Commit Graph

102 Commits

Author SHA1 Message Date
John McCall a0296f7987 Remember the "found declaration" for an overload candidate, which is the
entity (if applicable) which was actually looked up.  If a candidate was found
via a using declaration, this is the UsingShadowDecl;  otherwise, if
the candidate is template specialization, this is the template;  otherwise,
this is the function.

The point of this exercise is that "found declarations" are the entities
we do access control for, not their underlying declarations.  Broadly speaking,
this patch fixes access control for using declarations.

There is a *lot* of redundant code calling into the overload-resolution APIs;
we really ought to clean that up.

llvm-svn: 98945
2010-03-19 07:35:19 +00:00
John McCall bbbbe4eaaf Split C++ friend declarations into their own header/implementation file.
I'm expecting this portion of the AST to grow and change, and I'd like to
be able to do that with minimal recompilation.  If this proves unnecessary
when access control is fully-implemented, I'll fold the classes back into
DeclCXX.h.

llvm-svn: 98249
2010-03-11 07:50:04 +00:00
Douglas Gregor 9a28e84b32 Keep an explicit stack of function and block scopes, each element of
which has the label map, switch statement stack, etc. Previously, we
had a single set of maps in Sema (for the function) along with a stack
of block scopes. However, this lead to funky behavior with nested
functions, e.g., in the member functions of local classes.

The explicit-stack approach is far cleaner, and we retain a 1-element
cache so that we're not malloc/free'ing every time we enter a
function. Fixes PR6382.

Also, tweaked the unused-variable warning suppression logic to look at
errors within a given Scope rather than within a given function. The
prior code wasn't looking at the right number-of-errors count when
dealing with blocks, since the block's count would be deallocated
before we got to ActOnPopScope. This approach works with nested
blocks/functions, and gives tighter error recovery.

llvm-svn: 97518
2010-03-01 23:15:13 +00:00
Douglas Gregor 44272ca356 Add some spacing in the code-completion results for a return statement
llvm-svn: 96567
2010-02-18 04:06:48 +00:00
John McCall bc077cf589 Thread a source location into the template-argument deduction routines. There
may be some other places that could take advantage of this new information,
but I haven't really looked yet.

llvm-svn: 95600
2010-02-08 23:07:23 +00:00
Douglas Gregor b92a1565c3 Implement the lvalue-to-rvalue conversion where needed. The
lvalue-to-rvalue conversion adjusts lvalues of qualified, non-class
type to rvalue expressions of the unqualified variant of that
type. For example, given:

  const int i;
  (void)(i + 17);

the lvalue-to-rvalue conversion for the subexpression "i" will turn it
from an lvalue expression (a DeclRefExpr) with type 'const int' into
an rvalue expression with type 'int'. Both C and C++ mandate this
conversion, and somehow we've slid through without implementing it. 

We now have both DefaultFunctionArrayConversion and
DefaultFunctionArrayLvalueConversion, and which gets used depends on
whether we do the lvalue-to-rvalue conversion or not. Generally, we do
the lvalue-to-rvalue conversion, but there are a few notable
exceptions:
  - the left-hand side of a '.' operator
  - the left-hand side of an assignment
  - a C++ throw expression
  - a subscript expression that's subscripting a vector

Making this change exposed two issues with blocks:
  - we were deducing const-qualified return types of non-class type
  from a block return, which doesn't fit well
  - we weren't always setting the known return type of a block when it
  was provided with the ^return-type syntax

Fixes the current Clang-on-Clang compile failure and PR6076.

llvm-svn: 95167
2010-02-03 00:27:59 +00:00
John McCall b89836b6db Pass access specifiers around in overload resolution.
llvm-svn: 94485
2010-01-26 01:37:31 +00:00
Douglas Gregor ff59f676d0 Teach code-completion to deal with calls to functions without prototypes.
llvm-svn: 94076
2010-01-21 15:46:19 +00:00
Douglas Gregor fc59ce1ea4 Switch a few callers of MaybeAddResult over to AddResult, when the
declarations we're adding do not need any name-hiding checks.

llvm-svn: 93431
2010-01-14 16:14:35 +00:00
Douglas Gregor 2b8162b7d0 Switch code-completion's ivar lookup over to LookupVisibleDecls,
eliminating yet one more ResultBuilder::MaybeAddResult caller.

llvm-svn: 93430
2010-01-14 16:08:12 +00:00
Douglas Gregor 78a210145b Start migrating code-completion results from
ResultBuilder::MaybeAddResult over to ResultBuilder::AddResult.

llvm-svn: 93429
2010-01-14 16:01:26 +00:00
Douglas Gregor 09bbc651f4 Switch the remaining code completions over to LookupVisibleDecls,
after adding the ability to determine whether our lookup is a
base-class lookup. Eliminate CollectMemberLookupResults, since it is
no longer used (yay).

llvm-svn: 93428
2010-01-14 15:47:35 +00:00
Douglas Gregor 200c99dd1e Move code completion for qualified name lookup (foo::) to
LookupVisibleDecls. Also, a function does not hide another function.

llvm-svn: 93421
2010-01-14 03:35:48 +00:00
Douglas Gregor a6e2edcc3a Eliminate the code-completion-specifier CollectLookupResults in favor
of the more general LookupVisibleDecls.

llvm-svn: 93419
2010-01-14 03:27:13 +00:00
Douglas Gregor 6ae4c5278c Simplify the code-completion logic for nested-name-specifiers: rather
than traversing visible declarations twice, only perform one traversal
and recognize nested-name-specifiers as special.

llvm-svn: 93418
2010-01-14 03:21:49 +00:00
Douglas Gregor 69a8e083c5 When providing completions for a member access expression in C++,
provided nested-name-specifier results for base classes (only), rather
than everything that could possibly be a nested-name-specifier.

llvm-svn: 93398
2010-01-14 01:17:14 +00:00
Douglas Gregor c580c5205f Switch code-completion for ordinary names over to the new(ish)
LookupVisibleDecls, unifying the name lookup mechanisms used by code
completion and typo correction. Aside from the software-engineering
improvements, this makes code-completion see through using directives
and see ivars when performing unqualified name lookup in an
Objective-C instance method.

llvm-svn: 93397
2010-01-14 01:09:38 +00:00
Douglas Gregor e0717ab39c More refactoring of ResultBuilder::MaybeAddResult. No intended
functionality change.

llvm-svn: 93386
2010-01-14 00:41:07 +00:00
Douglas Gregor 7c208616b4 Refactor the "is this declaration interesting" logic in
code-completion's ResultBuilder::MaybeAddResult for later reuse.

llvm-svn: 93379
2010-01-14 00:20:49 +00:00
Douglas Gregor f98e6a28e4 Banish the notion of a "rank" for code-completion results, since we
are no longer using it for anything. No intended functionality change.

llvm-svn: 93376
2010-01-13 23:51:12 +00:00
Douglas Gregor 52ce62f069 Improve the sorting of code-completion results. We now always sort by
the "typed" text, first, then take into account
nested-name-specifiers, name hiding, etc. This means that the
resulting sort is actually alphabetical :)

llvm-svn: 93370
2010-01-13 23:24:38 +00:00
Douglas Gregor 48d462573d Code-completion for @public, @protected, @private, @package.
llvm-svn: 93361
2010-01-13 21:54:15 +00:00
Douglas Gregor f193416359 Whenever completing ordinary names for an Objective-C source, also
provide completions for @ keywords. Previously, we only provided
@-completions after an @ was actually typed, which is useful but
probably not the common case.

Also, make sure a few Objective-C 2.0 completions only show up when
Objective-C 2.0 support is enabled (the default).

llvm-svn: 93354
2010-01-13 21:24:21 +00:00
Douglas Gregor 6a80393385 Use horizontal-space markers in code-completion results rather than
embedding single space characters. <rdar://problem/7485503>

llvm-svn: 93231
2010-01-12 06:38:28 +00:00
Douglas Gregor 504a6ae83e Improve code completion by introducing patterns for the various C and
C++ grammatical constructs that show up in top-level (namespace-level)
declarations, member declarations, template declarations, statements,
expressions, conditions, etc. For example, we now provide a pattern
for

  static_cast<type>(expr)

when we can have an expression, or

  using namespace identifier;

when we can have a using directive.

Also, improves the results of code completion at the beginning of a
top-level declaration. Previously, we would see value names (function
names, global variables, etc.); now we see types, namespace names,
etc., but no values.

llvm-svn: 93134
2010-01-10 23:08:15 +00:00
Chris Lattner 1ca1d7adc2 remove extraneous #include
llvm-svn: 92310
2009-12-30 19:54:10 +00:00
Douglas Gregor 2d435306e5 Typo correction for type names when they appear in declarations, e.g., given
tring str2;

we produce the following diagnostic + fix-it:

typo.cpp:15:1: error: unknown type name 'tring'; did you mean 'string'?
  tring str2;
  ^~~~~
  string


To make this really useful, we'll need to introduce typo correction in
many more places (wherever we do name lookup), and implement
declaration-vs-expression heuristics that cope with typos
better. However, for now this will handle the simple cases where we
already get good "unknown type name" diagnostics.

The LookupVisibleDecls functions are intended to be used by code
completion as well as typo correction; that refactoring will happen
later.

llvm-svn: 92308
2009-12-30 17:04:44 +00:00
Benjamin Kramer d20ef75b91 Remove some dead variables clang-analyzer found.
llvm-svn: 92162
2009-12-25 15:43:36 +00:00
Douglas Gregor 04c5f973ee Objective-C methods can be variadic, too. Who knew.
llvm-svn: 91951
2009-12-23 00:21:46 +00:00
Douglas Gregor b3fa919c2d Extend code-completion results with the type of each result
llvm-svn: 91702
2009-12-18 18:53:37 +00:00
John McCall 5750077300 Shift things around so that it's easier to recover from a missing
function in a C++ call using an arbitrary call-expression type.
Actually exploit this to fix the recovery implemented earlier.

The diagnostic is still iffy, though.

llvm-svn: 91538
2009-12-16 12:17:52 +00:00
Chris Lattner 2f3da9b205 update to match LLVM API change:
Remove isPod() from DenseMapInfo, splitting it out to its own
isPodLike type trait.  This is a generally useful type trait for
more than just DenseMap, and we really care about whether something
acts like a pod, not whether it really is a pod.

llvm-svn: 91422
2009-12-15 07:26:51 +00:00
Jeffrey Yasskin 1615d45daa Un-namespace-qualify llvm_unreachable. It's a macro, so the qualification gave
no extra safety anyway.

llvm-svn: 91207
2009-12-12 05:05:38 +00:00
Douglas Gregor 3ef5952017 When code completion of an overload set fails, produce results for ordinary name lookup instead.
llvm-svn: 91141
2009-12-11 19:06:04 +00:00
Douglas Gregor 0f622367d3 Add a function's cv-qualifiers to the code-completion results as an
informative chunk.

llvm-svn: 91139
2009-12-11 18:44:16 +00:00
Douglas Gregor 800f2f0ff0 Teach code completion to instantiate templates when it needs to
llvm-svn: 91138
2009-12-11 18:28:39 +00:00
Douglas Gregor 7078839b0e Member function templates can occur after . or ->
llvm-svn: 91137
2009-12-11 18:14:22 +00:00
Douglas Gregor 99fe2ad8c8 Tweak code-completion results by suppressing class template
specializations and class template partial specializations (they're
never named directly). Also, member access expressions only refer to
value declarations (fields, functions, enumerators, etc.) and
Objective-C property declarations; filter out everything else.

llvm-svn: 91133
2009-12-11 17:31:05 +00:00
Douglas Gregor 83c49b56fe Class template (partial) specializations should not show up in code completion results
llvm-svn: 91125
2009-12-11 16:18:54 +00:00
Douglas Gregor 9225369891 When in an Objective-C instance method, super is a valid code-completion result
llvm-svn: 90758
2009-12-07 09:54:55 +00:00
Douglas Gregor bc7c5e471a Code completion for Objective-C @ keywords that are statements or expressions
llvm-svn: 90757
2009-12-07 09:51:25 +00:00
Douglas Gregor f48706c787 Code completion for Objective-C @ directives
llvm-svn: 90756
2009-12-07 09:27:33 +00:00
Douglas Gregor 94bb5e8d75 PointerUnion == PointerUnion does not do what I thought it did. Also, fix a thinko in a PointerUnion::get call.
llvm-svn: 90719
2009-12-06 21:27:58 +00:00
Douglas Gregor 05e7ca3659 Switch the std::multimap shadow map used in code completion to an
llvm::DenseMap, for a 20% performance improvement in the
Cocoa-big-list performance benchmark.

llvm-svn: 90715
2009-12-06 20:23:50 +00:00
Benjamin Kramer 4053e5db54 Avoid trashing two temporary strings.
llvm-svn: 90663
2009-12-05 10:22:15 +00:00
Benjamin Kramer ed60635d70 strcasecmp is unportable.
llvm-svn: 90662
2009-12-05 10:07:04 +00:00
Douglas Gregor 249d6826cd Improve the performance of code completion by 2.2x when completing for ordinary names with Cocoa.h included, by drastically improving the performance of our results sorting.
llvm-svn: 90661
2009-12-05 09:08:56 +00:00
Douglas Gregor f09935f16c Switch the clang-to-CIndex interface for code-completion to a binary format, for a massive speedup
llvm-svn: 90209
2009-12-01 05:55:20 +00:00
John McCall 132f3133ae Remove all of Sema's explicit uses of OverloadedFunctionDecl except for
those associated with TemplateNames.

llvm-svn: 90162
2009-11-30 22:55:35 +00:00
Benjamin Kramer b33a97c384 Remove useless c_str() calls in SemaCodeComplete.
llvm-svn: 90088
2009-11-29 20:18:50 +00:00