On 09/10/2012 03:56 PM, Dmitri Gribenko wrote:
Hello,
The attached patch adds briefComment property to CompletionString.
Nice.
In order to take advantage of this functionality, we also need to
expose CXCodeComplete_Flags enum. Is TranslationUnit class a good
place to put these constants?
The way you implemented this mirrors exactly the way we have implemented
the parsing flags. This means, it is definitely a reasonable solution.
However, looking at it, I got the feeling that there may be a nicer way.
What about not exposing those flags at all, but adding some optional
parameters to the complete function:
def codeComplete(self, path, line, column, unsaved_files=None,
include_macros=False, include_code_patterns=False,
include_brief_comments=False):
options = 0
if include_macros:
options += 1
if include_code_patterns:
options += 2
if include_brief_comments:
options += 4
This seems to be a nicer interface. C not having named parameters, does
not mean we cannot use them to enhance the python interface.
Index: bindings/python/clang/cindex.py
===================================================================
--- bindings/python/clang/cindex.py (revision 163506)
+++ bindings/python/clang/cindex.py (working copy)
@@ -1735,6 +1735,10 @@
res = conf.lib.clang_getCompletionAvailability(self.obj)
return availabilityKinds[res]
+ @property
+ def briefComment(self):
+ return conf.lib.clang_getCompletionBriefComment(self.obj).spelling
Until today all methods return directly a CXString, but don't call
'.spelling' on it. It seems to make sense to stay consistent here.
(You are right that returning a string directly could have some
advantages, but, if this is the case, we should fix it everywhere and
apply such a fix in a separate patch)
+ # Used to indicate that brief documentation comments should be included
+ # into the set of code completions returned from this translation unit.
+ INCLUDE_BRIEF_COMMENTS_IN_CODE_COMPLETION = 128
I would call it:
PARSE_INCLUDE_BRIEF_COMMENTS_IN_CODE_COMPLETION = 128
All elements of this enum are prefixed with PARSE_. For consistency, we
should continue to do so.
+ # Whether to include macros within the set of code completions returned.
+ CODE_COMPLETE_INCLUDE_MACROS = 1
+
+ # Whether to include code patterns for language constructs within the set
+ # of code completions, e.g., for loops.
+ CODE_COMPLETE_INCLUDE_CODE_PATTERNS = 2
+
+ # Whether to include brief documentation within the set of code completions
+ # returned.
+ CODE_COMPLETE_INCLUDE_BRIEF_COMMENTS = 4
As mentioned above, you may consider to not expose them at all, but to
use named parameters.
Cheers
Tobi
P.S.: This would be a great opportunity to get our first code completion
test case.
_______________________________________________
cfe-commits mailing list
[email protected]
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
