GNOME Bugzilla – Bug 575530
Accessing news.ycombinator.com with libsoup does not work
Last modified: 2009-03-16 16:30:25 UTC
With webkit-gtk-1.1.2 and epiphany-2.27.0 accessing http://news.ycombinator.com does not work. Log with ephy/webkit [snip] > GET / HTTP/1.1 > Soup-Debug-Timestamp: 1237194249 > Soup-Debug: SoupSessionAsync 1 (0x159bbe0), SoupMessage 32 (0x1cdeba0), SoupSocket 30 (0x1c71360) > Host: news.ycombinator.com > User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; et-ee) AppleWebKit/528.5+ (KHTML, like Gecko, Safari/528.5+) epiphany > Accept: application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5 [/snip] When checking with CURL, website answers: HTTP/1.0 200 OK Content-Type: text/html; charset=utf-8 Connection: close
This is a bug in the software running that website (Arc?). It's using plain LF as a delimiter between header lines instead of CRLF (and LF LF instead of CRLF CRLF between the header and body): read(13, "HTTP/1.0 200 OK\nContent-Type: text/html; charset=utf-8\nConnection: close\n\n"..., 8192) = 74 There's already a bug filed about making libsoup deal with that. Not sure when it will be fixed. (You could also try to get the website fixed :). *** This bug has been marked as a duplicate of 571283 ***