3 Thingiverse bulk downloader
13 import multiprocessing
15 from shutil
import copyfile
16 from bs4
import BeautifulSoup
18 URL_BASE
= "https://www.thingiverse.com"
19 URL_COLLECTION
= URL_BASE
+ "/ajax/thingcollection/list_collected_things"
20 USER_COLLECTION
= URL_BASE
+ "/ajax/user/designs"
22 ID_REGEX
= re
.compile(r
'"id":(\d*),')
23 TOTAL_REGEX
= re
.compile(r
'"total":(\d*),')
24 LAST_PAGE_REGEX
= re
.compile(r
'"last_page":(\d*),')
25 # This appears to be fixed at 12, but if it changes would screw the rest up.
26 PER_PAGE_REGEX
= re
.compile(r
'"per_page":(\d*),')
27 NO_WHITESPACE_REGEX
= re
.compile(r
'[-\s]+')
34 class State(enum
.Enum
):
37 ALREADY_DOWNLOADED
= enum
.auto()
41 """ Remove whitespace from a string """
42 return str(NO_WHITESPACE_REGEX
.sub('-', value
))
47 Normalizes string, converts to lowercase, removes non-alpha characters,
48 and converts spaces to hyphens.
50 value
= unicodedata
.normalize('NFKD', value
).encode(
51 'ascii', 'ignore').decode()
52 value
= str(re
.sub(r
'[^\w\s-]', '', value
).strip())
53 value
= str(NO_WHITESPACE_REGEX
.sub('-', value
))
56 class Downloader(multiprocessing
.Process
):
58 Class to handle downloading the things we have found to get.
61 def __init__(self
, thing_queue
, download_directory
):
62 multiprocessing
.Process
.__init
__(self
)
63 # TODO: add parameters
64 self
.thing_queue
= thing_queue
65 self
.download_directory
= download_directory
68 """ actual download loop.
71 thing_id
= self
.thing_queue
.get()
73 logging
.info("Shutting download queue")
74 self
.thing_queue
.task_done()
76 logging
.info("Handling id {}".format(thing_id
))
77 Thing(thing_id
).download(self
.download_directory
)
78 self
.thing_queue
.task_done()
86 """ Holds details of a group of things for download
87 This is effectively (although not actually) an abstract class
88 - use Collection or Designs instead.
91 def __init__(self
, quick
):
97 # Should we stop downloading when we hit a known datestamp?
99 # These should be set by child classes.
101 self
.download_dir
= None
102 self
.collection_url
= None
104 def _get_small_grouping(self
, req
):
105 """ Handle small groupings """
106 soup
= BeautifulSoup(req
.text
, features
='lxml')
107 links
= soup
.find_all('a', {'class': 'card-img-holder'})
108 self
.things
= [x
['href'].split(':')[1] for x
in links
]
109 self
.total
= len(self
.things
)
114 """ retrieve the things of the grouping. """
116 # We've already done it.
119 # Check for initialisation:
121 logging
.error("No URL set - object not initialised properly?")
122 raise ValueError("No URL set - object not initialised properly?")
124 # Get the internal details of the grouping.
125 logging
.debug("Querying {}".format(self
.url
))
126 c_req
= requests
.get(self
.url
)
127 total
= TOTAL_REGEX
.search(c_req
.text
)
129 # This is a small (<13) items grouping. Pull the list from this req.
130 return self
._get
_small
_grouping
(c_req
)
131 self
.total
= total
.groups()[0]
132 self
.req_id
= ID_REGEX
.search(c_req
.text
).groups()[0]
133 self
.last_page
= int(LAST_PAGE_REGEX
.search(c_req
.text
).groups()[0])
134 self
.per_page
= PER_PAGE_REGEX
.search(c_req
.text
).groups()[0]
136 'base_url': self
.url
,
141 for current_page
in range(1, self
.last_page
+ 1):
142 parameters
['page'] = current_page
143 req
= requests
.post(self
.collection_url
, parameters
)
144 soup
= BeautifulSoup(req
.text
, features
='lxml')
145 links
= soup
.find_all('a', {'class': 'card-img-holder'})
146 self
.things
+= [x
['href'].split(':')[1] for x
in links
]
151 """ Downloads all the files in a collection """
155 if not self
.download_dir
:
157 "No download_dir set - invalidly initialised object?")
159 base_dir
= os
.getcwd()
161 os
.mkdir(self
.download_dir
)
162 except FileExistsError
:
163 logging
.info("Target directory {} already exists. Assuming a resume."
164 .format(self
.download_dir
))
165 logging
.info("Downloading {} thing(s).".format(self
.total
))
166 for idx
, thing
in enumerate(self
.things
):
167 logging
.info("Downloading thing {}".format(idx
))
168 RC
= Thing(thing
).download(self
.download_dir
)
169 if self
.quick
and RC
==State
.ALREADY_DOWNLOADED
:
170 logging
.info("Caught up, stopping.")
174 class Collection(Grouping
):
175 """ Holds details of a collection. """
177 def __init__(self
, user
, name
, directory
, quick
):
178 Grouping
.__init
__(self
, quick
)
181 self
.url
= "{}/{}/collections/{}".format(
182 URL_BASE
, self
.user
, strip_ws(self
.name
))
183 self
.download_dir
= os
.path
.join(directory
,
184 "{}-{}".format(slugify(self
.user
), slugify(self
.name
)))
185 self
.collection_url
= URL_COLLECTION
188 class Designs(Grouping
):
189 """ Holds details of all of a users' designs. """
191 def __init__(self
, user
, directory
, quick
):
192 Grouping
.__init
__(self
, quick
)
194 self
.url
= "{}/{}/designs".format(URL_BASE
, self
.user
)
195 self
.download_dir
= os
.path
.join(
196 directory
, "{} designs".format(slugify(self
.user
)))
197 self
.collection_url
= USER_COLLECTION
201 """ An individual design on thingiverse. """
203 def __init__(self
, thing_id
):
204 self
.thing_id
= thing_id
205 self
.last_time
= None
207 self
._needs
_download
= True
210 self
.download_dir
= None
212 def _parse(self
, base_dir
):
213 """ Work out what, if anything needs to be done. """
217 url
= "{}/thing:{}/files".format(URL_BASE
, self
.thing_id
)
219 req
= requests
.get(url
)
220 except requests
.exceptions
.ConnectionError
as error
:
221 logging
.error("Unable to connect for thing {}: {}".format(
222 self
.thing_id
, error
))
226 soup
= BeautifulSoup(self
.text
, features
='lxml')
228 #code.interact(local=dict(globals(), **locals()))
230 self
.title
= slugify(soup
.find_all('h1')[0].text
.strip())
233 "No title found for thing {}".format(self
.thing_id
))
234 self
.title
= self
.thing_id
236 if req
.status_code
== 404:
238 "404 for thing {} - DMCA or invalid number?".format(self
.thing_id
))
241 if req
.status_code
> 299:
243 "bad status code {} for thing {} - try again later?".format(req
.status_code
, self
.thing_id
))
246 self
.download_dir
= os
.path
.join(base_dir
, self
.title
)
248 logging
.debug("Parsing {} ({})".format(self
.thing_id
, self
.title
))
250 if not os
.path
.exists(self
.download_dir
):
255 timestamp_file
= os
.path
.join(self
.download_dir
, 'timestamp.txt')
256 if not os
.path
.exists(timestamp_file
):
257 # Old download from before
259 "Old-style download directory found. Assuming update required.")
264 with
open(timestamp_file
, 'r') as timestamp_handle
:
265 self
.last_time
= timestamp_handle
.readlines()[0]
266 logging
.info("last downloaded version: {}".format(self
.last_time
))
267 except FileNotFoundError
:
268 # Not run on this thing before.
270 "Old-style download directory found. Assuming update required.")
271 self
.last_time
= None
275 # OK, so we have a timestamp, lets see if there is anything new to get
276 file_links
= soup
.find_all('a', {'class': 'file-download'})
277 for file_link
in file_links
:
278 timestamp
= file_link
.find_all('time')[0]['datetime']
279 logging
.debug("Checking {} (updated {})".format(
280 file_link
["title"], timestamp
))
281 if timestamp
> self
.last_time
:
283 "Found new/updated file {}".format(file_link
["title"]))
284 self
._needs
_download
= True
287 # Got here, so nope, no new files.
288 self
._needs
_download
= False
291 def download(self
, base_dir
):
292 """ Download all files for a given thing.
293 Returns True iff the thing is now downloaded (not iff it downloads the thing!)
296 self
._parse
(base_dir
)
300 "Unable to parse {} - aborting download".format(self
.thing_id
))
303 if not self
._needs
_download
:
304 print("{} - {} already downloaded - skipping.".format(self
.thing_id
, self
.title
))
305 return State
.ALREADY_DOWNLOADED
307 # Have we already downloaded some things?
308 timestamp_file
= os
.path
.join(self
.download_dir
, 'timestamp.txt')
310 if os
.path
.exists(self
.download_dir
):
311 if not os
.path
.exists(timestamp_file
):
312 # edge case: old style dir w/out timestamp.
314 "Old style download dir found for {}".format(self
.title
))
316 target_dir
= "{}_old".format(self
.download_dir
)
317 while os
.path
.exists(target_dir
):
318 prev_count
= prev_count
+ 1
319 target_dir
= "{}_old_{}".format(self
.download_dir
, prev_count
)
320 os
.rename(self
.download_dir
, target_dir
)
322 prev_dir
= "{}_{}".format(self
.download_dir
, slugify(self
.last_time
))
323 os
.rename(self
.download_dir
, prev_dir
)
325 # Get the list of files to download
326 soup
= BeautifulSoup(self
.text
, features
='lxml')
327 file_links
= soup
.find_all('a', {'class': 'file-download'})
333 if not self
.last_time
:
334 # If we don't have anything to copy from, then it is all new.
335 new_file_links
= file_links
337 new_last_time
= file_links
[0].find_all('time')[0]['datetime']
340 code
.interact(local
=dict(globals(), **locals()))
342 for file_link
in file_links
:
343 timestamp
= file_link
.find_all('time')[0]['datetime']
344 logging
.debug("Found file {} from {}".format(
345 file_link
["title"], timestamp
))
346 if timestamp
> new_last_time
:
347 new_last_time
= timestamp
349 for file_link
in file_links
:
350 timestamp
= file_link
.find_all('time')[0]['datetime']
351 logging
.debug("Checking {} (updated {})".format(
352 file_link
["title"], timestamp
))
353 if timestamp
> self
.last_time
:
354 new_file_links
.append(file_link
)
356 old_file_links
.append(file_link
)
357 if not new_last_time
or timestamp
> new_last_time
:
358 new_last_time
= timestamp
360 logging
.debug("new timestamp {}".format(new_last_time
))
362 # OK. Time to get to work.
363 logging
.debug("Generating download_dir")
364 os
.mkdir(self
.download_dir
)
365 # First grab the cached files (if any)
366 logging
.info("Copying {} unchanged files.".format(len(old_file_links
)))
367 for file_link
in old_file_links
:
368 old_file
= os
.path
.join(prev_dir
, file_link
["title"])
369 new_file
= os
.path
.join(self
.download_dir
, file_link
["title"])
371 logging
.debug("Copying {} to {}".format(old_file
, new_file
))
372 copyfile(old_file
, new_file
)
373 except FileNotFoundError
:
375 "Unable to find {} in old archive, redownloading".format(file_link
["title"]))
376 new_file_links
.append(file_link
)
378 # Now download the new ones
379 files
= [("{}{}".format(URL_BASE
, x
['href']), x
["title"])
380 for x
in new_file_links
]
381 logging
.info("Downloading {} new files of {}".format(
382 len(new_file_links
), len(file_links
)))
384 for url
, name
in files
:
385 file_name
= os
.path
.join(self
.download_dir
, name
)
386 logging
.debug("Downloading {} from {} to {}".format(
387 name
, url
, file_name
))
388 data_req
= requests
.get(url
)
389 with
open(file_name
, 'wb') as handle
:
390 handle
.write(data_req
.content
)
391 except Exception as exception
:
392 logging
.error("Failed to download {} - {}".format(name
, exception
))
393 os
.rename(self
.download_dir
, "{}_failed".format(self
.download_dir
))
397 image_dir
= os
.path
.join(self
.download_dir
, 'images')
398 imagelinks
= soup
.find_all('span', {'class': 'gallery-slider'})[0] \
399 .find_all('div', {'class': 'gallery-photo'})
400 logging
.info("Downloading {} images.".format(len(imagelinks
)))
403 for imagelink
in imagelinks
:
404 url
= next(filter(None, [imagelink
[x
] for x
in ['data-full',
407 'data-thumb']]), None)
410 "Unable to find any urls for {}".format(imagelink
))
413 filename
= os
.path
.basename(url
)
414 if filename
.endswith('stl'):
415 filename
= "{}.png".format(filename
)
416 image_req
= requests
.get(url
)
417 with
open(os
.path
.join(image_dir
, filename
), 'wb') as handle
:
418 handle
.write(image_req
.content
)
419 except Exception as exception
:
420 print("Failed to download {} - {}".format(filename
, exception
))
421 os
.rename(self
.download_dir
, "{}_failed".format(self
.download_dir
))
424 # instructions are good too.
425 logging
.info("Downloading readme")
427 readme_txt
= soup
.find('meta', property='og:description')[
429 with
open(os
.path
.join(self
.download_dir
, 'readme.txt'), 'w') as readme_handle
:
430 readme_handle
.write("{}\n".format(readme_txt
))
431 except (TypeError, KeyError) as exception
:
432 logging
.warning("No readme? {}".format(exception
))
433 except IOError as exception
:
434 logging
.warning("Failed to write readme! {}".format(exception
))
436 # Best get some licenses
437 logging
.info("Downloading license")
439 license_txt
= soup
.find('div', {'class': 'license-text'}).text
441 with
open(os
.path
.join(self
.download_dir
, 'license.txt'), 'w') as license_handle
:
442 license_handle
.write("{}\n".format(license_txt
))
443 except AttributeError as exception
:
444 logging
.warning("No license? {}".format(exception
))
445 except IOError as exception
:
446 logging
.warning("Failed to write license! {}".format(exception
))
449 # Now write the timestamp
450 with
open(timestamp_file
, 'w') as timestamp_handle
:
451 timestamp_handle
.write(new_last_time
)
452 except Exception as exception
:
453 print("Failed to write timestamp file - {}".format(exception
))
454 os
.rename(self
.download_dir
, "{}_failed".format(self
.download_dir
))
456 self
._needs
_download
= False
457 logging
.debug("Download of {} finished".format(self
.title
))
461 def do_batch(batch_file
, download_dir
, quick
):
462 """ Read a file in line by line, parsing each as a set of calls to this script."""
463 with
open(batch_file
) as handle
:
466 logging
.info("Handling instruction {}".format(line
))
467 command_arr
= line
.split()
468 if command_arr
[0] == "thing":
470 "Handling batch thing instruction: {}".format(line
))
471 Thing(command_arr
[1]).download(download_dir
)
473 if command_arr
[0] == "collection":
475 "Handling batch collection instruction: {}".format(line
))
476 Collection(command_arr
[1], command_arr
[2],
477 download_dir
, quick
).download()
479 if command_arr
[0] == "user":
481 "Handling batch collection instruction: {}".format(line
))
482 Designs(command_arr
[1], download_dir
, quick
).download()
484 logging
.warning("Unable to parse current instruction. Skipping.")
488 """ Entry point for script being run as a command. """
489 parser
= argparse
.ArgumentParser()
490 parser
.add_argument("-l", "--log-level", choices
=[
491 'debug', 'info', 'warning'], default
='info', help="level of logging desired")
492 parser
.add_argument("-d", "--directory",
493 help="Target directory to download into")
494 parser
.add_argument("-f", "--log-file",
495 help="Place to log debug information to")
496 parser
.add_argument("-q", "--quick", action
="store_true",
497 help="Assume date ordering on posts")
499 subparsers
= parser
.add_subparsers(
500 help="Type of thing to download", dest
="subcommand")
501 collection_parser
= subparsers
.add_parser(
502 'collection', help="Download one or more entire collection(s)")
503 collection_parser
.add_argument(
504 "owner", help="The owner of the collection(s) to get")
505 collection_parser
.add_argument(
506 "collections", nargs
="+", help="Space seperated list of the name(s) of collection to get")
507 thing_parser
= subparsers
.add_parser(
508 'thing', help="Download a single thing.")
509 thing_parser
.add_argument(
510 "things", nargs
="*", help="Space seperated list of thing ID(s) to download")
511 user_parser
= subparsers
.add_parser(
512 "user", help="Download all things by one or more users")
513 user_parser
.add_argument(
514 "users", nargs
="+", help="A space seperated list of the user(s) to get the designs of")
515 batch_parser
= subparsers
.add_parser(
516 "batch", help="Perform multiple actions written in a text file")
517 batch_parser
.add_argument(
518 "batch_file", help="The name of the file to read.")
519 subparsers
.add_parser("version", help="Show the current version")
521 args
= parser
.parse_args()
522 if not args
.subcommand
:
525 if not args
.directory
:
526 args
.directory
= os
.getcwd()
528 logger
= logging
.getLogger()
529 formatter
= logging
.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s')
530 logger
.setLevel(logging
.DEBUG
)
531 console_handler
= logging
.StreamHandler()
532 console_handler
.setLevel(args
.log_level
.upper())
534 logger
.addHandler(console_handler
)
536 file_handler
= logging
.FileHandler(args
.log_file
)
537 file_handler
.setLevel(logging
.DEBUG
)
538 file_handler
.setFormatter(formatter
)
539 logger
.addHandler(file_handler
)
543 thing_queue
= multiprocessing
.JoinableQueue()
544 logging
.debug("starting {} downloader(s)".format(DOWNLOADER_COUNT
))
545 downloaders
= [Downloader(thing_queue
, args
.directory
) for _
in range(DOWNLOADER_COUNT
)]
546 for downloader
in downloaders
:
550 if args
.subcommand
.startswith("collection"):
551 for collection
in args
.collections
:
552 Collection(args
.owner
, collection
, args
.directory
, args
.quick
).download()
553 if args
.subcommand
== "thing":
554 for thing
in args
.things
:
555 thing_queue
.put(thing
)
556 if args
.subcommand
== "user":
557 for user
in args
.users
:
558 Designs(user
, args
.directory
, args
.quick
).download()
559 if args
.subcommand
== "version":
560 print("thingy_grabber.py version {}".format(VERSION
))
561 if args
.subcommand
== "batch":
562 do_batch(args
.batch_file
, args
.directory
, args
.quick
)
564 # Stop the downloader processes
565 for downloader
in downloaders
:
566 thing_queue
.put(None)
568 if __name__
== "__main__":