP_THROW_NEW macro to the build system. This macro should be used when defining a local |operator new| function. It will be set to an empty throw function by default on all platforms except mac CW builds. This will allow us to not crash on systems that expect |operator new| to throw an exception by default and will suppress compiler warnings about how |operator new| should throw an exception instead of returning NULL.
Bug 149032 r=timeless sr=scc
2002-07-02 20:25:30 +00:00
dbaron@fas.harvard.edu e990a4631d Convert users of nsAReadable[C]String and nsAWritable[C]String typedefs to [const] nsA[C]String. b=131899 r=scc sr=jag a=asa 2002-03-24 00:16:18 +00:00
jst@netscape.com 4df5e8e48b Fixing blocker bug 106527 (as part of bug 69468), allow creation of atoms for empty strings but don't crash when creating atoms for strings that have no internal buffer. r=dbaron@netscape.com, sr=brendan@mozilla.org 2001-10-24 23:36:58 +00:00
jst@netscape.com 197bfa1337 Fixing bug 69468. Finally adding support for string nullness to XPConnect. r=jaggernaut@netscape.com, shaver@mozilla.org, sr=jband@netscape.com 2001-10-24 07:15:06 +00:00
dbaron@fas.harvard.edu 41aa292546 Reduce the performance cost of (threadsafely) refcounting many atoms by creating a second implementation of nsIAtom (derived from the first) that does not refcount and gets destroyed at XPCOM shutdown. Normal atoms are converted to permanent ones without loss of pointer identity when a permanent atom is requested and a normal one already exists. b=92141 r=waterson sr=brendan 2001-10-20 23:19:07 +00:00
dbaron@fas.harvard.edu d0430dd000 Making debugging code be |#ifdef DEBUG|. b=91036 r=harishd sr=jst 2001-10-16 05:31:36 +00:00
dbaron@fas.harvard.edu d52d3d978e Replace uses of overloaded |fputs| and most uses of |ns[C]String::ToCString|. b=104763 r=jag rs=scc 2001-10-16 03:53:44 +00:00
rjesup@wgate.com 095dd7186f Bug 103330: Use Assign() instead of SetLength(0); Append(). r=jag, sr=alecf 2001-10-08 20:07:04 +00:00
gerv@gerv.net 58d409fba9 Relicensing Round 1, Take 2. Most C-like NPL files -> NPL/GPL/LGPL. Bug 98089. 2001-09-28 20:14:13 +00:00
waterson@netscape.com 92768eba98 Bug 95296. Use nsCRT::HashCode() instead of bad lossy slow hash routine. sr=brendan 2001-08-16 03:24:45 +00:00
jst@netscape.com 519fe64e10 scc checking in from jst's account. bug #75220: sr=jst, rs=brendan. renaming |nsPromiseC?Concatenation| to |nsDependentC?Concatenation|; |nsPromiseC?Substring| to |nsDependentC?Substring|; |nsLiteralC?String| and |nsLocalC?String| to |nsDependentC?String|, as these new names better reflect clients obligations to instances. 2001-05-19 11:27:30 +00:00
scc@mozilla.org 4e56e9386c landing string branch; see bug #73786 2001-04-02 19:40:52 +00:00
disttsc@bart.nl f25f68afff Fix bad nsPromiseFlatString usage in nsAtomTable, bug=71759, r=bryner, sr=shaver 2001-03-14 07:41:08 +00:00
sspitzer@netscape.com 7be0795135 fix for #58238. sr=bienvenu. fix crash when growing the hash table
that contained the list of atoms.  see bug report for all the details.
thanks to bienvenu, scc, brendan, and hyatt for the help.
2000-12-18 19:58:20 +00:00
warren@netscape.com a98f484246 Bug 47207 - removing printf / logging changes. Sticking in big toe. r=valeski,sr=waterson 2000-10-27 06:18:39 +00:00
scc@mozilla.org 19ee7fdf68 bug #49817: fix |normalize_forward|, |normalize_backward|; make |Begin(Read|Writ)ing| call |normalize_forward|; pass iterators by reference. r=mjudge 2000-09-02 04:10:44 +00:00
jst@netscape.com 5eb0a62024 Fixing AIX and SunOS 5.7 bustage. r=scc&vidur 2000-08-24 22:17:25 +00:00
jst@netscape.com 37c60f9ad7 Attempting to fix AIX and OS2 bustage, code written by scc, r=jst 2000-08-23 23:36:58 +00:00
jst@netscape.com 3ba32f670e DOM string changes. All nsString& in DOM interfaces (and interfaces needed by DOM implementations) have been changed to nsAReadableString& and nsAWritableString&. String implementation additions (sanctioned by scc) to support DOM needs. Bug 49091. r=vidur,jst,scc 2000-08-23 17:27:06 +00:00
warren@netscape.com faab19500d Fix for hash code performance problem discovered by bienvenu. 'Sampling' hash code was statistically evil. 2000-08-20 21:29:10 +00:00
warren@netscape.com 8df9a9bce0 Bug 46711. Removed nsAutoString travisty from nsStringKey. Introduced nsCStringKey. Made them both share the underlying string when possible. r=waterson 2000-08-10 06:19:37 +00:00
mkaply@us.ibm.com 25cfcb8e6a # 34082
r= dougt@netscape.com
OS/2 bring up continues - Adding PR_CALLBACKs for VisualAge Compiler
2000-05-02 22:38:04 +00:00
scc@netscape.com 0cbd1727d3 making string conversions explicit 2000-04-01 00:39:02 +00:00