1 # GNU MediaGoblin -- federated, autonomous media hosting
2 # Copyright (C) 2011 Free Software Foundation, Inc
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 mongokit
import Document
21 from mediagoblin
import util
22 from mediagoblin
.auth
import lib
as auth_lib
23 from mediagoblin
import mg_globals
24 from mediagoblin
.db
import migrations
25 from mediagoblin
.db
.util
import ASCENDING
, DESCENDING
, ObjectId
26 from mediagoblin
.util
import Pagination
27 from mediagoblin
.util
import DISPLAY_IMAGE_FETCHING_ORDER
41 A user of MediaGoblin.
44 - username: The username of this user, should be unique to this instance.
45 - email: Email address of this user
46 - created: When the user was created
47 - plugin_data: a mapping of extra plugin information for this User.
48 Nothing uses this yet as we don't have plugins, but someday we
50 - pw_hash: Hashed version of user's password.
51 - email_verified: Whether or not the user has verified their email or not.
52 Most parts of the site are disabled for users who haven't yet.
53 - status: whether or not the user is active, etc. Currently only has two
54 values, 'needs_email_verification' or 'active'. (In the future, maybe
55 we'll change this to a boolean with a key of 'active' and have a
56 separate field for a reason the user's been disabled if that's
57 appropriate... email_verified is already separate, after all.)
58 - verification_key: If the user is awaiting email verification, the user
59 will have to provide this key (which will be encoded in the presented
60 URL) in order to confirm their email as active.
61 - is_admin: Whether or not this user is an administrator or not.
62 - url: this user's personal webpage/website, if appropriate.
63 - bio: biography of this user (plaintext, in markdown)
64 - bio_html: biography of the user converted to proper HTML.
66 __collection__
= 'users'
71 'created': datetime
.datetime
,
72 'plugin_data': dict, # plugins can dump stuff here.
74 'email_verified': bool,
76 'verification_key': unicode,
79 'bio' : unicode, # May contain markdown
80 'bio_html': unicode, # May contain plaintext, or HTML
81 'fp_verification_key': unicode, # forgotten password verification key
82 'fp_token_expire': datetime
.datetime
85 required_fields
= ['username', 'created', 'pw_hash', 'email']
88 'created': datetime
.datetime
.utcnow
,
89 'email_verified': False,
90 'status': u
'needs_email_verification',
91 'verification_key': lambda: unicode(uuid
.uuid4()),
94 def check_login(self
, password
):
96 See if a user can login with this password
98 return auth_lib
.bcrypt_check_password(
99 password
, self
['pw_hash'])
102 class MediaEntry(Document
):
104 Record of a piece of media.
107 - uploader: A reference to a User who uploaded this.
109 - title: Title of this work
111 - slug: A normalized "slug" which can be used as part of a URL to retrieve
112 this work, such as 'my-works-name-in-slug-form' may be viewable by
113 'http://mg.example.org/u/username/m/my-works-name-in-slug-form/'
114 Note that since URLs are constructed this way, slugs must be unique
115 per-uploader. (An index is provided to enforce that but code should be
116 written on the python side to ensure this as well.)
118 - created: Date and time of when this piece of work was uploaded.
120 - description: Uploader-set description of this work. This can be marked
121 up with MarkDown for slight fanciness (links, boldness, italics,
124 - description_html: Rendered version of the description, run through
125 Markdown and cleaned with our cleaning tool.
127 - media_type: What type of media is this? Currently we only support
130 - media_data: Extra information that's media-format-dependent.
131 For example, images might contain some EXIF data that's not appropriate
132 to other formats. You might store it like:
134 mediaentry['media_data']['exif'] = {
135 'manufacturer': 'CASIO',
137 'exposure_time': .659}
139 Alternately for video you might store:
141 # play length in seconds
142 mediaentry['media_data']['play_length'] = 340
144 ... so what's appropriate here really depends on the media type.
146 - plugin_data: a mapping of extra plugin information for this User.
147 Nothing uses this yet as we don't have plugins, but someday we
150 - tags: A list of tags. Each tag is stored as a dictionary that has a key
151 for the actual name and the normalized name-as-slug, so ultimately this
153 [{'name': 'Gully Gardens',
154 'slug': 'gully-gardens'},
155 {'name': 'Castle Adventure Time?!",
156 'slug': 'castle-adventure-time'}]
158 - state: What's the state of this file? Active, inactive, disabled, etc...
159 But really for now there are only two states:
160 "unprocessed": uploaded but needs to go through processing for display
161 "processed": processed and able to be displayed
163 - queued_media_file: storage interface style filepath describing a file
164 queued for processing. This is stored in the mg_globals.queue_store
167 - queued_task_id: celery task id. Use this to fetch the task state.
169 - media_files: Files relevant to this that have actually been processed
170 and are available for various types of display. Stored like:
171 {'thumb': ['dir1', 'dir2', 'pic.png'}
173 - attachment_files: A list of "attachment" files, ones that aren't
174 critical to this piece of media but may be usefully relevant to people
175 viewing the work. (currently unused.)
177 - fail_error: path to the exception raised
180 __collection__
= 'media_entries'
183 'uploader': ObjectId
,
186 'created': datetime
.datetime
,
187 'description': unicode, # May contain markdown/up
188 'description_html': unicode, # May contain plaintext, or HTML
189 'media_type': unicode,
190 'media_data': dict, # extra data relevant to this media_type
191 'plugin_data': dict, # plugins can dump stuff here.
195 # For now let's assume there can only be one main file queued
197 'queued_media_file': [unicode],
198 'queued_task_id': unicode,
200 # A dictionary of logical names to filepaths
203 # The following should be lists of lists, in appropriate file
205 'attachment_files': list,
207 # If things go badly in processing things, we'll store that
209 'fail_error': unicode,
210 'fail_metadata': dict}
213 'uploader', 'created', 'media_type', 'slug']
216 'created': datetime
.datetime
.utcnow
,
217 'state': u
'unprocessed'}
219 def get_comments(self
):
220 return self
.db
.MediaComment
.find({
221 'media_entry': self
['_id']}).sort('created', DESCENDING
)
223 def get_display_media(self
, media_map
, fetch_order
=DISPLAY_IMAGE_FETCHING_ORDER
):
225 Find the best media for display.
228 - media_map: a dict like
229 {u'image_size': [u'dir1', u'dir2', u'image.jpg']}
230 - fetch_order: the order we should try fetching images in
233 (media_size, media_path)
235 media_sizes
= media_map
.keys()
237 for media_size
in DISPLAY_IMAGE_FETCHING_ORDER
:
238 if media_size
in media_sizes
:
239 return media_map
[media_size
]
241 def main_mediafile(self
):
244 def generate_slug(self
):
245 self
['slug'] = util
.slugify(self
['title'])
247 duplicate
= mg_globals
.database
.media_entries
.find_one(
248 {'slug': self
['slug']})
251 self
['slug'] = "%s-%s" % (self
['_id'], self
['slug'])
253 def url_for_self(self
, urlgen
):
255 Generate an appropriate url for ourselves
257 Use a slug if we have one, else use our '_id'.
259 uploader
= self
.uploader()
263 'mediagoblin.user_pages.media_home',
264 user
=uploader
['username'],
268 'mediagoblin.user_pages.media_home',
269 user
=uploader
['username'],
270 media
=unicode(self
['_id']))
272 def url_to_prev(self
, urlgen
):
274 Provide a url to the previous entry from this user, if there is one
276 cursor
= self
.db
.MediaEntry
.find({'_id' : {"$gt": self
['_id']},
277 'uploader': self
['uploader'],
278 'state': 'processed'}).sort(
279 '_id', ASCENDING
).limit(1)
281 return urlgen('mediagoblin.user_pages.media_home',
282 user
=self
.uploader()['username'],
283 media
=unicode(cursor
[0]['slug']))
285 def url_to_next(self
, urlgen
):
287 Provide a url to the next entry from this user, if there is one
289 cursor
= self
.db
.MediaEntry
.find({'_id' : {"$lt": self
['_id']},
290 'uploader': self
['uploader'],
291 'state': 'processed'}).sort(
292 '_id', DESCENDING
).limit(1)
295 return urlgen('mediagoblin.user_pages.media_home',
296 user
=self
.uploader()['username'],
297 media
=unicode(cursor
[0]['slug']))
300 return self
.db
.User
.find_one({'_id': self
['uploader']})
302 def get_fail_exception(self
):
304 Get the exception that's appropriate for this error
306 if self
['fail_error']:
307 return util
.import_component(self
['fail_error'])
310 class MediaComment(Document
):
312 A comment on a MediaEntry.
315 - media_entry: The media entry this comment is attached to
316 - author: user who posted this comment
317 - created: when the comment was created
318 - content: plaintext (but markdown'able) version of the comment's content.
319 - content_html: the actual html-rendered version of the comment displayed.
320 Run through Markdown and the HTML cleaner.
323 __collection__
= 'media_comments'
326 'media_entry': ObjectId
,
328 'created': datetime
.datetime
,
330 'content_html': unicode}
333 'media_entry', 'author', 'created', 'content']
336 'created': datetime
.datetime
.utcnow
}
338 def media_entry(self
):
339 return self
.db
.MediaEntry
.find_one({'_id': self
['media_entry']})
342 return self
.db
.User
.find_one({'_id': self
['author']})
351 def register_models(connection
):
353 Register all models in REGISTER_MODELS with this connection.
355 connection
.register(REGISTER_MODELS
)