GNOME Bugzilla – Bug 575153
libsoup escapes hostnames, breaks IDNs
Last modified: 2009-12-19 14:44:30 UTC
Hi, It looks like libsoup doesn't have IDN support. Trying to fetch http://www.märchen.de/ fails, seems to be because it escapes the hostname: NET: downloading http://www.märchen.de NET: download status code: 0 NET: source after download: >>>http://www.m%C3%A4rchen.de/<<<
IDN not working is bug 548287. It would fail even if it didn't escape the hostname. :) however, i believe the SoupURI handling of non-ASCII names is currently somewhere between "unexpected" and "broken".
i think its already fixed in 2.28.0 with gresolver?
IDNs work but I didn't do any work on SoupURI, so it's probably not doing the right thing in all cases...
duping to the general URI-parsing nitpicking bug. (the page linked there gives specific recommendations about hostname encoding) *** This bug has been marked as a duplicate of bug 590524 ***