Python module to implement API-like functionality for the FurAffinity.net website.
Project description
Fur Affinity API
Python module to implement API-like functionality for the FurAffinity.net website
Requirements
Python 3.8+ is necessary to run this module.
Usage
The API is comprised of a main class FAAPI
, two submission classes Submission
and SubmissionPartial
, a journal class Journal
, and a user class User
.
Once FAAPI
is initialized its method can be used to crawl FA and return machine-readable objects.
import faapi
import json
cookies = [
{"name": "a", "value": "38565475-3421-3f21-7f63-3d341339737"},
{"name": "b", "value": "356f5962-5a60-0922-1c11-65003b703038"},
]
api = faapi.FAAPI(cookies)
sub, sub_file = api.get_submission(12345678,get_file=True)
print(sub.id, sub.title, sub.author, f"{len(sub_file)/1024:02f}KiB")
with open(f"{sub.id}.json", "w") as f:
f.write(json.dumps(sub))
with open(sub.file_url.split("/")[-1], "wb") as f:
f.write(sub_file)
gallery, _ = api.gallery("user_name", 1)
with open("user_name-gallery.json", "w") as f:
f.write(json.dumps(gallery))
robots.txt
At init, the FAAPI
object downloads the robots.txt file from FA to determine the Crawl-delay
value set therein. If not set, a value of 1 second is used.
To respect this value, the default behaviour of of the FAAPI
object is to wait when a get request is made if the last request was performed more recently then the crawl delay value.
See under FAAPI for more details on this behaviour.
Furthermore, any get operation that points to a disallowed path from robots.txt will raise an exception. This check should not be circumvented and the developer of this module does not take responsibility for violations of the TOS of Fur Affinity.
Cookies
To access protected pages, cookies from an active session are needed. These cookies must be given to the FAAPI object as a list of dictionaries, each containing a name
and a value
field. The cookies list should look like the following random example:
cookies = [
{"name": "a", "value": "38565475-3421-3f21-7f63-3d341339737"},
{"name": "b", "value": "356f5962-5a60-0922-1c11-65003b703038"},
]
To access session cookies, consult the manual of the browser used to login.
Note: it is important to not logout of the session the cookies belong to, otherwise they will no longer work.
Note: as of 2020-02-21 only cookies a
and b
are needed.
User Agent
FAAPI
attaches a User-Agent
header to every request. The user agent string is generated at startup in the following format: faapi/{package version} Python/{python version} {system name}/{system release}
.
Objects
FAAPI
This is the main object that handles all the calls to scrape pages and get submissions.
It holds 6 different fields:
cookies: List[dict] = []
cookies passed at initsession: CloudflareScraper
cfscrape session used for get requestsrobots: Dict[str, List[str]]
robots.txt valuescrawl_delay: float
crawl delay from robots.txt, else 1last_get: float
time of last get (not UNIX time, usestime.perf_counter
for more precision)raise_for_delay: bool = False
if set toTrue
, raises an exception if a get call is made before enough time has passed
Init
__init__(cookies: List[dict] = None)
The class init has a single optional argument cookies
necessary to read logged-in-only pages.
The cookies can be omitted and the API will still be able to access public pages.
Note: Cookies must be in the format mentioned above in #Cookies.
Methods & Properties
load_cookies(cookies: List[Dict[str, Any]])
Load new cookies in the object and remake theCloudflareScraper
session.connection_status -> bool
Returns the status of the connection.get(path: str, **params) -> requests.Response
This returns a response object containing the result of the get operation on the given url with the optional**params
added to it (url provided is considered as path from 'https://www.furaffinity.net/').get_parse(path: str, **params) -> Optional[bs4.BeautifulSoup]
Similar toget()
but returns the parsed HTML from the normal get operation. If the GET request encountered an error, anHTTPError
exception is raised. If the response is not ok, thenNone
is returned.get_submission(submission_id: int, get_file: bool = False) -> Tuple[Submission, Optional[bytes]]
Given a submission ID, it returns aSubmission
object containing the various metadata of the submission itself and abytes
object with the submission file ifget_file
is passed asTrue
.get_submission_file(submission: Submission) -> Optional[bytes]
Given a submission object, it downloads its file and returns it as abytes
object.get_user(user: str) -> User
Given a username, it returns aUser
object containing information regarding the user.gallery(user: str, page: int = 1) -> Tuple[List[SubmissionPartial], int]
Returns the list of submissions found on a specific gallery page and the number of the next page. The returned page number is set to 0 if it is the last page.scraps(user: str, page: int = 1) -> -> Tuple[List[SubmissionPartial], int]
Returns the list of submissions found on a specific scraps page and the number of the next page. The returned page number is set to 0 if it is the last page.favorites(user: str, page: str = "") -> Tuple[List[SubmissionPartial], str]
Downloads a user's favorites page. Because of how favorites pages work on FA, thepage
argument (and the one returned) are strings. If the favorites page is the last then an empty string is returned as next page. An empty page value as argument is equivalent to page 1.
Note: favorites page "numbers" do not follow any scheme and are only generated server-side.journals(user: str, page: int = 1) -> -> Tuple[List[Journal], int]
Returns the list of submissions found on a specific journals page and the number of the next page. The returned page number is set to 0 if it is the last page.search(q: str = "", page: int = 0, **params) -> Tuple[List[SubmissionPartial], int, int, int, int]
Parses FA search given the query (and optional other params) and returns the submissions found and the next page together with basic search statistics: the number of the first submission in the page (0-indexed), the number of the last submission in the page (0-indexed), and the total number of submissions found in the search. For example if the the last three returned integers are 0, 47 and 437, then the page contains submissions 1 through 48 of a search that has found a total of 437 submissions.
Note: as of October 2020 the "/search" path is disallowed by Fur Affinity's robots.txt.watchlist_to(self, user: str) -> List[User]
Given a username, returns a list ofUser
objects for each user that is watching the given user.watchlist_by(self, user: str) -> List[User]
Given a username, returns a list ofUser
objects for each user that is watched by the given user.user_exists(user: str) -> int
Checks if the passed user exists - i.e. if there is a page under that name - and returns an int result.- 0 okay
- 1 account disabled
- 2 system error
- 3 unknown error
- 4 request error
submission_exists(submission_id: int) -> int
Checks if the passed submissions exists - i.e. if there is a page with that ID - and returns an int result.- 0 okay
- 1 account disabled
- 2 system error
- 3 unknown error
- 4 request error
journal_exists(journal_id: int) -> int
Checks if the passed journal exists - i.e. if there is a page under that ID - and returns an int result.- 0 okay
- 1 account disabled
- 2 system error
- 3 unknown error
- 4 request error
Journal
This object contains information gathered when parsing a journals page or a specific journal page. It contains the following fields:
id: int
journal idtitle: str
journal titledate: str
upload date in YYYY-MM-DD formatauthor: str
journal authorcontent: str
journal contentmentions: List[str]
the users mentioned in the content (if they were mentioned as links, e.g. :iconusername:, @username, etc.)journal_item: Union[bs4.element.Tag, bs4.BeautifulSoup]
the journal tag/page used to parse the object fields
Journal
objects can be directly casted to a dict object or iterated through.
Init
__init__(journal_item: Union[bs4.element.Tag, bs4.BeautifulSoup] = None)
Journal
takes one optional parameters: a journal section tag from a journals page or a parsed journal page. Parsing is then performed based on the class of the passed object.
Methods
parse(journal_item: Union[bs4.element.Tag, bs4.BeautifulSoup] = None)
Parses the stored journal tag/page for information. Ifjournal_item
is passed, it overwrites the existingjournal_item
value.
SubmissionPartial
This lightweight submission object is used to contain the information gathered when parsing gallery, scraps, favorites and search pages. It contains only the following fields:
id: int
submission idtitle: str
submission titleauthor: str
submission authorrating: str
submission rating [general, mature, adult]type: str
submission type [text, image, etc...]submission_figure: bs4.element.Tag
the figure tag used to parse the object fields
SubmissionPartial
objects can be directly casted to a dict object or iterated through.
Init
__init__(submission_figure: bs4.element.Tag)
SubmissionPartial
init needs a figure tag taken from a parsed page.
Methods
parse(submission_figure: bs4.element.Tag)
Parses the stored submission figure tag for information. Ifsubmission_figure
is passed, it overwrites the existingsubmission_figure
value.
Submission
The main class that parses and holds submission metadata.
id: int
submission idtitle: str
submission titleauthor: str
submission authordate: str
upload date in YYYY-MM-DD formattags: List[str]
tags listcategory: str
category *species: str
species *gender: str
gender *rating: str
rating *description: str
the description as an HTML formatted stringmentions: List[str]
the users mentioned in the description (if they were mentioned as links, e.g. :iconusername:, @username, etc.)folder: str
the submission folder (gallery or scraps)file_url: str
the url to the submission filesubmission_page: bs4.BeautifulSoup
the submission page used to parse the object fields
* these are extracted exactly as they appear on the submission page
Submission
objects can be directly casted to a dict object and iterated through.
Init
__init__(submission_page: bs4.BeautifulSoup = None)
To initialise the object, an optional bs4.BeautifulSoup
object is needed containing the parsed HTML of a submission page.
If no submission_page
is passed then the object fields will remain at their default - empty - value.
Methods
parse(submission_page: bs4.BeautifulSoup = None)
Parses the stored submission page for metadata. Ifsubmission_page
is passed, it overwrites the existingsubmission_page
value.
User
A small class that holds a user's full information.
name: str
display name with capital letters and extra characters such as "_"status: str
user status (~, !, etc.)profile: str
profile text in HTML formatuser_page: bs4.BeautifulSoup
the user page used to parse the object fields
Init
__init__(user_page: bs4.BeautifulSoup = None)
To initialise the object, an optional bs4.BeautifulSoup
object is needed containing the parsed HTML of a submission page.
If no user_page
is passed then the object fields will remain at their default - empty - value.
Methods
parse(user_page: bs4.BeautifulSoup = None)
Parses the stored user page for metadata. Ifuser_page
is passed, it overwrites the existinguser_page
value.
Contributing
All contributions and suggestions are welcome!
The only requirement is that any merge request must be sent to the GitLab project as the one on GitHub is only a mirror: GitLab/FALocalRepo
If you have suggestions for fixes or improvements, you can open an issue with your idea, see #Issues for details.
Issues
If any problem is encountered during usage of the program, an issue can be opened on the project's pages on GitLab (preferred) or GitHub (mirror repository).
Issues can also be used to suggest improvements and features.
When opening an issue for a problem, please copy the error message and describe the operation in progress when the error occurred.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.