Commit | Line | Data |
---|---|---|
975060c9 OM |
1 | #!/usr/bin/env python3 |
2 | """ | |
3 | Thingiverse bulk downloader | |
4 | """ | |
5 | ||
6 | import re | |
4a98996b | 7 | import sys |
975060c9 OM |
8 | import os |
9 | import argparse | |
10 | import unicodedata | |
11 | import requests | |
fa2f3251 | 12 | import logging |
6a777954 | 13 | import multiprocessing |
7b84ba6d | 14 | import enum |
fb28c59b | 15 | import datetime |
3c82f75b | 16 | from shutil import copyfile |
975060c9 | 17 | from bs4 import BeautifulSoup |
b497d705 OM |
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 | |
d194b140 | 25 | import atexit |
9828dabe | 26 | import py7zr |
8ed15058 OM |
27 | import glob |
28 | import shutil | |
975060c9 | 29 | |
ae598d73 OM |
30 | SEVENZIP_FILTERS = [{'id': py7zr.FILTER_LZMA2}] |
31 | ||
8ed15058 OM |
32 | # I don't think this is exported by datetime |
33 | DEFAULT_DATETIME_FORMAT = '%Y-%m-%d %H:%M:%S' | |
3ac180ed OM |
34 | # Windows cannot handle : in filenames |
35 | SAFE_DATETIME_FORMAT = '%Y-%m-%d %H.%M.%S' | |
8ed15058 | 36 | |
975060c9 OM |
37 | URL_BASE = "https://www.thingiverse.com" |
38 | URL_COLLECTION = URL_BASE + "/ajax/thingcollection/list_collected_things" | |
948bd56f | 39 | USER_COLLECTION = URL_BASE + "/ajax/user/designs" |
975060c9 OM |
40 | |
41 | ID_REGEX = re.compile(r'"id":(\d*),') | |
42 | TOTAL_REGEX = re.compile(r'"total":(\d*),') | |
43 | LAST_PAGE_REGEX = re.compile(r'"last_page":(\d*),') | |
44 | # This appears to be fixed at 12, but if it changes would screw the rest up. | |
45 | PER_PAGE_REGEX = re.compile(r'"per_page":(\d*),') | |
dd8c35f4 OM |
46 | NO_WHITESPACE_REGEX = re.compile(r'[-\s]+') |
47 | ||
6a777954 | 48 | DOWNLOADER_COUNT = 1 |
7b84ba6d | 49 | RETRY_COUNT = 3 |
6a777954 | 50 | |
65bd8b43 OM |
51 | MAX_PATH_LENGTH = 250 |
52 | ||
ae598d73 | 53 | VERSION = "0.9.0" |
dbdb1782 | 54 | |
8ed15058 | 55 | TIMESTAMP_FILE = "timestamp.txt" |
b497d705 OM |
56 | |
57 | #BROWSER = webdriver.PhantomJS('./phantomjs') | |
58 | options = Options() | |
fb28c59b | 59 | options.add_argument("--headless") |
b497d705 OM |
60 | BROWSER = webdriver.Firefox(options=options) |
61 | ||
62 | BROWSER.set_window_size(1980, 1080) | |
63 | ||
64 | ||
65 | @dataclass | |
66 | class FileLink: | |
67 | name: str | |
ae598d73 OM |
68 | last_update: datetime.datetime |
69 | link: str | |
70 | ||
71 | class FileLinks: | |
72 | def __init__(self, initial_links=[]): | |
73 | self.links = [] | |
74 | self.last_update = None | |
75 | for link in initial_links: | |
76 | self.append(link) | |
77 | ||
78 | def __iter__(self): | |
79 | return iter(self.links) | |
80 | ||
81 | def __getitem__(self, item): | |
82 | return self.links[item] | |
83 | ||
84 | def __len__(self): | |
85 | return len(self.links) | |
86 | ||
87 | def append(self, link): | |
88 | try: | |
89 | self.last_update = max(self.last_update, link.last_update) | |
90 | except TypeError: | |
91 | self.last_update = link.last_update | |
92 | self.links.append(link) | |
8ed15058 | 93 | |
b497d705 | 94 | |
7b84ba6d OM |
95 | class State(enum.Enum): |
96 | OK = enum.auto() | |
97 | FAILED = enum.auto() | |
98 | ALREADY_DOWNLOADED = enum.auto() | |
99 | ||
8ed15058 OM |
100 | def rename_unique(dir_name, target_dir_name): |
101 | """ Move a directory sideways to a new name, ensuring it is unique. | |
65bd8b43 | 102 | """ |
8ed15058 | 103 | target_dir = target_dir_name |
65bd8b43 OM |
104 | inc = 0 |
105 | while os.path.exists(target_dir): | |
8ed15058 | 106 | target_dir = "{}_{}".format(target_dir_name, inc) |
65bd8b43 OM |
107 | inc += 1 |
108 | os.rename(dir_name, target_dir) | |
8ed15058 OM |
109 | return target_dir |
110 | ||
111 | ||
112 | def fail_dir(dir_name): | |
113 | """ When a download has failed, move it sideways. | |
114 | """ | |
115 | return rename_unique(dir_name,"{}_failed".format(dir_name)) | |
65bd8b43 OM |
116 | |
117 | ||
118 | def truncate_name(file_name): | |
119 | """ Ensure the filename is not too long for, well windows basically. | |
120 | """ | |
121 | path = os.path.abspath(file_name) | |
122 | if len(path) <= MAX_PATH_LENGTH: | |
123 | return path | |
124 | to_cut = len(path) - (MAX_PATH_LENGTH + 3) | |
125 | base, extension = os.path.splitext(path) | |
126 | inc = 0 | |
127 | new_path = "{}_{}{}".format(base, inc, extension) | |
128 | while os.path.exists(new_path): | |
129 | new_path = "{}_{}{}".format(base, inc, extension) | |
130 | inc += 1 | |
131 | return new_path | |
132 | ||
133 | ||
dd8c35f4 OM |
134 | def strip_ws(value): |
135 | """ Remove whitespace from a string """ | |
136 | return str(NO_WHITESPACE_REGEX.sub('-', value)) | |
975060c9 | 137 | |
dbdb1782 | 138 | |
975060c9 OM |
139 | def slugify(value): |
140 | """ | |
d194b140 OM |
141 | Normalise string, removes invalid for filename charactersr |
142 | and converts string to lowercase. | |
975060c9 | 143 | """ |
d194b140 | 144 | value = unicodedata.normalize('NFKC', value).lower().strip() |
65bd8b43 OM |
145 | value = re.sub(r'[\\/<>:\?\*\|"]', '', value) |
146 | value = re.sub(r'\.*$', '', value) | |
147 | return value | |
975060c9 | 148 | |
b497d705 OM |
149 | class PageChecker(object): |
150 | def __init__(self): | |
151 | self.log = [] | |
152 | self.title = None | |
153 | self.file_count = None | |
154 | self.files = None | |
fb28c59b OM |
155 | self.images = None |
156 | self.license = None | |
b497d705 OM |
157 | |
158 | ||
159 | def __call__(self, _): | |
160 | try: | |
161 | self.log.append("call") | |
162 | if self.title is None: | |
163 | # first find the name | |
164 | name = EC._find_element(BROWSER, (By.CSS_SELECTOR, "[class^=ThingPage__modelName]")) | |
165 | if name is None: | |
166 | return False | |
167 | self.title = name.text | |
168 | ||
169 | if self.file_count is None: | |
170 | # OK. Do we know how many files we have to download? | |
171 | metrics = EC._find_elements(BROWSER, (By.CSS_SELECTOR, "[class^=MetricButton]")) | |
172 | self.log.append("got some metrics: {}".format(len(metrics))) | |
173 | cur_count = int([x.text.split("\n")[0] for x in metrics if x.text.endswith("\nThing Files")][0]) | |
174 | self.log.append(cur_count) | |
175 | if cur_count == 0: | |
176 | return False | |
177 | self.file_count = cur_count | |
178 | ||
179 | self.log.append("looking for {} files".format(self.file_count)) | |
180 | fileRows = EC._find_elements(BROWSER, (By.CSS_SELECTOR, "[class^=ThingFile__fileRow]")) | |
181 | self.log.append("found {} files".format(len(fileRows))) | |
fb28c59b OM |
182 | if len(fileRows) < self.file_count: |
183 | return False | |
184 | ||
185 | self.log.append("Looking for images") | |
186 | # By this point _should_ have loaded all the images | |
187 | self.images = EC._find_elements(BROWSER, (By.CSS_SELECTOR, "[class^=thumb]")) | |
188 | self.license = EC._find_element(BROWSER, (By.CSS_SELECTOR, "[class^=License__licenseText]")).text | |
189 | self.log.append("found {} images".format(len(self.images))) | |
190 | self.files = fileRows | |
191 | return True | |
b497d705 OM |
192 | except Exception: |
193 | return False | |
194 | ||
195 | ||
196 | ||
197 | ||
6a777954 OM |
198 | class Downloader(multiprocessing.Process): |
199 | """ | |
200 | Class to handle downloading the things we have found to get. | |
201 | """ | |
202 | ||
ae598d73 | 203 | def __init__(self, thing_queue, download_directory, compress): |
6a777954 OM |
204 | multiprocessing.Process.__init__(self) |
205 | # TODO: add parameters | |
206 | self.thing_queue = thing_queue | |
207 | self.download_directory = download_directory | |
ae598d73 | 208 | self.compress = compress |
6a777954 OM |
209 | |
210 | def run(self): | |
211 | """ actual download loop. | |
212 | """ | |
213 | while True: | |
214 | thing_id = self.thing_queue.get() | |
215 | if thing_id is None: | |
216 | logging.info("Shutting download queue") | |
217 | self.thing_queue.task_done() | |
218 | break | |
219 | logging.info("Handling id {}".format(thing_id)) | |
ae598d73 | 220 | Thing(thing_id).download(self.download_directory, self.compress) |
6a777954 OM |
221 | self.thing_queue.task_done() |
222 | return | |
223 | ||
7b84ba6d | 224 | |
6a777954 OM |
225 | |
226 | ||
dbdb1782 | 227 | |
3522a3bf | 228 | class Grouping: |
d66f1f78 | 229 | """ Holds details of a group of things for download |
3c82f75b OM |
230 | This is effectively (although not actually) an abstract class |
231 | - use Collection or Designs instead. | |
232 | """ | |
dbdb1782 | 233 | |
ae598d73 | 234 | def __init__(self, quick, compress): |
975060c9 OM |
235 | self.things = [] |
236 | self.total = 0 | |
237 | self.req_id = None | |
238 | self.last_page = 0 | |
239 | self.per_page = None | |
7b84ba6d OM |
240 | # Should we stop downloading when we hit a known datestamp? |
241 | self.quick = quick | |
ae598d73 | 242 | self.compress = compress |
948bd56f | 243 | # These should be set by child classes. |
3522a3bf OM |
244 | self.url = None |
245 | self.download_dir = None | |
948bd56f | 246 | self.collection_url = None |
975060c9 | 247 | |
3522a3bf OM |
248 | def _get_small_grouping(self, req): |
249 | """ Handle small groupings """ | |
975060c9 | 250 | soup = BeautifulSoup(req.text, features='lxml') |
dbdb1782 | 251 | links = soup.find_all('a', {'class': 'card-img-holder'}) |
975060c9 | 252 | self.things = [x['href'].split(':')[1] for x in links] |
fa2f3251 | 253 | self.total = len(self.things) |
975060c9 OM |
254 | |
255 | return self.things | |
256 | ||
3522a3bf OM |
257 | def get(self): |
258 | """ retrieve the things of the grouping. """ | |
975060c9 OM |
259 | if self.things: |
260 | # We've already done it. | |
261 | return self.things | |
262 | ||
3522a3bf OM |
263 | # Check for initialisation: |
264 | if not self.url: | |
fa2f3251 | 265 | logging.error("No URL set - object not initialised properly?") |
3522a3bf OM |
266 | raise ValueError("No URL set - object not initialised properly?") |
267 | ||
268 | # Get the internal details of the grouping. | |
fa2f3251 | 269 | logging.debug("Querying {}".format(self.url)) |
3522a3bf | 270 | c_req = requests.get(self.url) |
975060c9 OM |
271 | total = TOTAL_REGEX.search(c_req.text) |
272 | if total is None: | |
3522a3bf OM |
273 | # This is a small (<13) items grouping. Pull the list from this req. |
274 | return self._get_small_grouping(c_req) | |
975060c9 OM |
275 | self.total = total.groups()[0] |
276 | self.req_id = ID_REGEX.search(c_req.text).groups()[0] | |
277 | self.last_page = int(LAST_PAGE_REGEX.search(c_req.text).groups()[0]) | |
278 | self.per_page = PER_PAGE_REGEX.search(c_req.text).groups()[0] | |
279 | parameters = { | |
dbdb1782 OM |
280 | 'base_url': self.url, |
281 | 'page': '1', | |
282 | 'per_page': '12', | |
283 | 'id': self.req_id | |
975060c9 OM |
284 | } |
285 | for current_page in range(1, self.last_page + 1): | |
286 | parameters['page'] = current_page | |
948bd56f | 287 | req = requests.post(self.collection_url, parameters) |
975060c9 | 288 | soup = BeautifulSoup(req.text, features='lxml') |
dbdb1782 | 289 | links = soup.find_all('a', {'class': 'card-img-holder'}) |
975060c9 OM |
290 | self.things += [x['href'].split(':')[1] for x in links] |
291 | ||
292 | return self.things | |
293 | ||
294 | def download(self): | |
295 | """ Downloads all the files in a collection """ | |
296 | if not self.things: | |
3522a3bf OM |
297 | self.get() |
298 | ||
299 | if not self.download_dir: | |
dbdb1782 OM |
300 | raise ValueError( |
301 | "No download_dir set - invalidly initialised object?") | |
3522a3bf | 302 | |
975060c9 | 303 | base_dir = os.getcwd() |
975060c9 | 304 | try: |
3522a3bf | 305 | os.mkdir(self.download_dir) |
975060c9 | 306 | except FileExistsError: |
fa2f3251 | 307 | logging.info("Target directory {} already exists. Assuming a resume." |
dbdb1782 | 308 | .format(self.download_dir)) |
fa2f3251 | 309 | logging.info("Downloading {} thing(s).".format(self.total)) |
dbdb1782 | 310 | for idx, thing in enumerate(self.things): |
fb28c59b | 311 | logging.info("Downloading thing {} - {}".format(idx, thing)) |
ae598d73 | 312 | RC = Thing(thing).download(self.download_dir, self.compress) |
7b84ba6d OM |
313 | if self.quick and RC==State.ALREADY_DOWNLOADED: |
314 | logging.info("Caught up, stopping.") | |
315 | return | |
975060c9 | 316 | |
dbdb1782 | 317 | |
ae598d73 OM |
318 | |
319 | ||
320 | ||
3522a3bf OM |
321 | class Collection(Grouping): |
322 | """ Holds details of a collection. """ | |
dbdb1782 | 323 | |
ae598d73 OM |
324 | def __init__(self, user, name, directory, quick, compress): |
325 | Grouping.__init__(self, quick, compress) | |
3522a3bf OM |
326 | self.user = user |
327 | self.name = name | |
3c82f75b OM |
328 | self.url = "{}/{}/collections/{}".format( |
329 | URL_BASE, self.user, strip_ws(self.name)) | |
d66f1f78 | 330 | self.download_dir = os.path.join(directory, |
3c82f75b | 331 | "{}-{}".format(slugify(self.user), slugify(self.name))) |
948bd56f | 332 | self.collection_url = URL_COLLECTION |
3522a3bf | 333 | |
dbdb1782 | 334 | |
3522a3bf OM |
335 | class Designs(Grouping): |
336 | """ Holds details of all of a users' designs. """ | |
dbdb1782 | 337 | |
ae598d73 OM |
338 | def __init__(self, user, directory, quick, compress): |
339 | Grouping.__init__(self, quick, compress) | |
3522a3bf OM |
340 | self.user = user |
341 | self.url = "{}/{}/designs".format(URL_BASE, self.user) | |
dbdb1782 OM |
342 | self.download_dir = os.path.join( |
343 | directory, "{} designs".format(slugify(self.user))) | |
948bd56f | 344 | self.collection_url = USER_COLLECTION |
975060c9 | 345 | |
dbdb1782 | 346 | |
3c82f75b OM |
347 | class Thing: |
348 | """ An individual design on thingiverse. """ | |
dbdb1782 | 349 | |
3c82f75b OM |
350 | def __init__(self, thing_id): |
351 | self.thing_id = thing_id | |
352 | self.last_time = None | |
353 | self._parsed = False | |
354 | self._needs_download = True | |
355 | self.text = None | |
356 | self.title = None | |
357 | self.download_dir = None | |
ae598d73 OM |
358 | self.time_stamp = None |
359 | self._file_links = FileLinks() | |
975060c9 | 360 | |
3c82f75b OM |
361 | def _parse(self, base_dir): |
362 | """ Work out what, if anything needs to be done. """ | |
363 | if self._parsed: | |
364 | return | |
e36c2a07 | 365 | |
3c82f75b | 366 | url = "{}/thing:{}/files".format(URL_BASE, self.thing_id) |
e0e69fc6 | 367 | try: |
b497d705 | 368 | BROWSER.get(url) |
fb28c59b | 369 | wait = WebDriverWait(BROWSER, 60) |
b497d705 OM |
370 | pc = PageChecker() |
371 | wait.until(pc) | |
e0e69fc6 | 372 | except requests.exceptions.ConnectionError as error: |
8cdd1b54 OM |
373 | logging.error("Unable to connect for thing {}: {}".format( |
374 | self.thing_id, error)) | |
375 | return | |
fb28c59b OM |
376 | except selenium.common.exceptions.TimeoutException: |
377 | logging.error(pc.log) | |
378 | logging.error("Timeout trying to parse thing {}".format(self.thing_id)) | |
379 | return | |
e0e69fc6 | 380 | |
b497d705 | 381 | self.title = pc.title |
247c2cd5 OM |
382 | if not pc.files: |
383 | logging.error("No files found for thing {} - probably thingiverse being broken, try again later".format(self.thing_id)) | |
b497d705 | 384 | for link in pc.files: |
fb28c59b | 385 | logging.debug("Parsing link: {}".format(link.text)) |
b497d705 | 386 | link_link = link.find_element_by_xpath(".//a").get_attribute("href") |
fb28c59b OM |
387 | if link_link.endswith("/zip"): |
388 | # bulk link. | |
389 | continue | |
390 | try: | |
391 | link_title, link_details, _ = link.text.split("\n") | |
392 | except ValueError: | |
393 | # If it is a filetype that doesn't generate a picture, then we get an extra field at the start. | |
394 | _, link_title, link_details, _ = link.text.split("\n") | |
395 | ||
396 | #link_details will be something like '461 kb | Updated 06-11-2019 | 373 Downloads' | |
397 | #need to convert from M D Y to Y M D | |
1267e583 | 398 | link_date = [int(x) for x in link_details.split("|")[1].split()[-1].split("-")] |
fb28c59b | 399 | try: |
d194b140 | 400 | self._file_links.append(FileLink(link_title, datetime.datetime(link_date[2], link_date[0], link_date[1]), link_link)) |
fb28c59b OM |
401 | except ValueError: |
402 | logging.error(link_date) | |
403 | ||
404 | self._image_links=[x.find_element_by_xpath(".//img").get_attribute("src") for x in pc.images] | |
d194b140 | 405 | self._license = pc.license |
fb28c59b | 406 | self.pc = pc |
e0e69fc6 | 407 | |
e0e69fc6 | 408 | |
8ed15058 OM |
409 | self.slug = "{} - {}".format(self.thing_id, slugify(self.title)) |
410 | self.download_dir = os.path.join(base_dir, self.slug) | |
411 | ||
412 | self._handle_old_directory(base_dir) | |
3c82f75b | 413 | |
fa2f3251 | 414 | logging.debug("Parsing {} ({})".format(self.thing_id, self.title)) |
8ed15058 | 415 | latest, self.last_time = self._find_last_download(base_dir) |
fa2f3251 | 416 | |
8ed15058 | 417 | if not latest: |
3b497b1a M |
418 | # Not yet downloaded |
419 | self._parsed = True | |
420 | return | |
3c82f75b | 421 | |
3c82f75b | 422 | |
8ed15058 | 423 | logging.info("last downloaded version: {}".format(self.last_time)) |
3c82f75b OM |
424 | |
425 | # OK, so we have a timestamp, lets see if there is anything new to get | |
ae598d73 OM |
426 | try: |
427 | if self._file_links.last_update > self.last_time: | |
dbdb1782 | 428 | logging.info( |
ae598d73 | 429 | "Found new/updated files {}".format(self._file_links.last_update)) |
3c82f75b OM |
430 | self._needs_download = True |
431 | self._parsed = True | |
432 | return | |
ae598d73 OM |
433 | except TypeError: |
434 | logging.warning("No files found for {}.".format(self.thing_id)) | |
b497d705 | 435 | |
3c82f75b | 436 | # Got here, so nope, no new files. |
3c82f75b OM |
437 | self._needs_download = False |
438 | self._parsed = True | |
439 | ||
8ed15058 OM |
440 | def _handle_old_directory(self, base_dir): |
441 | """ Deal with any old directories from previous versions of the code. | |
442 | """ | |
443 | old_dir = os.path.join(base_dir, slugify(self.title)) | |
444 | if os.path.exists(old_dir): | |
445 | logging.warning("Found old style download_dir. Moving.") | |
446 | rename_unique(old_dir, self.download_dir) | |
447 | ||
448 | def _handle_outdated_directory(self, base_dir): | |
449 | """ Move the current download directory sideways if the thing has changed. | |
450 | """ | |
451 | if not os.path.exists(self.download_dir): | |
452 | # No old directory to move. | |
453 | return None | |
454 | timestamp_file = os.path.join(self.download_dir, TIMESTAMP_FILE) | |
455 | if not os.path.exists(timestamp_file): | |
456 | # Old form of download directory | |
457 | target_dir_name = "{} - old".format(self.download_dir) | |
458 | else: | |
3ac180ed | 459 | target_dir_name = "{} - {}".format(self.download_dir, self.last_time.strftime(SAFE_DATETIME_FORMAT)) |
8ed15058 OM |
460 | return rename_unique(self.download_dir, target_dir_name) |
461 | ||
462 | def _find_last_download(self, base_dir): | |
463 | """ Look for the most recent previous download (if any) of the thing. | |
464 | """ | |
465 | logging.info("Looking for old things") | |
466 | ||
467 | # First the DL directory itself. | |
468 | timestamp_file = os.path.join(self.download_dir, TIMESTAMP_FILE) | |
469 | ||
470 | latest = None | |
471 | latest_time = None | |
472 | ||
473 | try: | |
474 | logging.debug("Checking for existing download in normal place.") | |
475 | with open(timestamp_file) as ts_fh: | |
476 | timestamp_text = ts_fh.read().strip() | |
477 | latest_time = datetime.datetime.strptime(timestamp_text, DEFAULT_DATETIME_FORMAT) | |
478 | latest = self.download_dir | |
479 | except FileNotFoundError: | |
480 | # No existing download directory. huh. | |
481 | pass | |
482 | except TypeError: | |
483 | logging.warning("Invalid timestamp file found in {}".format(self.download_dir)) | |
484 | ||
485 | # TODO: Maybe look for old download directories. | |
486 | ||
487 | ||
488 | # Now look for 7z files | |
489 | candidates = glob.glob(os.path.join(base_dir, "{}*.7z".format(self.thing_id))) | |
490 | # +3 to allow for ' - ' | |
491 | leading_length =len(self.slug)+3 | |
492 | for path in candidates: | |
493 | candidate = os.path.basename(path) | |
494 | try: | |
495 | logging.debug("Examining '{}' - '{}'".format(candidate, candidate[leading_length:-3])) | |
3ac180ed | 496 | candidate_time = datetime.datetime.strptime(candidate[leading_length:-3], SAFE_DATETIME_FORMAT) |
8ed15058 OM |
497 | except ValueError: |
498 | logging.warning("There was an error finding the date in {}. Ignoring.".format(candidate)) | |
499 | continue | |
500 | try: | |
501 | if candidate_time > latest_time: | |
502 | latest_time = candidate_time | |
503 | latest = candidate | |
504 | except TypeError: | |
505 | latest_time = candidate_time | |
506 | latest = candidate | |
507 | logging.info("Found last old thing: {} / {}".format(latest,latest_time)) | |
508 | return (latest, latest_time) | |
509 | ||
510 | ||
511 | ||
ae598d73 | 512 | def download(self, base_dir, compress): |
7b84ba6d OM |
513 | """ Download all files for a given thing. |
514 | Returns True iff the thing is now downloaded (not iff it downloads the thing!) | |
515 | """ | |
3c82f75b OM |
516 | if not self._parsed: |
517 | self._parse(base_dir) | |
518 | ||
e0e69fc6 | 519 | if not self._parsed: |
8cdd1b54 OM |
520 | logging.error( |
521 | "Unable to parse {} - aborting download".format(self.thing_id)) | |
7b84ba6d | 522 | return State.FAILED |
e0e69fc6 | 523 | |
3c82f75b | 524 | if not self._needs_download: |
7b84ba6d OM |
525 | print("{} - {} already downloaded - skipping.".format(self.thing_id, self.title)) |
526 | return State.ALREADY_DOWNLOADED | |
3c82f75b | 527 | |
247c2cd5 OM |
528 | if not self._file_links: |
529 | print("{} - {} appears to have no files. Thingiverse acting up again?".format(self.thing_id, self.title)) | |
530 | return State.FAILED | |
531 | ||
3c82f75b | 532 | # Have we already downloaded some things? |
8ed15058 | 533 | renamed_dir = self._handle_outdated_directory(base_dir) |
3c82f75b OM |
534 | |
535 | # Get the list of files to download | |
3c82f75b OM |
536 | |
537 | new_file_links = [] | |
538 | old_file_links = [] | |
ae598d73 | 539 | self.time_stamp = None |
3c82f75b OM |
540 | |
541 | if not self.last_time: | |
542 | # If we don't have anything to copy from, then it is all new. | |
b497d705 OM |
543 | logging.debug("No last time, downloading all files") |
544 | new_file_links = self._file_links | |
ae598d73 | 545 | self.time_stamp = new_file_links[0].last_update |
b497d705 OM |
546 | |
547 | for file_link in new_file_links: | |
ae598d73 OM |
548 | self.time_stamp = max(self.time_stamp, file_link.last_update) |
549 | logging.debug("New timestamp will be {}".format(self.time_stamp)) | |
3c82f75b | 550 | else: |
ae598d73 | 551 | self.time_stamp = self.last_time |
b497d705 OM |
552 | for file_link in self._file_links: |
553 | if file_link.last_update > self.last_time: | |
3c82f75b | 554 | new_file_links.append(file_link) |
ae598d73 | 555 | self.time_stamp = max(self.time_stamp, file_link.last_update) |
3c82f75b OM |
556 | else: |
557 | old_file_links.append(file_link) | |
3c82f75b | 558 | |
ae598d73 | 559 | logging.debug("new timestamp {}".format(self.time_stamp)) |
3c82f75b OM |
560 | |
561 | # OK. Time to get to work. | |
fa2f3251 | 562 | logging.debug("Generating download_dir") |
3c82f75b | 563 | os.mkdir(self.download_dir) |
b497d705 | 564 | filelist_file = os.path.join(self.download_dir, "filelist.txt") |
d194b140 | 565 | with open(filelist_file, 'w', encoding="utf-8") as fl_handle: |
b497d705 OM |
566 | for fl in self._file_links: |
567 | base_link = fl.link | |
568 | try: | |
569 | fl.link=requests.get(fl.link, allow_redirects=False).headers['location'] | |
fb28c59b OM |
570 | except Exception: |
571 | # Sometimes Thingiverse just gives us the direct link the first time. Not sure why. | |
572 | pass | |
b497d705 | 573 | |
fb28c59b | 574 | fl_handle.write("{},{},{}, {}\n".format(fl.link, fl.name, fl.last_update, base_link)) |
b497d705 OM |
575 | |
576 | ||
3c82f75b | 577 | # First grab the cached files (if any) |
fa2f3251 | 578 | logging.info("Copying {} unchanged files.".format(len(old_file_links))) |
3c82f75b | 579 | for file_link in old_file_links: |
8ed15058 | 580 | old_file = os.path.join(renamed_dir, file_link.name) |
65bd8b43 | 581 | new_file = truncate_name(os.path.join(self.download_dir, file_link.name)) |
3c82f75b | 582 | try: |
fa2f3251 | 583 | logging.debug("Copying {} to {}".format(old_file, new_file)) |
3c82f75b OM |
584 | copyfile(old_file, new_file) |
585 | except FileNotFoundError: | |
dbdb1782 OM |
586 | logging.warning( |
587 | "Unable to find {} in old archive, redownloading".format(file_link["title"])) | |
3c82f75b OM |
588 | new_file_links.append(file_link) |
589 | ||
590 | # Now download the new ones | |
dbdb1782 | 591 | logging.info("Downloading {} new files of {}".format( |
b497d705 | 592 | len(new_file_links), len(self._file_links))) |
3c82f75b | 593 | try: |
b497d705 | 594 | for file_link in new_file_links: |
65bd8b43 | 595 | file_name = truncate_name(os.path.join(self.download_dir, file_link.name)) |
dbdb1782 | 596 | logging.debug("Downloading {} from {} to {}".format( |
b497d705 OM |
597 | file_link.name, file_link.link, file_name)) |
598 | data_req = requests.get(file_link.link) | |
3c82f75b OM |
599 | with open(file_name, 'wb') as handle: |
600 | handle.write(data_req.content) | |
601 | except Exception as exception: | |
b497d705 | 602 | logging.error("Failed to download {} - {}".format(file_link.name, exception)) |
65bd8b43 | 603 | fail_dir(self.download_dir) |
7b84ba6d | 604 | return State.FAILED |
3c82f75b | 605 | |
b497d705 | 606 | |
b497d705 | 607 | # People like images. But this doesn't work yet. |
680039fe | 608 | image_dir = os.path.join(self.download_dir, 'images') |
fb28c59b | 609 | logging.info("Downloading {} images.".format(len(self._image_links))) |
680039fe OM |
610 | try: |
611 | os.mkdir(image_dir) | |
fb28c59b OM |
612 | for imagelink in self._image_links: |
613 | filename = os.path.basename(imagelink) | |
680039fe OM |
614 | if filename.endswith('stl'): |
615 | filename = "{}.png".format(filename) | |
fb28c59b | 616 | image_req = requests.get(imagelink) |
65bd8b43 | 617 | with open(truncate_name(os.path.join(image_dir, filename)), 'wb') as handle: |
680039fe OM |
618 | handle.write(image_req.content) |
619 | except Exception as exception: | |
620 | print("Failed to download {} - {}".format(filename, exception)) | |
65bd8b43 | 621 | fail_dir(self.download_dir) |
7b84ba6d | 622 | return State.FAILED |
680039fe | 623 | |
fb28c59b | 624 | """ |
4f75dd69 OM |
625 | # instructions are good too. |
626 | logging.info("Downloading readme") | |
627 | try: | |
8cdd1b54 OM |
628 | readme_txt = soup.find('meta', property='og:description')[ |
629 | 'content'] | |
630 | with open(os.path.join(self.download_dir, 'readme.txt'), 'w') as readme_handle: | |
4f75dd69 OM |
631 | readme_handle.write("{}\n".format(readme_txt)) |
632 | except (TypeError, KeyError) as exception: | |
633 | logging.warning("No readme? {}".format(exception)) | |
634 | except IOError as exception: | |
635 | logging.warning("Failed to write readme! {}".format(exception)) | |
636 | ||
fb28c59b | 637 | """ |
4f75dd69 OM |
638 | # Best get some licenses |
639 | logging.info("Downloading license") | |
640 | try: | |
fb28c59b | 641 | if self._license: |
65bd8b43 | 642 | with open(truncate_name(os.path.join(self.download_dir, 'license.txt')), 'w', encoding="utf-8") as license_handle: |
fb28c59b | 643 | license_handle.write("{}\n".format(self._license)) |
4f75dd69 OM |
644 | except IOError as exception: |
645 | logging.warning("Failed to write license! {}".format(exception)) | |
fb28c59b | 646 | |
3c82f75b OM |
647 | try: |
648 | # Now write the timestamp | |
8ed15058 | 649 | with open(os.path.join(self.download_dir,TIMESTAMP_FILE), 'w', encoding="utf-8") as timestamp_handle: |
ae598d73 | 650 | timestamp_handle.write(self.time_stamp.__str__()) |
3c82f75b OM |
651 | except Exception as exception: |
652 | print("Failed to write timestamp file - {}".format(exception)) | |
65bd8b43 | 653 | fail_dir(self.download_dir) |
7b84ba6d | 654 | return State.FAILED |
3c82f75b | 655 | self._needs_download = False |
fa2f3251 | 656 | logging.debug("Download of {} finished".format(self.title)) |
ae598d73 OM |
657 | if not compress: |
658 | return State.OK | |
659 | ||
660 | ||
661 | thing_dir = "{} - {} - {}".format(self.thing_id, | |
662 | slugify(self.title), | |
3ac180ed | 663 | self.time_stamp.strftime(SAFE_DATETIME_FORMAT)) |
ae598d73 OM |
664 | file_name = os.path.join(base_dir, |
665 | "{}.7z".format(thing_dir)) | |
666 | logging.debug("Compressing {} to {}".format( | |
667 | self.title, | |
668 | file_name)) | |
ae598d73 | 669 | with py7zr.SevenZipFile(file_name, 'w', filters=SEVENZIP_FILTERS) as archive: |
ae598d73 OM |
670 | archive.writeall(self.download_dir, thing_dir) |
671 | logging.debug("Compression of {} finished.".format(self.title)) | |
8ed15058 OM |
672 | shutil.rmtree(self.download_dir) |
673 | logging.debug("Removed temporary download dir of {}.".format(self.title)) | |
7b84ba6d | 674 | return State.OK |
975060c9 | 675 | |
dbdb1782 | 676 | |
ae598d73 OM |
677 | |
678 | ||
679 | def do_batch(batch_file, download_dir, quick, compress): | |
1ab49020 OM |
680 | """ Read a file in line by line, parsing each as a set of calls to this script.""" |
681 | with open(batch_file) as handle: | |
682 | for line in handle: | |
683 | line = line.strip() | |
cf280385 M |
684 | if not line: |
685 | # Skip empty lines | |
686 | continue | |
1ab49020 OM |
687 | logging.info("Handling instruction {}".format(line)) |
688 | command_arr = line.split() | |
689 | if command_arr[0] == "thing": | |
dbdb1782 OM |
690 | logging.debug( |
691 | "Handling batch thing instruction: {}".format(line)) | |
ae598d73 | 692 | Thing(command_arr[1]).download(download_dir, compress) |
1ab49020 OM |
693 | continue |
694 | if command_arr[0] == "collection": | |
dbdb1782 OM |
695 | logging.debug( |
696 | "Handling batch collection instruction: {}".format(line)) | |
697 | Collection(command_arr[1], command_arr[2], | |
ae598d73 | 698 | download_dir, quick, compress).download() |
1ab49020 OM |
699 | continue |
700 | if command_arr[0] == "user": | |
dbdb1782 OM |
701 | logging.debug( |
702 | "Handling batch collection instruction: {}".format(line)) | |
ae598d73 | 703 | Designs(command_arr[1], download_dir, quick, compress).download() |
1ab49020 OM |
704 | continue |
705 | logging.warning("Unable to parse current instruction. Skipping.") | |
706 | ||
dbdb1782 | 707 | |
975060c9 OM |
708 | def main(): |
709 | """ Entry point for script being run as a command. """ | |
710 | parser = argparse.ArgumentParser() | |
dbdb1782 OM |
711 | parser.add_argument("-l", "--log-level", choices=[ |
712 | 'debug', 'info', 'warning'], default='info', help="level of logging desired") | |
713 | parser.add_argument("-d", "--directory", | |
714 | help="Target directory to download into") | |
4f94efc8 OM |
715 | parser.add_argument("-f", "--log-file", |
716 | help="Place to log debug information to") | |
7b84ba6d OM |
717 | parser.add_argument("-q", "--quick", action="store_true", |
718 | help="Assume date ordering on posts") | |
ae598d73 OM |
719 | parser.add_argument("-c", "--compress", action="store_true", |
720 | help="Compress files") | |
721 | ||
7b84ba6d | 722 | |
dbdb1782 OM |
723 | subparsers = parser.add_subparsers( |
724 | help="Type of thing to download", dest="subcommand") | |
725 | collection_parser = subparsers.add_parser( | |
b7bfef68 | 726 | 'collection', help="Download one or more entire collection(s)") |
dbdb1782 | 727 | collection_parser.add_argument( |
b7bfef68 | 728 | "owner", help="The owner of the collection(s) to get") |
dbdb1782 | 729 | collection_parser.add_argument( |
b7bfef68 | 730 | "collections", nargs="+", help="Space seperated list of the name(s) of collection to get") |
dbdb1782 OM |
731 | thing_parser = subparsers.add_parser( |
732 | 'thing', help="Download a single thing.") | |
8cdd1b54 OM |
733 | thing_parser.add_argument( |
734 | "things", nargs="*", help="Space seperated list of thing ID(s) to download") | |
dbdb1782 | 735 | user_parser = subparsers.add_parser( |
b7bfef68 | 736 | "user", help="Download all things by one or more users") |
8cdd1b54 OM |
737 | user_parser.add_argument( |
738 | "users", nargs="+", help="A space seperated list of the user(s) to get the designs of") | |
dbdb1782 OM |
739 | batch_parser = subparsers.add_parser( |
740 | "batch", help="Perform multiple actions written in a text file") | |
741 | batch_parser.add_argument( | |
742 | "batch_file", help="The name of the file to read.") | |
680039fe | 743 | subparsers.add_parser("version", help="Show the current version") |
4a98996b | 744 | |
975060c9 | 745 | args = parser.parse_args() |
4a98996b OM |
746 | if not args.subcommand: |
747 | parser.print_help() | |
748 | sys.exit(1) | |
d66f1f78 OM |
749 | if not args.directory: |
750 | args.directory = os.getcwd() | |
4f94efc8 OM |
751 | |
752 | logger = logging.getLogger() | |
753 | formatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s') | |
754 | logger.setLevel(logging.DEBUG) | |
755 | console_handler = logging.StreamHandler() | |
756 | console_handler.setLevel(args.log_level.upper()) | |
757 | ||
758 | logger.addHandler(console_handler) | |
759 | if args.log_file: | |
760 | file_handler = logging.FileHandler(args.log_file) | |
761 | file_handler.setLevel(logging.DEBUG) | |
762 | file_handler.setFormatter(formatter) | |
763 | logger.addHandler(file_handler) | |
fa2f3251 | 764 | |
6a777954 OM |
765 | |
766 | # Start downloader | |
767 | thing_queue = multiprocessing.JoinableQueue() | |
768 | logging.debug("starting {} downloader(s)".format(DOWNLOADER_COUNT)) | |
ae598d73 | 769 | downloaders = [Downloader(thing_queue, args.directory, args.compress) for _ in range(DOWNLOADER_COUNT)] |
6a777954 OM |
770 | for downloader in downloaders: |
771 | downloader.start() | |
772 | ||
773 | ||
4a98996b | 774 | if args.subcommand.startswith("collection"): |
b7bfef68 | 775 | for collection in args.collections: |
ae598d73 | 776 | Collection(args.owner, collection, args.directory, args.quick, args.compress).download() |
4a98996b | 777 | if args.subcommand == "thing": |
b7bfef68 | 778 | for thing in args.things: |
6a777954 | 779 | thing_queue.put(thing) |
3522a3bf | 780 | if args.subcommand == "user": |
b7bfef68 | 781 | for user in args.users: |
ae598d73 | 782 | Designs(user, args.directory, args.quick, args.compress).download() |
db8066ec OM |
783 | if args.subcommand == "version": |
784 | print("thingy_grabber.py version {}".format(VERSION)) | |
1ab49020 | 785 | if args.subcommand == "batch": |
ae598d73 | 786 | do_batch(args.batch_file, args.directory, args.quick, args.compress) |
1ab49020 | 787 | |
6a777954 OM |
788 | # Stop the downloader processes |
789 | for downloader in downloaders: | |
790 | thing_queue.put(None) | |
975060c9 | 791 | |
d194b140 OM |
792 | atexit.register(BROWSER.quit) |
793 | ||
0930777e OM |
794 | if __name__ == "__main__": |
795 | multiprocessing.freeze_support() | |
975060c9 | 796 | main() |