Last modified: 2012-05-03 02:42:39 UTC
In the above URL, we see how I had to remove the >$ unicode U+3000 >U+3000 IDEOGRAPHIC SPACE >UTF-8: e3 80 80 UTF-16BE: 3000 Decimal:   > >Category: Zs (Separator, Space) >Bidi: WS (Whitespace) >Decomposition: <wide> 0020 from >交通部有做總表:http://www.highwaybus.nat.gov.tw/work/permission_report.htm 。 lest it get appended into the link, and not treated just like the ASCII space next to it. Shouldn't the parser treat both types of spaces the same in this situation? Can Asian users reasonably be expected to always remember to terminate in ASCII spaces? I wonder how Bugzilla will treat that line, 5 lines above this. See also bug 1414.
And indeed, http://taizhongbus.jidanni.org/index.php?title=使用者討論:Msnhinet8&diff=5160&oldid=5111 finds me shaving an U+3002 IDEOGRAPHIC FULL STOP off of http://www.wikia.com/wiki/Wikia_(%E6%AD%A3)%E3%80%82 to get the correct http://www.wikia.com/wiki/Wikia_(%E6%AD%A3)
*** Bug 25409 has been marked as a duplicate of this bug. ***
Updated to cover all unicode characters in the 'separator, space' category. See r93291 which is pending review.
This was reviewed and should be deployed with 1.19