Last modified: 2012-09-26 22:59:12 UTC
Normally it's easy to get the raw text of an article adding "?action=raw" to the normal URL, like http://de.wikipedia.org/wiki/Nekrolog_2003?action=raw equal to the info page http://de.wikipedia.org/wiki/Nekrolog_2003?action=info This does not work if the page name contains a dot: http://de.wikipedia.org/wiki/Nekrolog_1._Quartal_2006?action=raw (I get "Forbidden: Invalid file extension found in the path info or query string.") The info page works: http://de.wikipedia.org/wiki/Nekrolog_1._Quartal_2006?action=info I know that there are other possibilites to get the raw text, but it think this should work, too.
To be exhaustive, it should be noted the canonical URL http//de.wikipedia.org/w/index.php?title=Nekrolog_1._Quartal_2006&action=raw works.
This is a protection against content type autodetection on older versions of Internet Explorer. Use the canonical URL so the .blah isn't at the end of the URL path.
This is not a bug, it is on purpose and by design for security reasons. Dots in that position can cause Internet Explorer to assume HTML parsing, raw content must only be served when the extension is controlled (i.e. .php). Aside from that, you shouldn't be constructing urls like this yourself. If you like to write your own urls, do follow the format the software uses, because the software will never create a url that follows the structure of one action combined with another (e.g. view action url, like /wiki/Page_name with action query). Which, for actions not otherwise specified (by default none), is index.php?title=&action= and that works. I'd say this is an instance of DontDoThat, the software doesn't lead there so its not a problem.
