I will agree that neither of those are ideal - but thankfully they're both sufficiently obscure enough that they won't cause immediate problems.
Could you file a bug on them? http://dev.jquery.com/ Thanks! --John On Sun, Jan 4, 2009 at 9:31 AM, Balazs Endresz <balazs.endr...@gmail.com> wrote: > > After some more testing there seems to be some issues with this > approach as it fails if you have a selector like this: > input[name=".types["] > If the value contains a dot and an opening bracket then the class > selector regex will match the string after the dot. > > Also it won't match .foo in this expression: > .foo #bar\] > > I guess the second one can be fixed somehow but the first one seems to > be pretty hard to resolve -- at least for me :) > > Cheers, > Balazs > > On Jan 4, 12:08 am, "John Resig" <jere...@gmail.com> wrote: >> This should be fixed now. Thanks for the tips, >> Balazs!http://github.com/jeresig/sizzle/commit/c757d135949d0a4b15a9fd6d72907... >> >> --John >> >> On Sat, Jan 3, 2009 at 10:35 AM, Balazs Endresz >> >> <balazs.endr...@gmail.com> wrote: >> >> > Hi, >> > a couple of months ago I made little patch to Sizzle >> >http://groups.google.com/group/jquery-dev/browse_thread/thread/d74e5a... >> > to enable selection by data and css properties inside an attribute >> > selector: >> >http://jsbin.com/otoqo/edit >> >> > It uses a negative lookahead to prevent other selectors to match >> > inside square brackets, so it "accidentally" fixes some of the bugs >> > reported here. But the interesting thing is that I ran a few >> > Slickspeed tests to see its impact on speed and quite surprisingly it >> > turned out to be even 1-2% faster on FF3 and in IE7 the difference is >> > below 1 percent. I only tested with 1.3b1 and it passes all the jQuery >> > selector tests as well! >> >> > On Jan 3, 3:25 pm, "John Resig" <jere...@gmail.com> wrote: >> >> The primary issue is that Sizzle randomly grabs portions of a selector >> >> in a specific order. This helps to improve speed but it harms complex >> >> cases like this. I'm working on a patch to resolve it. >> >> >> --John >> >> >> On Sat, Jan 3, 2009 at 12:56 AM, chuyeow <chuy...@gmail.com> wrote: >> >> >> > Hi Olaf, >> >> >> > This bug has been reported before but there is no patch for it yet. >> >> >> > So far I've been able to narrow it down to selectors with periods in >> >> > them being incorrectly recognized as a class selector, but do not have >> >> > enough regex-fu to know how to solve the bug yet. >> >> >> > The class selector regex in Sizzle: /\.((?:[\w\u0128-\uFFFF_-]|\\.)+)/ >> >> >> > Running this regex against "a[href$='.jpg']" (or any similar non-class >> >> > selector with a period in them): >> >> > "a[href$='.jpg/']".match(/\.((?:[\w\u0128-\uFFFF_-]|\\.)+)/) >> >> > => returns matches = [.jpg, jpg] which is wrong >> >> >> > Hopefully this helps anyone also looking at the bug. >> >> >> > Cheers, >> >> > Chu Yeow >> >> >> > On Dec 30 2008, 8:19 pm, Olaf <olaf.bo...@t-online.de> wrote: >> >> >> Hi, I test 1.3.b1 and this Selector work any more: >> >> >> >> <script type="text/javascript"> >> >> >> //<![CDATA[ >> >> >> $(document).ready(function() { >> >> >> // no more with 1.3b1 >> >> >> // $("#content a[href$='.jpg']").lightBox(); >> >> >> $("#content a").lightBox();}); >> >> >> >> //]]> >> >> >> </script> >> >> >> >> What is to do? Thanks. > > > --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "jQuery Development" group. To post to this group, send email to jquery-dev@googlegroups.com To unsubscribe from this group, send email to jquery-dev+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/jquery-dev?hl=en -~----------~----~----~----~------~----~------~--~---
