| 1 | # GNU MediaGoblin -- federated, autonomous media hosting |
| 2 | # Copyright (C) 2011, 2012 MediaGoblin contributors. See AUTHORS. |
| 3 | # |
| 4 | # This program is free software: you can redistribute it and/or modify |
| 5 | # it under the terms of the GNU Affero General Public License as published by |
| 6 | # the Free Software Foundation, either version 3 of the License, or |
| 7 | # (at your option) any later version. |
| 8 | # |
| 9 | # This program is distributed in the hope that it will be useful, |
| 10 | # but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 11 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
| 12 | # GNU Affero General Public License for more details. |
| 13 | # |
| 14 | # You should have received a copy of the GNU Affero General Public License |
| 15 | # along with this program. If not, see <http://www.gnu.org/licenses/>. |
| 16 | from __future__ import unicode_literals |
| 17 | import logging |
| 18 | import re |
| 19 | |
| 20 | from mediagoblin import meddleware |
| 21 | |
| 22 | _log = logging.getLogger(__name__) |
| 23 | |
| 24 | class TrimWhiteSpaceMeddleware(meddleware.BaseMeddleware): |
| 25 | _setup_plugin_called = 0 |
| 26 | RE_MULTI_WHITESPACE = re.compile(b'(\s)\s+', re.M) |
| 27 | |
| 28 | def process_response(self, request, response): |
| 29 | """Perform very naive html tidying by removing multiple whitespaces""" |
| 30 | # werkzeug.BaseResponse has no content_type attr, this comes via |
| 31 | # werkzeug.wrappers.CommonRequestDescriptorsMixin (part of |
| 32 | # wrappers.Response) |
| 33 | if getattr(response ,'content_type', None) != 'text/html': |
| 34 | return |
| 35 | |
| 36 | # This is a tad more complex than needed to be able to handle |
| 37 | # response.data and response.body, depending on whether we have |
| 38 | # a werkzeug Resonse or a webob one. Let's kill webob soon! |
| 39 | if hasattr(response, 'body') and not hasattr(response, 'data'): |
| 40 | # Old-style webob Response object. |
| 41 | # TODO: Remove this once we transition away from webob |
| 42 | resp_attr = 'body' |
| 43 | else: |
| 44 | resp_attr = 'data' |
| 45 | # Don't flatten iterator to list when we fudge the response body |
| 46 | # (see werkzeug.Response documentation) |
| 47 | response.implicit_sequence_conversion = False |
| 48 | |
| 49 | # Set the tidied text. Very naive tidying for now, just strip all |
| 50 | # subsequent whitespaces (this preserves most newlines) |
| 51 | setattr(response, resp_attr, re.sub( |
| 52 | TrimWhiteSpaceMeddleware.RE_MULTI_WHITESPACE, br'\1', |
| 53 | getattr(response, resp_attr))) |
| 54 | |
| 55 | @classmethod |
| 56 | def setup_plugin(cls): |
| 57 | """Set up this meddleware as a plugin during 'setup' hook""" |
| 58 | global _log |
| 59 | if cls._setup_plugin_called: |
| 60 | _log.info('Trim whitespace plugin was already set up.') |
| 61 | return |
| 62 | |
| 63 | _log.debug('Trim whitespace plugin set up.') |
| 64 | cls._setup_plugin_called += 1 |
| 65 | |
| 66 | # Append ourselves to the list of enabled Meddlewares |
| 67 | meddleware.ENABLED_MEDDLEWARE.append( |
| 68 | '{0}:{1}'.format(cls.__module__, cls.__name__)) |
| 69 | |
| 70 | |
| 71 | hooks = { |
| 72 | 'setup': TrimWhiteSpaceMeddleware.setup_plugin |
| 73 | } |