| """FB Comments Crawler — Gradio Space UI. |
| |
| Crawl 100% of comments (with pagination + replies) from a Facebook post you |
| have permission to read (see README), and download CSV/JSON with commenter UID. |
| |
| The OAuth login step (getting a Page access token) must be done on your local |
| machine because Facebook redirects the browser to localhost. This Space accepts |
| the token you already have and does the crawling. |
| """ |
|
|
| import json |
| import os |
| import tempfile |
| from pathlib import Path |
|
|
| import gradio as gr |
| import pandas as pd |
|
|
| from fb_comments import ( |
| CommentCrawler, |
| CommentCrawlerError, |
| RateLimitError, |
| parse_post_identifier, |
| resolve_post_id, |
| ) |
|
|
|
|
| |
| |
| |
| def crawl(token: str, url_or_id: str, include_replies: bool): |
| """Run the crawl; return (csv_path, json_path, df, status_text).""" |
| token = (token or "").strip() |
| url_or_id = (url_or_id or "").strip() |
| if not token: |
| return None, None, None, "❌ Token requi. Get a Page access token first (see tab Help)." |
| if not url_or_id: |
| return None, None, None, "❌ Post URL/id requi." |
|
|
| try: |
| post_id = parse_post_identifier(url_or_id) |
| except Exception as e: |
| return None, None, None, f"❌ Invalid post URL/id: {e}" |
|
|
| cc = CommentCrawler(token) |
| try: |
| post_id = resolve_post_id(post_id, cc) |
| except Exception: |
| pass |
|
|
| try: |
| comments = list(cc.comments(post_id, include_replies=bool(include_replies))) |
| except RateLimitError as e: |
| return None, None, None, f"❌ Rate limited: {e}" |
| except CommentCrawlerError as e: |
| return None, None, None, f"❌ API error: {e} — verify token is a Page token of a page you manage." |
| except Exception as e: |
| return None, None, None, f"❌ Unexpected: {e}" |
|
|
| if not comments: |
| return None, None, None, "No comments (or post not readable with this token)." |
|
|
| rows = [] |
| for c in comments: |
| author = c.get("from") or {} |
| rows.append({ |
| "comment_id": c.get("id"), |
| "parent_id": (c.get("parent") or {}).get("id"), |
| "name": author.get("name"), |
| "uid": author.get("id"), |
| "message": c.get("message") or "", |
| "created_time": c.get("created_time"), |
| "like_count": c.get("like_count"), |
| }) |
| df = pd.DataFrame(rows) |
| out_dir = tempfile.gettempdir() |
| base = f"comments_{len(rows)}" |
| csv_path = str(Path(out_dir) / f"{base}.csv") |
| json_path = str(Path(out_dir) / f"{base}.json") |
| df.to_csv(csv_path, index=False, encoding="utf-8-sig") |
| with open(json_path, "w", encoding="utf-8") as fh: |
| json.dump(rows, fh, ensure_ascii=False, indent=2) |
| return csv_path, json_path, df, f"OK: {len(rows)} comments (top-level + replies) collected." |
|
|
|
|
| |
| |
| |
| HELP = """ |
| ### How to get a Page access token |
| 1. Create a FB app at developers.facebook.com/apps → **Facebook Login**, keep **Development** mode. |
| 2. Add `http://localhost:8080/callback` to **Valid OAuth Redirect URIs** + enable Web OAuth. |
| 3. Add your account as **Admin** under Roles. (No App Review needed for pages you own.) |
| 4. On your own machine run: |
| ``` |
| pip install . # install fb-comments-crawler |
| fbcomments login --app-id <APP_ID> --app-secret <SECRET> --port 8080 |
| ``` |
| → open the printed URL, log in as page admin → token saved to `token.json`. |
| 5. Paste the `page_access_token` value into the box here. |
| |
| > You can only read comments on posts of **Pages you manage** (permission |
| > `pages_read_engagement`). Posts of other pages return error #283. |
| The **uid** column holds the **PSID** (Page-Scoped user ID) — the modern stable FB user identifier. |
| """ |
|
|
|
|
| |
| |
| |
| with gr.Blocks(title="FB Comments Crawler") as demo: |
| gr.Markdown("# FB Comments Crawler\nPull 100% of comments (with UID) from a FB post you manage, download CSV/JSON.") |
|
|
| with gr.Tab("Crawl"): |
| token = gr.Textbox(label="Page access token", type="password", |
| placeholder="EAAG... paste token from token.json") |
| url = gr.Textbox(label="Post URL or id", |
| placeholder="https://www.facebook.com/MyPage/posts/pfbid0...") |
| replies = gr.Checkbox(label="Include reply comments", value=True) |
| btn = gr.Button("Crawl", variant="primary") |
|
|
| status = gr.Markdown("Ready.") |
| df_out = gr.Dataframe(label="Comments preview") |
| csv_file = gr.File(label="Download CSV") |
| json_file = gr.File(label="Download JSON") |
|
|
| btn.click(fn=crawl, inputs=[token, url, replies], |
| outputs=[csv_file, json_file, df_out, status]) |
|
|
| with gr.Tab("Help"): |
| gr.Markdown(HELP) |
|
|
| demo.queue() |
|
|
| if __name__ == "__main__": |
| demo.launch() |
