diff options
| author | francis <francis> | 2008-11-10 18:18:30 +0000 |
|---|---|---|
| committer | francis <francis> | 2008-11-10 18:18:30 +0000 |
| commit | 405cd94fb7995e307cb74aa1b75055a95155a7c6 (patch) | |
| tree | f5695c70553c82d8450186012ebc67b56f76e1c5 | |
| parent | 8880863dd52593633494aee20e289fe0bbd9c6a1 (diff) | |
Use proper conversion function for HTML main parts.
| -rw-r--r-- | app/models/incoming_message.rb | 14 |
1 files changed, 4 insertions, 10 deletions
diff --git a/app/models/incoming_message.rb b/app/models/incoming_message.rb index fcd80857e..7fc8a640d 100644 --- a/app/models/incoming_message.rb +++ b/app/models/incoming_message.rb @@ -19,7 +19,7 @@ # Copyright (c) 2007 UK Citizens Online Democracy. All rights reserved. # Email: francis@mysociety.org; WWW: http://www.mysociety.org/ # -# $Id: incoming_message.rb,v 1.166 2008-11-10 18:08:30 francis Exp $ +# $Id: incoming_message.rb,v 1.167 2008-11-10 18:18:30 francis Exp $ # TODO # Move some of the (e.g. quoting) functions here into rblib, as they feel @@ -608,14 +608,8 @@ class IncomingMessage < ActiveRecord::Base text = main_part.body text_charset = main_part.charset if main_part.content_type == 'text/html' - # XXX could use better HTML to text conversion than this! - # (it only matters for emails without a text part, so not a massive deal - # e.g. http://www.whatdotheyknow.com/request/35/response/177 ) - text.gsub!(/<br[^>]+>/, "\n") - text.gsub!(/<p[^>]+>/, "\n\n") - text.gsub!(/<div[^>]+>/, "\n\n") - text.gsub!(/<\/?[^>]*>/, "") - text = HTMLEntities.decode_entities(text) + # e.g. http://www.whatdotheyknow.com/request/35/response/177 + text = IncomingMessage.get_attachment_text_internal_one_file(main_part.content_type, text) end end @@ -864,7 +858,7 @@ class IncomingMessage < ActiveRecord::Base text += child.read() + "\n\n" end elsif content_type == 'text/html' - IO.popen("/usr/bin/lynx -force_html -dump " + tempfile.path, "r") do |child| + IO.popen("/usr/bin/lynx -display_charset=UTF-8 -force_html -dump " + tempfile.path, "r") do |child| text += child.read() + "\n\n" end elsif content_type == 'application/vnd.ms-excel' |