mediagoblin/db/mongo/models.py

   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
  17 import datetime
  18
  19 from mongokit import Document
  20
  21 from mediagoblin import mg_globals
  22 from mediagoblin.db.mongo import migrations
  23 from mediagoblin.db.mongo.util import ASCENDING, DESCENDING, ObjectId
  24 from mediagoblin.tools.pagination import Pagination
  25 from mediagoblin.tools import url
  26 from mediagoblin.db.mixin import UserMixin, MediaEntryMixin
  27
  28 ###################
  29 # Custom validators
  30 ###################
  31
  32 ########
  33 # Models
  34 ########
  35
  36
  37 class User(Document, UserMixin):
  38     """
  39     A user of MediaGoblin.
  40
  41     Structure:
  42      - username: The username of this user, should be unique to this instance.
  43      - email: Email address of this user
  44      - created: When the user was created
  45      - plugin_data: a mapping of extra plugin information for this User.
  46        Nothing uses this yet as we don't have plugins, but someday we
  47        might... :)
  48      - pw_hash: Hashed version of user's password.
  49      - email_verified: Whether or not the user has verified their email or not.
  50        Most parts of the site are disabled for users who haven't yet.
  51      - status: whether or not the user is active, etc.  Currently only has two
  52        values, 'needs_email_verification' or 'active'.  (In the future, maybe
  53        we'll change this to a boolean with a key of 'active' and have a
  54        separate field for a reason the user's been disabled if that's
  55        appropriate... email_verified is already separate, after all.)
  56      - verification_key: If the user is awaiting email verification, the user
  57        will have to provide this key (which will be encoded in the presented
  58        URL) in order to confirm their email as active.
  59      - is_admin: Whether or not this user is an administrator or not.
  60      - url: this user's personal webpage/website, if appropriate.
  61      - bio: biography of this user (plaintext, in markdown)
  62     """
  63     __collection__ = 'users'
  64     use_dot_notation = True
  65
  66     structure = {
  67         'username': unicode,
  68         'email': unicode,
  69         'created': datetime.datetime,
  70         'plugin_data': dict,  # plugins can dump stuff here.
  71         'pw_hash': unicode,
  72         'email_verified': bool,
  73         'status': unicode,
  74         'verification_key': unicode,
  75         'is_admin': bool,
  76         'url': unicode,
  77         'bio': unicode,      # May contain markdown
  78         'fp_verification_key': unicode,  # forgotten password verification key
  79         'fp_token_expire': datetime.datetime,
  80         }
  81
  82     required_fields = ['username', 'created', 'pw_hash', 'email']
  83
  84     default_values = {
  85         'created': datetime.datetime.utcnow,
  86         'email_verified': False,
  87         'status': u'needs_email_verification',
  88         'is_admin': False}
  89
  90
  91 class MediaEntry(Document, MediaEntryMixin):
  92     """
  93     Record of a piece of media.
  94
  95     Structure:
  96      - uploader: A reference to a User who uploaded this.
  97
  98      - title: Title of this work
  99
 100      - slug: A normalized "slug" which can be used as part of a URL to retrieve
 101        this work, such as 'my-works-name-in-slug-form' may be viewable by
 102        'http://mg.example.org/u/username/m/my-works-name-in-slug-form/'
 103        Note that since URLs are constructed this way, slugs must be unique
 104        per-uploader.  (An index is provided to enforce that but code should be
 105        written on the python side to ensure this as well.)
 106
 107      - created: Date and time of when this piece of work was uploaded.
 108
 109      - description: Uploader-set description of this work.  This can be marked
 110        up with MarkDown for slight fanciness (links, boldness, italics,
 111        paragraphs...)
 112
 113      - description_html: Rendered version of the description, run through
 114        Markdown and cleaned with our cleaning tool.
 115
 116      - media_type: What type of media is this?  Currently we only support
 117        'image' ;)
 118
 119      - media_data: Extra information that's media-format-dependent.
 120        For example, images might contain some EXIF data that's not appropriate
 121        to other formats.  You might store it like:
 122
 123          mediaentry.media_data['exif'] = {
 124              'manufacturer': 'CASIO',
 125              'model': 'QV-4000',
 126              'exposure_time': .659}
 127
 128        Alternately for video you might store:
 129
 130          # play length in seconds
 131          mediaentry.media_data['play_length'] = 340
 132
 133        ... so what's appropriate here really depends on the media type.
 134
 135      - plugin_data: a mapping of extra plugin information for this User.
 136        Nothing uses this yet as we don't have plugins, but someday we
 137        might... :)
 138
 139      - tags: A list of tags.  Each tag is stored as a dictionary that has a key
 140        for the actual name and the normalized name-as-slug, so ultimately this
 141        looks like:
 142          [{'name': 'Gully Gardens',
 143            'slug': 'gully-gardens'},
 144           {'name': 'Castle Adventure Time?!",
 145            'slug': 'castle-adventure-time'}]
 146
 147      - state: What's the state of this file?  Active, inactive, disabled, etc...
 148        But really for now there are only two states:
 149         "unprocessed": uploaded but needs to go through processing for display
 150         "processed": processed and able to be displayed
 151
 152      - license: URI for media's license.
 153
 154      - queued_media_file: storage interface style filepath describing a file
 155        queued for processing.  This is stored in the mg_globals.queue_store
 156        storage system.
 157
 158      - queued_task_id: celery task id.  Use this to fetch the task state.
 159
 160      - media_files: Files relevant to this that have actually been processed
 161        and are available for various types of display.  Stored like:
 162          {'thumb': ['dir1', 'dir2', 'pic.png'}
 163
 164      - attachment_files: A list of "attachment" files, ones that aren't
 165        critical to this piece of media but may be usefully relevant to people
 166        viewing the work.  (currently unused.)
 167
 168      - fail_error: path to the exception raised
 169      - fail_metadata:
 170     """
 171     __collection__ = 'media_entries'
 172     use_dot_notation = True
 173
 174     structure = {
 175         'uploader': ObjectId,
 176         'title': unicode,
 177         'slug': unicode,
 178         'created': datetime.datetime,
 179         'description': unicode,  # May contain markdown/up
 180         'description_html': unicode,  # May contain plaintext, or HTML
 181         'media_type': unicode,
 182         'media_data': dict,  # extra data relevant to this media_type
 183         'plugin_data': dict,  # plugins can dump stuff here.
 184         'tags': [dict],
 185         'state': unicode,
 186         'license': unicode,
 187
 188         # For now let's assume there can only be one main file queued
 189         # at a time
 190         'queued_media_file': [unicode],
 191         'queued_task_id': unicode,
 192
 193         # A dictionary of logical names to filepaths
 194         'media_files': dict,
 195
 196         # The following should be lists of lists, in appropriate file
 197         # record form
 198         'attachment_files': list,
 199
 200         # If things go badly in processing things, we'll store that
 201         # data here
 202         'fail_error': unicode,
 203         'fail_metadata': dict}
 204
 205     required_fields = [
 206         'uploader', 'created', 'media_type', 'slug']
 207
 208     default_values = {
 209         'created': datetime.datetime.utcnow,
 210         'state': u'unprocessed'}
 211
 212     def get_comments(self, ascending=False):
 213         if ascending:
 214             order = ASCENDING
 215         else:
 216             order = DESCENDING
 217
 218         return self.db.MediaComment.find({
 219                 'media_entry': self._id}).sort('created', order)
 220
 221     def generate_slug(self):
 222         self.slug = url.slugify(self.title)
 223
 224         duplicate = mg_globals.database.media_entries.find_one(
 225             {'slug': self.slug})
 226
 227         if duplicate:
 228             self.slug = "%s-%s" % (self._id, self.slug)
 229
 230     def url_to_prev(self, urlgen):
 231         """
 232         Provide a url to the previous entry from this user, if there is one
 233         """
 234         cursor = self.db.MediaEntry.find({'_id': {"$gt": self._id},
 235                                           'uploader': self.uploader,
 236                                           'state': 'processed'}).sort(
 237                                                     '_id', ASCENDING).limit(1)
 238         for media in cursor:
 239             return media.url_for_self(urlgen)
 240
 241     def url_to_next(self, urlgen):
 242         """
 243         Provide a url to the next entry from this user, if there is one
 244         """
 245         cursor = self.db.MediaEntry.find({'_id': {"$lt": self._id},
 246                                           'uploader': self.uploader,
 247                                           'state': 'processed'}).sort(
 248                                                     '_id', DESCENDING).limit(1)
 249
 250         for media in cursor:
 251             return media.url_for_self(urlgen)
 252
 253     @property
 254     def get_uploader(self):
 255         return self.db.User.find_one({'_id': self.uploader})
 256
 257
 258 class MediaComment(Document):
 259     """
 260     A comment on a MediaEntry.
 261
 262     Structure:
 263      - media_entry: The media entry this comment is attached to
 264      - author: user who posted this comment
 265      - created: when the comment was created
 266      - content: plaintext (but markdown'able) version of the comment's content.
 267      - content_html: the actual html-rendered version of the comment displayed.
 268        Run through Markdown and the HTML cleaner.
 269     """
 270
 271     __collection__ = 'media_comments'
 272     use_dot_notation = True
 273
 274     structure = {
 275         'media_entry': ObjectId,
 276         'author': ObjectId,
 277         'created': datetime.datetime,
 278         'content': unicode,
 279         'content_html': unicode}
 280
 281     required_fields = [
 282         'media_entry', 'author', 'created', 'content']
 283
 284     default_values = {
 285         'created': datetime.datetime.utcnow}
 286
 287     def media_entry(self):
 288         return self.db.MediaEntry.find_one({'_id': self['media_entry']})
 289
 290     @property
 291     def get_author(self):
 292         return self.db.User.find_one({'_id': self['author']})
 293
 294
 295 REGISTER_MODELS = [
 296     MediaEntry,
 297     User,
 298     MediaComment]
 299
 300
 301 def register_models(connection):
 302     """
 303     Register all models in REGISTER_MODELS with this connection.
 304     """
 305     connection.register(REGISTER_MODELS)