New --ignore-config flag to specify a rules file

#11

README.md

@@ -49,15 +49,16 @@ You can also install **w3c-html-validator** globally (`--global`) and then run i
 
 ### 3. CLI flags
 Command-line flags:
-| Flag         | Description                                                     | Value      |
-| ------------ | --------------------------------------------------------------- | ---------- |
-| `--continue` | Report messages but do not throw an error if validation failed. | N/A        |
-| `--delay`    | Debounce pause in milliseconds between each file validation.    | **number** |
-| `--exclude`  | Comma separated list of strings to match in paths to skip.      | **string** |
-| `--ignore`   | Skip messages containing a string or matching a RegEx.          | **string** |
-| `--note`     | Place to add a comment only for humans.                         | **string** |
-| `--quiet`    | Suppress messages for successful validations.                   | N/A        |
-| `--trim`     | Truncate validation messages to not exceed a maximum length.    | **number** |
+| Flag              | Description                                                     | Value      |
+| ----------------- | --------------------------------------------------------------- | ---------- |
+| `--continue`      | Report messages but do not throw an error if validation failed. | N/A        |
+| `--delay`         | Debounce pause in milliseconds between each file validation.    | **number** |
+| `--exclude`       | Comma separated list of strings to match in paths to skip.      | **string** |
+| `--ignore`        | Skip messages containing a string or matching a regex.          | **string** |
+| `--ignore-config` | File listing regex patterns of messages to skip.                | **string** |
+| `--note`          | Place to add a comment only for humans.                         | **string** |
+| `--quiet`         | Suppress messages for successful validations.                   | N/A        |
+| `--trim`          | Truncate validation messages to not exceed a maximum length.    | **number** |
 
 ### 4. Example CLI usage
 Examples:
@@ -71,10 +72,13 @@ Examples:
    Allow the ugly slashes of self-closing tags despite XHTML being a hideous scourge on the web.
 
    - `html-validator docs '--ignore=/^Duplicate ID/'`<br>
-   Use a RegEx (regular expression) to skip all validation messages that start with "Duplicate ID".
+   Use a regex (regular expression) to skip all validation messages that start with "Duplicate ID".
 
    - `html-validator docs '--ignore=/^Duplicate ID|^Section lacks|^Element .blockquote. not allowed/'`<br>
-   Use a RegEx with "or" operators (`|`) to skip multiple validation messages.
+   Use a regex with "or" operators (`|`) to skip multiple validation messages.
+
+   - `html-validator docs --ignore-config=config/regex-patterns.txt`<br>
+   Similar to the pervious command but regex patterns are stored in a configuration file (see the _Ignore Configuration File_ section below).
 
    - `html-validator --quiet`<br>
    Suppress "pass" status messages.
@@ -85,6 +89,20 @@ Examples:
    - `html-validator docs --trim=30 --continue`<br>
    Truncate validation messages to 30 characters and do not abort CI if validation fails.
 
+### 5. Ignore Configuration File
+The optional `--ignore-config=FILENAME` flag specifies a configuration file with one regex pattern per line.&nbsp;
+Empty lines and lines starting with a double slash (`//`) are considered comments.
+Validation messages matching any of the regex patterns will be skipped.
+
+Example configuration file with 3 regex patterns:
+```config
+// Ignore Config for w3c-html-validator
+
+^Duplicate ID
+^Element .blockquote. not allowed
+^Element .style. not allowed
+```
+
 ## D) Application Code and Testing Frameworks
 In addition to the CLI interface, the **w3c-html-validator** package can also be imported and called directly in ESM and TypeScript projects.
 

bin/cli.js

@@ -26,12 +26,13 @@ import { w3cHtmlValidator } from '../dist/w3c-html-validator.js';
 import fs from 'fs';
 
 // Parameters and flags
-const validFlags = ['continue', 'delay', 'exclude', 'ignore', 'note', 'quiet', 'trim'];
-const cli =        cliArgvUtil.parse(validFlags);
-const files =      cli.params;
-const ignore =     cli.flagMap.ignore ?? null;
-const delay =      Number(cli.flagMap.delay) || 500;  //default half second debounce pause
-const trim =       Number(cli.flagMap.trim) || null;
+const validFlags = ['continue', 'delay', 'exclude', 'ignore', 'ignore-config', 'note', 'quiet', 'trim'];
+const cli =          cliArgvUtil.parse(validFlags);
+const files =        cli.params;
+const ignore =       cli.flagMap.ignore ?? null;
+const ignoreConfig = cli.flagMap.ignoreConfig ?? null;
+const delay =        Number(cli.flagMap.delay) || 500;  //default half second debounce pause
+const trim =         Number(cli.flagMap.trim) || null;
 
 // Validator
 const keep =         (filename) => !filename.includes('node_modules/');
@@ -55,9 +56,19 @@ const reporterOptions = {
    quiet:          cli.flagOn.quiet,
    maxMessageLen:  trim,
    };
-const isRegex =      /^\/.*\/$/;  //starts and ends with a slash indicating it's a regex
-const skip =         isRegex.test(ignore) ? new RegExp(ignore.slice(1, -1)) : ignore;
+const getIgnoreMessages = () => {
+   const toLines =    (text) => text.replace(/\r/g, '').split('\n').map(line => line.trim());
+   const readLines =  (file) => toLines(fs.readFileSync(file).toString());
+   const notComment = (line) => line.length > 1 && !line.startsWith('//');
+   const lines =      ignoreConfig ? readLines(ignoreConfig).filter(notComment) : [];
+   const regexes =    lines.map(line => new RegExp(line));
+   const isRegex =    /^\/.*\/$/;  //starts and ends with a slash indicating it's a regex
+   if (isRegex.test(ignore))
+      regexes.push(new RegExp(ignore.slice(1, -1)));
+   return regexes.length ? regexes : ignore;
+   };
 const handleReport = (report) => w3cHtmlValidator.reporter(report, reporterOptions);
-const options =      (filename) => ({ filename: filename, ignoreMessages: skip });
+const options =      (filename) => ({ filename: filename, ignoreMessages: getIgnoreMessages() });
 const getReport =    (filename) => w3cHtmlValidator.validate(options(filename)).then(handleReport);
-filenames.forEach((filename, i) => globalThis.setTimeout(() => getReport(filename), i * delay));
+const processFile =  (filename, i) => globalThis.setTimeout(() => getReport(filename), i * delay);
+filenames.forEach(processFile);

spec/ignore-config.txt

@@ -0,0 +1,5 @@
+// Ignore Config for w3c-html-validator
+
+^Duplicate ID
+^Element .blockquote. not allowed
+^Element .style. not allowed

spec/mocha.spec.js

@@ -412,4 +412,10 @@ describe('Executing the CLI', () => {
       assertDeepStrictEqual(actual, expected);
       });
 
+   it('skips validation message matching --ignore and --ignore-config regex patterns', () => {
+      const actual =   run('html-validator spec/html "--ignore=/^Section lacks heading/" --ignore-config=spec/ignore-config.txt');
+      const expected = null;
+      assertDeepStrictEqual(actual, expected);
+      });
+
    });

w3c-html-validator.ts

@@ -8,12 +8,12 @@ import request from 'superagent';
 
 // Type Declarations
 export type ValidatorSettings = {
-   html:           string,              //example: '<!doctype html><html><head><title>Home</title></html>''
-   filename:       string,              //example: 'docs/index.html'
-   website:        string,              //example: 'https://pretty-print-json.js.org/'
-   checkUrl:       string,              //default: 'https://validator.w3.org/nu/'
-   ignoreLevel:    'info' | 'warning',  //skip unwanted messages ('warning' also skips 'info')
-   ignoreMessages: string | RegExp,     //matcher to skip unwanted messages
+   html:           string,                      //example: '<!doctype html><html><head><title>Home</title></html>''
+   filename:       string,                      //example: 'docs/index.html'
+   website:        string,                      //example: 'https://pretty-print-json.js.org/'
+   checkUrl:       string,                      //default: 'https://validator.w3.org/nu/'
+   ignoreLevel:    'info' | 'warning',          //skip unwanted messages ('warning' also skips 'info')
+   ignoreMessages: string | RegExp | RegExp[],  //matcher to skip unwanted messages
    output:         ValidatorResultsOutput,
    };
 export type ValidatorResultsMessage = {
@@ -74,8 +74,8 @@ const w3cHtmlValidator = {
          throw Error('[w3c-html-validator] Invalid ignoreLevel option: ' + settings.ignoreLevel);
       if (settings.output !== 'json' && settings.output !== 'html')
          throw Error('[w3c-html-validator] Option "output" must be "json" or "html".');
-      const mode = settings.html ? 'html' : settings.filename ? 'filename' : 'website';
-      const readFile = (filename: string) => fs.readFileSync(filename, 'utf-8').replace(/\r/g, '');
+      const mode =      settings.html ? 'html' : settings.filename ? 'filename' : 'website';
+      const readFile =  (filename: string) => fs.readFileSync(filename, 'utf-8').replace(/\r/g, '');
       const inputHtml = settings.html ?? (settings.filename ? readFile(settings.filename) : null);
       const makePostRequest = () => request.post(settings.checkUrl)
          .set('Content-Type', 'text/html; encoding=utf-8')
@@ -85,25 +85,26 @@ const w3cHtmlValidator = {
       const w3cRequest = inputHtml ? makePostRequest() : makeGetRequest();
       w3cRequest.set('User-Agent', 'W3C HTML Validator ~ github.com/center-key/w3c-html-validator');
       w3cRequest.query({ out: settings.output });
-      const json = settings.output === 'json';
+      const json =    settings.output === 'json';
       const success = '<p class="success">';
       const titleLookup = {
          html:     'HTML String (characters: ' + inputHtml?.length + ')',
          filename: settings.filename,
          website:  settings.website,
          };
+      const noRegex = !settings.ignoreMessages || typeof settings.ignoreMessages === 'string';
+      const regexes = noRegex ? [] : <RegExp[]>[settings.ignoreMessages].flat();
       const filterMessages = (response: request.Response): request.Response => {
          const aboveInfo = (subType: ValidatorResultsMessageSubType): boolean =>
             settings.ignoreLevel === 'info' && !!subType;
          const aboveIgnoreLevel = (message: ValidatorResultsMessage): boolean =>
             !settings.ignoreLevel || message.type !== 'info' || aboveInfo(message.subType);
          const skipSubstr = (title: string) =>
             typeof settings.ignoreMessages === 'string' && title.includes(settings.ignoreMessages);
-         const skipRegEx = (title: string) =>
-            settings.ignoreMessages?.constructor.name === 'RegExp' &&
-            (<RegExp>settings.ignoreMessages).test(title);
+         const skipRegex = (title: string) =>
+            regexes.some(regex => regex.test(title));
          const isImportant = (message: ValidatorResultsMessage): boolean =>
-            aboveIgnoreLevel(message) && !skipSubstr(message.message) && !skipRegEx(message.message);
+            aboveIgnoreLevel(message) && !skipSubstr(message.message) && !skipRegex(message.message);
          if (json)
             response.body.messages = response.body.messages?.filter(isImportant) ?? [];
          return response;