steamarchiver

A set of utilities to preserve Steam content, including tools to archive app information, archive depots in SteamPipe CDN content and extract them, and download versions of the Steam client.

Send feedback to the mailing list.

The official repository is hosted on Sourcehut, and mirrored on GitHub. Contributions welcome: send patches to the mailing list or pull requests on GitHub (patches are preferred).

Special credit goes to SolsticeGameStudios for maintaining the current fork of the ValvePython library used by steamarchiver.

Usage

For the Python scripts, install the requirements:

pip3 install -r requirements.txt

Note: If you have previously run this command, you will need to manually uninstall the steam python libraries before rerunning. This ensures you get a tested Git Commit of the Steam Python Library.

pip3 uninstall steam

Many scripts require authentication, which can be passed on the command line using the -u [username] flag. -p [password] will allow you to pass the password on the command line as well; if you do not do this, you will be interactively prompted for a password.

Usage for the Python scripts:

• depot_archiver.py downloads depots (the logical groupings of game content files Steam delivers). You can give it an appid, in which case it'll download the latest depots for that app, or you can give it a specific appid, depotid, and manifest number (manifests are specific versions of depots.) It will download the manifest, appinfo, and encrypted depot chunks. Ownership of an app is required to download its manifest; you can only download an app without logging in if its manifest has previously been downloaded. Most (non-dedicated server, non-Valve) free apps are not available to anonymous users; you will still need to log into an account to download free games!
  • A CSV File can be used as the input data set. The file needs to start with 4 headers in the following order. It is highly recommended to make the CSV using a text editor, such as Notepad. Excel has issues with the Manifest IDs due to them being long numbers. Excel's default action is to convert it and cut out some of the digits, breaking the list.
    • AppID
    • DepotID
    • ManifestID
    • Branch
  • If you are running the Steam Client on the same PC when attempting to download your files, it is prefered to include the --login-id argument, otherwise Steam will disconnect from your client. By Default, Login ID is generated based on your PC's private IP Address, so having 2 connections to steam from the same PC will conflict. Using this Argument with a random set of numbers will fix this
  • Additionally can take the output generated from workshop_scraper as input for the -wf argument. Mutually exclusive from downloading game depots
• depot_validator.py verifies every chunk in a depot folder to ensure none of the files are corrupted or bad.
• get_depot_keys.py logs into a Steam account and dumps all the depot keys it has access to, which can be used to decrypt downloaded depots. To get the key for a depot, your account must own a package that includes access to the depot, and it must also own a package that includes an app which specifies the depot should be installed; you cannot get keys for depots you have access to that aren't included in any released apps, since that's how Steam prevents you from decrypting preloaded game content. Keys will be saved to depot_keys.txt
• depot_extractor.py extracts downloaded depots. It requires the key but can work completely offline.
• list_downloaded_manifests.py can be used to verify if a particular depot/manifest has been downloaded, or it can list out the manifests used by branches of an app and check if they've been downloaded.
• get_appinfo.py downloads the latest appinfo for the specified apps, or for all publicly visible apps if run with no arguments.
• update_appinfo.py attempts to download appinfo for only the apps that have changes since the last time get_appinfo was run.
• get_client.py downloads the manifest for the Steam client and downloads all needed packages. When run without arguments it'll download the release version of the Steam client for Win32; you can also specify a different channel, e.g. get_client.py steam_client_ubuntu12, get_client.py steam_client_publicbeta_osx, get_client.py steam_cmd_linux, get_client.py steamchina_win32, etc.
• unpack_sis.py unpacks a Steam game backup or retail master (which consists of a sku.sis manifest, csd files containing depot data, and cdm files containing metadata about the locations of chunks in the csd.) Unpacking with this script is free, but for actually extracting the data afterward with depot_extractor you need to have a downloaded copy of the manifest (you can get one by running depot_archiver in dry run mode, using the -d flag) and the decryption key.
• pack_sis.py goes the other way, packing depot chunks from a depots/ subfolder into .csd and .csm archive files. This is useful for transferring (as moving two files is usually much faster than moving a couple thousand), or for restoring a backup using the Steam client (requires specifying a depot manifest so an sku.sis file can be generated.)
• diff_manifests.py displays the difference between two manifests of the same depot, showing which files were added/changed/removed, the download size of the depot, and the change in on-disk size after download.
• steam_websocket_mitm.py is an mitmproxy script to inspect Steam's WebSockets network traffic. This is only really useful for debugging.
• login.py runs an interactive login for testing purposes.
• • This script does not need to be run directly for downloading purposes.
• chunkstore.py loads a .csd/.csm and lists the depot ID, encryption, and number of chunks without unpacking anything.
• workshop_scraper.py is used to scrape the workshop page for an app.
  • This can export to a csv or json file.
  • There is a limitation in how many items and pages can be viewed depending on the workshop filter, up to 50,010 items (1667 pages).
  • use --date-start and --date-end in yyyy-mm-dd format to narrow the filtering down. Can go back as far as October of 2011, when Steam Workshop first released.
  • Some pages display with less than 30 items, which will cause the script to end prematurely. Use --force to have it continue going until it reaches the final calculated page.

The folder steamlancache contains an HTTP server (written in Golang) that you can use as a LAN cache for Steam to speed up downloads and automatically archive all the content you download in the formats these scripts expect. Just run the server with the folder the Python scripts are in as the working directory, and set a DNS record to point "lancache.steamcontent.com" to the IP of the server.

For help finding appids, depotids, and manifestids, check out https://steamdb.info (requires logging in with a non-limited Steam account to view older manifests)

Example

Get all the depot keys available to anonymous users:

python3 get_depot_keys.py

Log into a specific account and get all its depot keys:

python3 get_depot_keys.py -u [username] -p [password]

Download all the depots for Team Fortress 2:

python3 depot_archiver.py -a 440

Download the Team Fortress 2 Linux client binaries that were released at 2022-08-11 22:29:49:

python3 depot_archiver.py -a 440 232253 5841585021586447253

Download using a CSV file as the data set for apps, depots, and manifests.

python3 depot_archiver.py -csv "<path>/<to>/<csv>"

Download and save dumped manifest as readable JSON file for Team Fortress 2 (Recommend use with DryRun "-d").

python3 depot_archiver --debug-manifest -a 440 232253 5841585021586447253

or

python3 depot_archiver --debug-manifest -csv "<path>/<to>/<csv>"

Extract those binaries:

python3 depot_extractor.py 232253 5841585021586447253 bdbeae4f56fa865d8df2f76623d3346fcd7e56df6dee13b0f23e4a0fe160a446

(Note: the key for the above command was found in depot_keys.txt, in this line:)

232253		bdbeae4f56fa865d8df2f76623d3346fcd7e56df6dee13b0f23e4a0fe160a446	TF2 Linux client

Scrape RimWorld's Workshop:

python3 workshop_scraper.py scrape 294100 -o RimWorld

License

Copyright 2021-2024 Benjamin Lowry

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at

   http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.