I don't disagree, but it is also not easy to do. Do you consider that a
precondition for this patch?
================
Comment at: lib/Format/Format.cpp:300
@@ +299,3 @@
+ // previous result unless we have hit the optimatization (and thus
+ // returned UINT_MAX) and are now computing for a lower StopAt.
+ unsigned SavedResult = I->second.first;
----------------
Manuel Klimek wrote:
> s/lower/higher/
No longer applicable
================
Comment at: lib/Format/Format.cpp:298
@@ +297,3 @@
+ if (I != Memory.end()) {
+ // If this state has already been examined, we can safely return the
+ // previous result unless we have hit the optimatization (and thus
----------------
Manuel Klimek wrote:
> I think it would be easier for me to read if this were written positively,
> instead of with "unless".
How about this?
http://llvm-reviews.chandlerc.com/D172
_______________________________________________
cfe-commits mailing list
[email protected]
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
