- DBooru
- Stable branch
- Rare updates
- Lots of changes per update
- Unstable
- Lots of commits
- Lots of updates
- New features
- Broken old features
- Rare commit messages
- If there a commit message then "Minor changes"*
*Minor changes may include removal of half of all code
- Search in downloaded images
- By tags
- By image dimensions
- Viewing images
- Loading images from *booru
- Even with proxy (socks5 only)
- Downloading images
- Exporting images
- Tag predictions
- Change page with ← or → arrows
- Scroll through images with ← or → arrows
¯\_(ツ)_/¯
To be deprecated
- Python 3.5+ or PyPy3 5.9.0+
- pip
- requests
- requests[security]
- pysocks
- idna
- cryptography
- ffmpeg or pillow
- flask
- jinja2
- werkzeug
- markupsafe
- click
- itsdangerous
- termcolor
- mysql-connector-python
Or you can just type
pip install --user -r requirements.txtin terminal
- Run
python DBooru_web.pyorpypy3 DBooru_web.py - Change default settings
or
- Set modules (line 11) in settings_file.py
- Configure modules (placed in dermod/sitesupport)
- (Optionally) Change other settings (View Settings_file.py)
- Run
python main.pyorpypy3 main.py - Type in "get images"
- Wait for images to download
- Use search
- ???
- PROFIT
- Run
python DBooru_web.pyorpypy3 DBooru_web.py - Press
Update DBbutton - Wait for images to download
- Use search
- ???
- PROFIT
Modules allows you to configure DBooru for using with multiple *booru without rewriting half of code
Open issue with site name and module description OR Write your own module
It's not too hard (if site uses api ofc), you can use included modules as example.
Module must contain:
- Configurable options
- Site domain
- Search query
- Search endpoint
- Paging parameter
- Empty page delimiter
- Additional parameters (such as API-key parameter)
- Hard limit
- Parser Configuration
Configurable options
>> username = "NAME GOES HERE"
>> apikey = "KEY"
Must have # Do not change values after user-configurable options
Site Domain
>> domain = "https://example.com"
Search query [Can be configurable]
>> query = "likedby:{}".format(username) [Preferred variant]
>> query = "sky, ponies" [Well, if you gonna download pictures with this tags only, then it's ok]
Search endpoint (Varies depending on site)
>> endpoint = "/search?q={}".format(query)
>> endpoint = "/tags/{}".format(query)
Paging parameter (Varies depending on site)
>> paginator = "&page={}"
>> paginator = "&p={}"
>> paginator = "/{}"
In case of overly specific pagination can be implemented as class (check gelbooru module)
>> class paginator:
>> def format(page):
>> return "/pg/{}".format(page*25)
Empty page delimiter (Varies depending on site)
> RegExp matching page with no images
>> empty_page = "\[\]$"
>> empty_page = "{images:\[\]"
Sleep time (to get rid of rate limiting)
> Value must be int or float
>> slp = 1
>> slp = 0.5
Additional params
> Such as api keys
>> params = '&key={}&username={}'.format(apikey, username)
Hard limit (for when your life is not limited enough)
>Use if api does not allow requesting pages after N
>tip: this is not mandatory, but if you don't want to be banned specify this
>> hard_limit = 750 [notice, this time we're not using quotes]
Parser Configuration
May use whatever python functions/packages you want to use (check gelbooru module)
MUST start with and MUST contain parse(self, string) method
>> class Module:
>> def __init__(self):
>> self.tags = []
>> self.ids = []
>> self.links = []
>> self.form = []
>> self.height = []
>> self.width = []
Parse() method
Response from server passed with string argument
If page number required `pg_num` added to input arguments
Note: pg_num receives raw page number (1, 2, 3, ...)
>> def parse(self, string):
>> # here goes parsing
>> # ...
>> # parsed values must be added to lists
>> for i in parsed_values:
>> self.tags.append(i[0]) # tags MUST be delimited by `,,`
>> self.ids.append(i[4])
>> self.links.append(i[3])
>> self.form.append(i[5])
>> self.height.append(i[2])
>> self.width.append(i[1])
Enter this commands if prompt starts with DB>
| Command | Description |
|---|---|
| help | Shows in-app help |
| get images | Downloads images that you liked/favorited on *booru |
| get images --force | Downloads images without checking if file exists |
| total | Prints amount of entries in local DB |
| count <tag> | Prints amount of entries tagged with <tag> |
| show <image_id> | Opens image in image viewer or browser if no viewers found |
| quit (exit) | Closes app |
| <anything> | Uses input as list of tags and searches for it |
Enter this commands if prompt starts with Search@DB>
| Command | Description |
|---|---|
| show <image_id> | Opens image in image viewer or browser if no viewers found |
| quit (exit) | Closes app |
| back | Returns to main menu |
| <any_number> | Show page numbered as <any_number> |
| <anything> | Uses input as list of tags and searches for it |
| Endpoint | Method | Parameters (Body for POST) | Description | Returns |
|---|---|---|---|---|
| "/" | GET | Main page | HTML-page | |
| "/search" | GET | q=search_query page=int | Search images | HTML-page |
| "/image/str" | GET | View image with tags | HTML-page | |
| "/dl/filename" | GET | Browser-friendly download method | Image | |
| "/raw/filename" | GET | Raw image data | Image | |
| "/predict" | GET | phrase=search_query | Tries to predict search query | Plain text data |
| "/next/str" | GET | Tries to get id of next (newer) image | Redirect (302) to /image/* | |
| "/previous/str | GET | Tries to get id of previous(older) image | Redirect (302) to /image/* | |
| "/thumbnail/filename" | GET | Makes thumbnail (500px) of image (returns full image if thumbnailer disabled) | Image | |
| "/json/search" | GET | q=query page=int | Searches images and returns json result of search | JSON |
| "/random" | GET | Redirects to random image | Redirect (302) to /image/* | |
| "/random/tags | GET | Redirects to random image tagged with tags | Redirect (302) to /image/* | |
| "/update" | GET | Updates DB (Same as CLI: get images) | Returns 200 code when update started successfully or 409 in case when there's already update in progress | |
| "/remove/prefix_id" | DELETE | Used to remove image and its DB entry | ||
| "/reload/prefix_id" | PATCH | Triggers image redownloading | ||
| "/settings" | GET | Settings page | HTML-page | |
| "/settings/<option>" | POST | opt_type=variable_type&<option>_new_opt=value | Endpoint for changing settings_file.py variables | Redirect (302) to /settings#<option>_form |
| "/settings/<modulename>/<option>" | POST | <option>_new_opt=value | Endpoint for changing module's variables | Redirect (302) to /settings#<modulename>_form |
- Wildcard through
*(asterisk)
Example: "ti
*" will return images where at least one tag starts withti
- Tags must be separated by
,(comma)
Example: "safe
,princess luna"
- Tags are not case sensitive
Example: "SaFe" == "SafE" == "safe"
- Search for "tag" will only return images tagged with "tag" not "tag*"
Example: "
safe" returns images with tag "safe" and doesn't returns "safezone"
- Searching multiple tags will return images matching all the tags
Example: "
safe, princess luna" will return images tagged with both "safe" and "princess luna"
- Exclude tags by placing
-(hyphen-minus) before tag
Example: "
-safe" returns all images not tagged with "safe"
- OR queries will return images tagged with tag1 or tag2
Example: "
(fluttershy|applejack)" returns images tagged withfluttershy,applejackor both tags
- OR queries could be mixed with AND queries
Example: "
safe,(princess luna|changeling)" returns images taggedprincess luna, safeorchangeling, safe
- Exclusion is not supported for OR queries. It's just pointless.
- Rules 1,3,4,5 work almost the same for exclude
Works only for filtering searches
Example:
width=100works, while-width=100doesn't work
heightwidthratio
=or==means equal to <value>!=or<>means not equal to <value>>means bigger than <value><means less than <value>>=means bigger or equal to <value><=means less or equal to <value>
Example: 'safe, width>100" will return images tagged with 'safe' tag and image with width bigger than 100px
| Option | Format | Description |
|---|---|---|
| modules | List (['String', 'String']) | Enables modules |
| suppress_errors | Bool (True/False) | Prints errors and stacktrace in case of happening |
| ssl_verify | Bool (True/False) or String ("Path") | Enable/Disable ssl verification or set custom CA Cert |
| enable_proxy | Bool (True/False) | Enables/Disables proxy for requests |
| proxy_type | String | Sets proxy type |
| proxy_ip | String (IP) | Sets proxy IP |
| proxy_port | String (Port) | Sets proxy port |
| BASE_DIR | Function | Magic for flask to work |
| web_ip | String (IP) | Set IP to bind Web interface |
| web_port | Integer (port) | Sets port to bind Web interface |
| thumbnailer | String (One of "ffmpeg", "PIL", "disabled") | Defines tool to make thumbnails or not to make them at all |
| conv_format | String (ffmpeg output format) | Format to use when making thumbnails |
| ffmpeg_args | String (ffmpeg parameters) | For situations when you think that default settings suck |
| gif_to_webp | Bool (True/False) | Creates webp thumbnails for gifs |
| disable_mobile | Bool (True/False) | Should tag prediction be disabled on mobile |
| predict_tags | Integer (number) | How many tags to show when predicting input |
| showing_imgs | Integer (number) | How many images to show per page |
| showing_tags | Integer (number) | How many tags to show per image (CLI-only) |
| images_path | String (Path) | Where to store loaded images |
| export_path | String (Path) | Where to store exported images |
| time_wait | Integer (seconds) | How long thread can stay alive |
| ids_file | String (Path/Filename) | Name for tempfile (No need to change) |
| db_name | String | Defines name for database file In case of using mysql defines schema name |
| use_mysql | Bool (True/False) | Defines if mysql connection should be used |
| mysql_user | String | Defines mysql user |
| mysql_password | String | Defines mysql password |
| mysql_host | String | Defines mysql host |
| mysql_port | Integer (Port) | Defines mysql port |
| thread_cap | Integer (number) | Defines maximum running threads before blocking creating new threads |
| sleep_time | Integer (seconds) | Defines time to wait before creating new thread after thread cap is reached |
| enable_polling | Bool | Setting for enabling/disabling polling for changes in settings_file.py |
| polling_time | Integer (seconds) | Defines time (in seconds) between checks for changes in settings_file.py |
| first_run | Bool | Defines if settings page should be opened on first run |
| touch_controls | Bool | Enables swipe controls on touchscreens |
DBooru can automatically generate /settings entries for variables in settings_file.py
Options block must start with two newlines and block title on the third
Every text line must start with #, except for variable declaration
Lines with no keywords will be interpreted as description
By default examples are NOT shown on page
Keywords must be followed by : , except for # Require example
| Keyword | Description | Notes |
|---|---|---|
| Example | Defines example | Can be used more than once |
| Format | Defines variable type | Should be one of string, bool, int. Defaults to string. |
| Options | Defines available options for variable | Must be formatted as Python list |
| Require example | Requires examples to be shown on /settings page | May or may not end with : |
# Will show examples and dropdown with options
# Require example
# Example: var = "something"
# ExaMplE: var = "nothing"
# Format: var = string
# Options: ['module', 'thing', 'more things']
var = "done"
# Example block
# Won't show example
# Input field will be provided
# description line
# description line
# Format: test_vat = string
# Example: test_var = "test"
# description line
test_var = "test passed"
# var2 desc
# does thing
# Format: var2 = bool
# ...
var2 = TrueAllows creating alias to tag, so you can find one tag using alias
Note: Aliasing alias won't work
aliases = {
"alias1": "aliased tag1",
"alias2": "aliased tag1",
"alias3": "aliased tag2"
}
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent