================
@@ -149,6 +149,63 @@ class TextTokenRetokenizer {
addToken();
}
+ /// Check if this line starts with @par or \par
+ bool startsWithParCommand() {
+ unsigned Offset = 1;
+
+ /// Skip all whitespace characters at the beginning.
+ /// This needs to backtrack because Pos has already advanced past the
+ /// actual \par or @par command by the time this function is called.
+ while (isWhitespace(*(Pos.BufferPtr - Offset)))
+ Offset++;
----------------
hdoc wrote:
In this case we actually have to read the buffer backwards to find the `@par`,
past `Pos.BufferStart` so it's a bit inconvenient to construct the `StringRef`
which starts at some unknown position before `Pos.BufferStart`. In theory
someone could have put 400 spaces between `@par` and the comment text, and we'd
have to accommodate that when constructing the `StringRef`.
I think the solution as it's written right now handles that well enough. I
could make it count spaces or look up the previous token containing the `@par`
and compute the correct StringRef for it, but I think this is good enough.
https://github.com/llvm/llvm-project/pull/91100
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
