LGTM. Can the ARM compiler cope with the templates?
http://codereview.chromium.org/42280/diff/1/3 File src/dateparser.cc (right): http://codereview.chromium.org/42280/diff/1/3#newcode85 Line 85: There should be two blank lines between methods. Please put them back. http://codereview.chromium.org/42280/diff/1/4 File src/dateparser.h (right): http://codereview.chromium.org/42280/diff/1/4#newcode235 Line 235: template <typename Char> Since the templates force us to move this to a .h file we should probably move it to an -inl.h file. http://codereview.chromium.org/42280/diff/1/4#newcode302 Line 302: // TODO(lrn) Is this correct? I should be possible to find the answer to this and write a test that verifies it. http://codereview.chromium.org/42280/diff/1/6 File test/mjsunit/date-parse.js (right): http://codereview.chromium.org/42280/diff/1/6#newcode44 Line 44: assertTrue(d > 0 && !isNaN(d), string); This should be two assertions. http://codereview.chromium.org/42280/diff/1/6#newcode49 Line 49: assertTrue(array.length == 2, "array [" + array + "] length != 2"); This should be assertEquals(2, array.length, ...) http://codereview.chromium.org/42280 --~--~---------~--~----~------------~-------~--~----~ v8-dev mailing list [email protected] http://groups.google.com/group/v8-dev -~----------~----~----~----~------~----~------~--~---
