Join GitHub today
GitHub is home to over 31 million developers working together to host and review code, manage projects, and build software together.Sign up
Add semicolons to avoid unicode feature-detect generating typos #1823
BTW: I am serving my pages as polyglot XHTML5 (serving "application/xhtml+xml" to browsers that can use it) so this particular bug shows up in firebug as a syntax error in the js and as a subsequent well-formedness error in the xhtml. See attached partial-screenshot.
I think the answer to your question is "yes, and yes", because although the unicode-test would still work fine in svg-content (where backslash-notation is used), at a glance the typo would seem to always cause a negative result (often false) to the offsetWidth test in non-svg content (where the html-entities-without-semicolons are used). In xhtml mode the test-characters are apparently not rendered at all due to the well-formedness error, and I guess in non-xml-html they would be rendered as the actual strings "ᝣ" and "☆" (either way they would be wrong) ...but I am not familiar with the code so take that as the quick impression that it is.
added a commit
this pull request
Jan 18, 2016
sorry it took so long to get this merged. I was trying to get xhtml5 testing built into the automated tests, but it isn't working in phantom. For whatever reason, the vml detect just won't work inside of a XHTML5 page, it ends up throwing a