[clangd] textDocument/documentLink to support include statements with macro argument (#137550)

See issue clangd/clangd#2375

The clangd textDocument/documentLink request does not support include
statements with macro argument and fails by returning an invalid range.

This PR adds support for macro argument by detecting what form of
include statement is currently being processed and returning a range
accordingly

First time contributing to LLVM, so if you have advice, questions or if
I messed up procedure please tell me !

---------

Co-authored-by: Adrien Garbani <[email protected]>

Author: frayien

clang-tools-extra/clangd/XRefs.cpp

@@ -859,17 +859,47 @@ std::vector<DocumentLink> getDocumentLinks(ParsedAST &AST) {
   for (auto &Inc : AST.getIncludeStructure().MainFileIncludes) {
     if (Inc.Resolved.empty())
       continue;
+
+    // Get the ___location of the # symbole of the "#include ..." statement
     auto HashLoc = SM.getComposedLoc(SM.getMainFileID(), Inc.HashOffset);
+
+    // get the # Token itself, std::next to get the "include" token and the
+    // first token after (aka "File Token")
     const auto *HashTok = AST.getTokens().spelledTokenContaining(HashLoc);
     assert(HashTok && "got inclusion at wrong offset");
     const auto *IncludeTok = std::next(HashTok);
     const auto *FileTok = std::next(IncludeTok);
-    // FileTok->range is not sufficient here, as raw lexing wouldn't yield
-    // correct tokens for angled filenames. Hence we explicitly use
-    // Inc.Written's length.
-    auto FileRange =
-        syntax::FileRange(SM, FileTok->___location(), Inc.Written.length())
-            .toCharRange(SM);
+
+    // The File Token can either be of kind :
+    // "less" if using the "#include <h-char-sequence> new-line" syntax
+    // "string_literal" if using the "#include "q-char-sequence" new-line"
+    // syntax something else (most likely "identifier") if using the "#include
+    // pp-tokens new-line" syntax (#include with macro argument)
+
+    CharSourceRange FileRange;
+
+    if (FileTok->kind() == tok::TokenKind::less) {
+      // FileTok->range would only include the '<' char. Hence we explicitly use
+      // Inc.Written's length.
+      FileRange =
+          syntax::FileRange(SM, FileTok->___location(), Inc.Written.length())
+              .toCharRange(SM);
+    } else if (FileTok->kind() == tok::TokenKind::string_literal) {
+      // FileTok->range includes the quotes for string literals so just return
+      // it.
+      FileRange = FileTok->range(SM).toCharRange(SM);
+    } else {
+      // FileTok is the first Token of a macro spelling
+
+      // Report the range of the first token (as it should be the macro
+      // identifier)
+      // We could use the AST to find the last spelled token of the macro and
+      // report a range spanning the full macro expression, but it would require
+      // using token-buffers that are deemed too unstable and crash-prone
+      // due to optimizations in cland
+
+      FileRange = FileTok->range(SM).toCharRange(SM);
+    }
 
     Result.push_back(
         DocumentLink({halfOpenToRange(SM, FileRange),

clang-tools-extra/clangd/unittests/XRefsTests.cpp

@@ -2784,14 +2784,24 @@ TEST(GetNonLocalDeclRefs, All) {
 
 TEST(DocumentLinks, All) {
   Annotations MainCpp(R"cpp(
+      #define HEADER_AA "faa.h"
+      #define HEADER_BB "fbb.h"
+      #define GET_HEADER(X) HEADER_ ## X 
+
       #/*comments*/include /*comments*/ $foo[["foo.h"]] //more comments
       int end_of_preamble = 0;
       #include $bar[[<bar.h>]]
+      #include $AA[[GET_HEADER]](AA) // Some comment !
+      # /* What about */ \
+      include /* multiple line */ \
+      $BB[[GET_HEADER]]( /* statements ? */ \
+      BB /* :) */ )
     )cpp");
 
   TestTU TU;
   TU.Code = std::string(MainCpp.code());
-  TU.AdditionalFiles = {{"foo.h", ""}, {"bar.h", ""}};
+  TU.AdditionalFiles = {
+      {"faa.h", ""}, {"fbb.h", ""}, {"foo.h", ""}, {"bar.h", ""}};
   TU.ExtraArgs = {"-isystem."};
   auto AST = TU.build();
 
@@ -2801,7 +2811,11 @@ TEST(DocumentLinks, All) {
           DocumentLink({MainCpp.range("foo"),
                         URIForFile::canonicalize(testPath("foo.h"), "")}),
           DocumentLink({MainCpp.range("bar"),
-                        URIForFile::canonicalize(testPath("bar.h"), "")})));
+                        URIForFile::canonicalize(testPath("bar.h"), "")}),
+          DocumentLink({MainCpp.range("AA"),
+                        URIForFile::canonicalize(testPath("faa.h"), "")}),
+          DocumentLink({MainCpp.range("BB"),
+                        URIForFile::canonicalize(testPath("fbb.h"), "")})));
 }
 
 } // namespace