The design document states that the lexer-only approach is very limited. This patch is based purely on the lexer. How is this going to be extended to handle ambiguous constructs that require an AST? Is there an intention to invoke the parser and resynchronize the AST with the token stream somehow?
================ Comment at: unittests/Format/Makefile:15 @@ +14,3 @@ +USEDLIBS = clangFormat.a clangTooling.a clangFrontend.a clangSerialization.a \ + clangDriver.a clangParse.a clangRewriteCore.a + clangRewriteFrontend.a clangSema.a clangAnalysis.a clangEdit.a \ ---------------- Missing backslash at the end of the line? ================ Comment at: lib/Format/Format.cpp:9 @@ +8,3 @@ +//===----------------------------------------------------------------------===// +// +// This is EXPERIMENTAL code under heavy development. It is not in a state yet, ---------------- Doxygen and \file this comment please. http://llvm-reviews.chandlerc.com/D80 _______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
