The tokenise object is now in a separate project and so composer is now required.

hexydec · hexydec · commit 73565f2e6fcb · 2021-04-18T21:30:58.000+01:00
Rewired all code to use the tokenise object from the new namespace.
Updated minify.md to reflect API changes.
Updated readme.md to add installation instructions and link to the online minifier.
Updated tag::html() to improve how attributes are encoded, now only &amp; and &lt; plus the quote specified as the quote style is encoded.
Expanded the functionality of the minimal setting in tag::html() to switch to single quotes when there are double quotes in the attribute value to make the encoded output smaller.
Updated tests.
diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.yml
@@ -8,13 +8,14 @@ jobs:
 
     steps:
       - uses: actions/checkout@v2
+      - uses: php-actions/composer@v5
 
       - name: PHPUnit Tests
         uses: php-actions/phpunit@v2
         with:
           php_version: 7.4
           php_extensions: mbstring xdebug
-          bootstrap: src/autoload.php
+          bootstrap: vendor/autoload.php
           args: --coverage-text --coverage-clover=coverage.xml tests
         env:
           XDEBUG_MODE: coverage
diff --git a/composer.json b/composer.json
@@ -7,11 +7,13 @@
         {
             "name": "Will Earp",
             "email": "will@hexydec.com",
-			"homepage": "https://github.com/hexydec/htmldoc"
+			"homepage": "https://github.com/hexydec"
         }
     ],
-    "minimum-stability": "beta",
+    "minimum-stability": "stable",
     "require": {
+        "php": ">=7.3",
+		"hexydec/tokenise": "@dev",
 		"hexydec/cssdoc": "@dev",
 		"hexydec/jslite": "@dev"
 	},
diff --git a/composer.lock b/composer.lock
diff --git a/docs/api/minify.md b/docs/api/minify.md
@@ -39,11 +39,16 @@ An optional array contains a list of configuration parameters to configure the m
 			<td><code>true</code></td>
 		</tr>
 		<tr>
-			<td rowspan="2"><code>comments</code></td>
-			<td rowspan="2">Array</td>
-			<td colspan="2">Remove comments, set to false to preserve comments</td>
+			<td rowspan="3"><code>comments</code></td>
+			<td rowspan="3">Array</td>
+			<td colspan="2">Options for removing comments</td>
 			<td><code>Array()</code></td>
 		</tr>
+		<tr>
+			<td><code>comments</code></td>
+			<td>Whether to remove comments</td>
+			<td><code>true</code></td>
+		</tr>
 		<tr>
 			<td><code>ie</code></td>
 			<td>Whether to preserve Internet Explorer specific comments</td>
@@ -76,8 +81,8 @@ An optional array contains a list of configuration parameters to configure the m
 			<td><code>true</code></td>
 		</tr>
 		<tr>
-			<td rowspan="8"><code>attributes</code></td>
-			<td rowspan="8">Array</td>
+			<td rowspan="9"><code>attributes</code></td>
+			<td rowspan="9">Array</td>
 			<td colspan="2">Minify attributes</td>
 			<td><code>Array()</code></td>
 		</tr>
diff --git a/docs/how-to-use.md b/docs/how-to-use.md
@@ -98,7 +98,7 @@ if ($doc->load($html)) {
 
 The `minify()` method can also accept an array of minification options to change what is minified and what is not, this can be useful for example for minification of HTML for emails.
 
-To see all the available options [see the API documentation](api/minify.md).
+To see all the available options [API documentation for the `minify()` method](api/minify.md).
 
 ## Outputting Documents
 
diff --git a/readme.md b/readme.md
@@ -34,6 +34,8 @@ if ($doc->load($html) {
 }
 ```
 
+You can test out the minifier online at [https://hexydec.com/htmldoc/](https://hexydec.com/htmldoc/), or run the supplied index.php file after installation.
+
 To extract data from an HTML document:
 
 ```php
@@ -56,6 +58,18 @@ if ($doc->open($url) {
 
 ```
 
+## Installation
+
+The easiest way to get up and running is to use composer:
+
+```
+$ composer install
+```
+
+HTMLdoc requires [\hexydec\token\tokenise](https://github.com/hexydec/tokenise) to run, which you can install manually if not using composer. Optionally you can also install [CSSdoc](https://github.com/hexydec/cssdoc) and [JSlite](https://github.com/hexydec/jslite) to perform inline CSS and Javascript minification respectively.
+
+All these dependencies will be installed through composer.
+
 ## Documentation
 
 - [How it works](docs/how-it-works.md)
diff --git a/src/htmldoc/htmldoc.php b/src/htmldoc/htmldoc.php
@@ -1,6 +1,7 @@
 <?php
 declare(strict_types = 1);
 namespace hexydec\html;
+use \hexydec\tokens\tokenise;
 
 class htmldoc extends config implements \ArrayAccess, \Iterator {
 
diff --git a/src/htmldoc/tokenise.php b/src/htmldoc/tokenise.php
diff --git a/src/htmldoc/tokens/comment.php b/src/htmldoc/tokens/comment.php
@@ -1,6 +1,7 @@
 <?php
 declare(strict_types = 1);
 namespace hexydec\html;
+use \hexydec\tokens\tokenise;
 
 class comment implements token {
 
diff --git a/src/htmldoc/tokens/doctype.php b/src/htmldoc/tokens/doctype.php
@@ -1,6 +1,7 @@
 <?php
 declare(strict_types = 1);
 namespace hexydec\html;
+use \hexydec\tokens\tokenise;
 
 class doctype implements token {
 
diff --git a/src/htmldoc/tokens/interfaces/token.php b/src/htmldoc/tokens/interfaces/token.php
@@ -1,6 +1,7 @@
 <?php
 declare(strict_types = 1);
 namespace hexydec\html;
+use \hexydec\tokens\tokenise;
 
 /**
  *
diff --git a/src/htmldoc/tokens/script.php b/src/htmldoc/tokens/script.php
@@ -1,6 +1,7 @@
 <?php
 declare(strict_types = 1);
 namespace hexydec\html;
+use \hexydec\tokens\tokenise;
 
 class script implements token {
 
diff --git a/src/htmldoc/tokens/style.php b/src/htmldoc/tokens/style.php
@@ -1,6 +1,7 @@
 <?php
 declare(strict_types = 1);
 namespace hexydec\html;
+use \hexydec\tokens\tokenise;
 
 class style implements token {
 
diff --git a/src/htmldoc/tokens/tag.php b/src/htmldoc/tokens/tag.php
@@ -1,6 +1,7 @@
 <?php
 declare(strict_types = 1);
 namespace hexydec\html;
+use \hexydec\tokens\tokenise;
 
 class tag implements token {
 
@@ -608,16 +609,24 @@ public function html(array $options = []) : string {
 		foreach ($this->attributes AS $key => $value) {
 			$html .= ' '.$key;
 			if ($value !== null || $options['xml']) {
-				$empty = in_array($value, Array(null, ''), true);
-				$quote = '"';
-				if ($options['quotestyle'] == 'single') {
-					$quote = "'";
-				} elseif (!$empty && $options['quotestyle'] == 'minimal' && strcspn($value, " =\"'`<>\n\r\t/") == strlen($value)) {
+				$empty = in_array($value, [null, ''], true);
+
+				// unquoted
+				if (!$options['xml'] && $options['quotestyle'] == 'minimal' && strcspn($value, " =\"'`<>\n\r\t/") == strlen($value)) {
 					$quote = '';
+
+				// single quotes || swap when minimal and there are double quotes in the string
+				} elseif ($options['quotestyle'] == 'single' || ($options['quotestyle'] == 'minimal' && mb_strpos($value, '"') !== false)) {
+					$quote = "'";
+					$value = str_replace(['&', "'", '<'], ['&amp;', '&#39;', '&lt;'], $value);
+
+				// double quotes
+				} else {
+					$quote = '"';
+					$value = str_replace(['&', '"', '<'], ['&amp;', '&quot;', '&lt;'], $value);
 				}
-				if (!$empty) {
-					$value = htmlspecialchars($value, ENT_HTML5 | ($options['quotestyle'] == 'single' ? ENT_QUOTES : ENT_COMPAT));
-				}
+
+				// compile
 				$html .= '='.$quote.$value.$quote;
 			}
 		}
diff --git a/src/htmldoc/tokens/text.php b/src/htmldoc/tokens/text.php
diff --git a/tests/htmldocTest.php b/tests/htmldocTest.php
