pr0dl

A command-line downloader for pr0gramm.com. Fetches media URLs from the API and downloads them in parallel with resume support.

Requirements

• Rust toolchain (https://rustup.rs)

Installation

git clone https://github.com/ryodari/pr0dl.git
cd pr0dl
cargo install --path .

Authentication

NSFW, NSFL and POL content requires a valid session. Log in to pr0gramm.com in your browser, then copy the pp and me cookie values from DevTools (Application → Cookies).

Pass them via flags or environment variables:

export PR0DL_PP="your_pp_cookie"
export PR0DL_ME="your_me_cookie"

Content Flags

Flags control which content categories are fetched. Add the values together to combine categories.

Flag	Category
1	SFW
2	NSFP
4	NSFW
8	NSFL
16	POL

31 (= 1+2+4+8+16) fetches everything. This matches what the pr0gramm web app uses.

Commands

run — Fetch and download in one step

pr0dl run --user USERNAME --flags 31 --output ./downloads

Options:
  --user <USER>         Download posts by this user
  --tags <TAGS>         Download posts matching these tags
  --collection <NAME>   Download a specific collection (e.g. "favoriten")
  --flags <FLAGS>       Content flags (default: 1)
  -o, --output <DIR>    Download directory (default: ./downloads)
  -j, --jobs <N>        Parallel download workers (default: 3)
  --pp <VALUE>          Session cookie (env: PR0DL_PP)
  --me <VALUE>          Session cookie (env: PR0DL_ME)
  --state <FILE>        Fetch progress file for resume (default: .fetch_state.json)
  --failed <FILE>       File to write failed URLs into (default: failed.txt)

fetch — Collect URLs only

Writes one URL per line to a file or stdout. Useful if you want to inspect or filter URLs before downloading.

pr0dl fetch --user USERNAME --flags 31 --output urls.txt

Same options as run, minus --jobs and --failed.

download — Download from a URL list

Downloads from a file previously created by fetch.

pr0dl download --input urls.txt --output ./downloads

Options:
  -i, --input <FILE>    File with one URL per line
  -o, --output <DIR>    Download directory (default: ./downloads)
  -j, --jobs <N>        Parallel download workers (default: 3)
  --failed <FILE>       File to write failed URLs into (default: failed.txt)

Examples

Download all posts by a user (all content categories):

pr0dl run --user Ryodari --flags 31 --output ./downloads

Download posts matching a tag search:

pr0dl run --tags "landscape photography" --flags 1 --output ./downloads

Download a user's favourites:

pr0dl run --user Ryodari --collection favoriten --flags 31 --output ./downloads

Download a specific named collection:

pr0dl run --user Ryodari --collection "meine-sammlung" --flags 31 --output ./downloads

Fetch URLs first, then download separately:

pr0dl fetch --user Ryodari --flags 31 --output urls.txt
pr0dl download --input urls.txt --output ./downloads -j 5

Retry failed downloads:

pr0dl download --input failed.txt --output ./downloads

Resume on Interruption

fetch and run save progress to .fetch_state.json after every page. If interrupted, re-running the same command resumes from where it left off. The state file is deleted automatically on successful completion.

Parallel Workers

The -j flag controls how many files are downloaded simultaneously. The default of 3 is conservative to avoid rate limiting. Values between 3 and 8 are reasonable depending on your connection. Higher values offer no benefit when the server or your bandwidth is the bottleneck, and increase the risk of receiving 429 responses.