markdown-link-check

Extracts links from markdown texts and checks whether each link is alive (200 OK) or dead. mailto: links are also validated.

Installation

To add the module to your project, run:

npm install --save-dev markdown-link-check

To install the command line tool globally, run:

npm install -g markdown-link-check

Run using Docker

Docker images are built with each release. Use the stable tag for the current stable release.

Add current directory with your README.md file as read only volume to docker run:

docker run -v .:/tmp:ro --rm -i ghcr.io/tcort/markdown-link-check:stable /tmp/README.md

Alternatively, if you wish to target a specific release, images are tagged with semantic versions (i.e. 3, 3.8, 3.8.3)

Run in a GitHub action

Please head on to github-action-markdown-link-check.

Run as a pre-commit hook

To run as a pre-commit hook:

- repo: https://github.com/tcort/markdown-link-check
  rev: ...
  hooks:
    - id: markdown-link-check
      args: [-q]

Run in a GitLab pipeline

linkchecker:
  stage: test
  image:
    name: ghcr.io/tcort/markdown-link-check:3.11.2
    entrypoint: ["/bin/sh", "-c"]
  script:
    - markdown-link-check ./docs
  rules:
    - changes:
      - "**/*.md"

Run in other tools

• Mega-Linter: Linters aggregator including markdown-link-check

API

markdownLinkCheck(markdown, [opts,] callback)

Given a string containing markdown formatted text and a callback, extract all of the links and check if they're alive or dead. Call the callback with (err, results)

Parameters:

• markdown string containing markdown formatted text.
• opts optional options object containing any of the following optional fields:
  • showProgressBar enable an ASCII progress bar.
  • timeout timeout in ms format. (e.g. "2000ms", 20s, 1m). Default 10s.
  • httpHeaders to apply URL specific headers, see example below.
  • ignorePatterns an array of objects holding regular expressions which a link is checked against and skipped for checking in case of a match. Example: [{ pattern: /foo/ }]
  • replacementPatterns an array of objects holding regular expressions which are replaced in a link with their corresponding replacement string. This behavior allows (for example) to adapt to certain platform conventions hosting the Markdown. The special replacement {{BASEURL}} can be used to dynamically link to the base folder (used from projectBaseUrl) (for example that / points to the root of your local repository). Example: [{ pattern: /^.attachments/, replacement: "file://some/conventional/folder/.attachments" }, { pattern: ^/, replacement: "{{BASEURL}}/"}]. You can add "global": true to use a global regular expression to replace all instances.
  • projectBaseUrl the URL to use for {{BASEURL}} replacement
  • ignoreDisable if this is true then disable comments are ignored.
  • retryOn429 if this is true then retry request when response is an HTTP code 429 after the duration indicated by retry-after header.
  • retryCount the number of retries to be made on a 429 response. Default 2.
  • fallbackRetryDelay the delay in ms format. (e.g. "2000ms", 20s, 1m) for retries on a 429 response when no retry-after header is returned or when it has an invalid value. Default is 60s.
  • aliveStatusCodes a list of HTTP codes to consider as alive. Example: [200,206]
  • reporters an array of reporter functions to use for outputting results. If not specified, default output will be generated (useful when using the API programmatically). Available built-in reporters can be imported from the command-line tool.
• callback function which accepts (err, results).
  • err an Error object when the operation cannot be completed, otherwise null.
  • results an array of objects with the following properties:
    • link the link provided as input
    • status a string set to either alive, ignored or dead.
    • statusCode the HTTP status code. Set to 0 if no HTTP status code was returned (e.g. when the server is down).
    • err any connection error that occurred, otherwise null.

Disable comments

You can write html comments to disable markdown-link-check for parts of the text.

disables markdown link check. reenables markdown link check. disables markdown link check for the next line. disables markdown link check for this line.

Examples

Module

Basic usage:

'use strict';

var markdownLinkCheck = require('markdown-link-check');

markdownLinkCheck('[example](https://example.com)', function (err, results) {
    if (err) {
        console.error('Error', err);
        return;
    }
    results.forEach(function (result) {
        console.log('%s is %s', result.link, result.status);
    });
});

With options, for example using URL specific headers:

'use strict';

var markdownLinkCheck = require('markdown-link-check');

markdownLinkCheck('[example](https://example.com)', { httpHeaders: [{ urls: ['https://example.com'], headers: { 'Authorization': 'Basic Zm9vOmJhcg==' }}] }, function (err, results) {
    if (err) {
        console.error('Error', err);
        return;
    }
    results.forEach(function (result) {
        console.log('%s is %s', result.link, result.status);
    });
});

Command Line Tool

The command line tool optionally takes 1 argument, the file name or http/https URL. If not supplied, the tool reads from standard input.

Check links from a markdown file hosted on the web

markdown-link-check https://github.com/tcort/markdown-link-check/blob/master/README.md

Check links from a local markdown file

markdown-link-check ./README.md

Check links from a local markdown folder (recursive)

This checks all files in folder ./docs with file extension *.md:

markdown-link-check ./docs

The files can also be searched for and filtered manually:

find . -name \*.md -print0 | xargs -0 -n1 markdown-link-check

Usage

Usage: markdown-link-check [options] [filenamesOrDirectorynamesOrUrls...]

Options:
  -V, --version           output the version number
  -p, --progress          show progress bar
  -c, --config [config]   apply a config file (JSON), holding e.g. url specific header configuration
  -q, --quiet             displays errors only
  -v, --verbose           displays detailed error information
  -i, --ignore     ignore input paths including an ignore path
  -a, --alive <code>      comma separated list of HTTP codes to be considered as alive
  -r, --retry             retry after the duration indicated in 'retry-after' header when HTTP code is 429
  --reporters      specify reporters to use
  --projectBaseUrl   the URL to use for {{BASEURL}} replacement
  --junit-output    output file for JUnit XML report (only used with junit reporter)
  -h, --help              display help for command

Reporters

markdown-link-check supports multiple output reporters to format the results of link checking:

Default Reporter

The default reporter outputs results to the console with colored status indicators:

markdown-link-check README.md

Output format:

  ✓ https://example.com/valid-link
  ✖ https://example.com/broken-link
  / https://example.com/ignored-link

  3 links checked.

  ERROR: 1 dead links found!

JUnit Reporter

The JUnit reporter generates XML output compatible with JUnit test result format, useful for CI/CD integration:

markdown-link-check --reporters junit README.md

To specify a custom output file:

markdown-link-check --reporters junit --junit-output results.xml README.md

If no output file is specified, the results are written to junit-results.xml by default.

Multiple Reporters

You can use multiple reporters simultaneously by specifying them in a comma-separated list:

markdown-link-check --reporters default,junit README.md

This will output to both the console (default reporter) and generate a JUnit XML file.

Config file format

config.json:

• ignorePatterns: An array of objects holding regular expressions which a link is checked against and skipped for checking in case of a match.
• replacementPatterns: An array of objects holding regular expressions which are replaced in a link with their corresponding replacement string. This behavior allows (for example) to adapt to certain platform conventions hosting the Markdown. The special replacement {{BASEURL}} can be used to dynamically link to the current working directory (for example that / points to the root of your current working directory). This parameter supports named regex groups the same way as string.replace method in node.
• httpHeaders: The headers are only applied to links where the link starts with one of the supplied URLs in the urls section.
• timeout timeout in ms format. (e.g. "2000ms", 20s, 1m). Default 10s.
• retryOn429 if this is true then retry request when response is an HTTP code 429 after the duration indicated by retry-after header.
• retryCount the number of retries to be made on a 429 response. Default 2.
• fallbackRetryDelay the delay in ms format. (e.g. "2000ms", 20s, 1m) for retries on a 429 response when no retry-after header is returned or when it has an invalid value. Default is 60s.
• aliveStatusCodes a list of HTTP codes to consider as alive.
• projectBaseUrl the URL to use for {{BASEURL}} replacement

Example:

{
  "projectBaseUrl":"${workspaceFolder}",
  "ignorePatterns": [
    {
      "pattern": "^https://example.net"
    }
  ],
  "replacementPatterns": [
    {
      "pattern": "^.attachments",
      "replacement": "file://some/conventional/folder/.attachments"
    },
    {
      "pattern": "^/",
      "replacement": "{{BASEURL}}/"
    },
    {
      "pattern": "%20",
      "replacement": "-",
      "global": true
    },
    {
      "pattern": "images/(?.*)",
      "replacement": "assets/$"
    }
  ],
  "httpHeaders": [
    {
      "urls": ["https://example.com"],
      "headers": {
        "Authorization": "Basic Zm9vOmJhcg==",
        "Foo": "Bar"
      }
    }
  ],
  "timeout": "20s",
  "retryOn429": true,
  "retryCount": 5,
  "fallbackRetryDelay": "30s",
  "aliveStatusCodes": [200, 206]
}

Testing

npm test

License

See LICENSE.md