<html><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; "><br><div><div>On Aug 26, 2008, at 8:37 PM, Daniel Dunbar wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite"><div dir="ltr">For those interested in this sort of thing, here are the prettified<br>gcov results of clang running on its test suite (71.6%):<br><br><a href="http://keeda.stanford.edu/~ddunbar/clang/coverage/">http://keeda.stanford.edu/~ddunbar/clang/coverage/</a><br> <br>And running on the test suite + LLVM's test suite (72.7%):<br><br><a href="http://keeda.stanford.edu/~ddunbar/clang/coverage.full/">http://keeda.stanford.edu/~ddunbar/clang/coverage.full/</a><br><br>Better than I expected, but it would be great to push those numbers<br> higher if anyone is bored. ;)<br></div></blockquote></div><br><div>Nice! Looking at liblex, it looks like we need to test headermaps (horrible apple feature), backtracking (new and shiny!), 6 of the 9 trigraphs, escaped newlines and trigraphs in identifiers, unterminated block comments, nul characters in the middle of file, the MS charize extension, some random expressions in the pp expr parsing, and some even more strange stuff...</div><div><br></div><div>-Chris</div></body></html>