Added ability to exclude urls. Closes #12

bin/hyperlink

@@ -29,6 +29,11 @@ var optimist = require('optimist'),
             describe: 'Crawl all HTML-pages linked with relative and root relative links. This stays inside your domain.',
             type: 'boolean'
         })
+        .options('exclude', {
+            describe: 'Url pattern to exclude from the build. Supports * wildcards. You can create multiple of these: --exclude *.php --exclude http://example.com/*.gif',
+            type: 'string',
+            demand: false
+        })
         .options('concurrency', {
             alias: 'c',
             describe: 'The maximum number of assets that can be loading at once (defaults to 100)',
@@ -46,6 +51,7 @@ if (commandLineOptions.h) {
 var urlTools = require('urltools'),
     // canonicalUrl = commandLineOptions.canonicalurl && urlTools.ensureTrailingSlash(commandLineOptions.canonicalurl),
     rootUrl = commandLineOptions.root && urlTools.urlOrFsPathToUrl(commandLineOptions.root, true),
+    excludePatterns = commandLineOptions.exclude && [].concat(commandLineOptions.exclude),
     inputUrls;
 
 var hyperlink = require('../lib/index');
@@ -72,6 +78,7 @@ if (commandLineOptions._.length > 0) {
 hyperlink({
     root: rootUrl,
     inputUrls: inputUrls,
+    excludePatterns: excludePatterns,
     recursive: commandLineOptions.recursive,
     verbose: commandLineOptions.verbose,
     concurrency: commandLineOptions.concurrency

lib/index.js

@@ -12,6 +12,10 @@ var TapRender = require('tap-render');
 
 var t = new TapRender();
 
+function regexEscape(pattern) {
+    return pattern.replace(/[\.\+\{\}\[\]\(\)\?\^\$]/g, '\\$&').replace(/\*/g, '.*?');
+}
+
 module.exports = function (options) {
     options = options || {};
 
@@ -20,6 +24,20 @@ module.exports = function (options) {
         root: options.root
     });
 
+    var excludePattern;
+
+    if (Array.isArray(options.excludePatterns)) {
+        excludePattern = new RegExp('^(:?' + options.excludePatterns.map(regexEscape).join('|') + ')');
+    }
+
+    function shouldSkip(url) {
+        if (excludePattern) {
+            return excludePattern.test(url);
+        }
+
+        return false;
+    }
+
     var relationsQuery = {
         type: query.not('HtmlAnchor'),
         crossorigin: false
@@ -33,9 +51,12 @@ module.exports = function (options) {
         redirects = redirects || [];
         relations = relations || [];
 
+        var skip = shouldSkip(url);
+
         if (status !== 200) {
             var invalidStatusReport = {
                 ok: false,
+                skip: skip,
                 name: 'should respond with HTTP status 200',
                 operator: 'error',
                 expected: [200, url].join(' '),
@@ -51,6 +72,7 @@ module.exports = function (options) {
 
         var report = {
             ok: true,
+            skip: skip,
             name: 'URI should have no redirects - ' + url,
             operator: 'noRedirects',
             expected: [200, url].join(' '),
@@ -116,6 +138,7 @@ module.exports = function (options) {
 
                 var insecureReport = {
                     ok: false,
+                    skip: skip,
                     name: 'URI should be secure - ' + url,
                     operator: 'mixed-content',
                     expected: insecureLogLine.replace(/\bhttps?:/g, 'https:'),
@@ -275,7 +298,7 @@ module.exports = function (options) {
         })
         // .writeStatsToStderr()
         .run(function () {
-            t.close();
-            process.exit(errorCount ? 1 : 0);
+            var results = t.close();
+            process.exit(results.fail ? 1 : 0);
         });
 };

package.json

@@ -44,7 +44,7 @@
     "lodash": "^4.16.4",
     "optimist": "^0.6.1",
     "request": "^2.75.0",
-    "tap-render": "^0.1.7",
+    "tap-render": "Munter/tap-render#0.1.7-patch1",
     "urltools": "^0.3.0"
   },
   "devDependencies": {