Link Verifier Changes (#81)

* The link verification change
* Include the option to exclude urls, add in tests for it
* Add in more tests, and info about them in the readme
---------

.github/workflows/test.yml

@@ -216,20 +216,61 @@ jobs:
           config: ./memory_statistics_config.json
           output: ./size_table_new.html
           check_against: ./size_table_expected.html
+
   test-link-verifier:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v2
+      - uses: actions/checkout@v3
+
       - name: Setup python environment
-        uses: actions/setup-python@v2
+        uses: actions/setup-python@v3
+
+      - env:
+          stepName: "Functional | Success | Link Verifier Works & Excludes Directory"
+        name: ${{ env.stepName }}
+        id: link-verifier-success-case
+        uses: ./link-verifier
         with:
-          python-version: '3.11.0'
-      - name: Test link verifier action
+          path: ./
+          exclude-dirs: complexity, formatting, badFiles
+          allowlist-file: link-verifier/badUrls.md
+
+      - env:
+          stepName: "Functional | Success | Link being Excluded"
+        name: ${{ env.stepName }}
+        id: link-verifier-exclude-url-success-case
         uses: ./link-verifier
         with:
           path: ./
-          exclude-dirs: complexity,formatting
-          include-file-types: .c,.html
+          exclude-dirs: complexity, formatting
+          exclude-urls: https://dummy-url.com/ota.bin, https://dummy-url.com/ota.com, https://dummy-url-three.com/ota.bin, https://github.com/FreeRTOS/A-Repo-That-Wins-You-The-Lottery
+
+      - env:
+          stepName: "Functional | Failure | Link Does Not Exist"
+        name: ${{ env.stepName }}
+        id: link-verifier-fail-case
+        continue-on-error: true
+        uses: ./link-verifier
+        with:
+          path: ./
+          exclude-dirs: complexity, formatting
+
+      - env:
+          stepName: Check Failure Test Case
+        name: ${{ env.stepName }}
+        id: check-failure-test-cases
+        shell: bash
+        run: |
+          # ${{ env.stepName }}
+          exitStatus=0
+          if [ "${{ steps.link-verifier-fail-case.outcome}}" = "failure" ]; then
+            echo -e "${{ env.bashPass }} Functional | Failure | Link Does Not Exist | Had Expected "failure" ${{ env.bashEnd }}"
+          else
+            echo -e "${{ env.bashFail }} Functional | Failure | Link Does Not Exist | Had Unexpected "success" ${{ env.bashEnd }}"
+            exitStatus=1
+          fi
+          exit $exitStatus
+
   test-manifest-verifier:
     runs-on: ubuntu-latest
     steps:

link-verifier/README.md

@@ -59,3 +59,8 @@ The `--user-agent` option allows specifying a specific user agent string to be u
 | `-k`, `--keep` | *None* | Option to keep temporary HTML files instead of deleting them. Only useful for debugging. |
 | `-v`, `--verbose` | *None* | Increase verbosity to print all files and links tested, instead of only errors. |
 | `-u`, `--user-agent`| A User-Agent string | User agent string to use for HTTP requests. |
+
+## Tests
+The various files that live inside of this directory are used in the PR checks
+for this repository. The relevant tests for this action can be found inside of
+[test.yml](../.github/workflows/test.yml)

link-verifier/action.yml

@@ -12,65 +12,112 @@ inputs:
   include-file-types:
     description: 'Comma-separated list of file type patters in repository to test. (Eg .c, .h)'
     required: false
-    default: '.c .h .dox'
+    default: '.c, .h, .dox, .md, .html'
   allowlist-file:
     description: 'Path to file containing allowlist of URLs.'
     required: false
     default: ''
+  exclude-urls:
+    description: 'Comma separated list of URLS not to check'
+    required: false
   user-agent:
     description: 'User agent string to use when making http requests.'
     required: false
     default: 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/99.0.4844.84 Safari/537.36'
 runs:
   using: "composite"
   steps:
-    - name: Install pandoc
+    - name: Setup Python for link verifier action
+      uses: actions/setup-python@v3
+
+    - env:
+        # The bash escape character is \033
+        # At time of writing, you can't add a global environment
+        # to an action file. If this gets changed please move this
+        bashPass:   \033[32;1mPASSED -
+        bashInfo:   \033[33;1mINFO -
+        bashFail:   \033[31;1mFAILED -
+        bashEnd:    \033[0m
+        stepName: Install Dependencies
+      name: ${{ env.stepName }}
+      shell: bash
       run: |
+        # ${{ env.stepName }}
+        echo -e "::group::${{ env.bashInfo }} ${{ env.stepName }} ${{ env.bashEnd }}"
+
         wget https://github.com/jgm/pandoc/releases/download/2.11/pandoc-2.11-1-amd64.deb -O pandoc.deb
         sudo dpkg -i pandoc.deb
+
         rm pandoc.deb
         sudo apt install debsums
+
         sudo debsums pandoc
-      shell: bash
-    - name: Install ghcurl
-      run: |
         sudo type -p curl >/dev/null || sudo apt install curl -y
-      shell: bash
-    - name: Install gh
-      run: |
         curl -fsSL https://cli.github.com/packages/githubcli-archive-keyring.gpg | sudo dd of=/usr/share/keyrings/githubcli-archive-keyring.gpg
         sudo chmod go+r /usr/share/keyrings/githubcli-archive-keyring.gpg
         echo "deb [arch=$(dpkg --print-architecture) signed-by=/usr/share/keyrings/githubcli-archive-keyring.gpg] https://cli.github.com/packages stable main" | sudo tee /etc/apt/sources.list.d/github-cli.list > /dev/null
+
         sudo apt update
         sudo apt install -y gh
-      shell: bash
-    - name: Install python dependencies
-      run: |
+        sudo apt install pandoc -y
         sudo apt-get install -y python3-setuptools python3-pip
+
         python3 -m pip install -r  $GITHUB_ACTION_PATH/requirements.txt
+
+        echo -e "::endgroup::"
+        echo -e "${{ env.bashPass }} ${{ env.stepName }} ${{ env.bashEnd}}"
+
+    - env:
+        bashPass: \033[32;1mPASSED -
+        bashInfo: \033[33;1mINFO -
+        bashFail: \033[31;1mFAILED -
+        bashEnd:  \033[0m
+        stepName: Check Links in Files
+      name: ${{ env.stepName }}
+      working-directory: ${{ inputs.path }}
       shell: bash
-    - name: Run link verifier script
       run: |
-        args="--verbose --test-markdown"
+        # ${{ env.stepName }}
+        echo -e "::group::${{ env.bashInfo }} ${{ env.stepName }} ${{ env.bashEnd }}"
+        args="--test-markdown"
         if [ -n "${{ inputs.exclude-dirs }}" ]; then
           dirs="${{ inputs.exclude-dirs }}"
           dirs="${dirs//,/ }"
           args+=" --exclude-dirs ${dirs}"
         fi
+
         if [ -n "${{ inputs.include-file-types }}" ]; then
           file_types="${{ inputs.include-file-types }}"
           file_types="${file_types//,/ }"
           args+=" --include-file-types ${file_types}"
         fi
+
         if [ -n "${{ inputs.allowlist-file }}" ]; then
-          allowlist_file="${{ inputs.allowlist-file }}"
-          allowlist_file="${allowlist_file//,/ }"
-          args+=" --allowlist-file ${allowlist_file}"
+          touch allowList.txt
+          cat ${{ inputs.allowlist-file }} >> allowList.txt
+        fi
+
+        if [[ "${{ inputs.exclude-urls }}" != "" ]]; then
+          touch allowList.txt
+          exclude_urls="${{ inputs.exclude-urls }}"
+          exclude_urls="${exclude_urls//,/ }"
+          for url in ${exclude_urls[@]}; do echo -e "$url" >> allowList.txt; done
         fi
-        echo "Running verify-links.py ${args} --user-agent \"${{ inputs.user-agent }}\""
-        if python3 ${GITHUB_ACTION_PATH}/verify-links.py ${args} --user-agent "${{ inputs.user-agent }}"; then
-          exit 0
+
+        if [ -n "${{ inputs.allowlist-file }}" ] || [ -n "${{ inputs.exclude-urls }}" ]; then
+          args+=" --allowlist-file allowList.txt"
+        fi
+
+        echo -e "${{ env.bashInfo }} Running: verify-links.py ${args} --user-agent \"${{ inputs.user-agent }}\" ${{ env.bashEnd }}"
+        set +e
+        python3 ${GITHUB_ACTION_PATH}/verify-links.py ${args} --user-agent "${{ inputs.user-agent }}";
+        exitStatus=$?
+        set -e
+
+        echo -e "::endgroup::"
+        if [ $exitStatus -eq 1 ]; then
+          echo -e "${{ env.bashFail }} ${{ env.stepName }} ${{ env.bashEnd }}"
         else
-          exit 1
+          echo -e "${{ env.bashPass }} ${{ env.stepName }} ${{ env.bashEnd }}"
         fi
-      shell: bash
+        exit $exitStatus

link-verifier/badUrls.md

@@ -0,0 +1,3 @@
+https://dummy-url.com/ota.bin
+https://dummy-url.com/ota.com
+https://dummy-url-three.com/ota.bin

link-verifier/fileTests/badFiles/fileWithLowercasemdIntheNameAndABrokenLink.md

@@ -0,0 +1,11 @@
+# I am a test file to make sure the regex name replacement only replaces files
+# that end in .md, not all files with md in their name.
+# I also exit to make sure we skip files that don't end in .[c, h, dox, md, html]
+# When using the default input
+# Here are two links that exist
+[verify-links.py](../../verify-links.py)
+[CI-CD-Github-Actions](https://github.com/FreeRTOS/CI-CD-Github-Actions]
+
+# Here's a link that doesn't exist that should cause us to fail if searched
+# But since we aren't searched by default this broken link should not cause an error
+[Incredible link](https://github.com/FreeRTOS/A-Repo-That-Wins-You-The-Lottery)

link-verifier/fileTests/badFiles/fileWithMDIntheNameAndABrokenLink.md

@@ -0,0 +1,11 @@
+# I am a test file to make sure the regex name replacement only replaces files
+# that end in .md, not all files with md in their name.
+# I also exit to make sure we skip files that don't end in .[c, h, dox, md, html]
+# When using the default input
+# Here are two links that exist
+[verify-links.py](../../verify-links.py)
+[CI-CD-Github-Actions](https://github.com/FreeRTOS/CI-CD-Github-Actions]
+
+# Here's a link that doesn't exist that should cause us to fail if searched
+# But since we aren't searched by default this broken link should not cause an error
+[Incredible link](https://github.com/FreeRTOS/A-Repo-That-Wins-You-The-Lottery)

link-verifier/fileTests/badFiles/fileWithMDIntheNameAndABrokenLink.txt

@@ -0,0 +1,11 @@
+# I am a test file to make sure the regex name replacement only replaces files
+# that end in .md, not all files with md in their name.
+# I also exit to make sure we skip files that don't end in .[c, h, dox, md, html]
+# When using the default input
+# Here are two links that exist
+[verify-links.py](../../verify-links.py)
+[CI-CD-Github-Actions](https://github.com/FreeRTOS/CI-CD-Github-Actions]
+
+# Here's a link that doesn't exist that should cause us to fail if searched
+# But since we aren't searched by default this broken link should not cause an error
+[Incredible link](https://github.com/FreeRTOS/A-Repo-That-Wins-You-The-Lottery)

link-verifier/fileTests/goodFiles/fileWithLowercasemdIntheName.md

@@ -0,0 +1,5 @@
+# I am a test file to make sure the regex name replacement only 
+# replaces files that end in .md, not all files with md in their name.
+# Here's a random link for it to test as well
+[verify-links.py](../../verify-links.py)
+[CI-CD-Github-Actions](https://github.com/FreeRTOS/CI-CD-Github-Actions)

link-verifier/fileTests/goodFiles/fileWithMDIntheName.md

@@ -0,0 +1,5 @@
+# I am a test file to make sure the regex name replacement only 
+# replaces files that end in .md, not all files with md in their name.
+# Here's a random link for it to test as well
+[verify-links.py](../../verify-links.py)
+[CI-CD-Github-Actions](https://github.com/FreeRTOS/CI-CD-Github-Actions)

link-verifier/fileTests/goodFiles/fileWithMDIntheName.txt

@@ -0,0 +1,8 @@
+# I am a test file to make sure the regex name replacement only replaces files
+# that end in .md, not all files with md in their name.
+# As this would cause the python script to fail
+
+# When using the default input
+# Here are two links that exist
+[verify-links.py](../../verify-links.py)
+[CI-CD-Github-Actions](https://github.com/FreeRTOS/CI-CD-Github-Actions]

link-verifier/verify-links.py

@@ -166,7 +166,7 @@ def parse_file(html_file):
     return HtmlFile(html_file)
 
 def html_name_from_markdown(filename):
-    md_pattern = re.compile('.md', re.IGNORECASE)
+    md_pattern = re.compile('\.md', re.IGNORECASE)
     return md_pattern.sub('.html', filename)
 
 def create_html(markdown_file):