f2f-scraper

Personal bulk-downloader for f2f.com (formerly f2f.net): posts, stories, and full chat conversations — including paid (fan) content you have access to.

Built to replace the broken Gertje823/f2f-downloader, which stopped working when f2f.net migrated to f2f.com and moved from HTML scraping to a cookie-authenticated JSON API.

⚠️ Private tool for archiving content you legally have access to with your own account. Respect creators' rights — do not redistribute. Using scrapers may violate the site's ToS and could get your account banned.

---

Features

f2f_scraper.py — posts & stories

• Full post feed for any creator (paginated via the API cursor)
• Images downloaded from signed CDN URLs
• Videos, including HLS streams: direct MP4 download, else HLS segment download + ffmpeg remux to MP4, else yt-dlp fallback
• Stories (24h posts) via --stories
• Comments, like/save counts and per-post metadata via -j
• Avatar + profile banner saved once per creator
• Resumable & idempotent: existing files are skipped on rerun
• Locked/paid posts you haven't paid for are skipped cleanly

f2f_chat_downloader.py — chat conversations

• Complete message history (cursor-paginated) for any creator chat
• All downloadable media: images + videos (same HLS pipeline)
• transcript.json — full history incl. raw API data + downloaded-file map
• transcript.txt — readable transcript with timestamps & sender
• Smart handling of edge cases:
  • Paywalled (PPV) messages → marked in transcript with price, poster preview saved
  • Deleted messages → marked, nothing fetched
  • Reply-quote messages → correctly ignored
• Rerun after purchasing a PPV message and it downloads automatically

run_all.py — one command, everything

• Auto-discovery: finds every creator you chat with, detects your own account and excludes it
• Subscription scope filter using the creator-profile flags (following, subscribed, cancelled):
  • --scope followed → only creators you currently follow (free) or are a fan of (paid) — excludes old chat partners
  • --scope fan → paid subscriptions only
  • --scope auto (default) → followed + fan, falls back to chats if the profile check is unavailable
  • --scope all → every chat creator (legacy behavior)
• Runs posts + stories, then chat, per creator; one failure doesn't stop the run
• Final SUMMARY table per creator (posts / chat → OK / FAIL / SKIP)
• --users a,b,c to limit the run to specific creators

---

Requirements

• Python 3.10+
• ffmpeg on your PATH (needed for HLS video conversion)
• Python packages:

pip install -r requirements.txt

(requirements.txt contains: requests, tqdm, yt-dlp.)

---

Cookie setup

The f2f.com API requires a logged-in session, passed via browser cookies.

1. The repo ships cookies-template.txt — it contains the cookie names only (no values): sessionid, csrftoken, shield_FPC, splash.
2. Copy it to cookies.txt in the same folder:

copy cookies-template.txt cookies.txt

3. Fill in the real values for your logged-in session, one name=value per line. Easiest way to get them:
   • Log in at https://f2f.com
   • Install the browser extension "Get cookies.txt LOCALLY" (Chrome/Edge) and export — or copy each value from DevTools (F12 → Application → Cookies → https://f2f.com)

Cookies expire: if you see 401 Unauthorized, re-export and rerun.

---

Commands

One command for everything

Command	What it does
python run_all.py -c cookies.txt	Download everything for all creators you follow or are a fan of
python run_all.py -c cookies.txt --scope fan	Only paid (fan) subscriptions
python run_all.py -c cookies.txt --scope followed	Follow (free) + fan (paid) — same as auto when profiles resolve
python run_all.py -c cookies.txt --scope all	Every chat creator, no scope filter
python run_all.py -c cookies.txt --users kimholland	Limit the run to specific creators
python run_all.py -c cookies.txt --no-chat	Posts/stories only
python run_all.py -c cookies.txt --no-posts	Chats only
python run_all.py -c cookies.txt --no-stories	Skip stories
python run_all.py -c cookies.txt -o .\downloads	Choose the output root folder

Single creator

# posts + stories + JSON metadata for one creator
python f2f_scraper.py -u kimholland -j --stories -o downloads

# full chat (transcript + media) for one creator
python f2f_chat_downloader.py -u kimholland -c cookies.txt -o downloads

Useful flags:

• -c/--cookies — cookie file (default cookies.txt)
• -o/--outdir — output root directory
• -j — write per-post/story JSON metadata (scraper)
• --stories — include 24h stories (scraper)
• -n/--name-len — description characters used in filenames (default 30)
• --chat-uuid — address a chat by UUID instead of username (chat downloader)
• --no-media — transcript export only, skip media (chat downloader)

---

Output layout

downloads/
└── <creator>/
    ├── avatar.jpg, banner.png
    ├── posts/                    images, videos, JSON metadata
    ├── stories/                  24h posts
    └── chat/
        ├── transcript.json       full history + raw API data
        ├── transcript.txt        readable transcript
        ├── *_chat_*.{jpg,mp4}    downloaded media
        └── *_paywalled_preview_*.jpg   poster previews of PPV messages

---

Repository files

File	Purpose
run_all.py	Orchestrator: discovery, scope filter, runs both tools per creator
f2f_scraper.py	Post/story downloader (images, HLS video, comments, metadata)
f2f_chat_downloader.py	Chat transcript + media downloader
requirements.txt	Python dependencies (pip install -r requirements.txt)
cookies-template.txt	Cookie names template — copy to cookies.txt and fill in values
cookies.txt	Local only — your real session cookies; never committed
README.md	This file

---

Troubleshooting

Symptom	Fix
401 Unauthorized	Cookies expired → re-export into cookies.txt
[!] script not found	Keep all three .py files + cookies.txt in one folder
TabError: inconsistent use of tabs and spaces	VS Code: Ctrl+Shift+P → "Convert Indentation to Spaces"
HLS video fails	Ensure ffmpeg is on PATH; pip install -U yt-dlp for the fallback
Repository not found on git push	See "Pushing to GitHub" below
_unresolved_media.json appears in chat/	Contains raw API data of items that failed — inspect it to fix parsing
Downloaded locked content looks blurred	That's what the API serves for content you haven't unlocked

---

Notes

• Signed CDN URLs expire quickly; media must be downloaded in the run that lists it.
• Everything is idempotent — reruns only fetch new content.
• After buying a PPV chat message, rerun the chat downloader to fetch it.
• fan scope = paid subscription (full access); follow = free follow (public content only).

---

Disclaimer

Not affiliated with or endorsed by f2f.com. Use at your own risk. Do not use this tool to distribute copyrighted material.