add account import with cookies; update mocks

Makefile

@@ -41,3 +41,15 @@ changelog:
 
 test34:
 	docker build -f Dockerfile-test .
+
+update-mocks:
+	twscrape user_by_id --raw 2244994945 | jq > ./tests/mocked-data/user_by_id_raw.json
+	twscrape user_by_login --raw twitterdev | jq > ./tests/mocked-data/user_by_login_raw.json
+	twscrape followers --raw --limit 10 2244994945 | jq > ./tests/mocked-data/followers_raw.json
+	twscrape following --raw --limit 10  2244994945 | jq > ./tests/mocked-data/following_raw.json
+	twscrape tweet_details --raw 1649191520250245121 | jq > ./tests/mocked-data/tweet_details_raw.json
+	twscrape retweeters --raw --limit 10 1649191520250245121 | jq > ./tests/mocked-data/retweeters_raw.json
+	twscrape favoriters --raw --limit 10 1649191520250245121 | jq > ./tests/mocked-data/favoriters_raw.json
+	twscrape user_tweets --raw --limit 10 2244994945 | jq > ./tests/mocked-data/user_tweets_raw.json
+	twscrape user_tweets_and_replies --raw --limit 10 2244994945 | jq > ./tests/mocked-data/user_tweets_and_replies_raw.json
+	twscrape search --raw --limit 10 "elon musk lang:en" | jq > ./tests/mocked-data/search_raw.json

tests/mocked-data/user_by_id_raw.json

@@ -48,25 +48,25 @@
             }
           },
           "fast_followers_count": 0,
-          "favourites_count": 2128,
-          "followers_count": 583721,
-          "friends_count": 1941,
+          "favourites_count": 2152,
+          "followers_count": 587135,
+          "friends_count": 1945,
           "has_custom_timelines": true,
           "is_translator": false,
-          "listed_count": 2395,
+          "listed_count": 2459,
           "location": "127.0.0.1",
           "media_count": 815,
           "name": "Twitter Dev",
-          "normal_followers_count": 583721,
+          "normal_followers_count": 587135,
           "pinned_tweet_ids_str": [
-            "1641222782594990080"
+            "1661790253886177280"
           ],
           "possibly_sensitive": false,
           "profile_banner_url": "https://pbs.twimg.com/profile_banners/2244994945/1660405530",
           "profile_image_url_https": "https://pbs.twimg.com/profile_images/1445764922474827784/W2zEPN7U_normal.jpg",
           "profile_interstitial_type": "",
           "screen_name": "TwitterDev",
-          "statuses_count": 4076,
+          "statuses_count": 4080,
           "translator_type": "regular",
           "url": "https://t.co/9wI31m3ELF",
           "verified": false,
@@ -87,8 +87,14 @@
         },
         "smart_blocked_by": false,
         "smart_blocking": false,
-        "business_account": {}
+        "business_account": {},
+        "highlights_info": {
+          "can_highlight_tweets": true,
+          "highlighted_tweets": "0"
+        },
+        "creator_subscriptions_count": 0,
+        "has_hidden_likes_on_profile": false
       }
     }
   }
-}
+}

tests/mocked-data/user_by_login_raw.json

@@ -48,25 +48,25 @@
             }
           },
           "fast_followers_count": 0,
-          "favourites_count": 2128,
-          "followers_count": 583731,
-          "friends_count": 1941,
+          "favourites_count": 2152,
+          "followers_count": 587135,
+          "friends_count": 1945,
           "has_custom_timelines": true,
           "is_translator": false,
-          "listed_count": 2395,
+          "listed_count": 2459,
           "location": "127.0.0.1",
           "media_count": 815,
           "name": "Twitter Dev",
-          "normal_followers_count": 583731,
+          "normal_followers_count": 587135,
           "pinned_tweet_ids_str": [
-            "1641222782594990080"
+            "1661790253886177280"
           ],
           "possibly_sensitive": false,
           "profile_banner_url": "https://pbs.twimg.com/profile_banners/2244994945/1660405530",
           "profile_image_url_https": "https://pbs.twimg.com/profile_images/1445764922474827784/W2zEPN7U_normal.jpg",
           "profile_interstitial_type": "",
           "screen_name": "TwitterDev",
-          "statuses_count": 4076,
+          "statuses_count": 4080,
           "translator_type": "regular",
           "url": "https://t.co/9wI31m3ELF",
           "verified": false,
@@ -96,6 +96,7 @@
           }
         },
         "is_profile_translatable": false,
+        "has_hidden_likes_on_profile": false,
         "verification_info": {
           "reason": {
             "description": {
@@ -113,16 +114,22 @@
                   "from_index": 75,
                   "to_index": 85,
                   "ref": {
-                    "url": "https://help.twitter.com/en/rules-and-policies/profile-labels",
+                    "url": "https://twitter.com/i/twitter_blue_sign_up",
                     "url_type": "ExternalUrl"
                   }
                 }
               ]
-            }
+            },
+            "verified_since_msec": "1395425700472"
           }
         },
-        "business_account": {}
+        "highlights_info": {
+          "can_highlight_tweets": true,
+          "highlighted_tweets": "0"
+        },
+        "business_account": {},
+        "creator_subscriptions_count": 0
       }
     }
   }
-}
+}

tests/test_utils.py

@@ -0,0 +1,18 @@
+from twscrape.utils import parse_cookies
+
+
+def test_cookies_parse():
+    val = "abc=123; def=456; ghi=789"
+    assert parse_cookies(val) == {"abc": "123", "def": "456", "ghi": "789"}
+
+    val = '{"abc": "123", "def": "456", "ghi": "789"}'
+    assert parse_cookies(val) == {"abc": "123", "def": "456", "ghi": "789"}
+
+    val = '[{"name": "abc", "value": "123"}, {"name": "def", "value": "456"}, {"name": "ghi", "value": "789"}]'
+    assert parse_cookies(val) == {"abc": "123", "def": "456", "ghi": "789"}
+
+    val = "eyJhYmMiOiAiMTIzIiwgImRlZiI6ICI0NTYiLCAiZ2hpIjogIjc4OSJ9"
+    assert parse_cookies(val) == {"abc": "123", "def": "456", "ghi": "789"}
+
+    val = "W3sibmFtZSI6ICJhYmMiLCAidmFsdWUiOiAiMTIzIn0sIHsibmFtZSI6ICJkZWYiLCAidmFsdWUiOiAiNDU2In0sIHsibmFtZSI6ICJnaGkiLCAidmFsdWUiOiAiNzg5In1d"
+    assert parse_cookies(val) == {"abc": "123", "def": "456", "ghi": "789"}

twscrape/account.py

@@ -62,4 +62,7 @@ class Account(JSONTrait):
         client.headers["x-twitter-active-user"] = "yes"
         client.headers["x-twitter-client-language"] = "en"
 
+        if "ct0" in client.cookies:
+            client.headers["x-csrf-token"] = client.cookies["ct0"]
+
         return client

twscrape/accounts_pool.py

@@ -11,7 +11,7 @@ from .account import Account
 from .db import execute, fetchall, fetchone
 from .logger import logger
 from .login import login
-from .utils import utc_ts
+from .utils import parse_cookies, utc_ts
 
 
 class AccountInfo(TypedDict):
@@ -35,17 +35,17 @@ class AccountsPool:
         self._db_file = db_file
 
     async def load_from_file(self, filepath: str, line_format: str):
-        assert "username" in line_format, "username is required"
-        assert "password" in line_format, "password is required"
-        assert "email" in line_format, "email is required"
-        assert "email_password" in line_format, "email_password is required"
-
         line_delim = guess_delim(line_format)
         tokens = line_format.split(line_delim)
 
+        required = set(["username", "password", "email", "email_password"])
+        if not required.issubset(tokens):
+            raise ValueError(f"Invalid line format: {line_format}")
+
         with open(filepath, "r") as f:
             lines = f.read().split("\n")
             lines = [x.strip() for x in lines if x.strip()]
+
             for line in lines:
                 data = [x.strip() for x in line.split(line_delim)]
                 if len(data) < len(tokens):
@@ -53,7 +53,8 @@ class AccountsPool:
                     continue
 
                 data = data[: len(tokens)]
-                await self.add_account(**{k: v for k, v in zip(tokens, data)})
+                vals = {k: v for k, v in zip(tokens, data) if k != "_"}
+                await self.add_account(**vals)
 
     async def add_account(
         self,
@@ -63,6 +64,7 @@ class AccountsPool:
         email_password: str,
         user_agent: str | None = None,
         proxy: str | None = None,
+        cookies: str | None = None,
     ):
         qs = "SELECT * FROM accounts WHERE username = :username"
         rs = await fetchone(self._db_file, qs, {"username": username})
@@ -82,9 +84,13 @@ class AccountsPool:
             locks={},
             stats={},
             headers={},
-            cookies={},
+            cookies=parse_cookies(cookies) if cookies else {},
             proxy=proxy,
         )
+
+        if "ct0" in account.cookies:
+            account.active = True
+
         await self.save(account)
 
     async def delete_accounts(self, usernames: str | list[str]):

twscrape/api.py

@@ -7,12 +7,7 @@ from .queue_client import QueueClient
 from .utils import encode_params, find_obj, get_by_path, to_old_obj, to_old_rep
 
 SEARCH_FEATURES = {
-    "rweb_lists_timeline_redesign_enabled": True,
-    "creator_subscriptions_tweet_preview_api_enabled": True,
-    "responsive_web_twitter_article_tweet_consumption_enabled": False,
     "tweet_with_visibility_results_prefer_gql_limited_actions_policy_enabled": True,
-    "responsive_web_media_download_video_enabled": False,
-    "longform_notetweets_inline_media_enabled": True,
 }
 
 
@@ -109,7 +104,12 @@ class API:
     async def user_by_id_raw(self, uid: int, kv=None):
         op = OP_UserByRestId
         kv = {"userId": str(uid), "withSafetyModeUserFields": True, **(kv or {})}
-        return await self._gql_item(op, kv)
+        ft = {
+            "hidden_profile_likes_enabled": True,
+            "highlights_tweets_tab_ui_enabled": True,
+            "creator_subscriptions_tweet_preview_api_enabled": True,
+        }
+        return await self._gql_item(op, kv, ft)
 
     async def user_by_id(self, uid: int, kv=None):
         rep = await self.user_by_id_raw(uid, kv=kv)
@@ -121,7 +121,13 @@ class API:
     async def user_by_login_raw(self, login: str, kv=None):
         op = OP_UserByScreenName
         kv = {"screen_name": login, "withSafetyModeUserFields": True, **(kv or {})}
-        return await self._gql_item(op, kv)
+        ft = {
+            "highlights_tweets_tab_ui_enabled": True,
+            "hidden_profile_likes_enabled": True,
+            "creator_subscriptions_tweet_preview_api_enabled": True,
+            "subscriptions_verification_info_verified_since_enabled": True,
+        }
+        return await self._gql_item(op, kv, ft)
 
     async def user_by_login(self, login: str, kv=None):
         rep = await self.user_by_login_raw(login, kv=kv)

twscrape/constants.py

@@ -3,6 +3,18 @@ TOKEN = "Bearer AAAAAAAAAAAAAAAAAAAAANRILgAAAAAAnNwIzUejRCOuH5E6I8xnZz4puTs%3D1Z
 GQL_URL = "https://twitter.com/i/api/graphql"
 LOGIN_URL = "https://api.twitter.com/1.1/onboarding/task.json"
 
+OP_SearchTimeline = "L1VfBERtzc3VkBBT0YAYHA/SearchTimeline"
+OP_UserByRestId = "Lxg1V9AiIzzXEiP2c8dRnw/UserByRestId"
+OP_UserByScreenName = "oUZZZ8Oddwxs8Cd3iW3UEA/UserByScreenName"
+OP_TweetDetail = "NmCeCgkVlsRGS1cAwqtgmw/TweetDetail"
+OP_Followers = "FKV1jfu4AawGapl2KCZbQw/Followers"
+OP_Following = "sKlU5dd_nanz9P2CxBt2sg/Following"
+OP_Retweeters = "Gnw_Swm60cS-biSLn2OWNw/Retweeters"
+OP_Favoriters = "rUyh8HWk8IXv_fvVKj3QjA/Favoriters"
+OP_UserTweets = "x8SpjuBpqoww-edf0aUUKA/UserTweets"
+OP_UserTweetsAndReplies = "RB2KVuVBRZe4GW8KkoVF2A/UserTweetsAndReplies"
+OP_ListLatestTweetsTimeline = "2Vjeyo_L0nizAUhHe3fKyA/ListLatestTweetsTimeline"
+
 GQL_FEATURES = {
     "blue_business_profile_image_shape_enabled": True,
     "responsive_web_graphql_exclude_directive_enabled": True,
@@ -23,16 +35,9 @@ GQL_FEATURES = {
     "responsive_web_text_conversations_enabled": False,
     "longform_notetweets_rich_text_read_enabled": True,
     "responsive_web_enhance_cards_enabled": False,
+    "creator_subscriptions_tweet_preview_api_enabled": True,
+    "longform_notetweets_inline_media_enabled": True,
+    "responsive_web_media_download_video_enabled": False,
+    "rweb_lists_timeline_redesign_enabled": True,
+    "responsive_web_twitter_article_tweet_consumption_enabled": False,
 }
-
-OP_SearchTimeline = "L1VfBERtzc3VkBBT0YAYHA/SearchTimeline"
-OP_UserByRestId = "Lxg1V9AiIzzXEiP2c8dRnw/UserByRestId"
-OP_UserByScreenName = "oUZZZ8Oddwxs8Cd3iW3UEA/UserByScreenName"
-OP_TweetDetail = "NmCeCgkVlsRGS1cAwqtgmw/TweetDetail"
-OP_Followers = "FKV1jfu4AawGapl2KCZbQw/Followers"
-OP_Following = "sKlU5dd_nanz9P2CxBt2sg/Following"
-OP_Retweeters = "Gnw_Swm60cS-biSLn2OWNw/Retweeters"
-OP_Favoriters = "rUyh8HWk8IXv_fvVKj3QjA/Favoriters"
-OP_UserTweets = "x8SpjuBpqoww-edf0aUUKA/UserTweets"
-OP_UserTweetsAndReplies = "RB2KVuVBRZe4GW8KkoVF2A/UserTweetsAndReplies"
-OP_ListLatestTweetsTimeline = "2Vjeyo_L0nizAUhHe3fKyA/ListLatestTweetsTimeline"

twscrape/queue_client.py

@@ -112,14 +112,21 @@ class QueueClient:
         except json.JSONDecodeError:
             res: Any = {"_raw": rep.text}
 
-        fn = logger.info if rep.status_code == 200 else logger.warning
-        fn(f"{rep.status_code:3d} - {req_id(rep)}")
-
+        msg = "OK"
         if "errors" in res:
-            msg = "; ".join([x["message"] for x in res["errors"]])
-            if rep.status_code == 200 and "_Missing: No status found with that ID." in msg:
-                return  # ignore this error
+            msg = "; ".join([f'({x.get("code", -1)}) {x["message"]}' for x in res["errors"]])
 
+        fn = logger.info if rep.status_code == 200 else logger.warning
+        fn(f"{rep.status_code:3d} - {req_id(rep)} - {msg}")
+
+        if msg.startswith("The following features cannot be null"):
+            logger.error(f"Invalid request: {msg}")
+            exit(1)
+
+        if rep.status_code == 200 and "_Missing: No status found with that ID." in msg:
+            return  # ignore this error
+
+        if msg != "OK":
             raise ApiError(rep, res)
 
         rep.raise_for_status()

twscrape/utils.py

@@ -1,7 +1,8 @@
+import base64
 import json
 from collections import defaultdict
 from datetime import datetime, timezone
-from typing import Any, AsyncGenerator, Callable, TypedDict, TypeVar
+from typing import Any, AsyncGenerator, Callable, TypeVar
 
 from httpx import HTTPStatusError, Response
 
@@ -181,3 +182,26 @@ def print_table(rows: list[dict], hr_after=False):
     print("\n".join(lines))
     if hr_after:
         print("-" * max_len)
+
+
+def parse_cookies(val: str) -> dict[str, str]:
+    try:
+        val = base64.b64decode(val).decode()
+    except Exception:
+        pass
+
+    try:
+        try:
+            res = json.loads(val)
+            if isinstance(res, list):
+                return {x["name"]: x["value"] for x in res}
+            if isinstance(res, dict):
+                return res
+        except json.JSONDecodeError:
+            res = val.split("; ")
+            res = [x.split("=") for x in res]
+            return {x[0]: x[1] for x in res}
+    except Exception:
+        pass
+
+    raise ValueError(f"Invalid cookie value: {val}")