Support extracting the raw URL from the HTML href attribute

.github/workflows/test.yml

@@ -13,6 +13,7 @@ jobs:
           - '8.0'
           - '8.1'
           - '8.2'
+          - '8.3'
     name: php ${{ matrix.php-version }} on ${{ matrix.operating-system }}
     runs-on: ${{ matrix.operating-system }}
     steps:

README.md

@@ -70,7 +70,7 @@ You can also include the supplied `html2text.php` and use `$text = convert_html_
 | Option | Default | Description |
 |--------|---------|-------------|
 | **ignore_errors** | `false` | Set to `true` to ignore any XML parsing errors. |
-| **drop_links** | `false` | Set to `true` to not render links as `[http://foo.com](My Link)`, but rather just `My Link`. |
+| **drop_links** | `false` | Set to `true` to not render links as `[http://foo.com](My Link)`, but rather just `My Link`. Set to `'href'` to return the URL instead of the name.|
 | **char_set** | `'auto'` | Specify a specific character set.  Pass multiple character sets (comma separated) to detect encoding, default is ASCII,UTF-8 |
 
 Pass along options as a second argument to `convert`, for example:

src/Html2Text.php

@@ -477,7 +477,9 @@ private static function iterateOverNode(\DOMNode $node, ?string $prevName, bool
 					} else {
 						// replace it
 						if ($output) {
-							if ($options['drop_links']) {
+							if ($options['drop_links'] === 'href') {
+								$output = str_replace(' ', '+', $href);
+							} elseif ($options['drop_links']) {
 								$output = "$output";
 							} else {
 								$output = "[$output]($href)";

tests/Html2TextTest.php

@@ -38,7 +38,7 @@ function doTestWithResults(string $test, string $result, $options = []): void {
 	}
 
 	/** @return array<array<string>> */
-	public function providerFiles(): array {
+	public static function providerFiles(): array {
 		return [
 			['basic'],
 			['anchors'],
@@ -91,6 +91,14 @@ public function testAnchorsDropLinks(): void {
 		$this->doTestWithResults("anchors", "anchors.no-links", ['drop_links' => true]);
 	}
 
+	public function testBasicHRefLinks(): void {
+		$this->doTestWithResults("basic", "basic.href-links", ['drop_links' => 'href']);
+	}
+
+	public function testAnchorsHRefLinks(): void {
+		$this->doTestWithResults("anchors", "anchors.href-links", ['drop_links' => 'href']);
+	}
+
 	public function testWindows1252(): void {
 		$this->doTestWithResults("windows-1252-example", "windows-1252-example", ['char_set' => 'windows-1252']);
 	}

tests/txt/anchors.href-links.txt

@@ -0,0 +1,5 @@
+A document without any HTML open/closing tags.
+---------------------------------------------------------------
+We try and use the representation given by common browsers of the HTML document, so that it looks similar when converted to plain text. http://foo.com - or http://www.foo.com http://foo.com
+
+An anchor which will not appear

tests/txt/basic.href-links.txt

@@ -0,0 +1,15 @@
+Hello, World!
+
+This is some e-mail content. Even though it has whitespace and newlines, the e-mail converter will handle it correctly.
+
+Even mismatched tags.
+
+A div
+Another div
+A div
+within a div
+
+Another line
+Yet another line
+
+http://foo.com