url-parse - npm Package Compare versions

Comparing version 0.2.3 to 1.0.0

fuzzy.js

 'use strict';
 
 var URL = require('./')
-  , url = new URL();
+  , url = new URL('');
 
@@ -6,0 +6,0 @@ /**

index.js

@@ -5,8 +5,27 @@ 'use strict';
   , lolcation = require('./lolcation')
-  , qs = require('querystringify');
+  , qs = require('querystringify')
+  , relativere = /^\/(?!\/)/;
 
-var keys = ',,protocol,username,password,host,hostname,port,pathname,query,hash'.split(',')
-  , inherit = { protocol: 1, host: 1, hostname: 1 }
-  , relativere = /^\/(?!\/)/
-  , parts = keys.length;
+/**
+ * These are the parse instructions for the URL parsers, it informs the parser
+ * about:
+ *
+ * 0. The char it Needs to parse, if it's a string it should be done using
+ *    indexOf, RegExp using exec and NaN means set as current value.
+ * 1. The property we should set when parsing this value.
+ * 2. Indication if it's backwards or forward parsing, when set as number it's
+ *    the value of extra chars that should be split off.
+ * 3. Inherit from location if non existing in the parser.
+ * 4. `toLowerCase` the resulting value.
+ */
+var instructions = [
+  ['#', 'hash'],                        // Extract from the back.
+  ['?', 'query'],                       // Extract from the back.
+  ['//', 'protocol', 2, 1, 1],          // Extract from the front.
+  ['/', 'pathname'],                    // Extract from the back.
+  ['@', 'auth', 1],                     // Extract from the front.
+  [NaN, 'host', undefined, 1, 1],       // Set left over value.
+  [/\:(\d+)$/, 'port'],                 // RegExp the back.
+  [NaN, 'hostname', undefined, 1, 1]    // Set left over.
+];
 
@@ -29,42 +48,7 @@ /**
 
-  //
-  // Story time children:
-  //
-  // FireFox 34 has some problems with their Regular Expression engine and
-  // executing a RegExp can cause a `too much recursion` error. We initially fixed
-  // this by moving the Regular Expression in the URL constructor so it's created
-  // every single time. This fixed it for some URL's but the more complex the
-  // URL's get the easier it is to trigger. Complexer URL like:
-  //
-  //   https://www.mozilla.org/en-US/firefox/34.0/whatsnew/?oldversion=33.1
-  //
-  // Still triggered the recursion error. After talking with Chrome and FireFox
-  // engineers it seemed to be caused by:
-  //
-  //   https://code.google.com/p/v8/issues/detail?id=430
-  //
-  // As FireFox started using Chrome's RegExp engine. After testing various of
-  // workarounds I finally stumbled upon this gem, use new RegExp as it sometimes
-  // behaves different then a RegExp literal. The biggest problem with this
-  // FireFox problem is that it's super hard to reproduce as our "normal" test
-  // suite doesn't catch it. The only way to reproduce it was run the parser in
-  // jsperf.com (uses the benchmark module from npm) and supply it the URL
-  // mentioned above as URL to parse.
-  //
-  // Steps for compiling the new RegExp:
-  //
-  // 1. Take the regular RegExp as seen below.
-  // 2. Escape the RegExp using XRegExp.escape from http://xregexp.com/tests/
-  // 3. ??
-  // 4. Profit.
-  //
-  // RegExp source: /^(?:(?:(([^:\/#\?]+:)?(?:(?:\/\/)(?:(?:(?:([^:@\/#\?]+)(?:\:([^:@\/#\?]*))?)@)?(([^:\/#\?\]\[]+|\[[^\/\]@#?]+\])(?:\:([0-9]+))?))?)?)?((?:\/?(?:[^\/\?#]+\/+)*)(?:[^\?#]*)))?(\?[^#]+)?)(#.*)?/
-  //
-  var regexp = new RegExp('\^\(\?:\(\?:\(\(\[\^:\\/\#\\\?\]\+:\)\?\(\?:\(\?:\\/\\/\)\(\?:\(\?:\(\?:\(\[\^:@\\/\#\\\?\]\+\)\(\?:\\:\(\[\^:@\\/\#\\\?\]\*\)\)\?\)@\)\?\(\(\[\^:\\/\#\\\?\\\]\\\[\]\+\|\\\[\[\^\\/\\\]@\#\?\]\+\\\]\)\(\?:\\:\(\[0\-9\]\+\)\)\?\)\)\?\)\?\)\?\(\(\?:\\/\?\(\?:\[\^\\/\\\?\#\]\+\\/\+\)\*\)\(\?:\[\^\\\?\#\]\*\)\)\)\?\(\\\?\[\^\#\]\+\)\?\)\(\#\.\*\)\?')
-    , relative = relativere.test(address)
-    , bits = regexp.exec(address)
+  var relative = relativere.test(address)
+    , parse, instruction, index, key
     , type = typeof location
     , url = this
-    , i = 0
-    , key;
+    , i = 0;
 
@@ -93,13 +77,33 @@ //
 
-  for (; i < parts; key = keys[++i]) {
-    if (!key) continue;
+  for (; i < instructions.length; i++) {
+    instruction = instructions[i];
+    parse = instruction[0];
+    key = instruction[1];
 
-    url[key] = bits[i] || (key in inherit || ('port' === key && relative) ? location[key] || '' : '');
+    if (parse !== parse) {
+      url[key] = address;
+    } else if ('string' === typeof parse) {
+      if (~(index = address.indexOf(parse))) {
+        if ('number' === typeof instruction[2]) {
+          url[key] = address.slice(0, index);
+          address = address.slice(index + instruction[2]);
+        } else {
+          url[key] = address.slice(index);
+          address = address.slice(0, index);
+        }
+      }
+    } else if (index = parse.exec(address)) {
+      url[key] = index[1];
+      address = address.slice(0, address.length - index[0].length);
+    }
 
+    url[key] = url[key] || (instruction[3] || ('port' === key && relative) ? location[key] || '' : '');
+
     //
-    // The protocol, host, host name should always be lower cased even if they
-    // are supplied in uppercase. This way, when people generate an `origin`
-    // it be correct.
+    // Hostname, host and protocol should be lowercased so they can be used to
+    // create a proper `origin`.
     //
-    if (i === 2 || i === 5 || i === 6) url[key] = url[key].toLowerCase();
+    if (instruction[4]) {
+      url[key] = url[key].toLowerCase();
+    }
   }
@@ -125,2 +129,12 @@
   //
+  // Parse down the `auth` for the username and password.
+  //
+  url.username = url.password = '';
+  if (url.auth) {
+    instruction = url.auth.split(':');
+    url.username = instruction[0] || '';
+    url.password = instruction[1] || '';
+  }
+
+  //
   // The href is just the compiled result.
@@ -127,0 +141,0 @@ //

package.json

 {
   "name": "url-parse",
-  "version": "0.2.3",
+  "version": "1.0.0",
   "description": "Parse URL in node using the URL module and in the browser using the DOM",
@@ -10,3 +10,3 @@ "main": "index.js",
     "test-travis": "istanbul cover node_modules/.bin/_mocha --report lcovonly -- --reporter spec --ui bdd ./test.js",
-    "browserify": "mkdir dist && browserify index.js -o dist/url-parse.js --standalone URLParse",
+    "browserify": "mkdir -p dist && browserify index.js -o dist/url-parse.js --standalone URLParse",
     "phantomjs": "mochify --reporter spec --ui bdd ./test.js",
@@ -29,6 +29,6 @@ "testling": "testling -u"
   "devDependencies": {
-    "assume": "0.0.x",
-    "browserify": "7.0.x",
+    "assume": "1.0.x",
+    "browserify": "8.1.x",
     "istanbul": "0.3.x",
-    "mocha": "2.0.x",
+    "mocha": "2.1.x",
     "mochify": "2.1.x",
@@ -35,0 +35,0 @@ "pre-commit": "0.0.x",

README.md

@@ -13,3 +13,6 @@ # url-parse
 module still have a really small foot print as this module's main intention is
-to be bundled with client-side code.
+to be bundled with client-side code. The only problem however with a RegExp
+based solution is that it required a lot of lookups causing major problems in
+FireFox. So the last and the current solution was a pure string parsing
+solution which chops up the URL in smaller pieces.
 
@@ -16,0 +19,0 @@ In addition to URL parsing we also expose the bundled `querystringify` module.

test.js

@@ -78,9 +78,2 @@ describe('url-parse', function () {
 
-  it('does not lowercase the USER:PASS', function () {
-    var url = 'HTTP://USER:[e-mail]';
-
-    assume(parse(url).username).equals('USER');
-    assume(parse(url).password).equals('PASS');
-  });
-
   it('does not lowercase the path', function () {
@@ -102,8 +95,111 @@ var url = 'HTTP://X.COM/Y/Z';
 
-  it('accepts @ in pathnames', function () {
-    var url = 'http://mt0.google.com/vt/lyrs=m@114&hl=en&src=api&x=2&y=2&z=3&s=';
+  it('understands an / as pathname', function () {
+    var url = 'http://example.com:80/'
+      , parsed = parse(url);
 
-    assume(parse(url).pathname).equals('/vt/lyrs=m@114&hl=en&src=api&x=2&y=2&z=3&s=');
+    assume(parsed.port).equals('');
+    assume(parsed.username).equals('');
+    assume(parsed.password).equals('');
+    assume(parsed.pathname).equals('/');
+    assume(parsed.host).equals('example.com');
+    assume(parsed.hostname).equals('example.com');
+    assume(parsed.href).equals('http://example.com/');
   });
 
+  it('does not care about spaces', function () {
+    var url = 'http://x.com/path?that\'s#all, folks'
+      , parsed = parse(url);
+
+    assume(parsed.port).equals('');
+    assume(parsed.username).equals('');
+    assume(parsed.password).equals('');
+    assume(parsed.pathname).equals('/path');
+    assume(parsed.hash).equal('#all, folks');
+    assume(parsed.query).equal('?that\'s');
+    assume(parsed.host).equals('x.com');
+    assume(parsed.hostname).equals('x.com');
+  });
+
+  it('accepts + in the url', function () {
+    var url = 'http://x.y.com+a/b/c'
+      , parsed = parse(url);
+
+    assume(parsed.protocol).equals('http:');
+    assume(parsed.host).equals('x.y.com+a');
+    assume(parsed.hostname).equals('x.y.com+a');
+    assume(parsed.pathname).equals('/b/c');
+  });
+
+  describe('ip', function () {
+    // coap://
+    //
+    it('parses ipv6', function () {
+      var url = 'http://[1080:0:0:0:8:800:200C:417A]:61616/foo/bar?q=z'
+        , parsed = parse(url);
+
+      assume(parsed.port).equals('61616');
+      assume(parsed.query).equals('?q=z');
+      assume(parsed.protocol).equals('http:');
+      assume(parsed.hostname).equals('[1080:0:0:0:8:800:200c:417a]');
+      assume(parsed.pathname).equals('/foo/bar');
+      assume(parsed.href).equals('http://[1080:0:0:0:8:800:200c:417a]:61616/foo/bar?q=z');
+    });
+
+    it('parses ipv6 with auth', function () {
+      var url = 'http://user:password@[3ffe:2a00:100:7031::1]:8080'
+        , parsed = parse(url);
+
+      assume(parsed.username).equals('user');
+      assume(parsed.password).equals('password');
+      assume(parsed.host).equals('[3ffe:2a00:100:7031::1]:8080');
+      assume(parsed.hostname).equals('[3ffe:2a00:100:7031::1]');
+      assume(parsed.href).equals(url);
+    });
+
+    it('parses ipv4', function () {
+      var url = 'http://222.148.142.13:61616/foo/bar?q=z'
+        , parsed = parse(url);
+
+      assume(parsed.port).equals('61616');
+      assume(parsed.query).equals('?q=z');
+      assume(parsed.protocol).equals('http:');
+      assume(parsed.hostname).equals('222.148.142.13');
+      assume(parsed.pathname).equals('/foo/bar');
+      assume(parsed.href).equals(url);
+    });
+  });
+
+  describe('auth', function () {
+    it('does not lowercase the USER:PASS', function () {
+      var url = 'HTTP://USER:[e-mail]'
+        , parsed = parse(url);
+
+      assume(parsed.username).equals('USER');
+      assume(parsed.password).equals('PASS');
+      assume(parsed.protocol).equals('http:');
+      assume(parsed.host).equals('example.com');
+      assume(parsed.hostname).equals('example.com');
+    });
+
+    it('accepts @ in pathnames', function () {
+      var url = 'http://mt0.google.com/vt/lyrs=m@114&hl=en&src=api&x=2&y=2&z=3&s='
+        , parsed = parse(url);
+
+      assume(parsed.pathname).equals('/vt/lyrs=m@114&hl=en&src=api&x=2&y=2&z=3&s=');
+      assume(parsed.username).equals('');
+      assume(parsed.password).equals('');
+    });
+
+    it('does not require passwords for auth', function () {
+      var url = 'http://[e-mail]/'
+        , parsed = parse(url);
+
+      assume(parsed.password).equals('');
+      assume(parsed.pathname).equals('/');
+      assume(parsed.username).equals('user');
+      assume(parsed.protocol).equals('http:');
+      assume(parsed.hostname).equals('www.example.com');
+    });
+  });
+
   it('accepts multiple ???', function () {
@@ -166,3 +262,4 @@ var url = 'http://mt0.google.com/vt/lyrs=m@114???&hl=en&src=api&x=2&y=2&z=3&s=';
     it('does not inherit auth from source object', function () {
-      var data = parse('/foo', parse('http://foo:[e-mail]'));
+      var from = parse('http://foo:[e-mail]')
+        , data = parse('/foo', from);
 
@@ -169,0 +266,0 @@ assume(data.port).equals('');

New alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Fixed alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

No v1

Quality

Package is not semver >=1. This means it is not stable and does not support ^ ranges.

Found 1 instance in 1 package

Improved metrics

Total package byte prevSize

29162

increased by12.77%

Lines of code

588

increased by19.03%

Number of low quality alerts

1

decreased by-50%

Number of lines in readme file

119

increased by2.59%

No dependency changes