Skip to content

CONFIGURATION.md

Latest commit

 

History

History
163 lines (119 loc) · 6.43 KB

CONFIGURATION.md

File metadata and controls

163 lines (119 loc) · 6.43 KB
Raw

Configuration

Every setting, where it lives, and what it actually controls.

GUI settings

The panel used to be one "Browsers" slider for two methods that no longer share anything: fuckingfast opens no browser at all (pure HTTP via curl_cffi), datanodes is Chrome + Turnstile. It is split in three.

Common — both methods

Setting Range Default What it does
Extractors 2–32 16 how many extractions run in parallel
DL streams 2–48 48 concurrent download connections
Retries 0–5 3 extraction retries per URL (network retries are separate)

On stream count. Measured on the same line: 48 streams gave 31.9 MB/s aggregate at ~1.9 MB/s per file with ~17 active; 8 streams gave 62.2 MB/s aggregate at 8–13 MB/s per file with 5 active. The ceiling is your total bandwidth, divided by the number of streams — not a per-connection cap on the free tier. Fewer streams means more bandwidth per file and, in that comparison, a higher aggregate too.

fuckingfast.co — HTTP, no browser

No settings. The panel only reports status: curl_cffi active (green) or missing (red). If it is missing every link gets a 403, and you see it immediately instead of finding out from the log.

datanodes.to — Chrome + Turnstile

Setting Range Default What it does
Pages 1–8 8 tabs open at once on the one shared window/identity
Captcha s 30–600 30 how long a manual Turnstile solve may take
Chrome path autodetect chrome.exe to drive over CDP, with a file picker
API key string — datanodes premium key, masked in the field

None of this needs setx any more. Each value used to be read once at process start, so the only way to change it was an environment variable plus a restart. The GUI calls moon_extract.configure() immediately before a run: what is on screen is what runs. Environment variables still work as defaults.

The start banner reports what is in use per host:

▶  85 links  ·  16 extractors  ·  8 streams  ·  3 retries  ·  v2.0
   fuckingfast: direct HTTP   ·   datanodes: 8 pages, captcha 30s
   chrome: C:\Program Files\Google\Chrome\Application\chrome.exe

Environment variables

All optional. The GUI overrides them at run time; the CLI reads them directly.

Variable Default What it does
MOON_CHROME_PATH (auto) path to chrome.exe when autodetection misses it
MOON_CHROME_PROFILE %LOCALAPPDATA%\MoonDownloader\chrome-profile the dedicated profile
MOON_REAL_CHROME 1 0 = go back to Playwright's Chromium
MOON_CDP_PORT 9222 the debugging port
MOON_DN_LANES 3 pages open at once on the shared window (1–8)
MOON_DN_API_KEY (empty) datanodes key (direct_link needs premium)
MOON_DN_HEADLESS 0 1 = headless — the captcha will not solve
MOON_DN_CAPTCHA_WAIT 240 seconds to wait for a manual solve
MOON_DEBUG (off) 1 = trace every gate of the extraction

setx only affects new processes: after setting a variable you have to close and reopen the prompt (or relaunch start.bat from a fresh one), otherwise nothing appears to change.

MOON_DN_LANES does not open more windows. It bounds how many heavy pages (Turnstile + ads) stay open at once on the one shared window. Setting it very high is what broke Cloudflare in 14.6 — see ENGINEERING_NOTES.md.

The Chrome profile

datanodes runs on a dedicated profile, never your daily-driver one:

%LOCALAPPDATA%\MoonDownloader\chrome-profile

Chrome refuses --remote-debugging-port on a --user-data-dir that another Chrome process already has open, so pointing this at your everyday profile silently produces a browser you cannot attach to.

Chrome is launched with only these arguments — every extra flag is a signal:

--remote-debugging-port=9222
--user-data-dir=%LOCALAPPDATA%\MoonDownloader\chrome-profile
--no-first-run --no-default-browser-check
--disable-features=Translate,OptimizationHints
--disable-dev-shm-usage
--window-size=1440,960

All workers share one Chrome and one context, so there is one cf_clearance: solve the captcha once and Cloudflare stops re-challenging on later links instead of treating every worker as a new visitor. The profile survives between runs — the more you use it, the less you get challenged.

The datanodes API key

file/direct_link is premium-only. A free key authenticates fine (account/info → status:200) and then answers:

{"msg":"This function not allowed in API","status":403}

file/download, file/dl, file/link, file/url, file/get and file/direct all return Invalid operation. There is no free endpoint that hands out a direct link.

Leaving the key set is still worth it: the extractor tries it first, so the day the account goes premium extraction becomes instant — one JSON GET, no browser, no captcha.

Optional files

Place them next to the scripts:

File Purpose
proxies.txt proxy list — ip:port:user:pass or http://user:pass@ip:port. downloads only, see below
settings.json written by the GUI: settings, pasted links, language

What proxies cover

The pool is consulted in exactly one place — download_file in moon_download.py, where _PROXY_POOL.next() picks the entry the download session is built from. Nothing in moon_extract.py takes a proxy: the Chrome instance datanodes needs for its Turnstile challenge is launched with no --proxy-server, and the curl_cffi session fuckingfast uses is constructed without one. Both connect directly.

So a run with proxies.txt loaded has the file host seeing your own address for every page load and your proxy's for the bytes. Rotation is bandwidth cover, not identity cover — if you need the extraction half proxied too, that is not something this file can do for you today.

The practical symptom: put unreachable proxies in the list and pages still open, while downloads fail. That is the design working, not the proxy system failing.

Output files

File Contents
moontech_*.log human-readable performance report
moontech_*.json per-file metrics
moontech_cli_*.log / .json the same, from the CLI
output_links.txt extracted direct links (Links-only mode)
failed_links.txt URLs that failed every retry