Get exclude hostnames as user input

Handle an edge case scenario in subprocess errors
Update README.md

README.md

@@ -43,13 +43,17 @@ jobs:
 
 Use the options below to configure debug and fail state when broken links are found in the repository/wiki pages.
 
-| option  | requirement | description                                                                                                                                              |
-|---------|-------------|----------------------------------------------------------------------------------------------------------------------------------------------------------|
-| `owner` | optional    | Owner/Organization of the repository - Defaults to current owner/org name                                                                                |
-| `repo`  | optional    | Name of the repository - Defaults to current repository's name                                                                                           |
-| `fail`  | optional    | If `true` (Default) the build is failed if broken links are found.<br/>If `false` the build completes successfully and warnings are provided in the logs |
-| `debug` | optional    | If `true` (Default is `false`) debug level logging is enabled                                                                                            |
+| option             | requirement | description                                                                                                                                              |
+|--------------------|-------------|----------------------------------------------------------------------------------------------------------------------------------------------------------|
+| `owner`            | optional    | Owner/Organization of the repository - Defaults to current owner/org name                                                                                |
+| `repo`             | optional    | Name of the repository - Defaults to current repository's name                                                                                           |
+| `fail`             | optional    | If `true` (Default) the build is failed if broken links are found.<br/>If `false` the build completes successfully and warnings are provided in the logs |
+| `debug`            | optional    | If `true` (Default is `false`) debug level logging is enabled                                                                                            |
+| `excludeHostnames` | optional    | Comma separated list of hostnames to ignore when failed                                                                                                  |
 
+> `excludeHostnames` will perform a regex like lookup, so wildcards (*) are not required and spaces don't matter<br>
+
+> To exclude any URL with `amazon`/`amzn` in it simply specify,<br>`excludeHostnames: "amazon,amzn"`
 
 ## [Release Notes][release-notes]
 **Requirement**

action.yml

@@ -1,31 +1,36 @@
-name: 'none-shall-pass'
-description: 'Run validations against hyperlinks in all markdown files (including Wiki pages)'
+name: none-shall-pass
+description: Run validations against hyperlinks in all markdown files (including Wiki pages)
 inputs:
   owner:
-    description: "Owner/Organization of the repository"
+    description: Owner/Organization of the repository
     required: false
     default: ${{ github.repository_owner }}
   repo:
-    description: "Name of the repository"
+    description: Name of the repository
     required: false
     default: ${{ github.event.repository.name }}
   fail:
-    description: "Flag to fail"
+    description: Flag to fail
     required: false
     default: "true"
   debug:
-    description: "Debug flag"
+    description: Debug flag
     required: false
     default: "false"
+  excludeHostnames:
+    description: Hostnames for which the failure should be ignored
+    required: false
+    default: ""
 outputs: {}
 runs:
-  using: 'docker'
-  image: 'Dockerfile'
+  using: docker
+  image: Dockerfile
   args:
     - ${{ inputs.owner }}
     - ${{ inputs.repo }}
     - ${{ inputs.fail }}
     - ${{ inputs.debug }}
+    - ${{ inputs.excludeHostnames }}
 branding:
   icon: check-square
   color: gray-dark

src/__init__.py

@@ -1,2 +1,2 @@
 """Placeholder for the version number of the package."""
-version = "1.0.8"
+version = "2.0a"

src/connection.py

@@ -13,6 +13,8 @@
 
 from logger import LOGGER
 
+exclusions = ["localhost", "127.0.0.1", socket.gethostbyname('localhost')]
+
 
 def verify_url(hyperlink: Tuple[str, str], timeout: Tuple[int, int] = (3, 3)) -> None:
     """Make a GET request to the hyperlink for validation using requests module.
@@ -22,7 +24,7 @@ def verify_url(hyperlink: Tuple[str, str], timeout: Tuple[int, int] = (3, 3)) ->
         timeout: A tuple of connect timeout and read timeout.
 
     See Also:
-        - Ignores amazon and localhost URLs.
+        - Ignores localhost URLs.
         - Ignores URLs that return 429 status code.
         - Retries once with a longer connect and read timeout in case of a timeout error.
 
@@ -43,7 +45,7 @@ def verify_url(hyperlink: Tuple[str, str], timeout: Tuple[int, int] = (3, 3)) ->
         return verify_url(hyperlink, (10, 10))
     except requests.RequestException as error:
         LOGGER.debug(error)
-    if any(map(lambda keyword: keyword in url, ("amazon", "localhost", socket.gethostbyname('localhost')))):
-        LOGGER.warning("[%s] - '%s' - '%s' is broken", current_process().name, text, url)
+    if any(map(lambda keyword: keyword in url, exclusions)):
+        LOGGER.warning("[%s] - '%s' - '%s' is broken but excluded", current_process().name, text, url)
     else:
         raise ValueError(f"[{current_process().name}] - {text!r} - {url!r} is broken")

src/git.py

@@ -24,7 +24,7 @@ def run_command(cmd: str) -> bool:
         subprocess.run(cmd, stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL, shell=True, check=True)
         return True
     except (subprocess.CalledProcessError, subprocess.SubprocessError, Exception) as error:
-        if isinstance(error, subprocess.CalledProcessError):
+        if hasattr(error, 'output') and error.output:
             result = error.output.decode(encoding='UTF-8').strip()
             LOGGER.warning("[%d]: %s", error.returncode, result or error.__str__())
         else:

src/validator.py

@@ -14,11 +14,17 @@
 REPO = sys.argv[2]
 FAIL = sys.argv[3]
 DEBUG = sys.argv[4]
+EXCLUDE_HOSTNAMES = sys.argv[5]
+
+connection.exclusions.extend(list(filter(None, map(str.strip, EXCLUDE_HOSTNAMES.split(',')))))
+
 if DEBUG == "true":
     LOGGER.setLevel(level=logging.DEBUG)
 else:
     LOGGER.setLevel(level=logging.INFO)
 
+LOGGER.debug(connection.exclusions)
+
 
 def verify_hyperlinks_in_md(filename: str) -> None:
     """Get all hyperlinks in a markdown file and validate them in a pool of threads.