313 Commits

Author SHA1 Message Date
Steve Naroff
3434bebec9 Make sure @property is allowed within a category.
Bug submitted by Keith Bauer.

CookieJar:Desktop keith$ cat test.m
#import <WebKit/WebKit.h>

llvm-svn: 44007
2007-11-12 04:22:52 +00:00
Steve Naroff
d0bf516e70 Remove Sema::ObjcBuildMethodParameter().
Modify Sema::ParseParamDeclarator() to work for both ActOnStartOfFunctionDef() and ObjcActOnStartOfMethodDef().

llvm-svn: 44006
2007-11-12 03:44:46 +00:00
Steve Naroff
b313fc3203 Replace 2 method definition actions (ActOnFunctionDefBody, ActOnMethodDefBody) with 1 method definition action (ActOnFinishFunctionBody). I can't think of any reason that we would need two action hooks.
llvm-svn: 44000
2007-11-11 23:20:51 +00:00
Steve Naroff
cac26f4f5f This is the last 5% of the solution to teaching Sema::ActOnInstanceMessage() about private methods (r43989).
While the diff is large, the idea is very simple. When we parse method definitions (in an @implementation), we need to add them incrementally (rather than wait until the @end).

Other details...

- Renamed Sema::ActOnAddMethodsToObjcDecl() to Sema::ActOnAtEnd(). The methods are now optional arguments.
- Removed Parser::AllImplMethods (a nice cleanup).
- Added location info to ObjcImplementationDecl (since we will need it very soon:-)
- Modified message.m test to no longer allow the bogus diagnostic.

llvm-svn: 43995
2007-11-11 17:19:15 +00:00
Steve Naroff
22e078e013 Teach Sema::ActOnInstanceMessage() about private methods. That is, methods declared in an implementation (but not listed in the interface).
This commit is only 95% of the bug fix. The last piece to this puzzle is to add the method decls to the implementation incrementally (as we encounter them). At the moment, the methods aren't added until we see an @end (which is too late).

I will complete this later...

llvm-svn: 43989
2007-11-11 00:10:47 +00:00
Fariborz Jahanian
0bd56b7727 Fixed a bug which exposed the internally built type to user code.
llvm-svn: 43987
2007-11-10 22:00:55 +00:00
Fariborz Jahanian
6aafd4130b Minor code clean up in method def area.
llvm-svn: 43980
2007-11-10 17:40:57 +00:00
Fariborz Jahanian
85e1d0d807 Represent method definitions as separate AST nodes. Pretty print will come next.
llvm-svn: 43979
2007-11-10 16:31:34 +00:00
Fariborz Jahanian
7a017215eb Some code clean up in the form of name changes for functions which
process method definitions.

llvm-svn: 43967
2007-11-09 22:27:59 +00:00
Fariborz Jahanian
b5605179c4 Added class context to method declarations. Provide "interface *" type
to 'self' method of instance methods.

llvm-svn: 43957
2007-11-09 19:52:12 +00:00
Fariborz Jahanian
19d4dbd217 Insert invisble arguments to method definition header.
llvm-svn: 43948
2007-11-09 17:18:29 +00:00
Fariborz Jahanian
56ff146ec4 First patch toward rewriting of method definitions. This is work in progress.
llvm-svn: 43915
2007-11-08 23:49:49 +00:00
Fariborz Jahanian
f76f2b0c1c Patch for objc2's property ASTs, as well as pretty-priting the ASTs.
llvm-svn: 43778
2007-11-06 22:01:00 +00:00
Chris Lattner
45d561ad99 improve decl merging logic to be more correct with
functions.  Patch contributed by Nuno Lopes, thanks!

llvm-svn: 43757
2007-11-06 06:07:26 +00:00
Chris Lattner
3209725c68 Add better validation for array types when merging decls. Patch
contributed by Oliver Hunt, thanks!

llvm-svn: 43750
2007-11-06 04:28:31 +00:00
Fariborz Jahanian
ac73ff8868 Remaining work to collect objective-c's type qualifiers and use them to encode
method types.

llvm-svn: 43617
2007-11-01 17:18:37 +00:00
Devang Patel
32714064b1 Propagate bitfield info.
llvm-svn: 43613
2007-11-01 16:29:56 +00:00
Fariborz Jahanian
7a9c47480d 1) More additions for objective-c's qualifier type.
2) Fixed a test failure (which should have failed all along!).

llvm-svn: 43589
2007-10-31 23:53:01 +00:00
Fariborz Jahanian
95b6076c77 Fixed problem with rewriting stand-alone @implementation (with no matching @interface).
A new test case added.

llvm-svn: 43568
2007-10-31 18:48:14 +00:00
Steve Naroff
6d40db0dde Implement a more sensible strategy for ObjC built-in types (addressing a long standing FIXME in Sema::GetObjcIdType()).
This removes several gross hacks to work around the previous "lazy" behavior.

Two notes:
- MinimalActions still needs to be taught about the built-in types (This breaks one of the -noop test cases). I started this, then added a FIXME.
- I didn't convert Sema::GetObjcProtoType() yet.

llvm-svn: 43567
2007-10-31 18:42:27 +00:00
Fariborz Jahanian
d728e949bb After Anders check-in, we can now encode 'Class' type.
llvm-svn: 43556
2007-10-31 16:00:13 +00:00
Fariborz Jahanian
ca5338196f Added new type and bitfield fields in some decl types in preparation for objective-c's type qualifiers.
Added initialization of Class/SEMA types.

llvm-svn: 43534
2007-10-31 00:12:35 +00:00
Chris Lattner
0b11172e8c minor tweaks
llvm-svn: 43515
2007-10-30 20:57:56 +00:00
Fariborz Jahanian
18d7b30241 Revisited my last patch to be able to do encoding of ivar types with 'id'.
llvm-svn: 43507
2007-10-30 18:27:03 +00:00
Fariborz Jahanian
509d8d6fc6 Added type encoding for 'id' type.
llvm-svn: 43504
2007-10-30 17:06:23 +00:00
Steve Naroff
5448cf6d61 - Add location info to category/protocol AST's
- Rewrite categories.

llvm-svn: 43501
2007-10-30 13:30:57 +00:00
Steve Naroff
4cd61acc33 Remove a couple FIXME's for rewriting ObjC interfaces (which are now being rewritten properly).
llvm-svn: 43494
2007-10-30 03:43:13 +00:00
Steve Naroff
c54840433f More support for rewriting ObjC intefaces. Still some edge cases to handle...
llvm-svn: 43493
2007-10-30 02:23:23 +00:00
Steve Naroff
33a1e80dd1 This commit contains lot's of small tweaks to how we pass around and store SourceLocation's for interfaces/protocols/categories/implementations.
llvm-svn: 43475
2007-10-29 21:38:07 +00:00
Steve Naroff
161a92b976 Start rewriting ObjC interfaces. As a start, we comment out all the methods. This involved refining how the parser/AST passes/manages SourceLocations for ObjcMethodDecl's.
llvm-svn: 43404
2007-10-26 20:53:56 +00:00
Fariborz Jahanian
3339d0d14e This patch allows synthesis generation of ivar offset for legacy objective-c @implementation
decl without an @interface decl.

llvm-svn: 43403
2007-10-26 20:50:24 +00:00
Chris Lattner
8beb9dee0e Fix a crash on test/Sema/invalid-decl.c
llvm-svn: 43188
2007-10-19 20:10:30 +00:00
Steve Naroff
eee59ebb15 Add a FIXME to an assert.
Change a dyn_cast_or_null back to cast (which is more efficient).

llvm-svn: 43152
2007-10-18 22:17:45 +00:00
Fariborz Jahanian
c34409c281 Patch to rewrite ivar tables metadata for classes defined.
llvm-svn: 43151
2007-10-18 22:09:03 +00:00
Steve Naroff
8569d77349 Fix the following bug...
unsigned char asso_values[] = { 34 };
int legal2() { 
  return asso_values[0]; 
}

The code that creates the new constant array type was operating on the original type.

As a result, the constant type being generated was "unsigned char [1][]" (which is wrong).

The fix is to operate on the element type - in this case, the correct type is "unsigned char [1]"

I added this case to array-init.c, which clearly didn't catch this bogosity...

llvm-svn: 43112
2007-10-18 03:27:23 +00:00
Anders Carlsson
f94cd1ffe6 Generate code for static variables that don't have initializers. Also, report an error if a static initializer is not constant.
llvm-svn: 43058
2007-10-17 00:52:43 +00:00
Chris Lattner
a7c19feca2 Add a new Rewriter::getRangeSize method.
Rename SourceRange::Begin()/End() to getBegin()/getEnd() for
consistency with other code.
Start building the rewriter towards handling @encode.

llvm-svn: 43047
2007-10-16 22:36:42 +00:00
Fariborz Jahanian
ebac2cb235 Patch to diagnose duplicate method implementations.
llvm-svn: 43046
2007-10-16 21:52:23 +00:00
Steve Naroff
d275426449 Remove ObjcMethodDecl::getNumMethodParams/getMethodParamDecl, they aren't used/needed.
Change ObjcMethodDecl::getMethodType to getResultType, to match FunctionDecl.

llvm-svn: 43045
2007-10-16 21:36:54 +00:00
Steve Naroff
698495522d Bad cast...need to use dyn_cast_or_null. Also changed Sema::InitBuiltinVaListType (which had the same bug).
llvm-svn: 43023
2007-10-16 00:00:18 +00:00
Anders Carlsson
7e13ab88a2 Add code generation and sema checking for __builtin_va_arg.
llvm-svn: 43006
2007-10-15 20:28:48 +00:00
Fariborz Jahanian
d4b3015dd7 Several name lookup conflict detection fixes involving objective-c names.
llvm-svn: 43000
2007-10-15 19:16:57 +00:00
Steve Naroff
ca0ecfe048 Add category lookup (removing a couple FIXME's).
Changed ObjcInterfaceDecl::ListCategories->CategoryList.

llvm-svn: 42968
2007-10-14 18:27:41 +00:00
Steve Naroff
5811baf1a0 - Added Sema::AddFactoryMethodToGlobalPool and Sema::AddInstanceMethodToGlobalPool and DenseMaps. This will allow us to efficiently lookup a method from a selector given no type information (for the "id" data type).
- Fixed some funky "}
                    else {" indentation in Sema::ActOnAddMethodsToObjcDecl(). I'd prefer we stay away from this style...it wastes space and isn't any easier to read (from my perspective, at least:-)

- Changed Parser::ParseObjCInterfaceDeclList() to only call Action::ActOnAddMethodsToObjcDecl() when it actually has methods to add (since most interface have methods, this is a very minor cleanup).

llvm-svn: 42957
2007-10-14 00:58:41 +00:00
Anders Carlsson
24ebce6fca Generate code for va_start and va_end.
llvm-svn: 42939
2007-10-12 23:56:29 +00:00
Fariborz Jahanian
fd0312ed79 Patch to check for duplicate method decls in protocols.
llvm-svn: 42938
2007-10-12 23:43:31 +00:00
Fariborz Jahanian
ecfe4f1453 Check and diagnose that objective-c objects may not be statically allocated.
llvm-svn: 42936
2007-10-12 22:10:42 +00:00
Fariborz Jahanian
02fbb68e85 Fixed typo in comment.
llvm-svn: 42928
2007-10-12 19:53:08 +00:00
Fariborz Jahanian
c7afeebb12 Fixed a @compatible_alias bug. In the process, discovered unnecessary 2ndry lookup
ok class names and streamlined this logic to do the lookup once.

llvm-svn: 42926
2007-10-12 19:38:20 +00:00
Steve Naroff
b915146a5d Replace one FIXME with another. We handle protocols just fine now. The ObjC decl will only be 0 when we have an error on the ObjC decl. I would prefer we pass in a decl that is marked as invalid. I don't think this is critical to fix now, however I'd like us to be consistent. There are currently many places that don't mark the decl as invalid (which need to be fixed)...
llvm-svn: 42923
2007-10-12 18:49:25 +00:00