Committed with a somewhat beefed up test-parsing.cc
https://chromiumcodereview.appspot.com/9600009/diff/18001/src/handles.cc File src/handles.cc (right): https://chromiumcodereview.appspot.com/9600009/diff/18001/src/handles.cc#newcode952 src/handles.cc:952: while (failure) { On 2012/03/12 10:55:05, rossberg wrote:
Nit: this seems more natural as a do-while loop.
I made it into a do-while, but it's no better (same line count, flag must still be declared outside the loop, still have to test the flag twice). https://chromiumcodereview.appspot.com/9600009/diff/18001/src/handles.h File src/handles.h (right): https://chromiumcodereview.appspot.com/9600009/diff/18001/src/handles.h#newcode173 src/handles.h:173: On 2012/03/12 10:55:05, rossberg wrote:
Nit: accidental extra newline?
Done. https://chromiumcodereview.appspot.com/9600009/diff/18001/src/unicode.h File src/unicode.h (right): https://chromiumcodereview.appspot.com/9600009/diff/18001/src/unicode.h#newcode121 src/unicode.h:121: if (code == kNoPreviousCharacter) return false; On 2012/03/12 10:55:05, rossberg wrote:
I still think this is implied by the bit masking below, but see that
you may
want to make it more explicit/robust.
Yes, I think that would be too implicit. https://chromiumcodereview.appspot.com/9600009/diff/18001/src/unicode.h#newcode132 src/unicode.h:132: static const int kNoPreviousCharacter = -1; On 2012/03/12 10:55:05, rossberg wrote:
int32_t?
Done. https://chromiumcodereview.appspot.com/9600009/ -- v8-dev mailing list [email protected] http://groups.google.com/group/v8-dev
