| 1 | #!/usr/bin/env python3 |
| 2 | """ |
| 3 | Thingiverse bulk downloader |
| 4 | """ |
| 5 | |
| 6 | import re |
| 7 | import sys |
| 8 | import os |
| 9 | import argparse |
| 10 | import unicodedata |
| 11 | import requests |
| 12 | import logging |
| 13 | import multiprocessing |
| 14 | import enum |
| 15 | import datetime |
| 16 | from shutil import copyfile |
| 17 | from dataclasses import dataclass |
| 18 | import py7zr |
| 19 | import glob |
| 20 | import shutil |
| 21 | from io import StringIO |
| 22 | from html.parser import HTMLParser |
| 23 | |
| 24 | SEVENZIP_FILTERS = [{'id': py7zr.FILTER_LZMA2}] |
| 25 | |
| 26 | # I don't think this is exported by datetime |
| 27 | DEFAULT_DATETIME_FORMAT = '%Y-%m-%d %H:%M:%S' |
| 28 | # Windows cannot handle : in filenames |
| 29 | SAFE_DATETIME_FORMAT = '%Y-%m-%d %H.%M.%S' |
| 30 | |
| 31 | API_BASE = "https://api.thingiverse.com" |
| 32 | ACCESS_QP = "access_token={}" |
| 33 | PAGE_QP = "page={}" |
| 34 | API_USER_DESIGNS = API_BASE + "/users/{}/things/?" + ACCESS_QP |
| 35 | API_USER_COLLECTIONS = API_BASE + "/users/{}/collections/all?" + ACCESS_QP |
| 36 | |
| 37 | # Currently useless as it gives the same info as the matching element in API_USER_COLLECTIONS |
| 38 | API_COLLECTION = API_BASE + "/collections/{}/?" + ACCESS_QP |
| 39 | API_COLLECTION_THINGS = API_BASE + "/collections/{}/things/?" + ACCESS_QP |
| 40 | |
| 41 | API_THING_DETAILS = API_BASE + "/things/{}/?" + ACCESS_QP |
| 42 | API_THING_FILES = API_BASE + "/things/{}/files/?" + ACCESS_QP |
| 43 | API_THING_IMAGES = API_BASE + "/things/{}/images/?" + ACCESS_QP |
| 44 | API_THING_DOWNLOAD = "/download/?" + ACCESS_QP |
| 45 | |
| 46 | DOWNLOADER_COUNT = 1 |
| 47 | RETRY_COUNT = 3 |
| 48 | |
| 49 | MAX_PATH_LENGTH = 250 |
| 50 | |
| 51 | VERSION = "0.10.5" |
| 52 | |
| 53 | TIMESTAMP_FILE = "timestamp.txt" |
| 54 | |
| 55 | SESSION = requests.Session() |
| 56 | |
| 57 | |
| 58 | class MLStripper(HTMLParser): |
| 59 | """ Turns HTML markup into plain text |
| 60 | """ |
| 61 | |
| 62 | def error(self, message): |
| 63 | raise ValueError(message) |
| 64 | |
| 65 | def __init__(self): |
| 66 | super().__init__() |
| 67 | self.reset() |
| 68 | self.strict = False |
| 69 | self.convert_charrefs = True |
| 70 | self.text = StringIO() |
| 71 | |
| 72 | def handle_data(self, d): |
| 73 | self.text.write(d) |
| 74 | |
| 75 | def get_data(self): |
| 76 | return self.text.getvalue() |
| 77 | |
| 78 | @staticmethod |
| 79 | def strip_tags(html): |
| 80 | s = MLStripper() |
| 81 | s.feed(html) |
| 82 | return s.get_data() |
| 83 | |
| 84 | |
| 85 | @dataclass |
| 86 | class ThingLink: |
| 87 | thing_id: str |
| 88 | name: str |
| 89 | api_link: str |
| 90 | |
| 91 | |
| 92 | @dataclass |
| 93 | class FileLink: |
| 94 | name: str |
| 95 | last_update: datetime.datetime |
| 96 | link: str |
| 97 | |
| 98 | |
| 99 | @dataclass |
| 100 | class ImageLink: |
| 101 | name: str |
| 102 | link: str |
| 103 | |
| 104 | |
| 105 | class FileLinks: |
| 106 | def __init__(self, initial_links=None): |
| 107 | if initial_links is None: |
| 108 | initial_links = [] |
| 109 | self.links = [] |
| 110 | self.last_update = None |
| 111 | for link in initial_links: |
| 112 | self.append(link) |
| 113 | |
| 114 | def __iter__(self): |
| 115 | return iter(self.links) |
| 116 | |
| 117 | def __getitem__(self, item): |
| 118 | return self.links[item] |
| 119 | |
| 120 | def __len__(self): |
| 121 | return len(self.links) |
| 122 | |
| 123 | def append(self, link): |
| 124 | try: |
| 125 | self.last_update = max(self.last_update, link.last_update) |
| 126 | except TypeError: |
| 127 | self.last_update = link.last_update |
| 128 | self.links.append(link) |
| 129 | |
| 130 | |
| 131 | class State(enum.Enum): |
| 132 | OK = enum.auto() |
| 133 | FAILED = enum.auto() |
| 134 | ALREADY_DOWNLOADED = enum.auto() |
| 135 | |
| 136 | |
| 137 | def sanitise_url(url): |
| 138 | """ remove api keys from an url |
| 139 | """ |
| 140 | return re.sub(r'access_token=\w*', |
| 141 | 'access_token=***', |
| 142 | url) |
| 143 | |
| 144 | |
| 145 | def strip_time(date_obj): |
| 146 | """ Takes a datetime object and returns another with the time set to 00:00 |
| 147 | """ |
| 148 | return datetime.datetime.combine(date_obj.date(), datetime.time()) |
| 149 | |
| 150 | |
| 151 | def rename_unique(dir_name, target_dir_name): |
| 152 | """ Move a directory sideways to a new name, ensuring it is unique. |
| 153 | """ |
| 154 | target_dir = target_dir_name |
| 155 | inc = 0 |
| 156 | while os.path.exists(target_dir): |
| 157 | target_dir = "{}_{}".format(target_dir_name, inc) |
| 158 | inc += 1 |
| 159 | os.rename(dir_name, target_dir) |
| 160 | return target_dir |
| 161 | |
| 162 | |
| 163 | def fail_dir(dir_name): |
| 164 | """ When a download has failed, move it sideways. |
| 165 | """ |
| 166 | return rename_unique(dir_name, "{}_failed".format(dir_name)) |
| 167 | |
| 168 | |
| 169 | def truncate_name(file_name): |
| 170 | """ Ensure the filename is not too long for, well windows basically. |
| 171 | """ |
| 172 | path = os.path.abspath(file_name) |
| 173 | if len(path) <= MAX_PATH_LENGTH: |
| 174 | return path |
| 175 | base, extension = os.path.splitext(path) |
| 176 | inc = 0 |
| 177 | new_path = "{}_{}{}".format(base, inc, extension) |
| 178 | while os.path.exists(new_path): |
| 179 | new_path = "{}_{}{}".format(base, inc, extension) |
| 180 | inc += 1 |
| 181 | return new_path |
| 182 | |
| 183 | |
| 184 | def slugify(value): |
| 185 | """ |
| 186 | Normalise string, removes invalid for filename charactersr |
| 187 | and converts string to lowercase. |
| 188 | """ |
| 189 | logging.debug("Sluggyfying {}".format(value)) |
| 190 | value = unicodedata.normalize('NFKC', value).lower().strip() |
| 191 | value = re.sub(r'[\\/<>:?*|"]', '', value) |
| 192 | value = re.sub(r'\.*$', '', value) |
| 193 | return value.strip() |
| 194 | |
| 195 | |
| 196 | class Downloader(multiprocessing.Process): |
| 197 | """ |
| 198 | Class to handle downloading the things we have found to get. |
| 199 | """ |
| 200 | |
| 201 | def __init__(self, thing_queue, download_directory, compress, api_key): |
| 202 | multiprocessing.Process.__init__(self) |
| 203 | # TODO: add parameters |
| 204 | self.thing_queue = thing_queue |
| 205 | self.download_directory = download_directory |
| 206 | self.compress = compress |
| 207 | self.api_key = api_key |
| 208 | |
| 209 | def run(self): |
| 210 | """ actual download loop. |
| 211 | """ |
| 212 | while True: |
| 213 | thing_id = self.thing_queue.get() |
| 214 | if thing_id is None: |
| 215 | logging.info("Shutting download queue") |
| 216 | self.thing_queue.task_done() |
| 217 | break |
| 218 | thing = None |
| 219 | if isinstance(thing_id, str): |
| 220 | thing = Thing.from_thing_id(thing_id) |
| 221 | if isinstance(thing_id, ThingLink): |
| 222 | thing = Thing(thing_id) |
| 223 | if not thing: |
| 224 | logging.error("Don't know how to handle thing_id {}".format(thing_id)) |
| 225 | else: |
| 226 | logging.info("Handling id {}".format(thing_id)) |
| 227 | thing.download(self.download_directory, self.compress, self.api_key) |
| 228 | self.thing_queue.task_done() |
| 229 | return |
| 230 | |
| 231 | |
| 232 | class Grouping: |
| 233 | """ Holds details of a group of things for download |
| 234 | This is effectively (although not actually) an abstract class |
| 235 | - use Collection or Designs instead. |
| 236 | """ |
| 237 | |
| 238 | def __init__(self, quick, compress, api_key): |
| 239 | self.things = [] |
| 240 | self.total = 0 |
| 241 | self.req_id = None |
| 242 | self.last_page = 0 |
| 243 | self.per_page = None |
| 244 | # Should we stop downloading when we hit a known datestamp? |
| 245 | self.quick = quick |
| 246 | self.compress = compress |
| 247 | self.api_key = api_key |
| 248 | # These should be set by child classes. |
| 249 | self.url = None |
| 250 | self.download_dir = None |
| 251 | |
| 252 | @property |
| 253 | def get(self): |
| 254 | """ retrieve the things of the grouping. """ |
| 255 | if self.things: |
| 256 | # We've already done it. |
| 257 | return self.things |
| 258 | |
| 259 | # Check for initialisation: |
| 260 | if not self.url: |
| 261 | logging.error("No URL set - object not initialised properly?") |
| 262 | raise ValueError("No URL set - object not initialised properly?") |
| 263 | |
| 264 | # Get the internal details of the grouping. |
| 265 | logging.debug("Querying {}".format(sanitise_url(self.url))) |
| 266 | |
| 267 | # follow next links until all items are found |
| 268 | current_url = self.url |
| 269 | while current_url != None: |
| 270 | logging.info("requesting:{}".format(sanitise_url(current_url))) |
| 271 | current_req = SESSION.get(current_url) |
| 272 | current_url = current_req.links.get('next', {}).get('url') |
| 273 | if current_req.status_code != 200: |
| 274 | logging.error( |
| 275 | "Got unexpected code {} from url {}: {}".format(current_req.status_code, sanitise_url(self.url), |
| 276 | current_req.text)) |
| 277 | else: |
| 278 | current_json = current_req.json() |
| 279 | for thing in current_json: |
| 280 | logging.debug(thing) |
| 281 | self.things.append(ThingLink(thing['id'], thing['name'], thing['url'])) |
| 282 | logging.info("Found {} things.".format(len(self.things))) |
| 283 | return self.things |
| 284 | |
| 285 | def download(self): |
| 286 | """ Downloads all the files in a collection """ |
| 287 | if not self.things: |
| 288 | self.get |
| 289 | |
| 290 | if not self.download_dir: |
| 291 | raise ValueError( |
| 292 | "No download_dir set - invalidly initialised object?") |
| 293 | |
| 294 | try: |
| 295 | os.mkdir(self.download_dir) |
| 296 | except FileExistsError: |
| 297 | logging.info("Target directory {} already exists. Assuming a resume." |
| 298 | .format(self.download_dir)) |
| 299 | logging.info("Downloading {} thing(s).".format(self.total)) |
| 300 | for idx, thing in enumerate(self.things): |
| 301 | logging.info("Downloading thing {} - {}".format(idx, thing)) |
| 302 | return_code = Thing(thing).download(self.download_dir, self.compress, self.api_key) |
| 303 | if self.quick and return_code == State.ALREADY_DOWNLOADED: |
| 304 | logging.info("Caught up, stopping.") |
| 305 | return |
| 306 | |
| 307 | |
| 308 | class Collection(Grouping): |
| 309 | """ Holds details of a collection. """ |
| 310 | |
| 311 | def __init__(self, user, name, directory, quick, compress, api_key): |
| 312 | Grouping.__init__(self, quick, compress, api_key) |
| 313 | self.user = user |
| 314 | self.name = name |
| 315 | self.paginated = False |
| 316 | # need to figure out the the ID for the collection |
| 317 | collection_url = API_USER_COLLECTIONS.format(user, api_key) |
| 318 | try: |
| 319 | current_req = SESSION.get(collection_url) |
| 320 | except requests.exceptions.ConnectionError as error: |
| 321 | logging.error("Unable to connect for collections for user {}: {}".format( |
| 322 | self.user, error)) |
| 323 | return |
| 324 | if current_req.status_code != 200: |
| 325 | logging.error( |
| 326 | "Got unexpected code {} from url {}: {}".format(current_req.status_code, sanitise_url(collection_url), |
| 327 | current_req.text)) |
| 328 | return |
| 329 | collection_list = current_req.json() |
| 330 | try: |
| 331 | # case insensitive to retain parity with previous behaviour |
| 332 | collection = [x for x in collection_list if x['name'].casefold() == name.casefold()][0] |
| 333 | except IndexError: |
| 334 | logging.error("Unable to find collection {} for user {}".format(name, user)) |
| 335 | return |
| 336 | self.collection_id = collection['id'] |
| 337 | self.url = API_COLLECTION_THINGS.format(self.collection_id, api_key) |
| 338 | |
| 339 | self.download_dir = os.path.join(directory, |
| 340 | "{}-{}".format(slugify(self.user), slugify(self.name))) |
| 341 | |
| 342 | |
| 343 | class Designs(Grouping): |
| 344 | """ Holds details of all of a users' designs. """ |
| 345 | |
| 346 | def __init__(self, user, directory, quick, compress, api_key): |
| 347 | Grouping.__init__(self, quick, compress, api_key) |
| 348 | self.user = user |
| 349 | self.url = API_USER_DESIGNS.format(user, api_key) |
| 350 | self.download_dir = os.path.join( |
| 351 | directory, "{} designs".format(slugify(self.user))) |
| 352 | |
| 353 | |
| 354 | class Thing: |
| 355 | """ An individual design on thingiverse. """ |
| 356 | |
| 357 | def __init__(self, thing_link): |
| 358 | self.thing_id = thing_link.thing_id |
| 359 | self.name = thing_link.name |
| 360 | self.last_time = None |
| 361 | self._parsed = False |
| 362 | self._needs_download = True |
| 363 | self.text = None |
| 364 | self.download_dir = None |
| 365 | self.time_stamp = None |
| 366 | self._file_links = FileLinks() |
| 367 | self._image_links = [] |
| 368 | |
| 369 | @classmethod |
| 370 | def from_thing_id(cls, thing_id): |
| 371 | """ |
| 372 | Factory method that looks up a thing by ID and creates a Thing object for it |
| 373 | :param thing_id: to look up |
| 374 | :return: Thing or None |
| 375 | """ |
| 376 | return Thing(ThingLink(thing_id, "", "")) |
| 377 | |
| 378 | def _parse(self, base_dir, api_key): |
| 379 | """ Work out what, if anything needs to be done. """ |
| 380 | if self._parsed: |
| 381 | return |
| 382 | |
| 383 | # First get the broad details |
| 384 | url = API_THING_DETAILS.format(self.thing_id, api_key) |
| 385 | try: |
| 386 | current_req = SESSION.get(url) |
| 387 | except requests.exceptions.ConnectionError as error: |
| 388 | logging.error("Unable to connect for thing {}: {}".format( |
| 389 | self.thing_id, error)) |
| 390 | return |
| 391 | # Check for DMCA |
| 392 | if current_req.status_code == 403: |
| 393 | logging.error("Access to thing {} is forbidden".format(self.thing_id)) |
| 394 | return |
| 395 | if current_req.status_code != 200: |
| 396 | logging.error("Got unexpected code {} from url {}: {}".format(current_req.status_code, sanitise_url(url), |
| 397 | current_req.text)) |
| 398 | return |
| 399 | |
| 400 | thing_json = current_req.json() |
| 401 | try: |
| 402 | self._license = thing_json['license'] |
| 403 | except KeyError: |
| 404 | logging.warning("No license found for thing {}?".format(self.thing_id)) |
| 405 | |
| 406 | details = None |
| 407 | try: |
| 408 | details = thing_json['details'] |
| 409 | except KeyError: |
| 410 | logging.warning("No description found for thing {}?".format(self.thing_id)) |
| 411 | |
| 412 | if details: |
| 413 | try: |
| 414 | self._details = MLStripper.strip_tags(details) |
| 415 | except ValueError as e: |
| 416 | logging.warning("Unable to strip HTML from readme: {}".format(e)) |
| 417 | self._details = details |
| 418 | |
| 419 | if not self.name: |
| 420 | # Probably generated with factory method. |
| 421 | try: |
| 422 | self.name = thing_json['name'] |
| 423 | except KeyError: |
| 424 | logging.warning("No name found for thing {}?".format(self.thing_id)) |
| 425 | self.name = self.thing_id |
| 426 | |
| 427 | # Now get the file details |
| 428 | file_url = API_THING_FILES.format(self.thing_id, api_key) |
| 429 | |
| 430 | try: |
| 431 | current_req = SESSION.get(file_url) |
| 432 | except requests.exceptions.ConnectionError as error: |
| 433 | logging.error("Unable to connect for thing {}: {}".format( |
| 434 | self.thing_id, error)) |
| 435 | return |
| 436 | |
| 437 | if current_req.status_code != 200: |
| 438 | logging.error("Unexpected status code {} for {}: {}".format(current_req.status_code, sanitise_url(file_url), |
| 439 | current_req.text)) |
| 440 | return |
| 441 | |
| 442 | link_list = current_req.json() |
| 443 | |
| 444 | if not link_list: |
| 445 | logging.error("No files found for thing {} - probably thingiverse being broken, try again later".format( |
| 446 | self.thing_id)) |
| 447 | |
| 448 | for link in link_list: |
| 449 | logging.debug("Parsing link: {}".format(sanitise_url(link['url']))) |
| 450 | try: |
| 451 | datestamp = datetime.datetime.strptime(link['date'], DEFAULT_DATETIME_FORMAT) |
| 452 | self._file_links.append( |
| 453 | FileLink(link['name'], datestamp, link['url'] + API_THING_DOWNLOAD.format(api_key))) |
| 454 | except ValueError: |
| 455 | logging.error(link['date']) |
| 456 | |
| 457 | # Finally get the image links |
| 458 | image_url = API_THING_IMAGES.format(self.thing_id, api_key) |
| 459 | |
| 460 | try: |
| 461 | current_req = SESSION.get(image_url) |
| 462 | except requests.exceptions.ConnectionError as error: |
| 463 | logging.error("Unable to connect for thing {}: {}".format( |
| 464 | self.thing_id, error)) |
| 465 | return |
| 466 | |
| 467 | if current_req.status_code != 200: |
| 468 | logging.error( |
| 469 | "Unexpected status code {} for {}: {}".format(current_req.status_code, sanitise_url(image_url), |
| 470 | current_req.text)) |
| 471 | return |
| 472 | |
| 473 | image_list = current_req.json() |
| 474 | |
| 475 | if not image_list: |
| 476 | logging.warning( |
| 477 | "No images found for thing {} - probably thingiverse being iffy as this seems unlikely".format( |
| 478 | self.thing_id)) |
| 479 | |
| 480 | for image in image_list: |
| 481 | logging.debug("parsing image: {}".format(image)) |
| 482 | name = None |
| 483 | try: |
| 484 | name = slugify(image['name']) |
| 485 | # TODO: fallback to other types |
| 486 | url = [x for x in image['sizes'] if x['type'] == 'display' and x['size'] == 'large'][0]['url'] |
| 487 | except KeyError: |
| 488 | logging.warning("Missing image for {}".format(name)) |
| 489 | self._image_links.append(ImageLink(name, url)) |
| 490 | |
| 491 | self.slug = "{} - {}".format(self.thing_id, slugify(self.name)) |
| 492 | self.download_dir = os.path.join(base_dir, self.slug) |
| 493 | |
| 494 | self._handle_old_directory(base_dir) |
| 495 | |
| 496 | logging.debug("Parsing {} ({})".format(self.thing_id, self.name)) |
| 497 | latest, self.last_time = self._find_last_download(base_dir) |
| 498 | |
| 499 | if not latest: |
| 500 | # Not yet downloaded |
| 501 | self._parsed = True |
| 502 | return |
| 503 | |
| 504 | logging.info("last downloaded version: {}".format(self.last_time)) |
| 505 | |
| 506 | # OK, so we have a timestamp, lets see if there is anything new to get |
| 507 | # First off, are we comparing an old download that threw away the timestamp? |
| 508 | ignore_time = self.last_time == strip_time(self.last_time) |
| 509 | try: |
| 510 | # TODO: Allow for comparison at the exact time |
| 511 | files_last_update = self._file_links.last_update |
| 512 | if ignore_time: |
| 513 | logging.info("Dropping time from comparison stamp as old-style download dir") |
| 514 | files_last_update = strip_time(files_last_update) |
| 515 | |
| 516 | if files_last_update > self.last_time: |
| 517 | logging.info( |
| 518 | "Found new/updated files {}".format(self._file_links.last_update)) |
| 519 | self._needs_download = True |
| 520 | self._parsed = True |
| 521 | return |
| 522 | except TypeError: |
| 523 | logging.warning("No files found for {}.".format(self.thing_id)) |
| 524 | |
| 525 | # Got here, so nope, no new files. |
| 526 | self._needs_download = False |
| 527 | self._parsed = True |
| 528 | |
| 529 | def _handle_old_directory(self, base_dir): |
| 530 | """ Deal with any old directories from previous versions of the code. |
| 531 | """ |
| 532 | old_dir = os.path.join(base_dir, slugify(self.name)) |
| 533 | if os.path.exists(old_dir): |
| 534 | logging.warning("Found old style download_dir. Moving.") |
| 535 | rename_unique(old_dir, self.download_dir) |
| 536 | |
| 537 | def _handle_outdated_directory(self): |
| 538 | """ Move the current download directory sideways if the thing has changed. |
| 539 | """ |
| 540 | if not os.path.exists(self.download_dir): |
| 541 | # No old directory to move. |
| 542 | return None |
| 543 | timestamp_file = os.path.join(self.download_dir, TIMESTAMP_FILE) |
| 544 | if not os.path.exists(timestamp_file): |
| 545 | # Old form of download directory |
| 546 | target_dir_name = "{} - old".format(self.download_dir) |
| 547 | else: |
| 548 | target_dir_name = "{} - {}".format(self.download_dir, self.last_time.strftime(SAFE_DATETIME_FORMAT)) |
| 549 | return rename_unique(self.download_dir, target_dir_name) |
| 550 | |
| 551 | def _find_last_download(self, base_dir): |
| 552 | """ Look for the most recent previous download (if any) of the thing. |
| 553 | """ |
| 554 | logging.info("Looking for old things") |
| 555 | |
| 556 | # First the DL directory itself. |
| 557 | timestamp_file = os.path.join(self.download_dir, TIMESTAMP_FILE) |
| 558 | |
| 559 | latest = None |
| 560 | latest_time = None |
| 561 | |
| 562 | try: |
| 563 | logging.debug("Checking for existing download in normal place.") |
| 564 | with open(timestamp_file) as ts_fh: |
| 565 | timestamp_text = ts_fh.read().strip() |
| 566 | latest_time = datetime.datetime.strptime(timestamp_text, DEFAULT_DATETIME_FORMAT) |
| 567 | latest = self.download_dir |
| 568 | except FileNotFoundError: |
| 569 | # No existing download directory. huh. |
| 570 | pass |
| 571 | except TypeError: |
| 572 | logging.warning("Invalid timestamp file found in {}".format(self.download_dir)) |
| 573 | |
| 574 | # TODO: Maybe look for old download directories. |
| 575 | |
| 576 | # Now look for 7z files |
| 577 | candidates = glob.glob(os.path.join(base_dir, "{}*.7z".format(self.thing_id))) |
| 578 | # +3 to allow for ' - ' |
| 579 | leading_length = len(self.slug) + 3 |
| 580 | for path in candidates: |
| 581 | candidate = os.path.basename(path) |
| 582 | try: |
| 583 | logging.debug("Examining '{}' - '{}'".format(candidate, candidate[leading_length:-3])) |
| 584 | candidate_time = datetime.datetime.strptime(candidate[leading_length:-3], SAFE_DATETIME_FORMAT) |
| 585 | except ValueError: |
| 586 | logging.warning("There was an error finding the date in {}. Ignoring.".format(candidate)) |
| 587 | continue |
| 588 | try: |
| 589 | if candidate_time > latest_time: |
| 590 | latest_time = candidate_time |
| 591 | latest = candidate |
| 592 | except TypeError: |
| 593 | latest_time = candidate_time |
| 594 | latest = candidate |
| 595 | logging.info("Found last old thing: {} / {}".format(latest, latest_time)) |
| 596 | return latest, latest_time |
| 597 | |
| 598 | def download(self, base_dir, compress, api_key): |
| 599 | """ Download all files for a given thing. |
| 600 | Returns True iff the thing is now downloaded (not iff it downloads the thing!) |
| 601 | """ |
| 602 | if not self._parsed: |
| 603 | self._parse(base_dir, api_key) |
| 604 | |
| 605 | if not self._parsed: |
| 606 | logging.error( |
| 607 | "Unable to parse {} - aborting download".format(self.thing_id)) |
| 608 | return State.FAILED |
| 609 | |
| 610 | if not self._needs_download: |
| 611 | logging.info("{} - {} already downloaded - skipping.".format(self.thing_id, self.name)) |
| 612 | return State.ALREADY_DOWNLOADED |
| 613 | |
| 614 | if not self._file_links: |
| 615 | logging.error( |
| 616 | "{} - {} appears to have no files. Thingiverse acting up again?".format(self.thing_id, self.name)) |
| 617 | return State.FAILED |
| 618 | |
| 619 | # Have we already downloaded some things? |
| 620 | renamed_dir = self._handle_outdated_directory() |
| 621 | |
| 622 | # Get the list of files to download |
| 623 | |
| 624 | new_file_links = [] |
| 625 | old_file_links = [] |
| 626 | self.time_stamp = None |
| 627 | |
| 628 | if not self.last_time: |
| 629 | # If we don't have anything to copy from, then it is all new. |
| 630 | logging.debug("No last time, downloading all files") |
| 631 | new_file_links = self._file_links |
| 632 | self.time_stamp = new_file_links[0].last_update |
| 633 | |
| 634 | for file_link in new_file_links: |
| 635 | self.time_stamp = max(self.time_stamp, file_link.last_update) |
| 636 | logging.debug("New timestamp will be {}".format(self.time_stamp)) |
| 637 | else: |
| 638 | self.time_stamp = self.last_time |
| 639 | for file_link in self._file_links: |
| 640 | if file_link.last_update > self.last_time: |
| 641 | new_file_links.append(file_link) |
| 642 | self.time_stamp = max(self.time_stamp, file_link.last_update) |
| 643 | else: |
| 644 | old_file_links.append(file_link) |
| 645 | |
| 646 | logging.debug("new timestamp {}".format(self.time_stamp)) |
| 647 | |
| 648 | # OK. Time to get to work. |
| 649 | logging.debug("Generating download_dir") |
| 650 | os.mkdir(self.download_dir) |
| 651 | filelist_file = os.path.join(self.download_dir, "filelist.txt") |
| 652 | with open(filelist_file, 'w', encoding="utf-8") as fl_handle: |
| 653 | for fl in self._file_links: |
| 654 | fl_handle.write("{},{},{}\n".format(fl.link, fl.name, fl.last_update)) |
| 655 | |
| 656 | # First grab the cached files (if any) |
| 657 | logging.info("Copying {} unchanged files.".format(len(old_file_links))) |
| 658 | if renamed_dir: |
| 659 | for file_link in old_file_links: |
| 660 | try: |
| 661 | old_file = os.path.join(renamed_dir, file_link.name) |
| 662 | new_file = truncate_name(os.path.join(self.download_dir, file_link.name)) |
| 663 | logging.debug("Copying {} to {}".format(old_file, new_file)) |
| 664 | copyfile(old_file, new_file) |
| 665 | except FileNotFoundError: |
| 666 | logging.warning( |
| 667 | "Unable to find {} in old archive, redownloading".format(file_link.name)) |
| 668 | new_file_links.append(file_link) |
| 669 | except TypeError: |
| 670 | # Not altogether sure how this could occur, possibly with some combination of the old file types |
| 671 | logging.warning( |
| 672 | "Typeerror looking for {} in {}".format(file_link.name, renamed_dir)) |
| 673 | new_file_links.append(file_link) |
| 674 | |
| 675 | # Now download the new ones |
| 676 | logging.info("Downloading {} new files of {}".format( |
| 677 | len(new_file_links), len(self._file_links))) |
| 678 | try: |
| 679 | for file_link in new_file_links: |
| 680 | file_name = truncate_name(os.path.join(self.download_dir, file_link.name)) |
| 681 | logging.debug("Downloading {} from {} to {}".format( |
| 682 | file_link.name, file_link.link, file_name)) |
| 683 | data_req = SESSION.get(file_link.link) |
| 684 | if data_req.status_code != 200: |
| 685 | logging.error("Unexpected status code {} for {}".format(data_req.status_code, |
| 686 | sanitise_url(file_link.link))) |
| 687 | logging.debug("Unexpected status code {} for {}: {}".format(data_req.status_code, |
| 688 | sanitise_url(file_link.link), |
| 689 | data_req.text)) |
| 690 | fail_dir(self.download_dir) |
| 691 | return State.FAILED |
| 692 | |
| 693 | with open(file_name, 'wb') as handle: |
| 694 | handle.write(data_req.content) |
| 695 | except Exception as exception: |
| 696 | logging.error("Failed to download {} - {}".format(file_link.name, exception)) |
| 697 | fail_dir(self.download_dir) |
| 698 | return State.FAILED |
| 699 | |
| 700 | # People like images. |
| 701 | image_dir = os.path.join(self.download_dir, 'images') |
| 702 | logging.info("Downloading {} images.".format(len(self._image_links))) |
| 703 | try: |
| 704 | os.mkdir(image_dir) |
| 705 | for imagelink in self._image_links: |
| 706 | filename = os.path.join(image_dir, imagelink.name) |
| 707 | image_req = SESSION.get(imagelink.link) |
| 708 | if image_req.status_code != 200: |
| 709 | logging.error("Unexpected status code {} for {}: {}".format(image_req.status_code, |
| 710 | sanitise_url(imagelink.link), |
| 711 | image_req.text)) |
| 712 | fail_dir(self.download_dir) |
| 713 | return State.FAILED |
| 714 | with open(truncate_name(filename), 'wb') as handle: |
| 715 | handle.write(image_req.content) |
| 716 | except Exception as exception: |
| 717 | logging.error("Failed to download {} - {}".format(imagelink.name, exception)) |
| 718 | fail_dir(self.download_dir) |
| 719 | return State.FAILED |
| 720 | |
| 721 | # Best get some licenses |
| 722 | logging.info("writing license file") |
| 723 | try: |
| 724 | if self._license: |
| 725 | with open(truncate_name(os.path.join(self.download_dir, 'license.txt')), 'w', |
| 726 | encoding="utf-8") as license_handle: |
| 727 | license_handle.write("{}\n".format(self._license)) |
| 728 | except IOError as exception: |
| 729 | logging.warning("Failed to write license! {}".format(exception)) |
| 730 | |
| 731 | logging.info("writing readme") |
| 732 | try: |
| 733 | if self._details: |
| 734 | with open(truncate_name(os.path.join(self.download_dir, 'readme.txt')), 'w', |
| 735 | encoding="utf-8") as readme_handle: |
| 736 | readme_handle.write("{}\n".format(self._details)) |
| 737 | except IOError as exception: |
| 738 | logging.warning("Failed to write readme! {}".format(exception)) |
| 739 | |
| 740 | try: |
| 741 | # Now write the timestamp |
| 742 | with open(os.path.join(self.download_dir, TIMESTAMP_FILE), 'w', encoding="utf-8") as timestamp_handle: |
| 743 | timestamp_handle.write(self.time_stamp.__str__()) |
| 744 | except Exception as exception: |
| 745 | logging.error("Failed to write timestamp file - {}".format(exception)) |
| 746 | fail_dir(self.download_dir) |
| 747 | return State.FAILED |
| 748 | self._needs_download = False |
| 749 | logging.debug("Download of {} finished".format(self.name)) |
| 750 | if not compress: |
| 751 | return State.OK |
| 752 | |
| 753 | thing_dir = "{} - {} - {}".format(self.thing_id, |
| 754 | slugify(self.name), |
| 755 | self.time_stamp.strftime(SAFE_DATETIME_FORMAT)) |
| 756 | file_name = os.path.join(base_dir, |
| 757 | "{}.7z".format(thing_dir)) |
| 758 | logging.debug("Compressing {} to {}".format( |
| 759 | self.name, |
| 760 | file_name)) |
| 761 | with py7zr.SevenZipFile(file_name, 'w', filters=SEVENZIP_FILTERS) as archive: |
| 762 | archive.writeall(self.download_dir, thing_dir) |
| 763 | logging.debug("Compression of {} finished.".format(self.name)) |
| 764 | shutil.rmtree(self.download_dir) |
| 765 | logging.debug("Removed temporary download dir of {}.".format(self.name)) |
| 766 | return State.OK |
| 767 | |
| 768 | |
| 769 | def do_batch(batch_file, download_dir, quick, compress, api_key): |
| 770 | """ Read a file in line by line, parsing each as a set of calls to this script.""" |
| 771 | with open(batch_file) as handle: |
| 772 | for line in handle: |
| 773 | line = line.strip() |
| 774 | if not line: |
| 775 | # Skip empty lines |
| 776 | continue |
| 777 | logging.info("Handling instruction {}".format(line)) |
| 778 | command_arr = line.split() |
| 779 | if command_arr[0] == "thing": |
| 780 | logging.debug( |
| 781 | "Handling batch thing instruction: {}".format(line)) |
| 782 | Thing.from_thing_id(command_arr[1]).download(download_dir, compress, api_key) |
| 783 | continue |
| 784 | if command_arr[0] == "collection": |
| 785 | logging.debug( |
| 786 | "Handling batch collection instruction: {}".format(line)) |
| 787 | Collection(command_arr[1], command_arr[2], |
| 788 | download_dir, quick, compress, api_key).download() |
| 789 | continue |
| 790 | if command_arr[0] == "user": |
| 791 | logging.debug( |
| 792 | "Handling batch collection instruction: {}".format(line)) |
| 793 | Designs(command_arr[1], download_dir, quick, compress, api_key).download() |
| 794 | continue |
| 795 | logging.warning("Unable to parse current instruction. Skipping.") |
| 796 | |
| 797 | |
| 798 | def main(): |
| 799 | """ Entry point for script being run as a command. """ |
| 800 | parser = argparse.ArgumentParser() |
| 801 | parser.add_argument("-l", "--log-level", choices=[ |
| 802 | 'debug', 'info', 'warning'], default='info', help="level of logging desired") |
| 803 | parser.add_argument("-d", "--directory", |
| 804 | help="Target directory to download into") |
| 805 | parser.add_argument("-f", "--log-file", |
| 806 | help="Place to log debug information to") |
| 807 | parser.add_argument("-q", "--quick", action="store_true", |
| 808 | help="Assume date ordering on posts") |
| 809 | parser.add_argument("-c", "--compress", action="store_true", |
| 810 | help="Compress files") |
| 811 | parser.add_argument("-a", "--api-key", |
| 812 | help="API key for thingiverse") |
| 813 | |
| 814 | subparsers = parser.add_subparsers( |
| 815 | help="Type of thing to download", dest="subcommand") |
| 816 | collection_parser = subparsers.add_parser( |
| 817 | 'collection', help="Download one or more entire collection(s)") |
| 818 | collection_parser.add_argument( |
| 819 | "owner", help="The owner of the collection(s) to get") |
| 820 | collection_parser.add_argument( |
| 821 | "collections", nargs="+", help="Space seperated list of the name(s) of collection to get") |
| 822 | thing_parser = subparsers.add_parser( |
| 823 | 'thing', help="Download a single thing.") |
| 824 | thing_parser.add_argument( |
| 825 | "things", nargs="*", help="Space seperated list of thing ID(s) to download") |
| 826 | user_parser = subparsers.add_parser( |
| 827 | "user", help="Download all things by one or more users") |
| 828 | user_parser.add_argument( |
| 829 | "users", nargs="+", help="A space seperated list of the user(s) to get the designs of") |
| 830 | batch_parser = subparsers.add_parser( |
| 831 | "batch", help="Perform multiple actions written in a text file") |
| 832 | batch_parser.add_argument( |
| 833 | "batch_file", help="The name of the file to read.") |
| 834 | subparsers.add_parser("version", help="Show the current version") |
| 835 | |
| 836 | args = parser.parse_args() |
| 837 | if not args.subcommand: |
| 838 | parser.print_help() |
| 839 | sys.exit(1) |
| 840 | if not args.directory: |
| 841 | args.directory = os.getcwd() |
| 842 | |
| 843 | logger = logging.getLogger() |
| 844 | formatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s') |
| 845 | logger.setLevel(logging.DEBUG) |
| 846 | console_handler = logging.StreamHandler() |
| 847 | console_handler.setLevel(args.log_level.upper()) |
| 848 | |
| 849 | if args.api_key: |
| 850 | api_key = args.api_key |
| 851 | else: |
| 852 | try: |
| 853 | with open("api.key") as fh: |
| 854 | api_key = fh.read().strip() |
| 855 | except Exception as e: |
| 856 | logging.error("Either specify the api-key on the command line or in a file called 'api.key'") |
| 857 | logging.error("Exception: {}".format(e)) |
| 858 | return |
| 859 | |
| 860 | logger.addHandler(console_handler) |
| 861 | if args.log_file: |
| 862 | file_handler = logging.FileHandler(args.log_file) |
| 863 | file_handler.setLevel(logging.DEBUG) |
| 864 | file_handler.setFormatter(formatter) |
| 865 | logger.addHandler(file_handler) |
| 866 | |
| 867 | # Start downloader |
| 868 | thing_queue = multiprocessing.JoinableQueue() |
| 869 | logging.debug("starting {} downloader(s)".format(DOWNLOADER_COUNT)) |
| 870 | downloaders = [Downloader(thing_queue, args.directory, args.compress, api_key) for _ in range(DOWNLOADER_COUNT)] |
| 871 | for downloader in downloaders: |
| 872 | downloader.start() |
| 873 | |
| 874 | if args.subcommand.startswith("collection"): |
| 875 | for collection in args.collections: |
| 876 | Collection(args.owner, collection, args.directory, args.quick, args.compress, api_key).download() |
| 877 | if args.subcommand == "thing": |
| 878 | for thing in args.things: |
| 879 | thing_queue.put(thing) |
| 880 | if args.subcommand == "user": |
| 881 | for user in args.users: |
| 882 | Designs(user, args.directory, args.quick, args.compress, api_key).download() |
| 883 | if args.subcommand == "version": |
| 884 | print("thingy_grabber.py version {}".format(VERSION)) |
| 885 | if args.subcommand == "batch": |
| 886 | do_batch(args.batch_file, args.directory, args.quick, args.compress, api_key) |
| 887 | |
| 888 | # Stop the downloader processes |
| 889 | for _ in downloaders: |
| 890 | thing_queue.put(None) |
| 891 | |
| 892 | |
| 893 | if __name__ == "__main__": |
| 894 | multiprocessing.freeze_support() |
| 895 | main() |