aio-libs / aio-libs/aiohttp

add aiohttp.MozillaCookieJar to load/save cookies.txt files

Abierto
#7,998 4 comentarios 1 reacción 0 asignados Ver en GitHub
enhancement
Lenguaje dominante
Python
Estrellas
16.5k
Forks
2.4k
Merge medio
17 h 22 min
PR fusionados (30 d)
212

Descripción

### Is your feature request related to a problem?

`aiohttp.CookieJar` can only load/save cookies from/to `cookies.pickle` files
but not from/to `cookies.txt` files

### Describe the solution you'd like

there should be `aiohttp.MozillaCookieJar`

### Describe alternatives you've considered

write my own `class AiohttpMozillaCookieJar(aiohttp.CookieJar)`

AiohttpMozillaCookieJar.py

```py
# NOTE this is not-really tested... its just a draft

from http.cookiejar import MozillaCookieJar
from http.cookies import BaseCookie, Morsel, SimpleCookie
from collections import defaultdict
from pathlib import Path
from datetime import datetime
from http.cookiejar import Cookie
from typing import Union

from aiohttp import CookieJar

PathLike = Union[str, "os.PathLike[str]"]

class AiohttpMozillaCookieJar(CookieJar):
"""
load/save cookies from/to a cookies.txt file with aiohttp

convert between http.cookiejar.MozillaCookieJar and aiohttp.CookieJar

import aiohttp
from AiohttpMozillaCookieJar import AiohttpMozillaCookieJar
jar = AiohttpMozillaCookieJar()
cookies_txt_path = "cookies.txt"
jar.load(cookies_txt_path)
async with aiohttp.ClientSession(cookie_jar=jar) as session:
url = "..."
response = await aiohttp_session.get(url)
jar.save(cookies_txt_path)

author: Milan Hauth
license: MIT License
"""

def load(self, file_path):

file_path = Path(file_path)

jar_1 = MozillaCookieJar()
jar_1.load(file_path)

# Cookie in jar_1 -> Morsel in jar_2

# jar_1._cookies # dict
# jar_1._cookies[domain] # dict
# jar_1._cookies[domain][path] # dict
# jar_1._cookies[domain][path][name] # Cookie

# jar_2._cookies # collections.defaultdict(SimpleCookie)
# jar_2._cookies[(domain, path)] # SimpleCookie
# jar_2._cookies[(domain, path)][name] # Morsel

for cookie_1 in jar_1:

morsel_2 = Morsel()

domain = cookie_1.domain
path = cookie_1.path
name = cookie_1.name

if name.lower() in Morsel._reserved:
#print(f"illegal morsel name: {name}")
continue

for key in (
'path',
'comment',
'domain',
'secure',
'version',
):
if (value := getattr(cookie_1, key)) is not None:
morsel_2[key] = value

morsel_2._key = cookie_1.name
morsel_2._value = cookie_1.value
morsel_2._coded_value = cookie_1.value

try:
morsel_2['expires'] = datetime.fromtimestamp(cookie_1.expires).strftime(
"%a, %d %b %Y %H:%M:%S GMT"
)
except (
OSError, # Invalid argument (value of cookie.expires is invalid)
TypeError, # cookie.expires is None
):
pass

for key in ('HttpOnly', 'SameSite'):
if (value := cookie_1.get_nonstandard_attr(key, None)) is not None:
morsel_2[key] = value

self._cookies[(domain, path)][name] = morsel_2

def save(self, file_path: PathLike) -> None:

file_path = Path(file_path)

jar_1 = MozillaCookieJar()

# Morsel in jar_2 -> Cookie in jar_1

# jar_2._cookies # collections.defaultdict(SimpleCookie)
# jar_2._cookies[(domain, path)] # SimpleCookie
# jar_2._cookies[(domain, path)][name] # Morsel

# jar_1._cookies # dict
# jar_1._cookies[domain] # dict
# jar_1._cookies[domain][path] # dict
# jar_1._cookies[domain][path][name] # Cookie

for (domain, path), cookie_2 in self._cookies.items():

for name, morsel_2 in cookie_2.items():

try:
expires = self._expirations[(domain, path, name)].timestamp()
except KeyError:
try:
expires = datetime.strptime(morsel_2["expires"], "%a, %d %b %Y %H:%M:%S %Z").timestamp()
except ValueError:
# morsel_2["expires"] can be empty string
expires = None

cookie_1 = Cookie(
0, # version
name, # name
morsel_2.value, # value
None, # port
False, # port_specified
domain, # domain
False, # domain_specified
False, # domain_initial_dot
path, # path
False, # path_specified
morsel_2["secure"], # secure
expires, # expires
False, # discard
None, # comment
None, # comment_url
{}, # rest
)
if not domain in jar_1._cookies:
jar_1._cookies[domain] = dict()
if not path in jar_1._cookies[domain]:
jar_1._cookies[domain][path] = dict()
jar_1._cookies[domain][path][name] = cookie_1

jar_1.save(file_path)
```

test-aiohttp-cookies.py

```py
#!/usr/bin/env python

import sys
import asyncio
import aiohttp
from AiohttpMozillaCookieJar import AiohttpMozillaCookieJar
import json
import os

cookie_jar_backend = "pickle"
cookie_jar_backend = "txt"

cookie_jar_path = f"test-aiohttp-cookies.cookies.{cookie_jar_backend}"

cookie_jar_kwargs = dict()
#cookie_jar_kwargs["unsafe"] = True

cookie_jar = None
if cookie_jar_backend == "txt":
cookie_jar = AiohttpMozillaCookieJar(**cookie_jar_kwargs)
elif cookie_jar_backend == "pickle":
cookie_jar = aiohttp.CookieJar(**cookie_jar_kwargs)
else:
raise ValueError(f"cookie_jar_backend {cookie_jar_backend}")

def dump_cookie_jar(cookie_jar):

for k1 in cookie_jar._cookies:
domain, path = k1

#print(f"cookie_jar dir", dir(cookie_jar))

#print(f"cookie_jar _expirations", cookie_jar._expirations)
#print(f"cookie_jar _host_only_cookies", cookie_jar._host_only_cookies)
#print(f"cookie_jar _max_time", cookie_jar._max_time)
#print(f"cookie_jar _unsafe", cookie_jar._unsafe)

#print(f"cookie_jar._cookies: domain={domain} path={path}", cookie_jar._cookies[k1])
#print(f"cookie_jar._cookies: domain={domain} path={path} dir", dir(cookie_jar._cookies[k1]))
#print(f"cookie_jar._cookies: domain={domain} path={path} items", cookie_jar._cookies[k1].items())
for name, morsel in cookie_jar._cookies[k1].items():
print(f"cookie_jar._cookies: domain={domain} path={path} {name}={morsel.value}")

if False:
# only dump cookiejar
cookie_jar.load(cookie_jar_path)
dump_cookie_jar(cookie_jar)
sys.exit()

async def main():

session_kwargs = dict()
# not used?
session_kwargs["cookie_jar"] = cookie_jar
session_kwargs["headers"] = dict(session_header="1")

get_kwargs = dict()
get_kwargs["headers"] = dict(request_header="1")
# TypeError: ClientSession._request() got an unexpected keyword argument 'cookie_jar'
#get_kwargs["cookie_jar"] = cookie_jar

if os.path.exists(cookie_jar_path):
print(f"loading cookies from {cookie_jar_path}")
cookie_jar.load(cookie_jar_path)

async with aiohttp.ClientSession(**session_kwargs) as aiohttp_session:

# debug cookies
# FIXME aiohttp does not send cookies?!

print(f"aiohttp_session.cookie_jar._cookies:")
dump_cookie_jar(aiohttp_session.cookie_jar)

#print(f"aiohttp_session.cookie_jar._cookies httpbin.dev /cookies:\n" + str(aiohttp_session.cookie_jar._cookies[("httpbin.dev", "/cookies")]))

url_base = "http://httpbin.dev"
#url_base = "https://httpbin.dev"

# debug: set cookies
#url = f"{url_base}/cookies/set?foo=bar"
url = f"{url_base}/cookies/set?foo=bar&a=1&b=2"
response = await aiohttp_session.get(url, **get_kwargs)

# Cookie headers are not here
#print(f"response.request_info.headers: {response.request_info.headers}")

response_text = (await response.content.read()).decode("utf8")
#print(f"response_text: {response_text}")

# Set-Cookie headers are not here
#print(f"response.headers: {response.headers}")

print(f"aiohttp_session.cookie_jar._cookies:")
dump_cookie_jar(aiohttp_session.cookie_jar)

#print(f"aiohttp_session.cookie_jar._cookies httpbin.dev /cookies:\n" + str(aiohttp_session.cookie_jar._cookies[("httpbin.dev", "/cookies")]))

# NOTE cookies are not returned by /headers
url = f"{url_base}/headers"
response = await aiohttp_session.get(url, **get_kwargs)

print(f"response.request_info.headers.Cookie: {response.request_info.headers.get('Cookie')}")
#print(f"response.request_info dir: {dir(response.request_info)}")

response_status = response.status
# TODO debug request headers
#print(f"response dir: {dir(response)}")
#print(f"response_status: {response_status}")

# Set-Cookie headers are not here
#print(f"headers: {response.headers}")

response_content = response.content
response_text = (await response_content.read()).decode("utf8")
print(f"sent headers: {response_text}")

# debug: get cookies
url = f"{url_base}/cookies"
response = await aiohttp_session.get(url, **get_kwargs)

print(f"response.request_info.headers.Cookie: {response.request_info.headers.get('Cookie')}")
#print(f"response.request_info dir: {dir(response.request_info)}")

response_status = response.status
# TODO debug request headers
#print(f"response dir: {dir(response)}")
#print(f"response_status: {response_status}")

# Set-Cookie headers are not here
#print(f"headers: {response.headers}")

response_content = response.content
response_text = (await response_content.read()).decode("utf8")
print(f"sent cookies: {response_text}")

print(f"aiohttp_session.cookie_jar._cookies:")
dump_cookie_jar(aiohttp_session.cookie_jar)

#print(f"aiohttp_session.cookie_jar._cookies httpbin.dev /cookies:\n" + str(aiohttp_session.cookie_jar._cookies[("httpbin.dev", "/cookies")]))

print(f"saving cookies to {cookie_jar_path}")
aiohttp_session.cookie_jar.save(cookie_jar_path)

asyncio.get_event_loop().run_until_complete(main())
```

### Related component

Client

### Additional context

stackoverflow:

[How to use cookies exported from chrome in python aiohttp?](https://stackoverflow.com/questions/48168379/how-to-use-cookies-exported-from-chrome-in-python-aiohttp)

I'm trying to visit some website using aiohttp/python. Currently I can export the site's cookies with extensions in Google Chrome, formatted like

> .domain.name TRUE / FALSE 1547016401 cookies values

### Code of Conduct

- [X] I agree to follow the aio-libs Code of Conduct

Guía de contribución

Abrir la guía de contribución

Evaluación

Este issue todavía no se ha evaluado.

Recibe los nuevos issues en tu correo

Un resumen breve de issues de GitHub para principiantes.