-
-
Save jeffaudi/89bba20e839d99e4afab to your computer and use it in GitHub Desktop.
| import pytumblr | |
| import os | |
| import code | |
| import oauth2 as oauth | |
| from pprint import pprint | |
| import json | |
| import urllib | |
| import codecs | |
| # Number of likes to fetch in one request | |
| limit = 20 | |
| # Directory where to save the images | |
| directory = "tumblr-likes" | |
| # List of ident downloaded | |
| downloaded = [] | |
| try: | |
| # Authenticate via OAuth | |
| client = pytumblr.TumblrRestClient( | |
| 'CONSUMER_KEY', | |
| 'CONSUMER_SECRET', | |
| 'OAUTH_TOKEN', | |
| 'OAUTH_TOKEN_SECRET' | |
| ) | |
| # Get the info on the user | |
| info = client.info() | |
| # Get the content | |
| name = info["user"]["name"] | |
| number = int(info["user"]["likes"]) | |
| # Currently the Tumblr API returns no more than 1000 likes | |
| pages = min(number // limit, 50) | |
| # Display the number of likes and pages of 20 | |
| print "Tumblr user {0} has {1} likes".format(name, number) | |
| print "{0} pages will be fetched".format(pages) | |
| posts = 0 | |
| total = 0 | |
| for page in xrange(0, pages): | |
| # For testing | |
| #if page == 1: | |
| # break | |
| # Get the likes | |
| offset = page * limit | |
| likes = client.likes(offset=offset, limit=limit)["liked_posts"] | |
| # Parse the likes | |
| for liked in likes: | |
| # Only the photos | |
| if "photos" in liked: | |
| downloaded.append([liked["id"], liked["reblog_key"]]) | |
| photos = liked["photos"] | |
| count = 0 | |
| # Parse photos | |
| for photo in photos: | |
| # Get the original size | |
| url = photo["original_size"]["url"] | |
| imgname = url.split('/')[-1] | |
| # Store in a directory based on blog name | |
| blog_dir = directory + "/" + liked["blog_name"] | |
| if not os.path.isdir(blog_dir): | |
| os.mkdir(blog_dir) | |
| # Create a unique name | |
| filename = blog_dir + "/" + str(liked["id"]) + "-" | |
| # Add numbers if more than one image | |
| if count > 0: | |
| filename += str(count) + "-" | |
| filename += imgname | |
| # Check if image is already on local disk | |
| if (os.path.isfile(filename)): | |
| print "File already exists : " + imgname | |
| else: | |
| print "Downloading " + imgname + " from " + liked["blog_name"] | |
| urllib.urlretrieve(url, filename) | |
| count += 1 | |
| posts += 1 | |
| total += count | |
| elif "video_url" in liked: | |
| # Get the video name | |
| url = liked["video_url"] | |
| vidname = url.split('/')[-1] | |
| count = 0 | |
| # Create a unique name | |
| filename = directory + "/" + liked["blog_name"] + "-" + str(liked["id"]) + "-" + vidname | |
| # Check if video is already on local disk | |
| if (os.path.isfile(filename)): | |
| print "File already exists : " + vidname | |
| else: | |
| print "Downloading " + vidname + " from " + liked["blog_name"] | |
| urllib.urlretrieve(url, filename) | |
| count += 1 | |
| posts += 1 | |
| total += count | |
| elif "body" in liked: | |
| # Create a unique name | |
| filename = directory + "/" + liked["blog_name"] + "-" + str(liked["id"]) + ".htm" | |
| with codecs.open(filename, "w", "utf-8") as ds: | |
| ds.write('<!doctype html><html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"><title></title></head><body>') | |
| ds.write(liked["body"]) | |
| ds.write('</body></html>') | |
| else: | |
| # If not a photo or a video, dump the JSON | |
| with open(str(liked["id"]) + "-" + str(liked["blog_name"]) + ".json", "w") as f: | |
| json.dump(liked, f) | |
| with open("downloaded.json", "w") as f: | |
| json.dump(downloaded, f) | |
| # Display some stats | |
| print "Total posts parsed : " + str(posts) | |
| print "Total images or videos downloaded : " + str(total) | |
| except: | |
| print "Unexpected error:", sys.exc_info()[0] | |
| raw_input("Press Enter to close window...") | |
Thanks for this!!!
FYI, to get this working I had to...
-
Add
#!/usr/bin/env pythonto the start. -
Run
`which python` pip install pytumblr oauth2 pprint
-
Add
import sys. -
Manually create the directory name. (I don't know enough python to suggest a command to do this)
I would create a PR, but this is just a gist, so I'm commenting.
Checking for a directory is in the code:
if not os.path.isdir(blog_dir):
os.mkdir(blog_dir)
but it only checks for blog_dir.
Try to join paths like this: os.path.join(directory, file_name)
Hello, id really like to see this in action. Im getting the following error:
Tumblr user xxxxx has xxxxx likes
50 pages will be fetched
Unexpected error: <class 'NameError'>
Process finished with exit code 0
Im still new to Python but it appears the code gets to:
for page in xrange(0, pages):
Then moves to the except line.
Any thoughts?
Thx!
Confirmed working today 8/28/2018. Remember to fill this fields:
client = pytumblr.TumblrRestClient(
'CONSUMER_KEY',
'CONSUMER_SECRET',
'OAUTH_TOKEN',
'OAUTH_TOKEN_SECRET'
)
Hello, id really like to see this in action. Im getting the following error:
Tumblr user xxxxx has xxxxx likes 50 pages will be fetched Unexpected error: <class 'NameError'> Process finished with exit code 0Im still new to Python but it appears the code gets to:
for page in xrange(0, pages):
Then moves to the except line.Any thoughts?
Thx!
change 'xrange' to 'range' basically omit the letter x
According to the API docs you should be able to get all likes by using the before or after parameters instead of offset:
When using the offset parameter the maximum limit on the offset is 1000. If you would like to get more results than that use either before or after.
So I guess you could take the time stamp of the last like in the current request and request the next 20 likes before that timestamp.
, line 40 print "Tumblr user {0} has {1} likes".format(name, number) ^ SyntaxError: invalid syntax
On running I hit this error...
I have an issue where it worked for a while and downloaded some of my likes, probably about 1/5th. Now it says I have 6679 likes and 333 pages, but the total posts parsed and images downloaded are 0.
, line 40 print "Tumblr user {0} has {1} likes".format(name, number) ^ SyntaxError: invalid syntaxOn running I hit this error...
Me too. Very annoying.
, line 40 print "Tumblr user {0} has {1} likes".format(name, number) ^ SyntaxError: invalid syntax
On running I hit this error...Me too. Very annoying.
Comment it out like this:
#print "Tumblr user {0} has {1} likes".format(name, number)
I forked this script over at https://gist.github.com/spoonyfork/d06e524df5dbe2b547f7d0e95fc9c37d with changes to get it to run on python 3.7. I also included instructions for use.
I also had success using this program to backup even more likes: https://github.com/neuro-sys/tumblr-likes-downloader
How do we use this...?