Version: (using KDE Devel) Installed from: Compiled sources In an XHTML 1.1 file umlaut character entities are not displayed at all if that file ('s name) has no .html or .xhtml ending. Test case follows
Created attachment 2027 [details] From this document, the
Created attachment 2028 [details] The '
both test case files are identical, but they have different content types
Confirmed. And if I put Konqueror in Latin1 in the first file, the entity is displayed.
That's interesting. If I changed the charset in the file there would be no difference (i.e. no 'ü' displayed). That means a narrow condition for that behavior. BTW I discovered all this when looking at http://mozilla.linuxfaqs.de/css (in german). That's some kind of mozilla 'propaganda' site. They have a few cool testcases there. Among other problems, you can't see the umlauts in Konqueror.
The classical problem: as Konqueror is a HTML browser, not a XHTMl one, it does not handle the XML declaration. So the title is expected in ISO-8859-1 not in UTF-8, which is declared one line below it, incorrectly as it does not follow the HTML-compatibility guidelines of XHTML 1.0. Have a nice day!
Sorry but my report was not specifically because of the umlauts in the title but also because of them everywhere else. The string "CSS-Menü" should also appear in a paragraph inside the document. In Konqueror the "ü" (= ü) is missing. And I wonder why a character encoding (even a wrongly set one) would affect 'my' umlauts if they were given as HTML-entities in the source code. Instead they aren't displayed at all. Besides I couldn't see why something like <title>CSS-Menü</title> (as it appears in _both_ testcases) was against the HTML-compatibility guidelines. And if Konqi ignored the <?xml ... ?> line (if I understood correctly) I do even less understand why some HTML entities are not displayed.
BTW I can confirm this behavior for Konqi( or KHTML)'s CVS_HEAD 20031205. And please do not get distracted by the character encoding problems on this page. Some of the comments I made with Konqi (including this one: '
Good, if you are using CVS HEAD, I am stopping here, as it might also be a regression. But what may be also the case: you are serving the first test case as text/xml (not text/html, not apllication/xhtml+xml) so by default you only have the XML character references (lt, gt, amp, quot, apos.) Have a nice day!
Still reproducable for me with KDE 3.3.2. If I change encoding to utf-8 (from autodetect / western europe) as defined in the XML everything works as expected.
Bug still reproducible in svn trunk r793457.
Thank you for the bug report. As this report hasn't seen any changes in 10 years or more, we ask if you can please confirm that the issue still persists. If this bug is no longer persisting or relevant please change the status to resolved.