MarkupToPh::_fixWrongBuffer() - Code Metrics - matecat/subfiltering - Measure and Improve Code Quality continuously with Scrutinizer

MarkupToPh::_fixWrongBuffer() A
last analyzed 2025-09-26 09:21 UTC

↳ Parent: MarkupToPh

Complexity

Conditions	1
Paths	1

Size

Total Lines	4
Code Lines	2

Duplication

Lines	0
Ratio	0 %

Code Coverage

Tests	3
CRAP Score	1

Importance

Changes

Metric	Value
cc	1
eloc	2
nc	1
nop	1
dl	0
loc	4
ccs	3
cts	3
cp	1
crap	1
rs	10
c	0
b	0
f	0

<?php
/**
 * Created by PhpStorm.
 * @author domenico [email protected] / [email protected]
 * Date: 05/11/18
 * Time: 15.30
 *
 */

namespace Matecat\SubFiltering\Filters;

use Matecat\SubFiltering\Commons\AbstractHandler;
use Matecat\SubFiltering\Enum\ConstantEnum;
use Matecat\SubFiltering\Enum\CTypeEnum;
use Matecat\SubFiltering\Filters\Html\CallbacksHandler;
use Matecat\SubFiltering\Filters\Html\HtmlParser;

/**
 * Class HtmlToPh
 *
 * This class converts HTML tags within a string into placeholder tags (<ph>).
 * It uses an HtmlParser with a set of callbacks to process different parts of the HTML content.
 *
 * @author  domenico [email protected] / [email protected]
 * @package SubFiltering
 *
 */
class MarkupToPh extends AbstractHandler {

    use CallbacksHandler;

    protected bool $isHTML = false;

    /**
     * Handles plain text content. Returns the buffer unchanged.
     *
     * @param string $buffer The plain text buffer.
     *
     * @return string The original buffer.
     */
    protected function _finalizePlainText( string $buffer ): string {
        return $buffer;
    }

    /**
     * Handles and finalizes an HTML tag.
     *
     * This method decodes HTML entities within the tag's attributes while preserving the '<' and '>' characters of the tag itself.
     * This is necessary to correctly handle encoded attribute values. For example, an attribute like `href="...?a=1&amp;amp;b=2"`
     * becomes `href="...?a=1&amp;b=2"`.
     *
     * @param string $buffer The HTML tag string.
     *
     * @return string The generated <ph> placeholder tag.
     */
    protected function _finalizeMarkupTag( string $buffer ): string {
        // Decode attributes by locking < and > first
        // Because a HTML tag has it's attributes encoded and here we get lt and gt decoded but not other parts of the string
        // Ex:
        // incoming string: <a href="/users/settings?test=123&amp;amp;foobar=1" target="_blank">
        // this should be: <a href="/users/settings?test=123&amp;foobar=1" target="_blank"> with only one ampersand encoding
        //
        $buffer = str_replace( [ '<', '>' ], [ '#_lt_#', '#_gt_#' ], $buffer );
        $buffer = html_entity_decode( $buffer, ENT_NOQUOTES | 16 /* ENT_XML1 */, 'UTF-8' );
        $buffer = str_replace( [ '#_lt_#', '#_gt_#' ], [ '<', '>' ], $buffer );

        return $this->_finalizeTag( $buffer );

    }

    /**
     * Converts a generic tag string into a <ph> placeholder.
     * The original tag is stored in the 'equiv-text' attribute, base64 encoded.
     *
     * @param string $buffer The tag string to convert.
     *
     * @return string The resulting <ph> tag.
     */
    protected function _finalizeTag( string $buffer ): string {
        $isHTML       = $this->isHTML;
        $this->isHTML = false;

        return '<ph id="' . $this->getPipeline()->getNextId() . '" ctype="' . ( $isHTML ? CTypeEnum::HTML : CTypeEnum::XML ) . '" equiv-text="base64:' . base64_encode( htmlentities( $buffer, ENT_NOQUOTES | 16 /* ENT_XML1 */ ) ) . '"/>';
    }

    /**
     * "Fixes" a buffer that was incorrectly identified as a tag by escaping its angle brackets.
     *
     * @param string $buffer The string buffer.
     *
     * @return string The fixed string with escaped angle brackets.
     */
    protected function _fixWrongBuffer( string $buffer ): string {
        $buffer = str_replace( "<", "&lt;", $buffer );

        return str_replace( ">", "&gt;", $buffer );
    }

    /**
     * Finalizes a <script> tag by converting it into a placeholder.
     *
     * @param string $buffer The script tag string.
     *
     * @return string The generated <ph> placeholder tag.
     */
    protected function _finalizeScriptTag( string $buffer ): string {
        return $this->_finalizeTag( $buffer );
    }

    /**
     * Validates a given tag string based on specific criteria for HTML5 and XML tags.
     *
     * The method determines whether a given tag string is valid by:
     * 1. Ensuring there are no placeholder markers (e.g., `##LESSTHAN##`, `##GREATERTHAN##`).
     * 2. Matching against a comprehensive HTML5 tag and attribute structure using regex.
     * 3. Optionally performing a stricter validation for XML tag structures.
     *
     * @param string $buffer The string representation of a tag to be validated.
     *
     * @return bool Returns true if the tag is considered valid; false otherwise.
     */
    protected function _isTagValid( string $buffer ): bool {

        // This is a safeguard against misinterpreting partially processed strings.
        // During filtering, inner tags might be replaced by placeholders (e.g., ##LESSTHAN##).
        // If such placeholders exist within what looks like a tag, it means the tag's
        // content is not yet restored, so we must not treat it as a valid, final tag.
        // For example, an original string like '&lt;a href="<x/>"&gt;' could become
        // '<a href="##LESSTHAN##x/##GREATERTHAN##">', which should not be converted to a <ph> tag.
        if ( strpos( $buffer, ConstantEnum::LTPLACEHOLDER ) !== false || strpos( $buffer, ConstantEnum::GTPLACEHOLDER ) !== false ) {
            return false;
        }

        /**
         * Validates if the given buffer contains a valid HTML5 tag.
         *
         * This method uses a regular expression to match and validate HTML5 tags, including their attributes.
         * It supports a wide range of HTML5 elements and global attributes, ensuring that the buffer adheres
         * to the HTML5 specification.
         *
         * Features:
         * - Matches all valid HTML5 tags, including opening, closing, and self-closing tags.
         * - Handles global attributes such as id, class, style, data-* attributes, ARIA attributes, and event handlers.
         * - Supports attribute values in double quotes, single quotes, or unquoted.
         * - Robust to multiple attributes, whitespace, and Unicode characters.
         *
         * Example HTML matched by the regex:
         * - `<div class="example" data-info="123">Content</div>`
         * - `<img src="image.png" alt="Image" />`
         * - `<button onclick="alert('Click!')">Click me</button>`
         *
         * @see https://regex101.com/r/o546zS/2
         *
         * @param string $buffer The string to validate as an HTML5 tag.
         *
         * @return bool Returns true if the buffer contains a valid HTML5 tag; false otherwise.
         */
        if ( preg_match( '#</?(?:a|abbr|address|area|article|aside|audio|b|base|bdi|bdo|blockquote|body|br|button|canvas|caption|cite|code|col|colgroup|data|datalist|dd|del|details|dfn|dialog|div|dl|dt|em|embed|fieldset|figcaption|figure|footer|form|h1|h2|h3|h4|h5|h6|head|header|hr|html|i|iframe|img|input|ins|kbd|label|legend|li|link|main|map|mark|menu|meta|meter|nav|noscript|object|ol|optgroup|option|output|p|param|picture|pre|progress|q|rb|rp|rt|rtc|ruby|s|samp|script|section|select|slot|small|source|span|strong|style|sub|summary|sup|table|tbody|td|template|textarea|tfoot|th|thead|time|title|tr|track|u|ul|var|video|wbr)(?:\s+[:a-z0-9\-._]+(?:=(?:"[^"]*"|\'[^\']*\'|[^\s>]+))?)*\s*/?>#ui', $buffer ) ) {
            $this->isHTML = true;

            return true;
        }

        /**
         * Validates the general structure of an XML tag using a stricter regex.
         *
         * This validation ensures that the XML tag adheres to the following rules:
         * - The tag may optionally start with a '/' character.
         * - The tag name must NOT start with a number or a hyphen.
         * - The tag name can only contain alphanumeric characters, hyphens (-), dots (.), and underscores (_).
         * - The tag name must have at least one character.
         * - The tag must end with a letter, a digit, a single quote ('), a double quote ("), or a forward slash (/).
         * - Attributes must be defined with an equal sign and quoted values (either single or double quotes).
         *
         * Notes:
         * - Unicode letters in element and attribute names are not allowed.
         * - This validation is stricter than the HTML5 validation and is tailored for XML documents.
         * - For more details, see the XML specification: https://www.w3.org/TR/xml/#NT-Attribute
         *
         * @see https://regex101.com/r/hsk9KU/4
         *
         * @param string $buffer The string representation of the tag to validate.
         *
         * @return bool Returns true if the tag matches the stricter XML structure; false otherwise.
         */
        if ( preg_match( '#</?(?![0-9\-]+)[a-z0-9\-._:]+?(?:\s+[:a-z0-9\-._]+=(?:"[^"]*"|\'[^\']*\'))*\s*/?>#ui', $buffer ) ) {
            return true;
        }

        return false;

    }

    /**
     * Main transformation method.
     *
     * It instantiates an HtmlParser, registers this class as the callback handler,
     * and processes the input segment to convert HTML tags to placeholders.
     *
     * @param string $segment The input string segment to process.
     *
     * @return string The transformed segment.
     */
    public function transform( string $segment ): string {

        // restore < e >
        $segment = str_replace( "&lt;", "<", $segment );
        $segment = str_replace( "&gt;", ">", $segment );

        $parser = new HtmlParser();
        $parser->registerCallbacksHandler( $this );

        return $parser->transform( $segment );
    }

}

1		<?php
2		/**
3		* Created by PhpStorm.
4		* @author domenico [email protected] / [email protected]
5		* Date: 05/11/18
6		* Time: 15.30
7		*
8		*/
9
10		namespace Matecat\SubFiltering\Filters;
11
12		use Matecat\SubFiltering\Commons\AbstractHandler;
13		use Matecat\SubFiltering\Enum\ConstantEnum;
14		use Matecat\SubFiltering\Enum\CTypeEnum;
15		use Matecat\SubFiltering\Filters\Html\CallbacksHandler;
16		use Matecat\SubFiltering\Filters\Html\HtmlParser;
17
18		/**
19		* Class HtmlToPh
20		*
21		* This class converts HTML tags within a string into placeholder tags (<ph>).
22		* It uses an HtmlParser with a set of callbacks to process different parts of the HTML content.
23		*
24		* @author domenico [email protected] / [email protected]
25		* @package SubFiltering
26		*
27		*/
28		class MarkupToPh extends AbstractHandler {
29
30		use CallbacksHandler;
31
32		protected bool $isHTML = false;
33
34		/**
35		* Handles plain text content. Returns the buffer unchanged.
36		*
37		* @param string $buffer The plain text buffer.
38		*
39		* @return string The original buffer.
40		*/
41	96	protected function _finalizePlainText( string $buffer ): string {
42	96	return $buffer;
43		}
44
45		/**
46		* Handles and finalizes an HTML tag.
47		*
48		* This method decodes HTML entities within the tag's attributes while preserving the '<' and '>' characters of the tag itself.
49		* This is necessary to correctly handle encoded attribute values. For example, an attribute like `href="...?a=1&amp;b=2"`
50		* becomes `href="...?a=1&b=2"`.
51		*
52		* @param string $buffer The HTML tag string.
53		*
54		* @return string The generated <ph> placeholder tag.
55		*/
56	19	protected function _finalizeMarkupTag( string $buffer ): string {
57		// Decode attributes by locking < and > first
58		// Because a HTML tag has it's attributes encoded and here we get lt and gt decoded but not other parts of the string
59		// Ex:
60		// incoming string: <a href="/users/settings?test=123&amp;foobar=1" target="_blank">
61		// this should be: <a href="/users/settings?test=123&foobar=1" target="_blank"> with only one ampersand encoding
62		//
63	19	$buffer = str_replace( [ '<', '>' ], [ '#_lt_#', '#_gt_#' ], $buffer );
64	19	$buffer = html_entity_decode( $buffer, ENT_NOQUOTES \| 16 /* ENT_XML1 */, 'UTF-8' );
65	19	$buffer = str_replace( [ '#_lt_#', '#_gt_#' ], [ '<', '>' ], $buffer );
66
67	19	return $this->_finalizeTag( $buffer );
68
69		}
70
71		/**
72		* Converts a generic tag string into a <ph> placeholder.
73		* The original tag is stored in the 'equiv-text' attribute, base64 encoded.
74		*
75		* @param string $buffer The tag string to convert.
76		*
77		* @return string The resulting <ph> tag.
78		*/
79	21	protected function _finalizeTag( string $buffer ): string {
80	21	$isHTML = $this->isHTML;
81	21	$this->isHTML = false;
82
83	21	return '<ph id="' . $this->getPipeline()->getNextId() . '" ctype="' . ( $isHTML ? CTypeEnum::HTML : CTypeEnum::XML ) . '" equiv-text="base64:' . base64_encode( htmlentities( $buffer, ENT_NOQUOTES \| 16 /* ENT_XML1 */ ) ) . '"/>';
84		}
85
86		/**
87		* "Fixes" a buffer that was incorrectly identified as a tag by escaping its angle brackets.
88		*
89		* @param string $buffer The string buffer.
90		*
91		* @return string The fixed string with escaped angle brackets.
92		*/
93	13	protected function _fixWrongBuffer( string $buffer ): string {
94	13	$buffer = str_replace( "<", "<", $buffer );
95
96	13	return str_replace( ">", ">", $buffer );
97		}
98
99		/**
100		* Finalizes a <script> tag by converting it into a placeholder.
101		*
102		* @param string $buffer The script tag string.
103		*
104		* @return string The generated <ph> placeholder tag.
105		*/
106	3	protected function _finalizeScriptTag( string $buffer ): string {
107	3	return $this->_finalizeTag( $buffer );
108		}
109
110		/**
111		* Validates a given tag string based on specific criteria for HTML5 and XML tags.
112		*
113		* The method determines whether a given tag string is valid by:
114		* 1. Ensuring there are no placeholder markers (e.g., `##LESSTHAN##`, `##GREATERTHAN##`).
115		* 2. Matching against a comprehensive HTML5 tag and attribute structure using regex.
116		* 3. Optionally performing a stricter validation for XML tag structures.
117		*
118		* @param string $buffer The string representation of a tag to be validated.
119		*
120		* @return bool Returns true if the tag is considered valid; false otherwise.
121		*/
122	26	protected function _isTagValid( string $buffer ): bool {
123
124		// This is a safeguard against misinterpreting partially processed strings.
125		// During filtering, inner tags might be replaced by placeholders (e.g., ##LESSTHAN##).
126		// If such placeholders exist within what looks like a tag, it means the tag's
127		// content is not yet restored, so we must not treat it as a valid, final tag.
128		// For example, an original string like '<a href="<x/>">' could become
129		// '<a href="##LESSTHAN##x/##GREATERTHAN##">', which should not be converted to a <ph> tag.
130	26	if ( strpos( $buffer, ConstantEnum::LTPLACEHOLDER ) !== false \|\| strpos( $buffer, ConstantEnum::GTPLACEHOLDER ) !== false ) {
131	2	return false;
132		}
133
134		/**
135		* Validates if the given buffer contains a valid HTML5 tag.
136		*
137		* This method uses a regular expression to match and validate HTML5 tags, including their attributes.
138		* It supports a wide range of HTML5 elements and global attributes, ensuring that the buffer adheres
139		* to the HTML5 specification.
140		*
141		* Features:
142		* - Matches all valid HTML5 tags, including opening, closing, and self-closing tags.
143		* - Handles global attributes such as id, class, style, data-* attributes, ARIA attributes, and event handlers.
144		* - Supports attribute values in double quotes, single quotes, or unquoted.
145		* - Robust to multiple attributes, whitespace, and Unicode characters.
146		*
147		* Example HTML matched by the regex:
148		* - `<div class="example" data-info="123">Content</div>`
149		* - `<img src="image.png" alt="Image" />`
150		* - `<button onclick="alert('Click!')">Click me</button>`
151		*
152		* @see https://regex101.com/r/o546zS/2
153		*
154		* @param string $buffer The string to validate as an HTML5 tag.
155		*
156		* @return bool Returns true if the buffer contains a valid HTML5 tag; false otherwise.
157		*/
158	24	if ( preg_match( '#</?(?:a\|abbr\|address\|area\|article\|aside\|audio\|b\|base\|bdi\|bdo\|blockquote\|body\|br\|button\|canvas\|caption\|cite\|code\|col\|colgroup\|data\|datalist\|dd\|del\|details\|dfn\|dialog\|div\|dl\|dt\|em\|embed\|fieldset\|figcaption\|figure\|footer\|form\|h1\|h2\|h3\|h4\|h5\|h6\|head\|header\|hr\|html\|i\|iframe\|img\|input\|ins\|kbd\|label\|legend\|li\|link\|main\|map\|mark\|menu\|meta\|meter\|nav\|noscript\|object\|ol\|optgroup\|option\|output\|p\|param\|picture\|pre\|progress\|q\|rb\|rp\|rt\|rtc\|ruby\|s\|samp\|script\|section\|select\|slot\|small\|source\|span\|strong\|style\|sub\|summary\|sup\|table\|tbody\|td\|template\|textarea\|tfoot\|th\|thead\|time\|title\|tr\|track\|u\|ul\|var\|video\|wbr)(?:\s+[:a-z0-9\-._]+(?:=(?:"[^"]"\|\'[^\']\'\|[^\s>]+))?)\s/?>#ui', $buffer ) ) {
159	15	$this->isHTML = true;
160
161	15	return true;
162		}
163
164		/**
165		* Validates the general structure of an XML tag using a stricter regex.
166		*
167		* This validation ensures that the XML tag adheres to the following rules:
168		* - The tag may optionally start with a '/' character.
169		* - The tag name must NOT start with a number or a hyphen.
170		* - The tag name can only contain alphanumeric characters, hyphens (-), dots (.), and underscores (_).
171		* - The tag name must have at least one character.
172		* - The tag must end with a letter, a digit, a single quote ('), a double quote ("), or a forward slash (/).
173		* - Attributes must be defined with an equal sign and quoted values (either single or double quotes).
174		*
175		* Notes:
176		* - Unicode letters in element and attribute names are not allowed.
177		* - This validation is stricter than the HTML5 validation and is tailored for XML documents.
178		* - For more details, see the XML specification: https://www.w3.org/TR/xml/#NT-Attribute
179		*
180		* @see https://regex101.com/r/hsk9KU/4
181		*
182		* @param string $buffer The string representation of the tag to validate.
183		*
184		* @return bool Returns true if the tag matches the stricter XML structure; false otherwise.
185		*/
186	10	if ( preg_match( '#</?(?![0-9\-]+)[a-z0-9\-._:]+?(?:\s+[:a-z0-9\-._]+=(?:"[^"]"\|\'[^\']\'))\s/?>#ui', $buffer ) ) {
187	7	return true;
188		}
189
190	4	return false;
191
192		}
193
194		/**
195		* Main transformation method.
196		*
197		* It instantiates an HtmlParser, registers this class as the callback handler,
198		* and processes the input segment to convert HTML tags to placeholders.
199		*
200		* @param string $segment The input string segment to process.
201		*
202		* @return string The transformed segment.
203		*/
204	97	public function transform( string $segment ): string {
205
206		// restore < e >
207	97	$segment = str_replace( "<", "<", $segment );
208	97	$segment = str_replace( ">", ">", $segment );
209
210	97	$parser = new HtmlParser();
211	97	$parser->registerCallbacksHandler( $this );
212
213	97	return $parser->transform( $segment );
214		}
215
216		}

matecat / subfiltering

MarkupToPh::_fixWrongBuffer() A last analyzed 2025-09-26 09:21 UTC

Complexity

Size

Duplication

Code Coverage

Importance

Duplication Side-by-Side

Filter issues like

MarkupToPh::_fixWrongBuffer() A
last analyzed 2025-09-26 09:21 UTC