| 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 bs4 import BeautifulSoup |
| 18 | from dataclasses import dataclass |
| 19 | import selenium |
| 20 | from selenium import webdriver |
| 21 | from selenium.webdriver.common.by import By |
| 22 | from selenium.webdriver.support.ui import WebDriverWait |
| 23 | from selenium.webdriver.support import expected_conditions as EC |
| 24 | from selenium.webdriver.firefox.options import Options |
| 25 | import atexit |
| 26 | import py7zr |
| 27 | |
| 28 | SEVENZIP_FILTERS = [{'id': py7zr.FILTER_LZMA2}] |
| 29 | |
| 30 | URL_BASE = "https://www.thingiverse.com" |
| 31 | URL_COLLECTION = URL_BASE + "/ajax/thingcollection/list_collected_things" |
| 32 | USER_COLLECTION = URL_BASE + "/ajax/user/designs" |
| 33 | |
| 34 | ID_REGEX = re.compile(r'"id":(\d*),') |
| 35 | TOTAL_REGEX = re.compile(r'"total":(\d*),') |
| 36 | LAST_PAGE_REGEX = re.compile(r'"last_page":(\d*),') |
| 37 | # This appears to be fixed at 12, but if it changes would screw the rest up. |
| 38 | PER_PAGE_REGEX = re.compile(r'"per_page":(\d*),') |
| 39 | NO_WHITESPACE_REGEX = re.compile(r'[-\s]+') |
| 40 | |
| 41 | DOWNLOADER_COUNT = 1 |
| 42 | RETRY_COUNT = 3 |
| 43 | |
| 44 | MAX_PATH_LENGTH = 250 |
| 45 | |
| 46 | VERSION = "0.9.0" |
| 47 | |
| 48 | |
| 49 | #BROWSER = webdriver.PhantomJS('./phantomjs') |
| 50 | options = Options() |
| 51 | options.add_argument("--headless") |
| 52 | BROWSER = webdriver.Firefox(options=options) |
| 53 | |
| 54 | BROWSER.set_window_size(1980, 1080) |
| 55 | |
| 56 | |
| 57 | @dataclass |
| 58 | class FileLink: |
| 59 | name: str |
| 60 | last_update: datetime.datetime |
| 61 | link: str |
| 62 | |
| 63 | class FileLinks: |
| 64 | def __init__(self, initial_links=[]): |
| 65 | self.links = [] |
| 66 | self.last_update = None |
| 67 | for link in initial_links: |
| 68 | self.append(link) |
| 69 | |
| 70 | def __iter__(self): |
| 71 | return iter(self.links) |
| 72 | |
| 73 | def __getitem__(self, item): |
| 74 | return self.links[item] |
| 75 | |
| 76 | def __len__(self): |
| 77 | return len(self.links) |
| 78 | |
| 79 | def append(self, link): |
| 80 | try: |
| 81 | self.last_update = max(self.last_update, link.last_update) |
| 82 | except TypeError: |
| 83 | self.last_update = link.last_update |
| 84 | self.links.append(link) |
| 85 | |
| 86 | |
| 87 | class State(enum.Enum): |
| 88 | OK = enum.auto() |
| 89 | FAILED = enum.auto() |
| 90 | ALREADY_DOWNLOADED = enum.auto() |
| 91 | |
| 92 | |
| 93 | def fail_dir(dir_name): |
| 94 | """ When a download has failed, move it sideways. |
| 95 | """ |
| 96 | target_dir = "{}_failed".format(dir_name) |
| 97 | inc = 0 |
| 98 | while os.path.exists(target_dir): |
| 99 | target_dir = "{}_failed_{}".format(dir_name, inc) |
| 100 | inc += 1 |
| 101 | os.rename(dir_name, target_dir) |
| 102 | |
| 103 | |
| 104 | def truncate_name(file_name): |
| 105 | """ Ensure the filename is not too long for, well windows basically. |
| 106 | """ |
| 107 | path = os.path.abspath(file_name) |
| 108 | if len(path) <= MAX_PATH_LENGTH: |
| 109 | return path |
| 110 | to_cut = len(path) - (MAX_PATH_LENGTH + 3) |
| 111 | base, extension = os.path.splitext(path) |
| 112 | inc = 0 |
| 113 | new_path = "{}_{}{}".format(base, inc, extension) |
| 114 | while os.path.exists(new_path): |
| 115 | new_path = "{}_{}{}".format(base, inc, extension) |
| 116 | inc += 1 |
| 117 | return new_path |
| 118 | |
| 119 | |
| 120 | def strip_ws(value): |
| 121 | """ Remove whitespace from a string """ |
| 122 | return str(NO_WHITESPACE_REGEX.sub('-', value)) |
| 123 | |
| 124 | |
| 125 | def slugify(value): |
| 126 | """ |
| 127 | Normalise string, removes invalid for filename charactersr |
| 128 | and converts string to lowercase. |
| 129 | """ |
| 130 | value = unicodedata.normalize('NFKC', value).lower().strip() |
| 131 | value = re.sub(r'[\\/<>:\?\*\|"]', '', value) |
| 132 | value = re.sub(r'\.*$', '', value) |
| 133 | return value |
| 134 | |
| 135 | class PageChecker(object): |
| 136 | def __init__(self): |
| 137 | self.log = [] |
| 138 | self.title = None |
| 139 | self.file_count = None |
| 140 | self.files = None |
| 141 | self.images = None |
| 142 | self.license = None |
| 143 | |
| 144 | |
| 145 | def __call__(self, _): |
| 146 | try: |
| 147 | self.log.append("call") |
| 148 | if self.title is None: |
| 149 | # first find the name |
| 150 | name = EC._find_element(BROWSER, (By.CSS_SELECTOR, "[class^=ThingPage__modelName]")) |
| 151 | if name is None: |
| 152 | return False |
| 153 | self.title = name.text |
| 154 | |
| 155 | if self.file_count is None: |
| 156 | # OK. Do we know how many files we have to download? |
| 157 | metrics = EC._find_elements(BROWSER, (By.CSS_SELECTOR, "[class^=MetricButton]")) |
| 158 | self.log.append("got some metrics: {}".format(len(metrics))) |
| 159 | cur_count = int([x.text.split("\n")[0] for x in metrics if x.text.endswith("\nThing Files")][0]) |
| 160 | self.log.append(cur_count) |
| 161 | if cur_count == 0: |
| 162 | return False |
| 163 | self.file_count = cur_count |
| 164 | |
| 165 | self.log.append("looking for {} files".format(self.file_count)) |
| 166 | fileRows = EC._find_elements(BROWSER, (By.CSS_SELECTOR, "[class^=ThingFile__fileRow]")) |
| 167 | self.log.append("found {} files".format(len(fileRows))) |
| 168 | if len(fileRows) < self.file_count: |
| 169 | return False |
| 170 | |
| 171 | self.log.append("Looking for images") |
| 172 | # By this point _should_ have loaded all the images |
| 173 | self.images = EC._find_elements(BROWSER, (By.CSS_SELECTOR, "[class^=thumb]")) |
| 174 | self.license = EC._find_element(BROWSER, (By.CSS_SELECTOR, "[class^=License__licenseText]")).text |
| 175 | self.log.append("found {} images".format(len(self.images))) |
| 176 | self.files = fileRows |
| 177 | return True |
| 178 | except Exception: |
| 179 | return False |
| 180 | |
| 181 | |
| 182 | |
| 183 | |
| 184 | class Downloader(multiprocessing.Process): |
| 185 | """ |
| 186 | Class to handle downloading the things we have found to get. |
| 187 | """ |
| 188 | |
| 189 | def __init__(self, thing_queue, download_directory, compress): |
| 190 | multiprocessing.Process.__init__(self) |
| 191 | # TODO: add parameters |
| 192 | self.thing_queue = thing_queue |
| 193 | self.download_directory = download_directory |
| 194 | self.compress = compress |
| 195 | |
| 196 | def run(self): |
| 197 | """ actual download loop. |
| 198 | """ |
| 199 | while True: |
| 200 | thing_id = self.thing_queue.get() |
| 201 | if thing_id is None: |
| 202 | logging.info("Shutting download queue") |
| 203 | self.thing_queue.task_done() |
| 204 | break |
| 205 | logging.info("Handling id {}".format(thing_id)) |
| 206 | Thing(thing_id).download(self.download_directory, self.compress) |
| 207 | self.thing_queue.task_done() |
| 208 | return |
| 209 | |
| 210 | |
| 211 | |
| 212 | |
| 213 | |
| 214 | class Grouping: |
| 215 | """ Holds details of a group of things for download |
| 216 | This is effectively (although not actually) an abstract class |
| 217 | - use Collection or Designs instead. |
| 218 | """ |
| 219 | |
| 220 | def __init__(self, quick, compress): |
| 221 | self.things = [] |
| 222 | self.total = 0 |
| 223 | self.req_id = None |
| 224 | self.last_page = 0 |
| 225 | self.per_page = None |
| 226 | # Should we stop downloading when we hit a known datestamp? |
| 227 | self.quick = quick |
| 228 | self.compress = compress |
| 229 | # These should be set by child classes. |
| 230 | self.url = None |
| 231 | self.download_dir = None |
| 232 | self.collection_url = None |
| 233 | |
| 234 | def _get_small_grouping(self, req): |
| 235 | """ Handle small groupings """ |
| 236 | soup = BeautifulSoup(req.text, features='lxml') |
| 237 | links = soup.find_all('a', {'class': 'card-img-holder'}) |
| 238 | self.things = [x['href'].split(':')[1] for x in links] |
| 239 | self.total = len(self.things) |
| 240 | |
| 241 | return self.things |
| 242 | |
| 243 | def get(self): |
| 244 | """ retrieve the things of the grouping. """ |
| 245 | if self.things: |
| 246 | # We've already done it. |
| 247 | return self.things |
| 248 | |
| 249 | # Check for initialisation: |
| 250 | if not self.url: |
| 251 | logging.error("No URL set - object not initialised properly?") |
| 252 | raise ValueError("No URL set - object not initialised properly?") |
| 253 | |
| 254 | # Get the internal details of the grouping. |
| 255 | logging.debug("Querying {}".format(self.url)) |
| 256 | c_req = requests.get(self.url) |
| 257 | total = TOTAL_REGEX.search(c_req.text) |
| 258 | if total is None: |
| 259 | # This is a small (<13) items grouping. Pull the list from this req. |
| 260 | return self._get_small_grouping(c_req) |
| 261 | self.total = total.groups()[0] |
| 262 | self.req_id = ID_REGEX.search(c_req.text).groups()[0] |
| 263 | self.last_page = int(LAST_PAGE_REGEX.search(c_req.text).groups()[0]) |
| 264 | self.per_page = PER_PAGE_REGEX.search(c_req.text).groups()[0] |
| 265 | parameters = { |
| 266 | 'base_url': self.url, |
| 267 | 'page': '1', |
| 268 | 'per_page': '12', |
| 269 | 'id': self.req_id |
| 270 | } |
| 271 | for current_page in range(1, self.last_page + 1): |
| 272 | parameters['page'] = current_page |
| 273 | req = requests.post(self.collection_url, parameters) |
| 274 | soup = BeautifulSoup(req.text, features='lxml') |
| 275 | links = soup.find_all('a', {'class': 'card-img-holder'}) |
| 276 | self.things += [x['href'].split(':')[1] for x in links] |
| 277 | |
| 278 | return self.things |
| 279 | |
| 280 | def download(self): |
| 281 | """ Downloads all the files in a collection """ |
| 282 | if not self.things: |
| 283 | self.get() |
| 284 | |
| 285 | if not self.download_dir: |
| 286 | raise ValueError( |
| 287 | "No download_dir set - invalidly initialised object?") |
| 288 | |
| 289 | base_dir = os.getcwd() |
| 290 | try: |
| 291 | os.mkdir(self.download_dir) |
| 292 | except FileExistsError: |
| 293 | logging.info("Target directory {} already exists. Assuming a resume." |
| 294 | .format(self.download_dir)) |
| 295 | logging.info("Downloading {} thing(s).".format(self.total)) |
| 296 | for idx, thing in enumerate(self.things): |
| 297 | logging.info("Downloading thing {} - {}".format(idx, thing)) |
| 298 | RC = Thing(thing).download(self.download_dir, self.compress) |
| 299 | if self.quick and RC==State.ALREADY_DOWNLOADED: |
| 300 | logging.info("Caught up, stopping.") |
| 301 | return |
| 302 | |
| 303 | |
| 304 | |
| 305 | |
| 306 | |
| 307 | class Collection(Grouping): |
| 308 | """ Holds details of a collection. """ |
| 309 | |
| 310 | def __init__(self, user, name, directory, quick, compress): |
| 311 | Grouping.__init__(self, quick, compress) |
| 312 | self.user = user |
| 313 | self.name = name |
| 314 | self.url = "{}/{}/collections/{}".format( |
| 315 | URL_BASE, self.user, strip_ws(self.name)) |
| 316 | self.download_dir = os.path.join(directory, |
| 317 | "{}-{}".format(slugify(self.user), slugify(self.name))) |
| 318 | self.collection_url = URL_COLLECTION |
| 319 | |
| 320 | |
| 321 | class Designs(Grouping): |
| 322 | """ Holds details of all of a users' designs. """ |
| 323 | |
| 324 | def __init__(self, user, directory, quick, compress): |
| 325 | Grouping.__init__(self, quick, compress) |
| 326 | self.user = user |
| 327 | self.url = "{}/{}/designs".format(URL_BASE, self.user) |
| 328 | self.download_dir = os.path.join( |
| 329 | directory, "{} designs".format(slugify(self.user))) |
| 330 | self.collection_url = USER_COLLECTION |
| 331 | |
| 332 | |
| 333 | class Thing: |
| 334 | """ An individual design on thingiverse. """ |
| 335 | |
| 336 | def __init__(self, thing_id): |
| 337 | self.thing_id = thing_id |
| 338 | self.last_time = None |
| 339 | self._parsed = False |
| 340 | self._needs_download = True |
| 341 | self.text = None |
| 342 | self.title = None |
| 343 | self.download_dir = None |
| 344 | self.time_stamp = None |
| 345 | self._file_links = FileLinks() |
| 346 | |
| 347 | def _parse(self, base_dir): |
| 348 | """ Work out what, if anything needs to be done. """ |
| 349 | if self._parsed: |
| 350 | return |
| 351 | |
| 352 | url = "{}/thing:{}/files".format(URL_BASE, self.thing_id) |
| 353 | try: |
| 354 | BROWSER.get(url) |
| 355 | wait = WebDriverWait(BROWSER, 60) |
| 356 | pc = PageChecker() |
| 357 | wait.until(pc) |
| 358 | except requests.exceptions.ConnectionError as error: |
| 359 | logging.error("Unable to connect for thing {}: {}".format( |
| 360 | self.thing_id, error)) |
| 361 | return |
| 362 | except selenium.common.exceptions.TimeoutException: |
| 363 | logging.error(pc.log) |
| 364 | logging.error("Timeout trying to parse thing {}".format(self.thing_id)) |
| 365 | return |
| 366 | |
| 367 | self.title = pc.title |
| 368 | if not pc.files: |
| 369 | logging.error("No files found for thing {} - probably thingiverse being broken, try again later".format(self.thing_id)) |
| 370 | for link in pc.files: |
| 371 | logging.debug("Parsing link: {}".format(link.text)) |
| 372 | link_link = link.find_element_by_xpath(".//a").get_attribute("href") |
| 373 | if link_link.endswith("/zip"): |
| 374 | # bulk link. |
| 375 | continue |
| 376 | try: |
| 377 | link_title, link_details, _ = link.text.split("\n") |
| 378 | except ValueError: |
| 379 | # If it is a filetype that doesn't generate a picture, then we get an extra field at the start. |
| 380 | _, link_title, link_details, _ = link.text.split("\n") |
| 381 | |
| 382 | #link_details will be something like '461 kb | Updated 06-11-2019 | 373 Downloads' |
| 383 | #need to convert from M D Y to Y M D |
| 384 | link_date = [int(x) for x in link_details.split("|")[1].split()[-1].split("-")] |
| 385 | try: |
| 386 | self._file_links.append(FileLink(link_title, datetime.datetime(link_date[2], link_date[0], link_date[1]), link_link)) |
| 387 | except ValueError: |
| 388 | logging.error(link_date) |
| 389 | |
| 390 | self._image_links=[x.find_element_by_xpath(".//img").get_attribute("src") for x in pc.images] |
| 391 | self._license = pc.license |
| 392 | self.pc = pc |
| 393 | |
| 394 | |
| 395 | self.old_download_dir = os.path.join(base_dir, slugify(self.title)) |
| 396 | self.download_dir = os.path.join(base_dir, "{} - {}".format(self.thing_id, slugify(self.title))) |
| 397 | |
| 398 | logging.debug("Parsing {} ({})".format(self.thing_id, self.title)) |
| 399 | |
| 400 | if not os.path.exists(self.download_dir): |
| 401 | logging.info("Looking for old dir at {}".format(self.old_download_dir)) |
| 402 | if os.path.exists(self.old_download_dir): |
| 403 | logging.warning("Found previous style download directory. Moving it from {} to {}".format(self.old_download_dir, self.download_dir)) |
| 404 | os.rename(self.old_download_dir, self.download_dir) |
| 405 | else: |
| 406 | # Not yet downloaded |
| 407 | self._parsed = True |
| 408 | return |
| 409 | |
| 410 | timestamp_file = os.path.join(self.download_dir, 'timestamp.txt') |
| 411 | if not os.path.exists(timestamp_file): |
| 412 | # Old download from before |
| 413 | logging.warning( |
| 414 | "Old-style download directory found. Assuming update required.") |
| 415 | self._parsed = True |
| 416 | return |
| 417 | |
| 418 | try: |
| 419 | with open(timestamp_file, 'r') as timestamp_handle: |
| 420 | # add the .split(' ')[0] to remove the timestamp from the old style timestamps |
| 421 | last_bits = [int(x) for x in timestamp_handle.readlines()[0].split(' ')[0].split("-")] |
| 422 | logging.warning(last_bits) |
| 423 | if last_bits[0] == 0: |
| 424 | last_bits[0] = 1 |
| 425 | if last_bits[1] == 0: |
| 426 | last_bits[1] = 1 |
| 427 | if last_bits[2] == 0: |
| 428 | last_bits[2] = 1980 |
| 429 | try: |
| 430 | self.last_time = datetime.datetime(last_bits[0], last_bits[1], last_bits[2]) |
| 431 | except ValueError: |
| 432 | # This one appears to be M D Y |
| 433 | self.last_time = datetime.datetime(last_bits[2], last_bits[0], last_bits[1]) |
| 434 | |
| 435 | logging.info("last downloaded version: {}".format(self.last_time)) |
| 436 | except FileNotFoundError: |
| 437 | # Not run on this thing before. |
| 438 | logging.info( |
| 439 | "Old-style download directory found. Assuming update required.") |
| 440 | self.last_time = None |
| 441 | self._needs_download = True |
| 442 | self._parsed = True |
| 443 | return |
| 444 | |
| 445 | # OK, so we have a timestamp, lets see if there is anything new to get |
| 446 | try: |
| 447 | if self._file_links.last_update > self.last_time: |
| 448 | logging.info( |
| 449 | "Found new/updated files {}".format(self._file_links.last_update)) |
| 450 | self._needs_download = True |
| 451 | self._parsed = True |
| 452 | return |
| 453 | except TypeError: |
| 454 | logging.warning("No files found for {}.".format(self.thing_id)) |
| 455 | |
| 456 | # Got here, so nope, no new files. |
| 457 | self._needs_download = False |
| 458 | self._parsed = True |
| 459 | |
| 460 | def download(self, base_dir, compress): |
| 461 | """ Download all files for a given thing. |
| 462 | Returns True iff the thing is now downloaded (not iff it downloads the thing!) |
| 463 | """ |
| 464 | if not self._parsed: |
| 465 | self._parse(base_dir) |
| 466 | |
| 467 | if not self._parsed: |
| 468 | logging.error( |
| 469 | "Unable to parse {} - aborting download".format(self.thing_id)) |
| 470 | return State.FAILED |
| 471 | |
| 472 | if not self._needs_download: |
| 473 | print("{} - {} already downloaded - skipping.".format(self.thing_id, self.title)) |
| 474 | return State.ALREADY_DOWNLOADED |
| 475 | |
| 476 | if not self._file_links: |
| 477 | print("{} - {} appears to have no files. Thingiverse acting up again?".format(self.thing_id, self.title)) |
| 478 | return State.FAILED |
| 479 | |
| 480 | # Have we already downloaded some things? |
| 481 | timestamp_file = os.path.join(self.download_dir, 'timestamp.txt') |
| 482 | prev_dir = None |
| 483 | if os.path.exists(self.download_dir): |
| 484 | if not os.path.exists(timestamp_file): |
| 485 | # edge case: old style dir w/out timestamp. |
| 486 | logging.warning("Old style download dir found at {}".format(self.title)) |
| 487 | prev_count = 0 |
| 488 | target_dir = "{}_old".format(self.download_dir) |
| 489 | while os.path.exists(target_dir): |
| 490 | prev_count = prev_count + 1 |
| 491 | target_dir = "{}_old_{}".format(self.download_dir, prev_count) |
| 492 | os.rename(self.download_dir, target_dir) |
| 493 | else: |
| 494 | prev_dir = "{}_{}".format(self.download_dir, slugify(self.last_time.__str__())) |
| 495 | os.rename(self.download_dir, prev_dir) |
| 496 | |
| 497 | # Get the list of files to download |
| 498 | |
| 499 | new_file_links = [] |
| 500 | old_file_links = [] |
| 501 | self.time_stamp = None |
| 502 | |
| 503 | if not self.last_time: |
| 504 | # If we don't have anything to copy from, then it is all new. |
| 505 | logging.debug("No last time, downloading all files") |
| 506 | new_file_links = self._file_links |
| 507 | self.time_stamp = new_file_links[0].last_update |
| 508 | |
| 509 | for file_link in new_file_links: |
| 510 | self.time_stamp = max(self.time_stamp, file_link.last_update) |
| 511 | logging.debug("New timestamp will be {}".format(self.time_stamp)) |
| 512 | else: |
| 513 | self.time_stamp = self.last_time |
| 514 | for file_link in self._file_links: |
| 515 | if file_link.last_update > self.last_time: |
| 516 | new_file_links.append(file_link) |
| 517 | self.time_stamp = max(self.time_stamp, file_link.last_update) |
| 518 | else: |
| 519 | old_file_links.append(file_link) |
| 520 | |
| 521 | logging.debug("new timestamp {}".format(self.time_stamp)) |
| 522 | |
| 523 | # OK. Time to get to work. |
| 524 | logging.debug("Generating download_dir") |
| 525 | os.mkdir(self.download_dir) |
| 526 | filelist_file = os.path.join(self.download_dir, "filelist.txt") |
| 527 | with open(filelist_file, 'w', encoding="utf-8") as fl_handle: |
| 528 | for fl in self._file_links: |
| 529 | base_link = fl.link |
| 530 | try: |
| 531 | fl.link=requests.get(fl.link, allow_redirects=False).headers['location'] |
| 532 | except Exception: |
| 533 | # Sometimes Thingiverse just gives us the direct link the first time. Not sure why. |
| 534 | pass |
| 535 | |
| 536 | fl_handle.write("{},{},{}, {}\n".format(fl.link, fl.name, fl.last_update, base_link)) |
| 537 | |
| 538 | |
| 539 | # First grab the cached files (if any) |
| 540 | logging.info("Copying {} unchanged files.".format(len(old_file_links))) |
| 541 | for file_link in old_file_links: |
| 542 | old_file = os.path.join(prev_dir, file_link.name) |
| 543 | new_file = truncate_name(os.path.join(self.download_dir, file_link.name)) |
| 544 | try: |
| 545 | logging.debug("Copying {} to {}".format(old_file, new_file)) |
| 546 | copyfile(old_file, new_file) |
| 547 | except FileNotFoundError: |
| 548 | logging.warning( |
| 549 | "Unable to find {} in old archive, redownloading".format(file_link["title"])) |
| 550 | new_file_links.append(file_link) |
| 551 | |
| 552 | # Now download the new ones |
| 553 | logging.info("Downloading {} new files of {}".format( |
| 554 | len(new_file_links), len(self._file_links))) |
| 555 | try: |
| 556 | for file_link in new_file_links: |
| 557 | file_name = truncate_name(os.path.join(self.download_dir, file_link.name)) |
| 558 | logging.debug("Downloading {} from {} to {}".format( |
| 559 | file_link.name, file_link.link, file_name)) |
| 560 | data_req = requests.get(file_link.link) |
| 561 | with open(file_name, 'wb') as handle: |
| 562 | handle.write(data_req.content) |
| 563 | except Exception as exception: |
| 564 | logging.error("Failed to download {} - {}".format(file_link.name, exception)) |
| 565 | fail_dir(self.download_dir) |
| 566 | return State.FAILED |
| 567 | |
| 568 | |
| 569 | # People like images. But this doesn't work yet. |
| 570 | image_dir = os.path.join(self.download_dir, 'images') |
| 571 | logging.info("Downloading {} images.".format(len(self._image_links))) |
| 572 | try: |
| 573 | os.mkdir(image_dir) |
| 574 | for imagelink in self._image_links: |
| 575 | filename = os.path.basename(imagelink) |
| 576 | if filename.endswith('stl'): |
| 577 | filename = "{}.png".format(filename) |
| 578 | image_req = requests.get(imagelink) |
| 579 | with open(truncate_name(os.path.join(image_dir, filename)), 'wb') as handle: |
| 580 | handle.write(image_req.content) |
| 581 | except Exception as exception: |
| 582 | print("Failed to download {} - {}".format(filename, exception)) |
| 583 | fail_dir(self.download_dir) |
| 584 | return State.FAILED |
| 585 | |
| 586 | """ |
| 587 | # instructions are good too. |
| 588 | logging.info("Downloading readme") |
| 589 | try: |
| 590 | readme_txt = soup.find('meta', property='og:description')[ |
| 591 | 'content'] |
| 592 | with open(os.path.join(self.download_dir, 'readme.txt'), 'w') as readme_handle: |
| 593 | readme_handle.write("{}\n".format(readme_txt)) |
| 594 | except (TypeError, KeyError) as exception: |
| 595 | logging.warning("No readme? {}".format(exception)) |
| 596 | except IOError as exception: |
| 597 | logging.warning("Failed to write readme! {}".format(exception)) |
| 598 | |
| 599 | """ |
| 600 | # Best get some licenses |
| 601 | logging.info("Downloading license") |
| 602 | try: |
| 603 | if self._license: |
| 604 | with open(truncate_name(os.path.join(self.download_dir, 'license.txt')), 'w', encoding="utf-8") as license_handle: |
| 605 | license_handle.write("{}\n".format(self._license)) |
| 606 | except IOError as exception: |
| 607 | logging.warning("Failed to write license! {}".format(exception)) |
| 608 | |
| 609 | try: |
| 610 | # Now write the timestamp |
| 611 | with open(timestamp_file, 'w', encoding="utf-8") as timestamp_handle: |
| 612 | timestamp_handle.write(self.time_stamp.__str__()) |
| 613 | except Exception as exception: |
| 614 | print("Failed to write timestamp file - {}".format(exception)) |
| 615 | fail_dir(self.download_dir) |
| 616 | return State.FAILED |
| 617 | self._needs_download = False |
| 618 | logging.debug("Download of {} finished".format(self.title)) |
| 619 | if not compress: |
| 620 | return State.OK |
| 621 | |
| 622 | |
| 623 | thing_dir = "{} - {} - {}".format(self.thing_id, |
| 624 | slugify(self.title), |
| 625 | self.time_stamp) |
| 626 | file_name = os.path.join(base_dir, |
| 627 | "{}.7z".format(thing_dir)) |
| 628 | logging.debug("Compressing {} to {}".format( |
| 629 | self.title, |
| 630 | file_name)) |
| 631 | #with libarchive.file_writer(filename, 'lzma', '7z') as archive: |
| 632 | with py7zr.SevenZipFile(file_name, 'w', filters=SEVENZIP_FILTERS) as archive: |
| 633 | #with py7zr.SevenZipFile(file_name, 'w' ) as archive: |
| 634 | archive.writeall(self.download_dir, thing_dir) |
| 635 | logging.debug("Compression of {} finished.".format(self.title)) |
| 636 | return State.OK |
| 637 | |
| 638 | |
| 639 | |
| 640 | |
| 641 | def do_batch(batch_file, download_dir, quick, compress): |
| 642 | """ Read a file in line by line, parsing each as a set of calls to this script.""" |
| 643 | with open(batch_file) as handle: |
| 644 | for line in handle: |
| 645 | line = line.strip() |
| 646 | if not line: |
| 647 | # Skip empty lines |
| 648 | continue |
| 649 | logging.info("Handling instruction {}".format(line)) |
| 650 | command_arr = line.split() |
| 651 | if command_arr[0] == "thing": |
| 652 | logging.debug( |
| 653 | "Handling batch thing instruction: {}".format(line)) |
| 654 | Thing(command_arr[1]).download(download_dir, compress) |
| 655 | continue |
| 656 | if command_arr[0] == "collection": |
| 657 | logging.debug( |
| 658 | "Handling batch collection instruction: {}".format(line)) |
| 659 | Collection(command_arr[1], command_arr[2], |
| 660 | download_dir, quick, compress).download() |
| 661 | continue |
| 662 | if command_arr[0] == "user": |
| 663 | logging.debug( |
| 664 | "Handling batch collection instruction: {}".format(line)) |
| 665 | Designs(command_arr[1], download_dir, quick, compress).download() |
| 666 | continue |
| 667 | logging.warning("Unable to parse current instruction. Skipping.") |
| 668 | |
| 669 | |
| 670 | def main(): |
| 671 | """ Entry point for script being run as a command. """ |
| 672 | parser = argparse.ArgumentParser() |
| 673 | parser.add_argument("-l", "--log-level", choices=[ |
| 674 | 'debug', 'info', 'warning'], default='info', help="level of logging desired") |
| 675 | parser.add_argument("-d", "--directory", |
| 676 | help="Target directory to download into") |
| 677 | parser.add_argument("-f", "--log-file", |
| 678 | help="Place to log debug information to") |
| 679 | parser.add_argument("-q", "--quick", action="store_true", |
| 680 | help="Assume date ordering on posts") |
| 681 | parser.add_argument("-c", "--compress", action="store_true", |
| 682 | help="Compress files") |
| 683 | |
| 684 | |
| 685 | subparsers = parser.add_subparsers( |
| 686 | help="Type of thing to download", dest="subcommand") |
| 687 | collection_parser = subparsers.add_parser( |
| 688 | 'collection', help="Download one or more entire collection(s)") |
| 689 | collection_parser.add_argument( |
| 690 | "owner", help="The owner of the collection(s) to get") |
| 691 | collection_parser.add_argument( |
| 692 | "collections", nargs="+", help="Space seperated list of the name(s) of collection to get") |
| 693 | thing_parser = subparsers.add_parser( |
| 694 | 'thing', help="Download a single thing.") |
| 695 | thing_parser.add_argument( |
| 696 | "things", nargs="*", help="Space seperated list of thing ID(s) to download") |
| 697 | user_parser = subparsers.add_parser( |
| 698 | "user", help="Download all things by one or more users") |
| 699 | user_parser.add_argument( |
| 700 | "users", nargs="+", help="A space seperated list of the user(s) to get the designs of") |
| 701 | batch_parser = subparsers.add_parser( |
| 702 | "batch", help="Perform multiple actions written in a text file") |
| 703 | batch_parser.add_argument( |
| 704 | "batch_file", help="The name of the file to read.") |
| 705 | subparsers.add_parser("version", help="Show the current version") |
| 706 | |
| 707 | args = parser.parse_args() |
| 708 | if not args.subcommand: |
| 709 | parser.print_help() |
| 710 | sys.exit(1) |
| 711 | if not args.directory: |
| 712 | args.directory = os.getcwd() |
| 713 | |
| 714 | logger = logging.getLogger() |
| 715 | formatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s') |
| 716 | logger.setLevel(logging.DEBUG) |
| 717 | console_handler = logging.StreamHandler() |
| 718 | console_handler.setLevel(args.log_level.upper()) |
| 719 | |
| 720 | logger.addHandler(console_handler) |
| 721 | if args.log_file: |
| 722 | file_handler = logging.FileHandler(args.log_file) |
| 723 | file_handler.setLevel(logging.DEBUG) |
| 724 | file_handler.setFormatter(formatter) |
| 725 | logger.addHandler(file_handler) |
| 726 | |
| 727 | |
| 728 | # Start downloader |
| 729 | thing_queue = multiprocessing.JoinableQueue() |
| 730 | logging.debug("starting {} downloader(s)".format(DOWNLOADER_COUNT)) |
| 731 | downloaders = [Downloader(thing_queue, args.directory, args.compress) for _ in range(DOWNLOADER_COUNT)] |
| 732 | for downloader in downloaders: |
| 733 | downloader.start() |
| 734 | |
| 735 | |
| 736 | if args.subcommand.startswith("collection"): |
| 737 | for collection in args.collections: |
| 738 | Collection(args.owner, collection, args.directory, args.quick, args.compress).download() |
| 739 | if args.subcommand == "thing": |
| 740 | for thing in args.things: |
| 741 | thing_queue.put(thing) |
| 742 | if args.subcommand == "user": |
| 743 | for user in args.users: |
| 744 | Designs(user, args.directory, args.quick, args.compress).download() |
| 745 | if args.subcommand == "version": |
| 746 | print("thingy_grabber.py version {}".format(VERSION)) |
| 747 | if args.subcommand == "batch": |
| 748 | do_batch(args.batch_file, args.directory, args.quick, args.compress) |
| 749 | |
| 750 | # Stop the downloader processes |
| 751 | for downloader in downloaders: |
| 752 | thing_queue.put(None) |
| 753 | |
| 754 | atexit.register(BROWSER.quit) |
| 755 | |
| 756 | if __name__ == "__main__": |
| 757 | multiprocessing.freeze_support() |
| 758 | main() |