TumblrDownloader
Enables batch download of a Tumblr blog using the official API.
Usage
Basic
- Open a terminal pointed at the script folder
- Make sure the script is executable (
chmod +x)
Then run:
./tumblrdl.php -b blog_name
Replace blog_name by the desired blog name (e.g. brainmess).
Advanced
Several options are available, for detailed help and usage examples run the following command:
./tumblrdl.php -h (or --help)
Output:
Usage: ./tumblrdl.php -b <blog_name>
-b <blog_name> (or --blog <blogname>) *required* the blog name (e.g 'brainmess')
-o=<offset> (or --offset=<offset>) *optional* the number of posts to skip before starting download (e.g. 100)
-l=<limit> (or --limit=<limit>) *optional* the number of post parsed in each run - by default 20 (e.g. 50)
-d=<path> (or --directory=<path>) *optional* the path to the download directory - by default script directory (e.g. /Users/username/Desktop)
-u (or --unlimited) *optional* a flag to tell the script to download every photo available (might take a while ^^)
-c (or --continue) *optional* a flag to tell the script to continue even if it encounters existing files
--wallfilter *optional* apply filter for only downloading photos that might be used as wallpapers (checks: landscape, ratio, dimensions)
--blacklist=<extensions> *optional* extensions of files NOT to download (separated by commas (,) with no space)
--whitelist=<extensions> *optional* extensions of files to download (separated by commas (,) with no space)
-h (or --help) prints this help
Examples:
./tumblrdl.php -b brainmess -u (download every photo available)
./tumblrdl.php -b brainmess -l=50 (download the last 50 photos)
./tumblrdl.php -b brainmess --wallfilter --whitelist=jpg,png (download photos with extension jpg or png that are fit for being wallpapers)
./tumblrdl.php --blog brainmess --offset=100 --limit=50 --directory=/Users/username/Desktop (download 50 photos on the desktop by skipping the last 100 posts)
...
Notes:
- configuration resides in the config.json file
- the 'offset' and 'limit' refers to the post count, not the photo count (!) as there may be more than one photo in a post.
- download directory path must be absolute (/Users/username/Desktop instead of ~/Desktop)
- once the script encounters an already downloaded photo (test for an existing file) it will stop (except when -c or --continue option is used)
- if the original photo is not available, the script try an download the next available bigger size.
- photos are downloaded following this architecture path_to_download_directory/blog_name/yyyy/mm/yyyymmdd_basename.extension
- if you use filters (wallfilter, black/whitelist) you may end up with empty folders as they are created before the download (room for improvement).
- the script checks first if the file's extension is in the blacklist and then in the whitelist therefore if you both allow and deny an extension it will be denied.
- the script converts extension to lowercase so you don't have to worry whether it is JPG or jpg...
Configuration
In order for the script to work you need to register an app on the Tumblr API (it only takes seconds), then rename the config-example.json to config.json and replace the values:
- API_KEY
FAQ
See the Wiki FAQ page.