with something that doesn't do a very good job of optimising autorelease
operations away.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@33644 72102866-910b-0410-8b05-ffd578937521
always the word before isa.
Switch to using the correct kind of alignof (the ABI alignment in structures,
not the preferred alignment of the type - this stops us from allocating 4 bytes
of padding on x86-32 where there is no need for any)
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@33642 72102866-910b-0410-8b05-ffd578937521
Recent Clang trunk versions complain about it.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@33634 72102866-910b-0410-8b05-ffd578937521
Richard: I'm unsure about three of these, which were fixes in memset() calls in:
- NSConcreteMapTable.m
- NSConcreteHashTable.m
- Additions/NSData+GNUstepBase.m
Please can you check them? I think they are intended to zero the entire object
(rather than the first word), but the lack of comments makes me unsure.
Most changes were just tweaks to variable types. I've also removed some dead code from NSInvocation. This was small group of things that were marked for internal use only, but not actually referenced in the code anywhere.
Other improvements:
- NSArray / NSDictionary fixed up to use the 10.7 (ARC-friendly) prototypes.
- getObjects:andKeys: implemented for NSDictionary (10.5 method)
- NSPointerArray and NSHashTable now properly support weak objects.
- Tests for weak objects in collections.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@33621 72102866-910b-0410-8b05-ffd578937521
These are currently unused - the next commit will make NSMapTable use them.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@33612 72102866-910b-0410-8b05-ffd578937521
To disable zone support completely, NSAllocateObject() should ignore the zone and NSDeallocateObject() should skip the zone lookup.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@33610 72102866-910b-0410-8b05-ffd578937521
(-attributedStringForObjectValue:withDefaultAttributes:):
Guard against initializing an NSAttributedString with a nil string.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@33596 72102866-910b-0410-8b05-ffd578937521
passed to -initWithString:attributes: doesn't respond to -length, or if
it is nil. This is the same behaviour as OS X.
* Source/NSAttributedString.m (-init): Call -initWithString:attributes:
with @"" instead of nil, since passing nil now causes an exception
to be thrown.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@33595 72102866-910b-0410-8b05-ffd578937521
Also do some quite hacky (and not totally correct) things to try to make sure that we aren't confused into thinking that the first NSThread is the main thread, if it's created on a separate thread. Currently only supported on FreeBSD and OpenBSD.
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@33521 72102866-910b-0410-8b05-ffd578937521
These show deviation from OS X behaviour (the first test passes by accident - it's actually doing the wrong thing in two different ways that cancel each other out).
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@33520 72102866-910b-0410-8b05-ffd578937521
Patch by Sebastian Reitenbach <sebastia@l00-bugdead-prods.de>
git-svn-id: svn+ssh://svn.gna.org/svn/gnustep/libs/base/trunk@33515 72102866-910b-0410-8b05-ffd578937521
