Last modified: 2009-08-23 00:36:50 UTC
There was a double-encoded link in the page, formed like [[%E5%8B%92%E5%86%85%C2%B7%E6%88%88%E8%A5%BF%E5%B0%BC#.E4.BD.9C.E5.93.81.E5.88.97.E8.A1.A8|作品列表]], and in one edit, a user accidentally inserted a space into the path, formed [[%E5%8B%92%E5%86%85%C2%B7%E6%88%88%E8 %A5%BF%E5%B0%BC#.E4.BD.9C.E5.93.81.E5.88.97.E8.A1.A8|作品列表]]. And this link cause the whole page failed to display, except the categories. See http://zh.wikipedia.org/w/index.php?oldid=10944408&uselang=en and http://zh.wikipedia.org/w/index.php?oldid=10944786&uselang=en But for a not-encoded link, such as [[勒内·戈西尼]], no matter how many spaces is inserted, there is no problem. So please check this out, thanks. Best regards.
Additionally, comment <!-- bodytext --> <!-- NewPP limit report Preprocessor node count: 2286/1000000 Post-expand include size: 9740/2048000 bytes Template argument size: 4483/2048000 bytes Expensive parser function count: 4/500 --> can be found where the article content should be placed.
The space is inserted between bytes which make up a single UTF-8 character; the result is not a valid UTF-8 string. In development trunk (as of r55512) the system seems to correctly reject the link; our current deployment isn't as careful and lets the bad string through, where it can eventually trigger behavior in the regular expression library which rejects the bad string and results in wiping out the whole article in parsing. This issue is covered in bug 11143; duping the issue. *** This bug has been marked as a duplicate of bug 11143 ***