# fflogsapi
fflogsapi is a lazy Python 3 client for [FF Logs](https://www.fflogs.com/)' v2 API with query caching functionality.
[![Tests](https://github.com/halworsen/fflogsapi/actions/workflows/test.yml/badge.svg?branch=master)](https://github.com/halworsen/fflogsapi/actions/workflows/test.yml)
[![Linting](https://github.com/halworsen/fflogsapi/actions/workflows/lint.yml/badge.svg?branch=master)](https://github.com/halworsen/fflogsapi/actions/workflows/lint.yml)
[![codecov](https://codecov.io/gh/halworsen/fflogsapi/branch/master/graph/badge.svg?token=YTEGMDJOGL)](https://codecov.io/gh/halworsen/fflogsapi)
[![pypi](https://shields.io/pypi/v/fflogsapi)](https://pypi.org/project/fflogsapi/)
---
## Features
* Retrieve information from FF Logs' v2 GraphQL API
* Including private information only accessible through the user API
* Lazy evaluation
* Queries for data are not executed until the result is explicitly needed
* Query caching
* Requesting the same data twice will instead fetch the result from cache
* Customizable cache lifetime and options to ignore cached results
* Sensible interface to the API
* Iterate over things that are intuitively iterable
* Dataclasses for many objects exposed by the API
## Example usage
```python
from config import CLIENT_ID, CLIENT_SECRET
from fflogsapi.client import FFLogsClient
client = FFLogsClient(CLIENT_ID, CLIENT_SECRET)
report = client.get_report('rGARYmQwTKbahXz9')
for fight in report:
print(f'Fight #{fight.fight_id}:', fight.name(), f'- Kill: {fight.is_kill()}')
pot_table = fight.table(filters={'sourceAurasPresent': 'Medicated'})
potted_damage = 0
for damage in pot_table['damageDone']:
potted_damage += damage['total']
print(f'Damage done under pots: {potted_damage}')
if not fight.is_kill():
print(f'Percentage reached: {fight.percentage()}')
client.close()
client.save_cache()
```
```python
from config import CLIENT_ID, CLIENT_SECRET
from fflogsapi.client import FFLogsClient
client = FFLogsClient(CLIENT_ID, CLIENT_SECRET)
for page in client.reports(filters={ 'guildID': 80551 }):
print(f'Reports in page: {page.count()}')
for report in page:
print(report.title(), f'Duration: {report.duration()}')
client.close()
client.save_cache()
```
## User mode
The default mode of the client is 'client' mode, which uses the public API. This is by far the most
convenient method to use the client, and usually provides enough functionality for the majority of
use cases.
If you need access to private information, however, it is possible to use the client in user mode,
granting access to extra information such as private reports provided by the user API.
To use user mode, you must first specify `https://localhost:4433` as the redirect URL in your API
client on FF Logs. Then, provide the `mode='user'` kwarg to the client when instantiating it:
```python
client = FFLogsClient(CLIENT_ID, CLIENT_SECRET, mode='user')
```
This will have the client popup a browser window for the user for login, after which the client has
access to the user API. Note that the client will generate a self-signed certificate to serve
the redirect. Your browser will likely produce a warning about this, although it is safe to ignore.
If you wish to handle the user authentication flow yourself, you can still use the API client in
user mode by calling `set_auth_response` on the client **before using it**:
```python
# Your implementation of the user authentication flow here
...
client = FFLogsClient(CLIENT_ID, CLIENT_SECRET, mode='user')
client.set_auth_response(response)
# Start using the client
...
```