From 81c834f80becf46d9dbb8d1dea89960c6640d289 Mon Sep 17 00:00:00 2001 From: Gonzalo Odiard Date: Thu, 18 Nov 2010 21:02:06 +0000 Subject: Don't choke when downloading a URL containing non-UTF-8 characters (OLPC#8857) A URL can contain arbitrary characters; there's no guarantee they are encoded using UTF-8. Using the encoding of the page that contains the URL is as good a guess as any. Instead of throwing an exception and failing to download the file, we will now record a "garbled" file name in the Journal if there's an encoding mismatch. Tested-By: Simon Schampijer Reviewed-By: Sascha Silbe Acked-By: Lucian Branescu --- (limited to 'po') -- cgit v0.9.1