Tagbooru - Danbooru Downloader and File Tagger
What is this? Script that downloads images from danbooru and renames them to "ID - tags". Example: "411914 - ibuki suika k hiro touhou.jpg".
Update 12 May 2010: So I finally got around to finishing that tag database... It's 82mb of text which is far beyond what I imagined it to become. The new version (v0.2) on the right is the working version I use for my own local danbooru archive. As perl still sucks with their renaming module, I'm pretty much left to wait until it's capable of renaming files to unicode, though danbooru only contains a couple of non-alphabet tags (the biggest one being ➈).
Description The popular non-linear imageboard Danbooru (danbooru.donmai.us) houses images as far as the eye can see. Sometimes someone wishes to save a local copy of a bunch of images for offline viewing. As Danbooru's filenames are based on a combination of the tags and a HASH code, it says nothing about the characters or objects within the image itself. In other words, the tags are not included in the filename. This is not a limitation of the Danbooru script itself. This was purposely chosen by the system administrators to prevent massive abuse. This is why I ask you to sensibly use this script, especially considering the bandwidth usage.
This script attempts to include the tags in the filenames for as far as the filesystem allows (leaving some room for subdirectories). It was fully written in Perl, tagbooru is my first serious attempt at writing something useful that can be used by anyone.
Objectives
The entire reason I'm publishing this is because I wish for my script to have some reason for existing. Of course I'm open for corrections. As the script borrows quite a bit of code from lwp-download, the license is equal to the Perl license.
Download: http://code.google.com/p/tagbooru/
Hinweis für Windows User bzw auch Linux: Perl braucht ihr noch ;P
Windows: http://strawberryperl.com/
Linux: über den Packetmanager einfach Perl nachinstallieren falls ihr es nicht installiert habt
Zum starten (windows) cmd -> perl start.pl
What is this? Script that downloads images from danbooru and renames them to "ID - tags". Example: "411914 - ibuki suika k hiro touhou.jpg".
Update 12 May 2010: So I finally got around to finishing that tag database... It's 82mb of text which is far beyond what I imagined it to become. The new version (v0.2) on the right is the working version I use for my own local danbooru archive. As perl still sucks with their renaming module, I'm pretty much left to wait until it's capable of renaming files to unicode, though danbooru only contains a couple of non-alphabet tags (the biggest one being ➈).
Description The popular non-linear imageboard Danbooru (danbooru.donmai.us) houses images as far as the eye can see. Sometimes someone wishes to save a local copy of a bunch of images for offline viewing. As Danbooru's filenames are based on a combination of the tags and a HASH code, it says nothing about the characters or objects within the image itself. In other words, the tags are not included in the filename. This is not a limitation of the Danbooru script itself. This was purposely chosen by the system administrators to prevent massive abuse. This is why I ask you to sensibly use this script, especially considering the bandwidth usage.
This script attempts to include the tags in the filenames for as far as the filesystem allows (leaving some room for subdirectories). It was fully written in Perl, tagbooru is my first serious attempt at writing something useful that can be used by anyone.
Objectives
- The script has to be able to download all images tagged with a single or a combination of tags
- The script has to be able to download solely based on ID numbers
- The filenames must contain as many tags as the filesystem allows with some room for directory names left aside (default: 200 characters)
- Preferably, the original date as attributed to the file on the server must be preserved
The entire reason I'm publishing this is because I wish for my script to have some reason for existing. Of course I'm open for corrections. As the script borrows quite a bit of code from lwp-download, the license is equal to the Perl license.
Download: http://code.google.com/p/tagbooru/
Hinweis für Windows User bzw auch Linux: Perl braucht ihr noch ;P
Windows: http://strawberryperl.com/
Linux: über den Packetmanager einfach Perl nachinstallieren falls ihr es nicht installiert habt
Zum starten (windows) cmd -> perl start.pl
