Skip to content
This repository has been archived by the owner on Oct 6, 2023. It is now read-only.
/ scraper Public archive
forked from LLNL/scraper

Python library for getting metadata from source code hosting tools

License

Notifications You must be signed in to change notification settings

doecode/scraper

 
 

Repository files navigation

NOTICE:

This forked repository has been archived and is no longer actively maintained (original project: LLNL/scraper). Please visit DOE CODE for more information, or for any questions, please contact [email protected].

Scraper

Scraper is a tool for scraping and visualizing open source data from various code hosting platforms, such as: GitHub.com, GitHub Enterprise, GitLab.com, hosted GitLab, and Bitbucket Server.

Getting Started: Code.gov

Code.gov is a newly launched website of the US Federal Government to allow the People to access metadata from the governments custom developed software. This site requires metadata to function, and this Python library can help with that!

To get started, you will need a GitHub Personal Auth Token to make requests to the GitHub API. This should be set in your environment or shell rc file with the name GITHUB_API_TOKEN:

    $ export GITHUB_API_TOKEN=XYZ

    $ echo "export GITHUB_API_TOKEN=XYZ" >> ~/.bashrc

Additionally, to perform the labor hours estimation, you will need to install cloc into your environment. This is typically done with a Package Manager such as npm or homebrew.

Then to generate a code.json file for your agency, you will need a config.json file to coordinate the platforms you will connect to and scrape data from. An example config file can be found in demo.json. Once you have your config file, you are ready to install and run the scraper!

    # Install Scraper from a local copy of this repository
    $ pip install -e .
    # OR
    # Install Scraper from PyPI
    $ pip install llnl-scraper

    # Run Scraper with your config file ``config.json``
    $ scraper --config config.json

A full example of the resulting code.json file can be found here.

Config File Options

The configuration file is a json file that specifies what repository platforms to pull projects from as well as some settings that can be used to override incomplete or inaccurate data returned via the scraping.

The basic structure is:

{
    // REQUIRED
    "contact_email": "...",  // Used when the contact email cannot be found otherwise

    // OPTIONAL
    "agency": "...",         // Your agency abbreviation here
    "organization": "...",   // The organization within the agency
    "permissions": { ... },  // Object containing default values for usageType and exemptionText

    // Platform configurations, described in more detail below
    "GitHub": [ ... ],
    "GitLab": [ ... ],
    "Bitbucket": [ ... ],
}
"GitHub": [
    {
        "url": "https://github.com",  // GitHub.com or GitHub Enterprise URL to inventory
        "token": null,                // Private token for accessing this GitHub instance
        "public_only": true,          // Only inventory public repositories

        "orgs": [ ... ],    // List of organizations to inventory
        "repos": [ ... ],   // List of single repositories to inventory
        "exclude": [ ... ]  // List of organizations / repositories to exclude from inventory
    }
],
"GitLab": [
    {
        "url": "https://gitlab.com",  // GitLab.com or hosted GitLab instance URL to inventory
        "token": null,                // Private token for accessing this GitHub instance
        "fetch_languages": false,     // Include individual calls to API for language metadata. Very slow, so defaults to false. (eg, for 191 projects on internal server, 5 seconds for False, 12 minutes, 38 seconds for True)

        "orgs": [ ... ],    // List of organizations to inventory
        "repos": [ ... ],   // List of single repositories to inventory
        "exclude": [ ... ]  // List of groups / repositories to exclude from inventory
    }
]
"Bitbucket": [
    {
        "url": "https://bitbucket.internal",  // Base URL for a Bitbucket Server instance
        "username": "",                       // Username to authenticate with
        "password": "",                       // Password to authenticate with
        "token": "",                          // Token to authenticate with, if supplied username and password are ignored

        "exclude": [ ... ]  // List of projects / repositories to exclude from inventory
    }
]
"TFS": [
    {
        "url": "https://tfs.internal",  // Base URL for a Team Foundation Server (TFS) or Visual Studio Team Services (VSTS) or Azure DevOps instance
        "token": null,                  // Private token for accessing this TFS instance

        "exclude": [ ... ]  // List of projects / repositories to exclude from inventory
    }
]

License

Scraper is released under an MIT license. For more details see the LICENSE file.

LLNL-CODE-705597

About

Python library for getting metadata from source code hosting tools

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Python 99.9%
  • Makefile 0.1%