1 # GNU MediaGoblin -- federated, autonomous media hosting
2 # Copyright (C) 2011, 2012 MediaGoblin contributors. See AUTHORS.
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.
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.
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/>.
19 from os
.path
import splitext
23 from werkzeug
.utils
import secure_filename
24 from werkzeug
.datastructures
import FileStorage
26 from mediagoblin
import mg_globals
27 from mediagoblin
.tools
.response
import json_response
28 from mediagoblin
.tools
.text
import convert_to_tag_list_of_dicts
29 from mediagoblin
.tools
.federation
import create_activity
, create_generator
30 from mediagoblin
.db
.models
import Collection
, MediaEntry
, ProcessingMetaData
31 from mediagoblin
.processing
import mark_entry_failed
, get_entry_and_processing_manager
32 from mediagoblin
.processing
.task
import ProcessMedia
33 from mediagoblin
.notifications
import add_comment_subscription
34 from mediagoblin
.media_types
import sniff_media
35 from mediagoblin
.user_pages
.lib
import add_media_to_collection
38 _log
= logging
.getLogger(__name__
)
41 def check_file_field(request
, field_name
):
42 """Check if a file field meets minimal criteria"""
43 retval
= (field_name
in request
.files
44 and isinstance(request
.files
[field_name
], FileStorage
)
45 and request
.files
[field_name
].stream
)
47 _log
.debug("Form did not contain proper file field %s", field_name
)
51 def new_upload_entry(user
):
53 Create a new MediaEntry for uploading
57 entry
.license
= user
.license_preference
61 def get_upload_file_limits(user
):
63 Get the upload_limit and max_file_size for this user
65 if user
.upload_limit
is not None and user
.upload_limit
>= 0: # TODO: debug this
66 upload_limit
= user
.upload_limit
68 upload_limit
= mg_globals
.app_config
.get('upload_limit', None)
70 max_file_size
= mg_globals
.app_config
.get('max_file_size', None)
72 return upload_limit
, max_file_size
75 class UploadLimitError(Exception):
77 General exception for when an upload will be over some upload limit
82 class FileUploadLimit(UploadLimitError
):
84 This file is over the site upload limit
89 class UserUploadLimit(UploadLimitError
):
91 This file is over the user's particular upload limit
96 class UserPastUploadLimit(UploadLimitError
):
98 The user is *already* past their upload limit!
104 def submit_media(mg_app
, user
, submitted_file
, filename
,
105 title
=None, description
=None, collection_slug
=None,
106 license
=None, metadata
=None, tags_string
=u
"",
107 callback_url
=None, urlgen
=None,):
110 - mg_app: The MediaGoblinApp instantiated for this process
111 - user: the user object this media entry should be associated with
112 - submitted_file: the file-like object that has the
113 being-submitted file data in it (this object should really have
114 a .name attribute which is the filename on disk!)
115 - filename: the *original* filename of this. Not necessarily the
116 one on disk being referenced by submitted_file.
117 - title: title for this media entry
118 - description: description for this media entry
119 - collection_slug: collection for this media entry
120 - license: license for this media entry
121 - tags_string: comma separated string of tags to be associated
123 - callback_url: possible post-hook to call after submission
124 - urlgen: if provided, used to do the feed_url update and assign a public
125 ID used in the API (very important).
127 upload_limit
, max_file_size
= get_upload_file_limits(user
)
128 if upload_limit
and user
.uploaded
>= upload_limit
:
129 raise UserPastUploadLimit()
131 # If the filename contains non ascii generate a unique name
132 if not all(ord(c
) < 128 for c
in filename
):
133 filename
= six
.text_type(uuid
.uuid4()) + splitext(filename
)[-1]
135 # Sniff the submitted media to determine which
136 # media plugin should handle processing
137 media_type
, media_manager
= sniff_media(submitted_file
, filename
)
139 # create entry and save in database
140 entry
= new_upload_entry(user
)
141 entry
.media_type
= media_type
142 entry
.title
= (title
or six
.text_type(splitext(filename
)[0]))
144 entry
.description
= description
or u
""
146 entry
.license
= license
or None
148 entry
.media_metadata
= metadata
or {}
150 # Process the user's folksonomy "tags"
151 entry
.tags
= convert_to_tag_list_of_dicts(tags_string
)
153 # Generate a slug from the title
154 entry
.generate_slug()
156 queue_file
= prepare_queue_task(mg_app
, entry
, filename
)
159 queue_file
.write(submitted_file
)
161 # Get file size and round to 2 decimal places
162 file_size
= mg_app
.queue_store
.get_file_size(
163 entry
.queued_media_file
) / (1024.0 * 1024)
164 file_size
= float('{0:.2f}'.format(file_size
))
166 # Check if file size is over the limit
167 if max_file_size
and file_size
>= max_file_size
:
168 raise FileUploadLimit()
170 # Check if user is over upload limit
171 if upload_limit
and (user
.uploaded
+ file_size
) >= upload_limit
:
172 raise UserUploadLimit()
174 user
.uploaded
= user
.uploaded
+ file_size
177 entry
.file_size
= file_size
179 # Save now so we have this data before kicking off processing
182 # Various "submit to stuff" things, callbackurl and this silly urlgen
185 metadata
= ProcessingMetaData()
186 metadata
.media_entry
= entry
187 metadata
.callback_url
= callback_url
191 # Generate the public_id, this is very importent, especially relating
192 # to deletion, it allows the shell to be accessable post-delete!
193 entry
.get_public_id(urlgen
)
195 # Generate the feed URL
197 'mediagoblin.user_pages.atom_feed',
198 qualified
=True, user
=user
.username
)
202 add_comment_subscription(user
, entry
)
205 create_activity("post", entry
, entry
.actor
)
210 collection
= Collection
.query
.filter_by(slug
=collection_slug
,
211 actor
=user
.id).first()
213 add_media_to_collection(collection
, entry
)
215 # Pass off to processing
217 # (... don't change entry after this point to avoid race
218 # conditions with changes to the document via processing code)
219 run_process_media(entry
, feed_url
)
224 def prepare_queue_task(app
, entry
, filename
):
226 Prepare a MediaEntry for the processing queue and get a queue file
228 # We generate this ourselves so we know what the task id is for
231 # (If we got it off the task's auto-generation, there'd be
232 # a risk of a race condition when we'd save after sending
234 task_id
= six
.text_type(uuid
.uuid4())
235 entry
.queued_task_id
= task_id
237 # Now store generate the queueing related filename
238 queue_filepath
= app
.queue_store
.get_unique_filepath(
241 secure_filename(filename
)])
243 # queue appropriately
244 queue_file
= app
.queue_store
.get_file(
245 queue_filepath
, 'wb')
247 # Add queued filename to the entry
248 entry
.queued_media_file
= queue_filepath
253 def run_process_media(entry
, feed_url
=None,
254 reprocess_action
="initial", reprocess_info
=None):
255 """Process the media asynchronously
257 :param entry: MediaEntry() instance to be processed.
258 :param feed_url: A string indicating the feed_url that the PuSH servers
259 should be notified of. This will be sth like: `request.urlgen(
260 'mediagoblin.user_pages.atom_feed',qualified=True,
261 user=request.user.username)`
262 :param reprocess_action: What particular action should be run.
263 :param reprocess_info: A dict containing all of the necessary reprocessing
264 info for the given media_type"""
266 reprocess_info
= reprocess_info
or {}
267 entry
, manager
= get_entry_and_processing_manager(entry
.id)
270 manager
.workflow(entry
, manager
, feed_url
, reprocess_action
, reprocess_info
)
271 except BaseException
as exc
:
272 # The purpose of this section is because when running in "lazy"
273 # or always-eager-with-exceptions-propagated celery mode that
274 # the failure handling won't happen on Celery end. Since we
275 # expect a lot of users to run things in this way we have to
276 # capture stuff here.
278 # ... not completely the diaper pattern because the
279 # exception is re-raised :)
280 mark_entry_failed(entry
.id, exc
)
281 # re-raise the exception
285 def api_upload_request(request
, file_data
, entry
):
286 """ This handles a image upload request """
287 # Use the same kind of method from mediagoblin/submit/views:submit_start
288 entry
.title
= file_data
.filename
290 # This will be set later but currently we just don't have enough information
294 entry
.get_public_id(request
.urlgen
)
296 queue_file
= prepare_queue_task(request
.app
, entry
, file_data
.filename
)
298 queue_file
.write(request
.data
)
301 return json_response(entry
.serialize(request
))
303 def api_add_to_feed(request
, entry
):
304 """ Add media to Feed """
305 feed_url
= request
.urlgen(
306 'mediagoblin.user_pages.atom_feed',
307 qualified
=True, user
=request
.user
.username
310 add_comment_subscription(request
.user
, entry
)
313 activity
= create_activity(
317 generator
=create_generator(request
)
320 run_process_media(entry
, feed_url
)