Moved all clients of Diagnostics to use FullSourceLoc instead of SourceLocation.
Added many utility methods to FullSourceLoc to provide shorthand for:

    FullLoc.getManager().someMethod(FullLoc.getLocation());
    
instead we have:

    FullLoc.someMethod();
    
Modified TextDiagnostics (and related classes) to use this short-hand.

llvm-svn: 44957

llvm-svn: 44945

llvm-svn: 44941

llvm-svn: 44938

llvm-svn: 44925

int test(int x, long long y) {
  return x << y;
}

we now realize the type of the shift is int, not long long.

This fixes a fixme from june.

llvm-svn: 44912

this is passed to sema and ignored there, so the second part of the
string will not make it into the AST.  Passing to Fariborz to finish
Sema + AST construction.

llvm-svn: 44898

return a loc.

llvm-svn: 44890

SourceManager is passed by reference, allowing the SourceManager to be
associated with a specific translation unit, and not the entire execution
of the driver.

Modified all users of Diagnostics to comply with this new interface.

Integrated SourceManager as a member variable of TargetInfo. TargetInfo will
eventually be associated with a single translation unit (just like
SourceManager).

Made the SourceManager reference in ASTContext private. Provided accessor
getSourceManager() for clients to use instead. Modified clients to comply with
new interface.

llvm-svn: 44878

llvm-svn: 44872

Avoid the first person in a recently added comment. 

llvm-svn: 44843

Change err_undef_protocolref to warn_undef_protocolref (this is consistent with GCC).

llvm-svn: 44840

Explicitly set the string literal type from "char *" to "constant array of char".

At this point, I am fairly certain the front-end is correct. Unfortunately, the back-end is still unhappy.

That said, I've commented out the two lines in globalinit.c that are causing problems.

Chris, please have a look...thanks!

llvm-svn: 44823

Add support for initializing char arrays from string literals.

Adapted from a patch by Anders Carlsson.

llvm-svn: 44816

C99 6.5.6.

llvm-svn: 44746

type.

llvm-svn: 44685

Have Sema::CheckConstantInitList skip semantic analysis when the element type is a record (until we implement the FIXME). This removes a bogus error for the following code...

snarofflocal% cat bug.m

#import <Foundation/NSGeometry.h>

#define NUMHELICOPTERRECTS 5

static NSRect helicopterRects[NUMHELICOPTERRECTS] = {
    {{27, 0}, {18, 11}},	// Bottom
    {{0, 8}, {4, 11}},		// Tail
    {{0, 16}, {50, 1}},
    {{22, 5}, {18, 12}},	// Body
    {{0, 10}, {42, 3}}
};

llvm-svn: 44684

Lookup methods in the global pool even when a statically typed object's class interface isn't in scope!

As the comment in the code indicates, I'm not fond of this. Nevertheless, gcc compat is a goal.

Here is the case I'm talking about...

#import <Foundation/Foundation.h>

@interface AnyClass : NSObject
- (NSRect)rect;
@end

@class Helicopter;

static void func(Helicopter *obj) {
  NSRect r = [obj rect];
}

...before this patch, we would warn/error. With this patch, everything "just works".

llvm-svn: 44682

Rewrite 'super' within a class method. This required some minor tweaks to the front-end.

llvm-svn: 44673

"@class Protocol;"

llvm-svn: 44670

llvm-svn: 44666

to rewriter (my previous patch).

llvm-svn: 44665

crash because of this.

llvm-svn: 44648

llvm-svn: 44617

llvm-svn: 44605

Revert DidWarn change - It won't catch all cases anyway and GCC warns for every excess element. Maybe later we can add back the limit and make it smarter.

llvm-svn: 44603

Recognize CompoundLiteralExpr's as valid lvalue's.

Also updated a FIXME in Sema::CheckInitializer()...

llvm-svn: 44602

llvm-svn: 44576

Handle initializing vector elements correctly. Emit just one warning if there are excess initializers, instead of one per initializer.

llvm-svn: 44525

llvm-svn: 44516

llvm-svn: 44496

llvm-svn: 44463

GCC has an extension where the left hand side of the ? : operator can be omitted. Handle this in a few more places.

llvm-svn: 44462

llvm-svn: 44449

Yesterday, I simplified how we stream top-level decls.

After a discussion with Ted, we both came to the conclusion that adding a "HandleTopLevelDeclaration" hook to ASConsumer is far more elegant. The default implementation of HandleTopLevelDeclaration will be responsible for iterating over the ScopedDecl (which has a chain of the decls:-).

TODO: Once Ted adds HandleTopLevelDeclaration, make sure TagDecls are chainged appropriately...
llvm-svn: 44445

Update the initializer's type, in addition to the decl, if we've changed the type of the decl based on it.

llvm-svn: 44440

floating-point literals that are represented exactly by the APFloat in
FloatingLiteral. For such literals, we do not emit a warning since such checks are
often performed in real code to see if a variable has changed from its original
value. This heuristic clearly can lead to false negatives, but the hope is it will
significantly reduce false positives to help make the compiler flag more useful.

llvm-svn: 44424

the APFloat representing the parsed literal can represent the literal value
exactly.  This is useful when performing various semantic checks on the code,
and issuing appropriate warnings to users.

llvm-svn: 44423

Several fixes/simplifications surrounding how we stream top-level decl AST's.

The following code...

typedef struct cssm_data {} CSSM_DATA, *CSSM_DATA_PTR;

struct Y { int A; };

struct X { int A; } D; 

struct X E, F;

...now produces the following output...

> ../../Debug/bin/clang xx.c -ast-print
Read top-level tag decl: 'cssm_data'
typedef struct cssm_data CSSM_DATA;
typedef struct cssm_data *CSSM_DATA_PTR;
Read top-level tag decl: 'Y'
Read top-level tag decl: 'X'
Read top-level variable decl: 'D'
Read top-level variable decl: 'E'
Read top-level variable decl: 'F'

...which is much more accurate than the previous -ast-print output...

typedef struct cssm_data CSSM_DATA;
typedef struct cssm_data CSSM_DATA;
Read top-level variable decl: 'D'
Read top-level variable decl: 'E'
Read top-level variable decl: 'E'

llvm-svn: 44421

Bug pointed out by Michael Zolda, thanks!

llvm-svn: 44408