PyGelbooru is an unofficial and lightweight asynchronous library for the Gelbooru API.
This library requires Python 3.6 or above.
You can install the library through pip as follows,
pip install pygelbooru
The primary use for this library is, naturally, to search for images with specific tags.
This can be done as so:
from pygelbooru import Gelbooru
# API key/user ID is optional, but access may be limited without them
gelbooru = Gelbooru('API_KEY', 'USER_ID')
results = await gelbooru.search_posts(tags=['dog ears', '1girl'], exclude_tags=['nude'])
[<GelbooruImage(id=5105386, filename='b77e69be0a4b...dde071dc.jpeg', owner='anon2003')>,
<GelbooruImage(id=5105161, filename='bf169f891ebe...02bceb5e.jpeg', owner='cpee')>,
<GelbooruImage(id=5104148, filename='46df3ebe2d41...4316d218e.jpg', owner='danbooru')>,
<GelbooruImage(id=5104080, filename='e8eec23d151e...419293401.png', owner='anon2003')>,
<GelbooruImage(id=5103937, filename='5bf279f3c546...be3fc53c8.jpg', owner='danbooru')>,
...
Tags can contain spaces when passed as arguments, they will simply be reformated with underscores before being queried, so you don't need to reformat them yourself.
Results are returned as a list of GelbooruImage containers. When cast to a string, this will return the image_url,
str(results[0])
'https://img2.gelbooru.com/images/b7/7e/b77e69be0a4b581eac597527dde071dc.jpeg'
You can also pull other information returned by the API, https://github.com/FujiMakoto/pygelbooru/blob/master/pygelbooru/gelbooru.py#L32-L47
In addition to searching for a large list of images, PyGelbooru also provides a helper method for when you're really just after a single, random image that matches the specified tags.
This method will automatically pull a random image from the last 20,000 Gelbooru image submissions.
result = await gelbooru.random_post(tags=['cat ears', '1girl', 'cat hood', 'bell'], exclude_tags=['nude'])
<GelbooruImage(id=5106718, filename='bbbdfbf9e883...161753514.png', owner='6498')>
You can fetch post comments directly from the GelbooruImage container,
post = await gelbooru.get_post(5099841)
await post.get_comments()
[<GelbooruComment(id=2486074, author='Anonymous', created_at='2020-01-28 08:47')>]
Besides searching for images, you can also pull information on tags as follows,
await gelbooru.tag_list(name='dog ears')
<GelbooruTag(id=773, name='dog_ears', count=22578)>
# Use "name_pattern" to search for partial matches to a specified tag
await gelbooru.tag_list(name_pattern='%splatoon%', limit=4)
[<GelbooruTag(id=892683, name='splatoon_(series)', count=11353)>,
<GelbooruTag(id=759189, name='splatoon_2', count=3488)>,
<GelbooruTag(id=612372, name='aori_(splatoon)', count=2266)>,
<GelbooruTag(id=612374, name='hotaru_(splatoon)', count=2248)>]