klimek added inline comments.
================
Comment at: lib/Format/BreakableToken.cpp:471
+ WhitespaceManager &Whitespaces)
{
+ if (Tok.is(TT_LineComment)) {
+ // If this is the first line of a token, inform Whitespace Manager about
it.
----------------
Without looking into this in a lot of detail: this looks like you want a
BreakableComment base class, and have BreakableBlockComment and
BreakableLineCommentSection derive from it and implement this method.
Scanning it a bit, it seems like there is still overlap - perhaps it's also
possible to pull out a couple of smaller sized methods in the interface and
write the algorithm in terms of those? That could also make it easier to
understand in general (large method alarm ;)
================
Comment at: lib/Format/ContinuationIndenter.cpp:1174-1175
LineIndex != EndIndex; ++LineIndex) {
- if (!DryRun)
- Token->replaceWhitespaceBefore(LineIndex, Whitespaces);
+ Token->replaceWhitespaceBefore(LineIndex, RemainingTokenColumns,
+ RemainingSpace, DryRun, Whitespaces);
unsigned TailOffset = 0;
----------------
Nice that this whole section required so few changes.
Why do we need to call into this in DryRun mode now, though? Does it need to
keep state inside in DryRun?
https://reviews.llvm.org/D27754
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
