|
1
|
"""Twitter/X source connector -- stub requiring auth or gallery-dl.""" |
|
2
|
|
|
3
|
import logging |
|
4
|
from pathlib import Path |
|
5
|
from typing import List, Optional |
|
6
|
|
|
7
|
from video_processor.sources.base import BaseSource, SourceFile |
|
8
|
|
|
9
|
logger = logging.getLogger(__name__) |
|
10
|
|
|
11
|
|
|
12
|
class TwitterSource(BaseSource): |
|
13
|
""" |
|
14
|
Fetch Twitter/X posts and threads. |
|
15
|
|
|
16
|
Twitter API v2 requires authentication. This connector attempts to use |
|
17
|
gallery-dl as a fallback for public tweets. |
|
18
|
|
|
19
|
Auth options: |
|
20
|
- Set TWITTER_BEARER_TOKEN env var for API v2 access |
|
21
|
- Install gallery-dl for scraping public tweets: pip install gallery-dl |
|
22
|
""" |
|
23
|
|
|
24
|
def __init__(self, url: str): |
|
25
|
self.url = url |
|
26
|
self._bearer_token: Optional[str] = None |
|
27
|
|
|
28
|
def authenticate(self) -> bool: |
|
29
|
"""Check for Twitter API token or gallery-dl availability.""" |
|
30
|
import os |
|
31
|
|
|
32
|
self._bearer_token = os.environ.get("TWITTER_BEARER_TOKEN") |
|
33
|
if self._bearer_token: |
|
34
|
return True |
|
35
|
|
|
36
|
# Check for gallery-dl fallback |
|
37
|
try: |
|
38
|
import gallery_dl # noqa: F401 |
|
39
|
|
|
40
|
logger.info("Using gallery-dl for Twitter content extraction") |
|
41
|
return True |
|
42
|
except ImportError: |
|
43
|
pass |
|
44
|
|
|
45
|
logger.error( |
|
46
|
"Twitter source requires either:\n" |
|
47
|
" 1. TWITTER_BEARER_TOKEN env var (Twitter API v2)\n" |
|
48
|
" 2. gallery-dl installed: pip install gallery-dl\n" |
|
49
|
"Twitter API access: https://developer.twitter.com/en/portal/dashboard" |
|
50
|
) |
|
51
|
return False |
|
52
|
|
|
53
|
def list_videos( |
|
54
|
self, |
|
55
|
folder_id: Optional[str] = None, |
|
56
|
folder_path: Optional[str] = None, |
|
57
|
patterns: Optional[List[str]] = None, |
|
58
|
) -> List[SourceFile]: |
|
59
|
"""Return a single SourceFile for the tweet/thread.""" |
|
60
|
return [ |
|
61
|
SourceFile( |
|
62
|
name=self.url.split("/")[-1] or "tweet", |
|
63
|
id=self.url, |
|
64
|
mime_type="text/plain", |
|
65
|
) |
|
66
|
] |
|
67
|
|
|
68
|
def download(self, file: SourceFile, destination: Path) -> Path: |
|
69
|
"""Download tweet content as text.""" |
|
70
|
destination = Path(destination) |
|
71
|
destination.parent.mkdir(parents=True, exist_ok=True) |
|
72
|
text = self.fetch_text() |
|
73
|
destination.write_text(text, encoding="utf-8") |
|
74
|
logger.info(f"Saved Twitter content to {destination}") |
|
75
|
return destination |
|
76
|
|
|
77
|
def fetch_text(self) -> str: |
|
78
|
"""Extract tweet text via API or gallery-dl.""" |
|
79
|
if self._bearer_token: |
|
80
|
return self._fetch_via_api() |
|
81
|
|
|
82
|
try: |
|
83
|
return self._fetch_via_gallery_dl() |
|
84
|
except ImportError: |
|
85
|
raise RuntimeError( |
|
86
|
"No Twitter extraction method available. See authenticate() for setup." |
|
87
|
) |
|
88
|
|
|
89
|
def _fetch_via_api(self) -> str: |
|
90
|
"""Fetch tweet via Twitter API v2.""" |
|
91
|
import re |
|
92
|
|
|
93
|
import requests |
|
94
|
|
|
95
|
match = re.search(r"/status/(\d+)", self.url) |
|
96
|
if not match: |
|
97
|
raise ValueError(f"Could not extract tweet ID from: {self.url}") |
|
98
|
|
|
99
|
tweet_id = match.group(1) |
|
100
|
resp = requests.get( |
|
101
|
f"https://api.twitter.com/2/tweets/{tweet_id}", |
|
102
|
headers={"Authorization": f"Bearer {self._bearer_token}"}, |
|
103
|
params={"tweet.fields": "author_id,created_at,text"}, |
|
104
|
timeout=15, |
|
105
|
) |
|
106
|
resp.raise_for_status() |
|
107
|
data = resp.json().get("data", {}) |
|
108
|
return f"{data.get('text', '')}\n\nCreated: {data.get('created_at', 'unknown')}" |
|
109
|
|
|
110
|
def _fetch_via_gallery_dl(self) -> str: |
|
111
|
"""Use gallery-dl to extract tweet metadata.""" |
|
112
|
import json |
|
113
|
import subprocess |
|
114
|
|
|
115
|
result = subprocess.run( |
|
116
|
["gallery-dl", "--dump-json", self.url], |
|
117
|
capture_output=True, |
|
118
|
text=True, |
|
119
|
timeout=30, |
|
120
|
) |
|
121
|
if result.returncode != 0: |
|
122
|
raise RuntimeError(f"gallery-dl failed: {result.stderr}") |
|
123
|
|
|
124
|
items = json.loads(result.stdout) |
|
125
|
texts = [] |
|
126
|
for item in items if isinstance(items, list) else [items]: |
|
127
|
if isinstance(item, dict): |
|
128
|
texts.append(item.get("content", item.get("text", str(item)))) |
|
129
|
return "\n\n".join(texts) if texts else "No text content extracted." |
|
130
|
|