ioeric added inline comments.

================
Comment at: clangd/index/Index.h:275
+  /// Contextually relevant files (e.g. the file we're code-completing in).
+  /// Paths should be absolute.
+  std::vector<std::string> ProximityPaths;
----------------
sammccall wrote:
> Hmm, the returned results have URIs, not paths.
> 
> It seems cleaner that the query params be URIs too.
> However then we need a way to know which scheme the index expects.
> 
> It could expose this information, or the index could accept a path here and 
> do the conversion itself. This indexes need to be "thick" and deal with the 
> URI infrastructure, but may be the simplest way.
> 
> What's your thinking here?
> However then we need a way to know which scheme the index expects.
> It could expose this information, or the index could accept a path here and 
> do the conversion itself. This indexes need to be "thick" and deal with the 
> URI infrastructure, but may be the simplest way.
Yeah, I was thinking the same. Also index symbols can have different schemes, 
and we might not have enough information to know which scheme is preferable 
until we see a candidate. 


Repository:
  rCTE Clang Tools Extra

https://reviews.llvm.org/D47937



_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

Reply via email to