Use flint for link checking (#2654)

Signed-off-by: Gregor Zeitlinger <gregor.zeitlinger@grafana.com>

Author: zeitlinger

.github/config/lychee.toml

@@ -9,12 +9,9 @@ max_concurrency = 4
 # Check link anchors
 include_fragments = true
 
-remap = [
-  # workaround for https://github.com/lycheeverse/lychee/issues/1729
-  "https://github.com/(.*?)/(.*?)/blob/(.*?)/(.*#.*)$ https://raw.githubusercontent.com/$1/$2/$3/$4"
-]
-
 exclude = [
   # excluding links to pull requests and issues is done for performance
   "^https://github.com/open-telemetry/opentelemetry-java-contrib/(issues|pull)/\\d+$",
+  # groovy-lang.org is frequently unreachable / times out
+  "^https?://[^/]*groovy-lang\\.org",
 ]

.github/renovate.json5

@@ -266,6 +266,17 @@
     },
   ],
   customManagers: [
+    {
+      // keep SHA-pinned raw.githubusercontent.com URLs in mise.toml up to date
+      customType: 'regex',
+      datasourceTemplate: 'github-tags',
+      managerFilePatterns: [
+        '/^mise\\.toml$/',
+      ],
+      matchStrings: [
+        'https://raw\\.githubusercontent\\.com/(?<depName>[^/]+/[^/]+)/(?<currentDigest>[a-f0-9]{40})/.*#\\s*(?<currentValue>v\\S+)',
+      ],
+    },
     {
       customType: 'regex',
       datasourceTemplate: 'npm',

.github/workflows/reusable-link-check.yml

@@ -12,20 +12,20 @@ jobs:
     steps:
       - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
         with:
-          fetch-depth: 0 # needed for merge-base used in lint:links-in-modified-files
+          fetch-depth: 0 # needed for merge-base used in modified-files mode
 
       - uses: jdx/mise-action@5228313ee0372e111a38da051671ca30fc5a96db # v3.6.3
 
       - name: Link check for pull requests
         if: github.event_name == 'pull_request'
         env:
           GITHUB_TOKEN: ${{ github.token }}
-        run: |
-          mise run lint:local-links
-          mise run lint:links-in-modified-files --base origin/${{ github.base_ref }} --head ${{ github.event.pull_request.head.sha }}
+          GITHUB_HEAD_SHA: ${{ github.event.pull_request.head.sha }}
+          PR_HEAD_REPO: ${{ github.event.pull_request.head.repo.full_name }}
+        run: mise run lint:links
 
       - name: Link check for pushes and scheduled workflows
         if: github.event_name != 'pull_request'
         env:
           GITHUB_TOKEN: ${{ github.token }}
-        run: mise run lint:links
+        run: mise run lint:links --full

.mise/tasks/lint/links-in-modified-files.sh

@@ -11,7 +11,7 @@ To use:
 * Follow the [instructions](https://github.com/open-telemetry/opentelemetry-java/blob/main/sdk-extensions/incubator/README.md#declarative-configuration) to configure OpenTelemetry with declarative configuration.
 * Configure the `.tracer_provider.sampler` to include the `cel_based` sampler.
 
-Support is now available for the java agent, see an [example here](https://github.com/open-telemetry/opentelemetry-java-examples/blob/main/javaagent).
+Support is now available for the java agent, see an [example here](https://github.com/open-telemetry/opentelemetry-java-examples/tree/main/javaagent).
 
 ## Overview
 

.mise/tasks/lint/links.sh

@@ -165,7 +165,7 @@ Like with the JMX Gatherer, the selection of provided metrics to use is still do
 However, there is now two distinct sets of metrics to select from using the `otel.jmx.target.source` configuration option:
 
 - `legacy`: [metrics definitions](./src/main/resources) equivalent to JMX Gatherer definitions to help transition and preserve compatibility
-- `instrumentation`: [metrics definitions inherited from instrumentation](https://github.com/open-telemetry/opentelemetry-java-instrumentation/blob/main/instrumentation/jmx-metrics/library/src/main/resources/jmx/rules/), which is now used as a reference for JMX metrics, those also aim to provide better alignment with [metrics semantic conventions](https://opentelemetry.io/docs/specs/semconv/general/metrics/).
+- `instrumentation`: [metrics definitions inherited from instrumentation](https://github.com/open-telemetry/opentelemetry-java-instrumentation/tree/main/instrumentation/jmx-metrics/library/src/main/resources/jmx/rules/), which is now used as a reference for JMX metrics, those also aim to provide better alignment with [metrics semantic conventions](https://opentelemetry.io/docs/specs/semconv/general/metrics/).
 
 In both cases, the metrics definitions themselves are embedded in the JMX Scraper binary, thus they
 will only change if the release version of the JMX Scraper binary changes.

.mise/tasks/lint/local-links.sh

@@ -11,3 +11,8 @@ idiomatic_version_file_enable_tools = []
 windows_executable_extensions = ["sh"]
 windows_default_file_shell_args = "bash"
 use_file_shell_for_executable_tasks = true
+
+# Pick the tasks you need from flint (https://github.com/grafana/flint)
+[tasks."lint:links"]
+description = "Check for broken links in changed files + all local links"
+file = "https://raw.githubusercontent.com/grafana/flint/8a39d96e17327c54974623b252eb5260d67ed68a/tasks/lint/links.sh" # v0.9.1

cel-sampler/README.md

@@ -13,7 +13,7 @@ To use:
 * Follow the [instructions](https://github.com/open-telemetry/opentelemetry-java/blob/main/sdk-extensions/incubator/README.md#declarative-configuration) to configure OpenTelemetry with declarative configuration.
 * Configure the `.tracer_provider.sampler` to include the `rule_based_routing` sampler.
 
-Support is now available for the java agent, see an [example here](https://github.com/open-telemetry/opentelemetry-java-examples/blob/main/javaagent).
+Support is now available for the java agent, see an [example here](https://github.com/open-telemetry/opentelemetry-java-examples/tree/main/javaagent).
 
 Schema for `rule_based_routing` sampler: