danbooru.py
danbooru.py is a Python script-slash-tool for interfacing with Danbooru. It allows you to batch-download images from their official API. It does not use screen scraping and, as far as I can tell, does not violate their terms of service. However, Danbooru has been limiting the API access as of late, so a lot of the times the script doesn’t work. It’s supposed to be because of the load, I don’t know.
Due to the way the script is made, it is easy to make it work with other sites that use the same software as Danbooru, like moe.imuoto. You can download a quick and dirty version of the script that does so: moe.py. Eventually, I’ll pretty these things up and release a proper version and support more sites.
Code Preview
This is a slightly outdated preview of what the code looks like. I’m including it because it looks pretty, and also because there is documentation in it.
| 1 | #!/usr/bin/env python |
| 2 | |
| 3 | ''' |
| 4 | danbooru.py (http://untu.ms/danbooru/) |
| 5 | ====================================== |
| 6 | A content retrieval tool for danbooru (http://danbooru.donmai.us/). The |
| 7 | requirements are Python 2.5 (http://python.org/) and a little console-fu. |
| 8 | |
| 9 | usage examples |
| 10 | ============== |
| 11 | * danbooru.py negima “cat ears” |
| 12 | Download content tagged negima and cat_ears to the default folder |
| 13 | (negima+cat_ears) |
| 14 | * danbooru.py -x himm -r safe “sawatari izumi” |
| 15 | Download content tagged sawatari_izumi and rated safe (-r or –rating) to |
| 16 | a folder named himm (-f or –folder) |
| 17 | * danbooru.py -l 50 gif |
| 18 | Download content tagged as gif, limiting it to 50 posts (-l or –limit) |
| 19 | * danbooru.py -i -n -s 8 flash |
| 20 | Download content tagged flash, ignoring the youngest local file (or last |
| 21 | id, refers to the -i or –no-last-id option), ignore whether the file |
| 22 | exists in the local database (-n or –no-db), and use server 8 (-s or |
| 23 | –server option, use -L or –list to see a list of available servers) |
| 24 | * danbooru.py -c * -x * |
| 25 | Catalogue (-c or –catalogue) and rename (-x or –fix) all files in all |
| 26 | subfolders in the current path |
| 27 | * danbooru.py -h |
| 28 | View a list of available commands |
| 29 | |
| 30 | version history |
| 31 | =============== |
| 32 | * 0.2: 09.01.2007 |
| 33 | * 0.1: 01.01.2007 |
| 34 | |
| 35 | copyright |
| 36 | ========= |
| 37 | danbooru.py is made by Reinis Ivanovs (dabas@untu.ms) and is released |
| 38 | to the public domain. |
| 39 | ''' |
| 40 | |
| 41 | import re |
| 42 | import os |
| 43 | import urllib |
| 44 | import shelve |
| 45 | import sqlite3 |
| 46 | import pickle |
| 47 | |
| 48 | from glob import glob, iglob |
| 49 | from hashlib import md5 |
| 50 | from sys import platform, stderr |
| 51 | from time import time |
| 52 | from xml.dom import minidom |
| 53 | |
| 54 | # time.clock is more granual than time.time on win32 |
| 55 | if platform == 'win32': |
| 56 | from time import clock as xtime, sleep |
| 57 | else: |
| 58 | from time import time as xtime, sleep |
| 59 | |
| 60 | # A tender age |
| 61 | __version__ = '0.2' |
| 62 | __build__ = '119' |
| 63 | |
| 64 | case = lambda count, word: word if count == 1 else word + 's' |
| 65 | cases = lambda count, singular, plural: singular if count == 1 else plural |
| 66 | |
| 67 | |
| 68 | # Identify as danbooru.py/0.x (change this if you want to go ninja) |
| 69 | class Opener(urllib.FancyURLopener): |
| 70 | version = 'danbooru.py/%s' % (__version__,) |
| 71 | urllib._urlopener = Opener() |
| 72 | |
| 73 | |
| 74 | class ServerIdError(KeyError): |
| 75 | '''No such server ID''' |
| 76 | |
| 77 | |
| 78 | class Robot(dict): |
| 79 | api_url = 'http://danbooru.donmai.us/api/' |
| 80 | posts_path = 'find_posts?tags=%(tags)s%(last_id)s%(rating)s&limit=\ |
| 81 | %(limit)d&offset=%(offset)d' |
| 82 | last_id = '+after_id:%d' |
| 83 | rating_path = '+rating:%s' |
| 84 | servers_path = 'find_servers' |
| 85 | md5_path = 'find_posts?md5=%s' |
| 86 | settings_filename = os.path.join(os.path.expanduser('~'), '.danboorudata') |
| 87 | db_filename = os.path.join(os.path.expanduser('~'), '.danboorudb') |
| 88 | namepattern = re.compile(r'(?:\d+_)?([a-f\d]{32})') |
| 89 | idpattern = re.compile(r'(\d+)_[a-f\d]{32}') |
| 90 | logfile = 'error.log' |
| 91 | |
| 92 | def __init__(self, args, limit, offset, **kwargs): |
| 93 | for key, value in kwargs.iteritems(): |
| 94 | self[key] = value |
| 95 | self.end = lambda text, start: '%s (%.2fs)' % (text, time()-start) |
| 96 | self.tags = self.parse_tags(args) |
| 97 | self.settings = self.load_settings() |
| 98 | self.servers = self.load_servers() |
| 99 | self.db, self.cur = self.load_db() |
| 100 | self.folder = self.tags |
| 101 | self.limit = limit |
| 102 | self.offset = offset |
| 103 | self.dl = Downloader() |
| 104 | |
| 105 | def get_last_id(self, pathname): |
| 106 | '''Get the youngest file by its danbooru id''' |
| 107 | if self['refresh']: |
| 108 | return '' |
| 109 | filenames = self.get_filenames(pathname) |
| 110 | filenames = sorted(filenames)[::-1] |
| 111 | for item in filenames: |
| 112 | folder, name, ext = self.split_path(item) |
| 113 | match = re.match(self.idpattern, name) |
| 114 | if match: |
| 115 | id = int(match.group(1).lstrip('0')) |
| 116 | break |
| 117 | else: id = 1 |
| 118 | return self.last_id % (int(id),) |
| 119 | |
| 120 | def error(self, message): |
| 121 | print >> stderr, 'Error: %s' % (message,) |
| 122 | |
| 123 | def retrieve_content(self): |
| 124 | '''Start downloading''' |
| 125 | print 'Downloading to %s…' % (self.folder,) |
| 126 | if not os.path.exists(self.folder): |
| 127 | os.mkdir(self.folder) |
| 128 | last_id = self.get_last_id(self.folder) |
| 129 | step, limit, offset = 100, self.limit, self.offset |
| 130 | for i in xrange(offset, limit, step): |
| 131 | j = i+step if i+step < limit else limit |
| 132 | params = { 'tags': self.tags, 'last_id': last_id, 'limit': j, |
| 133 | 'offset': i, 'rating': self.rating_path % self['rating'] \ |
| 134 | if self['rating'] else ''} |
| 135 | path, start = self.posts_path % params, time() |
| 136 | url = self.api_url+path |
| 137 | print 'API:', path+'…', |
| 138 | data = self.get_data(url, 'post', 'id') |
| 139 | print self.end('done', start) |
| 140 | if self['nodb'] or not len(data): |
| 141 | print '%d posts returned' % (len(data),) |
| 142 | if not len(data): break |
| 143 | else: |
| 144 | before = len(data) |
| 145 | self.filter_data(data) |
| 146 | values = (before, case(before, 'post'), |
| 147 | len(data), cases(len(data), 'wasn\'t', 'weren\'t')) |
| 148 | print '%d %s returned, %d %s in the local database' % values |
| 149 | if self['simulate']: continue |
| 150 | if len(data): |
| 151 | for key, value in data.iteritems(): |
| 152 | self.get_post(key, value) |
| 153 | self.update_db(data) |
| 154 | self.db.commit() |
| 155 | try: |
| 156 | if before < step: break |
| 157 | except UnboundLocalError, e: |
| 158 | print e |
| 159 | else: |
| 160 | print 'Post limit (%d) met' % (limit,) |
| 161 | if not glob(os.path.join(self.folder, '*')): |
| 162 | print '%s is empty: removing' % (self.folder,) |
| 163 | os.rmdir(self.folder) |
| 164 | |
| 165 | def get_post(self, id, post): |
| 166 | '''Download an individual post''' |
| 167 | filename = post['file_name'] |
| 168 | # Figure out the local name (id is padded with zeroes) |
| 169 | localname = os.path.join(self.folder, '%07d_%s' % (id, filename)) |
| 170 | if os.path.exists(localname): |
| 171 | self.error('File already exists') |
| 172 | return |
| 173 | server = self.servers[self.server] |
| 174 | server = 'http://' + server['host'] + server['path'] |
| 175 | url = server + '/'.join((filename[0:2], filename[2:4], filename)) |
| 176 | print url |
| 177 | print self.dl.retrieve(url, localname, self.exit), \ |
| 178 | 'KiB retrieved in %s' % (self.folder,) |
| 179 | |
| 180 | def filter_data(self, data): |
| 181 | '''Filter out the data that already exists in the local db''' |
| 182 | values = '”,”'.join([str(key) for key in data.keys()]) |
| 183 | query = self.cur.execute(self.by_id_command % values) |
| 184 | for row in query: |
| 185 | id, = row |
| 186 | if id in data: |
| 187 | del data[id] |
| 188 | return data |
| 189 | |
| 190 | def log(self, message): |
| 191 | '''Unused''' |
| 192 | print >> open(self.logfile, 'a+'), message |
| 193 | |
| 194 | def use_server(self, id): |
| 195 | '''Set the server for this instance to use''' |
| 196 | if id not in self.servers: |
| 197 | raise ServerIdError, id |
| 198 | self.server = id |
| 199 | |
| 200 | def load_servers(self): |
| 201 | '''Load the servers from stored data''' |
| 202 | if 'servers' in self.settings: |
| 203 | servers = self.settings['servers'] |
| 204 | else: |
| 205 | servers = self.update_servers() |
| 206 | return servers |
| 207 | |
| 208 | # TODO: remove redundant parts and use the get_data() method instead |
| 209 | def update_servers(self): |
| 210 | '''Download and parse the server list from the api''' |
| 211 | if 'servers_fresh' in dir(self): |
| 212 | return self.servers |
| 213 | print 'Updating servers list…', |
| 214 | start = time() |
| 215 | url = self.api_url + self.servers_path |
| 216 | data = minidom.parse(urllib.urlopen(url)) |
| 217 | results = {} |
| 218 | for server in data.getElementsByTagName('server'): |
| 219 | attributes = dict(server.attributes.items()) |
| 220 | results[int(attributes.pop('id'))] = attributes |
| 221 | if not len(results): |
| 222 | print self.end('done', start) |
| 223 | self.error('danbooro seems to be down') |
| 224 | self.exit() |
| 225 | data.unlink() |
| 226 | self.servers_fresh = True |
| 227 | self.save_settings(servers=results) |
| 228 | print self.end('done', start) |
| 229 | return results |
| 230 | |
| 231 | def list_servers(self): |
| 232 | '''Print servers''' |
| 233 | print 'Listing servers…' |
| 234 | row = lambda id, host: '%s %s' % (str(id).rjust(2), host) |
| 235 | print row('ID', 'Host') |
| 236 | for id in self.servers: |
| 237 | print row(id, self.servers[id]['host']), '[default]' \ |
| 238 | if id == self.settings['default'] else '' |
| 239 | |
| 240 | def load_settings(self): |
| 241 | '''Connect to the persistent settings''' |
| 242 | return shelve.open(self.settings_filename) |
| 243 | |
| 244 | def save_settings(self, **kwargs): |
| 245 | '''Save and flush settings''' |
| 246 | for key in kwargs: |
| 247 | self.settings[key] = kwargs[key] |
| 248 | self.settings.sync() |
| 249 | |
| 250 | def get_data(self, url, elementname, keyname): |
| 251 | '''Fetch and parse data from the api (would be many lines longer if \ |
| 252 | this had to be actually spidered)''' |
| 253 | data = minidom.parse(urllib.urlopen(url)) |
| 254 | results = {} |
| 255 | for server in data.getElementsByTagName(elementname): |
| 256 | attributes = dict(server.attributes.items()) |
| 257 | results[int(attributes.pop(keyname))] = attributes |
| 258 | data.unlink() |
| 259 | return results |
| 260 | |
| 261 | def get_serverlist(self): |
| 262 | url, start = self.api_url + self.servers_path, time() |
| 263 | print 'Getting servers list…', |
| 264 | results = self.get_data(url, 'server', 'id') |
| 265 | print self.end(start) |
| 266 | return results |
| 267 | |
| 268 | def get_content_data(self, tags=None, limit=None, offset=None, hashes=None): |
| 269 | if tags and limit and offset: |
| 270 | url = self.api_url + self.posts_path % (tags, limit, offset) |
| 271 | count = limit - offset |
| 272 | elif hashes: |
| 273 | url = self.api_url + self.md5_path % ','.join(hashes) |
| 274 | count = len(hashes) |
| 275 | print 'Getting content data for %d %s…' % (count, case(count, 'file')), |
| 276 | start = time() |
| 277 | results = self.get_data(url, 'post', 'id') |
| 278 | print self.end('done', start) |
| 279 | if len(results) < count: |
| 280 | missing = count - len(results) |
| 281 | print '%d %s not found' % (missing, case(missing, 'file')) |
| 282 | return results |
| 283 | |
| 284 | def parse_tags(self, args): |
| 285 | '''Parse script arguments''' |
| 286 | tags = [urllib.quote(item.replace(' ', '_')).replace('%2B', '+') \ |
| 287 | for item in args] |
| 288 | tags = '+'.join(tags) |
| 289 | return tags |
| 290 | |
| 291 | def exit(self): |
| 292 | '''Say bye and report db changes''' |
| 293 | changes = self.db.total_changes |
| 294 | if changes: |
| 295 | print '%d %s to the local database in this session' % \ |
| 296 | (changes, case(changes, 'change')) |
| 297 | print 'Bye~!' |
| 298 | exit() |
| 299 | |
| 300 | def load_db(self): |
| 301 | '''Connect to the sqlite db''' |
| 302 | self.init_db_command ='''CREATE TABLE IF NOT EXISTS content \ |
| 303 | (id INTEGER PRIMARY KEY, md5 TEXT, tags TEXT, misc BLOB);''' |
| 304 | self.update_db_command ='''INSERT OR IGNORE into content \ |
| 305 | (id, md5, tags, misc) values (%d, “%s“, “%s“, “%s“);''' |
| 306 | self.by_md5_command ='''SELECT md5, id FROM content \ |
| 307 | WHERE md5 IN (“%s“);''' |
| 308 | self.by_id_command ='''SELECT id FROM content WHERE id IN (“%s“);''' |
| 309 | db = sqlite3.connect(self.db_filename) |
| 310 | db.text_factory = lambda text: unicode(text, 'utf-8', 'ignore') |
| 311 | cur = db.cursor() |
| 312 | cur.execute(self.init_db_command) |
| 313 | return db, cur |
| 314 | |
| 315 | def hash_in_filename(self, filename): |
| 316 | '''Try to avoid hashing the file''' |
| 317 | name, ext = os.path.splitext(os.path.basename(filename)) |
| 318 | results = re.search(self.namepattern, name) |
| 319 | return results.groups()[0] if results else None |
| 320 | |
| 321 | def filter_hashes(self, hashes): |
| 322 | '''Remove hashes that exist in the local database''' |
| 323 | values = '”,”'.join(hashes.values()) |
| 324 | query = self.cur.execute(self.by_md5_command % values) |
| 325 | for hash in query: |
| 326 | hash, id = hash |
| 327 | for key, value in hashes.copy().iteritems(): |
| 328 | if hash != value: continue |
| 329 | del hashes[key] |
| 330 | return hashes |
| 331 | |
| 332 | def get_hashes(self, names, source, filter=True): |
| 333 | '''Get hashes for files in a path''' |
| 334 | print 'Getting hashes for %d %s in %s…' % \ |
| 335 | (len(names), case(len(names), 'file'), source), |
| 336 | results, start = {}, time() |
| 337 | for item in names: |
| 338 | hash = self.hash_in_filename(item) \ |
| 339 | or md5(open(item, 'rb').read()).hexdigest() |
| 340 | results[item] = hash |
| 341 | if filter: |
| 342 | results = self.filter_hashes(results) |
| 343 | print self.end('done', start) |
| 344 | return results |
| 345 | |
| 346 | def catalogue_content(self, pathname): |
| 347 | '''Add files to the local database''' |
| 348 | print 'Starting to catalogue %s…' % (pathname,) |
| 349 | filenames = self.get_filenames(pathname) |
| 350 | hashes = self.get_hashes(filenames, pathname) |
| 351 | message = '%d of %d files already in local database' |
| 352 | print message % (len(filenames)-len(hashes), len(filenames)) |
| 353 | count, step = 0, 100 |
| 354 | for i in xrange(0, len(hashes), step): |
| 355 | data = self.get_content_data(hashes=hashes.values()[i:i+step]) |
| 356 | count += len(data) |
| 357 | if self['simulate']: continue |
| 358 | self.update_db(data) |
| 359 | self.db.commit() |
| 360 | print '%d %s added to database' % (count, cases(count, 'entry', 'entries')) |
| 361 | |
| 362 | def split_path(self, pathname): |
| 363 | '''Split the path in a tuple of three''' |
| 364 | folder, name = os.path.split(pathname) |
| 365 | name, ext = os.path.splitext(name) |
| 366 | return folder, name, ext |
| 367 | |
| 368 | def fix_filenames(self, pathname): |
| 369 | '''Rename files to id_hash''' |
| 370 | print 'Fixing filenames in %s…' % (pathname,) |
| 371 | filenames = self.get_filenames(pathname) |
| 372 | start, count = time(), 0 |
| 373 | hashes = self.get_hashes(filenames, pathname, filter=False) |
| 374 | values = '”,”'.join(hashes.values()) |
| 375 | query = self.cur.execute(self.by_md5_command % values) |
| 376 | query = dict(query.fetchall()) |
| 377 | for filename, hash in hashes.iteritems(): |
| 378 | if hash not in query.keys(): |
| 379 | continue |
| 380 | folder, oldname, ext = self.split_path(filename) |
| 381 | # Figure out the new name (id is padded with zeroes) |
| 382 | newname = '%07d_%s%s' % (query[hash], hash, ext) |
| 383 | newname = os.path.join(folder, newname) |
| 384 | if filename == newname: |
| 385 | continue |
| 386 | if os.path.exists(newname): |
| 387 | os.remove(filename) |
| 388 | else: |
| 389 | try: os.rename(filename, newname) |
| 390 | except WindowsError, e: |
| 391 | print e |
| 392 | count -= 1 |
| 393 | count += 1 |
| 394 | print '%d %s fixed' % (count, case(count, 'filename')) |
| 395 | |
| 396 | def expand_paths(self, source): |
| 397 | '''Does exactly what the name says''' |
| 398 | names = set() |
| 399 | for item in source.split(): |
| 400 | names.update(glob(item)) |
| 401 | return filter(os.path.isdir, names) |
| 402 | |
| 403 | def get_filenames(self, pathname): |
| 404 | '''Again, does just what the name says''' |
| 405 | names = glob(os.path.join(pathname, '*')) |
| 406 | return filter(os.path.isfile, names) |
| 407 | |
| 408 | def update_db(self, data): |
| 409 | '''Write data to the transaction (has to be committed to the db explicitly)''' |
| 410 | for key, value in data.iteritems(): |
| 411 | value['author'] = value['author'].encode('utf-8') |
| 412 | values = (key, value.pop('md5'), value.pop('tags'), pickle.dumps(value)) |
| 413 | try: |
| 414 | self.cur.execute(self.update_db_command % values) |
| 415 | except sqlite3.OperationalError, e: |
| 416 | print e |
| 417 | |
| 418 | |
| 419 | class Downloader(object): |
| 420 | '''Shows a progress bar for downloads. this is actually useful outside the |
| 421 | scope of danbooru.py''' |
| 422 | |
| 423 | before = .0 |
| 424 | history = [] |
| 425 | cycles = 0 |
| 426 | average = lambda self: sum(self.history) / (len(self.history) or 1) |
| 427 | |
| 428 | def __init__(self, width=55): |
| 429 | self.width = width |
| 430 | self.kibi = lambda bits: bits / 2 ** 10 |
| 431 | self.proc = lambda a, b: a / (b * 0.01) |
| 432 | |
| 433 | def retrieve(self, url, destination, callback=None): |
| 434 | self.size = 0 |
| 435 | xtime() |
| 436 | try: urllib.urlretrieve(url, destination, self.progress) |
| 437 | except KeyboardInterrupt: |
| 438 | print '\nDownload cancelled' |
| 439 | for i in range(5): |
| 440 | try: |
| 441 | os.remove(destination) |
| 442 | break |
| 443 | except: |
| 444 | sleep(.1) |
| 445 | else: raise |
| 446 | if callback: callback() |
| 447 | exit() |
| 448 | |
| 449 | return self.size |
| 450 | |
| 451 | def progress(self, blocks, blocksize, filesize): |
| 452 | self.cycles += 1 |
| 453 | bits = min(blocks*blocksize, filesize) |
| 454 | done = self.proc(bits, filesize) if bits != filesize else 100 |
| 455 | bar = self.bar(done) |
| 456 | if not self.cycles % 3 and bits != filesize: |
| 457 | now = xtime() |
| 458 | elapsed = now-self.before |
| 459 | if elapsed: |
| 460 | speed = self.kibi(blocksize * 3 / elapsed) |
| 461 | self.history.append(speed) |
| 462 | self.history = self.history[-4:] |
| 463 | self.before = now |
| 464 | average = round(sum(self.history[-4:]) / 4, 1) |
| 465 | self.size = self.kibi(bits) |
| 466 | print '\r[%s] %s KiB/s ' % (bar, str(average)), |
| 467 | |
| 468 | def bar(self, done): |
| 469 | span = self.width * done * 0.01 |
| 470 | offset = len(str(int(done))) - .99 |
| 471 | result = ('%d%%' % (done,)).center(self.width) |
| 472 | return result.replace(' ', '-', int(span - offset)) |
| 473 | |
| 474 | |
| 475 | def parse_options(): |
| 476 | '''Parse arguments passed to the script''' |
| 477 | help = { 'limit': 'set how many posts (not files) to get from the api \ |
| 478 | [default: %default]', |
| 479 | 'offset': 'set the position to start downloading from \ |
| 480 | [default: %default]', |
| 481 | 'server': 'which server to use (takes an index, see -L for a list of \ |
| 482 | available servers)', |
| 483 | 'refresh': 'allow retrieving posts older than the highest id \ |
| 484 | of the local files in the destination folder', |
| 485 | 'nodb': 'allow downloading posts that are already present \ |
| 486 | in the local database', |
| 487 | 'catalogue': 'add local files to the database \ |
| 488 | (queries the api with their hashes)', |
| 489 | 'fixnames': 'change filenames to |
| 490 | 'folder': 'override the download destination (default is same as tags)', |
| 491 | 'update': 'update the serverlist', |
| 492 | 'list': 'see a list of available servers', |
| 493 | 'set_default': 'set a default server', |
| 494 | 'rating': 'convenience shortcut to the rating: tag', |
| 495 | 'simulate': 'don\'t download files or add posts to the database', |
| 496 | } |
| 497 | usage = '%prog [-l NUM] [-o NUM] [-s NUM] [-r safe|questionable|explicit] \ |
| 498 | [-f PATH] [-i] [-n] [-c PATH] [-x PATH] [-u] [-L] [-d] |
| 499 | from optparse import OptionParser |
| 500 | parser = OptionParser(usage=usage, version='%s.%s' % (__version__, __build__), |
| 501 | description='A tool for retrieving content from danbooru.donmai.us') |
| 502 | parser.add_option('-l', '–limit', dest='limit', help=help['limit'], \ |
| 503 | metavar='NUM', default=1000, type='int') |
| 504 | parser.add_option('-o', '–offset', dest='offset', help=help['offset'], \ |
| 505 | metavar='NUM', default=0, type='int') |
| 506 | parser.add_option('-s', '–server', dest='server', help=help['server'], \ |
| 507 | metavar='NUM', default=None, type='int') |
| 508 | parser.add_option('-r', '–rating', dest='rating', help=help['rating'], \ |
| 509 | metavar='NAME', default=None, type='string') |
| 510 | parser.add_option('-f', '–folder', dest='folder', help=help['folder'], \ |
| 511 | metavar='PATH', default=None) |
| 512 | parser.add_option('-i', '–no-last-id', dest='refresh', \ |
| 513 | help=help['refresh'], action='store_true', default=False) |
| 514 | parser.add_option('-n', '–no-db', dest='nodb', help=help['nodb'], \ |
| 515 | action='store_true', default=False) |
| 516 | parser.add_option('-c', '–catalogue', dest='catalogue', \ |
| 517 | help=help['catalogue'], metavar='PATH', default=None) |
| 518 | parser.add_option('-x', '–fix', dest='fixnames', help=help['fixnames'], \ |
| 519 | metavar='PATH', default=None) |
| 520 | parser.add_option('-u', '–update', dest='update', help=help['update'], \ |
| 521 | action='store_true', default=False) |
| 522 | parser.add_option('-L', '–list', dest='list', help=help['list'], \ |
| 523 | action='store_true', default=False) |
| 524 | parser.add_option('-d', '–default', dest='set_default', \ |
| 525 | help=help['set_default'], metavar='ID', default=None, type='int') |
| 526 | parser.add_option('-e', '–simulate', dest='simulate', \ |
| 527 | help=help['simulate'], action='store_true', default=False) |
| 528 | options, args = parser.parse_args() |
| 529 | return options, args, parser |
| 530 | |
| 531 | |
| 532 | def main(): |
| 533 | '''Decide what to do based on the options returned by optparse''' |
| 534 | options, args, parser = parse_options() |
| 535 | robot = Robot(args, options.limit, options.offset, rating=options.rating, |
| 536 | refresh=options.refresh, nodb=options.nodb, simulate=options.simulate) |
| 537 | if options.rating: |
| 538 | values = ('safe', 'explicit', 'questionable') |
| 539 | if options.rating not in values: |
| 540 | parser.error('only %r are valid ratings' % (values,)) |
| 541 | else: |
| 542 | robot.rating = options.rating |
| 543 | if options.refresh: |
| 544 | print 'Using No Last ID mode…' |
| 545 | if options.folder: |
| 546 | robot.folder = options.folder |
| 547 | if options.catalogue: |
| 548 | for name in robot.expand_paths(options.catalogue): |
| 549 | robot.catalogue_content(name) |
| 550 | if options.fixnames: |
| 551 | for name in robot.expand_paths(options.fixnames): |
| 552 | robot.fix_filenames(name) |
| 553 | if options.update: |
| 554 | robot.update_servers() |
| 555 | if options.set_default: |
| 556 | server_id = options.set_default |
| 557 | robot.use_server(server_id) |
| 558 | robot.save_settings(default=server_id) |
| 559 | print 'Default server set to %d (%s)' % \ |
| 560 | (server_id, robot.servers[server_id]['host']) |
| 561 | if 'default' not in robot.settings: |
| 562 | server_id = robot.servers.keys()[0] |
| 563 | robot.use_server(server_id) |
| 564 | robot.save_settings(default=server_id) |
| 565 | if options.list: |
| 566 | robot.list_servers() |
| 567 | elif options.server: |
| 568 | robot.use_server(options.server) |
| 569 | else: |
| 570 | robot.use_server(robot.settings['default']) |
| 571 | if robot.tags: |
| 572 | if 'server' not in dir(robot): |
| 573 | robot.use_server(robot.settings['default']) |
| 574 | print 'Using server %d (%s)' % \ |
| 575 | (robot.server, robot.servers[robot.server]['host']) |
| 576 | robot.retrieve_content() |
| 577 | robot.exit() |
| 578 | |
| 579 | |
| 580 | if __name__ == '__main__': |
| 581 | main() |
13 Comments »
№ 1 Zealot 2009.08.11 @ 23:39
Hi Reinis Ivanovs !
I’m not english so excuse my mistakes.
First : thanks for this script,
I’m using this on Danbooru but since a few hours he don’t work !
When i’m using the script he tell me “0 post return” but when i go to Danbooru the tag work.
Can you fix this error ? Or it’s a error from Danbooru maybe ?
Thanks in advance !
PS : the script for moe.imouto work.
№ 2 Reinis Ivanovs 2009.08.17 @ 14:21
Sorry, it’s nothing I can fix. As it says above:
My suggestion is to simply wait and try again later.
№ 3 Pow 2009.08.20 @ 23:56
It was only limited for anonymous users and those with “member” status. “privileged” status and up doesn’t have this restriction.
№ 4 Zealot 2009.08.21 @ 2:49
The script is working again :D
Thanks for you work !
№ 5 Xor 2009.08.30 @ 0:31
Thanks for your script!
Afaik, Sankaku Complex is a danbooru site too? Do you think I could adapt your code for this one? Thanks again!
№ 6 Xor 2009.08.30 @ 1:13
done, works fine, thank you!
№ 7 Xor 2009.08.31 @ 21:45
I just cant’t make the database to work fine, I don’t understand how your get_data could work seeing the api.
This :md5_path = ‘find_posts?md5=%s’, for instance, I just cannot see it in the api documentation, and it doesn’t seem to work, thus get_data always return a 404…
№ 8 Reinis Ivanovs 2009.08.31 @ 23:34
I’m not immediately sure what your problem was, but the script is old, has accumulated some quick and dirty fixes, and danbooru’s API has changed
a bit. In any event, I replaced the URLs in moe.py and it seems to work with Sankaku Complex. Here you go.
№ 9 Xor 2009.09.01 @ 22:02
thanks but I finally managed to make the databasework correctly with the catalogue option. I tried to add EXIFs tags to the images with some library but it’s not working as of now.
well I modified a few things to make the catalogue work, such as changing md5_path = ‘find_posts?md5=%s’ which doesn’t seem to work anymore.
I also added a few functions to sort the images by rating, but I made it quite dirtily, as I’ve been using python since yesterday…
Your script is quite complex for a beginner, heh!
№ 10 Reinis Ivanovs 2009.09.04 @ 12:18
Here’s a version that someone sent me for Konachan.com.
№ 11 Malart 2009.10.27 @ 16:47
I just wanted to say that Danbooru seems to have implemented a mandatory free account to view images and the script doesn’t seem to work anymore… would there be a way around this? I’m not really a programmer… but would there be any way to be able to enter your account info in the script and have it log in to your account in order to download? Or am I being a noob? heh
№ 12 Reinis Ivanovs 2009.10.27 @ 16:50
Yeah, their API seems to provide a way to log in. I suppose I’ll update the script when I have more free time. Thanks for the heads up.
№ 13 Draker 2010.01.06 @ 4:49
i got it to work with logins.
file at
http://goldangel.tuxmania.org/danbooru.py
http://pastebin.com/f250e7483
RSS feed for comments on this post. TrackBack URI