38492668891

Committed 14 Oct 2026 12:52AM UTC coverage: 92.573% (-0.5%) from 93.115%

Build Type

push

github

Committed by

Commit Message

added upgrading.md

Coverage Stats

2306 of 2491 relevant lines covered (92.57%)

0.93 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

80.28

/src/Texy/Modules/HtmlModule.php

<?php declare(strict_types=1);

/**
 * This file is part of the Texy! (https://texy.deegee.dev)
 * Copyright (c) 2004 David Grudl (https://davidgrudl.com)
 */

namespace Texy\Modules;

use Texy;
use Texy\{HtmlElement, Patterns};
use function count, is_array, is_string;


/**
 * Processes HTML tags and comments in input text.
 */
final class HtmlModule extends Texy\Module
{
        /** pass HTML comments to output? */
        public bool $passComment = true;


        public function __construct(Texy\Texy $texy)
        {
                $this->texy = $texy;

                $texy->addHandler('htmlComment', $this->solveComment(...));
                $texy->addHandler('htmlTag', $this->solveTag(...));

                $texy->registerLinePattern(
                        $this->patternTag(...),
                        '#<(/?)([a-z][a-z0-9_:-]{0,50})((?:\s++[a-z0-9\_:-]++|=\s*+"[^"'
                        . Patterns::MARK
                        . ']*+"|=\s*+\'[^\''
                        . Patterns::MARK
                        . ']*+\'|=[^\s>'
                        . Patterns::MARK
                        . ']++)*)\s*+(/?)>#isu',
                        'html/tag',
                );

                $texy->registerLinePattern(
                        $this->patternComment(...),
                        '#<!--([^' . Patterns::MARK . ']*?)-->#is',
                        'html/comment',
                );
        }


        /**
         * Callback for: <!-- comment -->.
         * @param  string[]  $matches
         */
        public function patternComment(Texy\LineParser $parser, array $matches): HtmlElement|string|null
        {
                [, $mComment] = $matches;
                return $this->texy->invokeAroundHandlers('htmlComment', $parser, [$mComment]);
        }


        /**
         * Callback for: <tag attr="...">.
         * @param  string[]  $matches
         */
        public function patternTag(Texy\LineParser $parser, array $matches): ?string
        {
                [, $mEnd, $mTag, $mAttr, $mEmpty] = $matches;
                // [1] => /
                // [2] => tag
                // [3] => attributes
                // [4] => /

                $isStart = $mEnd !== '/';
                $isEmpty = $mEmpty === '/';
                if (!$isEmpty && str_ends_with($mAttr, '/')) { // uvizlo v $mAttr?
                        $mAttr = substr($mAttr, 0, -1);
                        $isEmpty = true;
                }

                // error - can't close empty element
                if ($isEmpty && !$isStart) {
                        return null;
                }

                // error - end element with atttrs
                $mAttr = trim(strtr($mAttr, "\n", ' '));
                if ($mAttr && !$isStart) {
                        return null;
                }

                $el = new HtmlElement($mTag);
                if ($isStart) {
                        $el->attrs = $this->parseAttributes($mAttr);
                }

                $res = $this->texy->invokeAroundHandlers('htmlTag', $parser, [$el, $isStart, $isEmpty]);

                if ($res instanceof HtmlElement) {
                        return $this->texy->protect($isStart ? $res->startTag() : $res->endTag(), $res->getContentType());
                }

                return $res;
        }


        /**
         * Finish invocation.
         */
        private function solveTag(
                Texy\HandlerInvocation $invocation,
                HtmlElement $el,
                bool $isStart,
                ?bool $forceEmpty = null,
        ): ?HtmlElement
        {
                $texy = $this->texy;

                // tag & attibutes
                $allowedTags = $texy->allowedTags; // speed-up
                if (!$allowedTags) {
                        return null; // all tags are disabled
                }

                // convert case
                $name = $el->getName();
                assert($name !== null);
                $lower = strtolower($name);
                if (isset($texy->getDTD()[$lower]) || $name === strtoupper($name)) {
                        // complete UPPER convert to lower
                        $name = $lower;
                        $el->setName($name);
                }

                if (is_array($allowedTags)) {
                        if (!isset($allowedTags[$name])) {
                                return null;
                        }
                } else { // allowedTags === Texy\Texy::ALL
                        if ($forceEmpty) {
                                $el->setName($name, empty: true);
                        }
                }

                // end tag? we are finished
                if (!$isStart) {
                        return $el;
                }

                $this->applyAttrs($el->attrs, is_array($allowedTags) ? $allowedTags[$name] : $texy::ALL);
                $this->applyClasses($el->attrs, $texy->getAllowedProps()[0]);
                $this->applyStyles($el->attrs, $texy->getAllowedProps()[1]);
                if (!$this->validateAttrs($el, $texy)) {
                        return null;
                }

                $el->validateAttrs($texy->getDTD());

                return $el;
        }


        /**
         * Finish invocation.
         */
        private function solveComment(Texy\HandlerInvocation $invocation, string $content): string
        {
                if (!$this->passComment) {
                        return '';
                }

                // sanitize comment
                $content = Texy\Regexp::replace($content, '#-{2,}#', ' - ');
                $content = trim($content, '-');

                return $this->texy->protect('<!--' . $content . '-->', Texy\Texy::CONTENT_MARKUP);
        }


        /**
         * @param  array<string, array<string|int|bool>|string|int|bool|null>  $attrs
         * @param  bool|string[]  $allowedAttrs
         */
        private function applyAttrs(array &$attrs, bool|array $allowedAttrs): void
        {
                if (!$allowedAttrs) {
                        $attrs = [];

                } elseif (is_array($allowedAttrs)) {
                        // skip disabled
                        $allowedAttrs = array_flip($allowedAttrs);
                        foreach ($attrs as $key => $foo) {
                                if (!isset($allowedAttrs[$key])) {
                                        unset($attrs[$key]);
                                }
                        }
                }
        }


        /**
         * @param  array<string, string|int|bool|array<string|int|bool>|null>  $attrs
         * @param  array<string, int>|bool  $allowedClasses
         */
        private function applyClasses(array &$attrs, bool|array $allowedClasses): void
        {
                if (!isset($attrs['class'])) {
                } elseif (is_array($allowedClasses)) {
                        $attrs['class'] = is_string($attrs['class']) ? explode(' ', $attrs['class']) : (array) $attrs['class'];
                        foreach ($attrs['class'] as $key => $value) {
                                if (!isset($allowedClasses[$value])) {
                                        unset($attrs['class'][$key]); // id & class are case-sensitive
                                }
                        }
                } elseif ($allowedClasses !== Texy\Texy::ALL) {
                        $attrs['class'] = null;
                }

                if (!isset($attrs['id'])) {
                } elseif (is_array($allowedClasses)) {
                        if (!is_string($attrs['id']) || !isset($allowedClasses['#' . $attrs['id']])) {
                                $attrs['id'] = null;
                        }

                } elseif ($allowedClasses !== Texy\Texy::ALL) {
                        $attrs['id'] = null;
                }
        }


        /**
         * @param  array<string, string|int|bool|array<string|int|bool>|null>  $attrs
         * @param  array<string, int>|bool  $allowedStyles
         */
        private function applyStyles(array &$attrs, bool|array $allowedStyles): void
        {
                if (!isset($attrs['style'])) {
                } elseif (is_array($allowedStyles)) {
                        if (is_string($attrs['style'])) {
                                $parts = explode(';', $attrs['style']);
                                $attrs['style'] = [];
                                foreach ($parts as $value) {
                                        if (count($pair = explode(':', $value, 2)) === 2) {
                                                $attrs['style'][trim($pair[0])] = trim($pair[1]);
                                        }
                                }
                        } else {
                                $attrs['style'] = (array) $attrs['style'];
                        }

                        foreach ($attrs['style'] as $key => $value) {
                                if (!isset($allowedStyles[strtolower((string) $key)])) { // CSS is case-insensitive
                                        unset($attrs['style'][$key]);
                                }
                        }
                } elseif ($allowedStyles !== Texy\Texy::ALL) {
                        $attrs['style'] = null;
                }
        }


        private function validateAttrs(HtmlElement $el, Texy\Texy $texy): bool
        {
                foreach (['src', 'href', 'name', 'id'] as $attr) {
                        if (isset($el->attrs[$attr])) {
                                $el->attrs[$attr] = is_string($el->attrs[$attr])
                                        ? trim($el->attrs[$attr])
                                        : '';
                                if ($el->attrs[$attr] === '') {
                                        unset($el->attrs[$attr]);
                                }
                        }
                }

                $name = $el->getName();
                if ($name === 'img') {
                        if (!isset($el->attrs['src'])) {
                                return false;
                        }

                        assert(is_string($el->attrs['src']));
                        if (!$texy->checkURL($el->attrs['src'], $texy::FILTER_IMAGE)) {
                                return false;
                        }

                        $texy->summary['images'][] = $el->attrs['src'];

                } elseif ($name === 'a') {
                        if (!isset($el->attrs['href']) && !isset($el->attrs['name']) && !isset($el->attrs['id'])) {
                                return false;
                        }

                        if (isset($el->attrs['href'])) {
                                assert(is_string($el->attrs['href']));
                                if ($texy->linkModule->forceNoFollow && str_contains($el->attrs['href'], '//')) {
                                        settype($el->attrs['rel'], 'array');
                                        $el->attrs['rel'][] = 'nofollow';
                                }

                                if (!$texy->checkURL($el->attrs['href'], $texy::FILTER_ANCHOR)) {
                                        return false;
                                }

                                $texy->summary['links'][] = $el->attrs['href'];
                        }

                } elseif (preg_match('#^h[1-6]#i', $name ?? '')) {
                        $texy->headingModule->TOC[] = [
                                'el' => $el,
                                'level' => (int) substr($name, 1),
                                'type' => 'html',
                        ];
                }

                return true;
        }


        /** @return array<string, string|bool> */
        private function parseAttributes(string $attrs): array
        {
                $matches = $res = [];
                preg_match_all(
                        '#([a-z0-9\_:-]+)\s*(?:=\s*(\'[^\']*\'|"[^"]*"|[^\'"\s]+))?()#isu',
                        $attrs,
                        $matches,
                        PREG_SET_ORDER,
                );

                foreach ($matches as $m) {
                        $key = strtolower($m[1]);
                        $value = $m[2];
                        if ($value === '') {
                                $res[$key] = true;
                        } elseif ($value[0] === '\'' || $value[0] === '"') {
                                $res[$key] = Texy\Helpers::unescapeHtml(substr($value, 1, -1));
                        } else {
                                $res[$key] = Texy\Helpers::unescapeHtml($value);
                        }
                }

                return $res;
        }
}

1	<?php declare(strict_types=1);
2
3	/**
4	* This file is part of the Texy! (https://texy.deegee.dev)
5	* Copyright (c) 2004 David Grudl (https://davidgrudl.com)
6	*/
7
8	namespace Texy\Modules;
9
10	use Texy;
11	use Texy\{HtmlElement, Patterns};
12	use function count, is_array, is_string;
13
14
15	/**
16	* Processes HTML tags and comments in input text.
17	*/
18	final class HtmlModule extends Texy\Module
19	{
20	/** pass HTML comments to output? */
21	public bool $passComment = true;
22
23
24	public function __construct(Texy\Texy $texy)
25	{
26	$this->texy = $texy;	1 ✔
27
28	$texy->addHandler('htmlComment', $this->solveComment(...));	1 ✔
29	$texy->addHandler('htmlTag', $this->solveTag(...));	1 ✔
30
31	$texy->registerLinePattern(	1 ✔
32	$this->patternTag(...),	1 ✔
33	'#<(/?)([a-z][a-z0-9_:-]{0,50})((?:\s++[a-z0-9\_:-]++\|=\s*+"[^"'
34	. Patterns::MARK
35	. ']+"\|=\s+\'[^\''
36	. Patterns::MARK
37	. ']*+\'\|=[^\s>'
38	. Patterns::MARK
39	. ']++))\s+(/?)>#isu',	1 ✔
40	'html/tag',	1 ✔
41	);
42
43	$texy->registerLinePattern(	1 ✔
44	$this->patternComment(...),	1 ✔
45	'#<!--([^' . Patterns::MARK . ']*?)-->#is',	1 ✔
46	'html/comment',	1 ✔
47	);
48	}	1 ✔
49
50
51	/**
52	* Callback for: <!-- comment -->.
53	* @param string[] $matches
54	*/
55	public function patternComment(Texy\LineParser $parser, array $matches): HtmlElement\|string\|null
56	{
57	[, $mComment] = $matches;	1 ✔
58	return $this->texy->invokeAroundHandlers('htmlComment', $parser, [$mComment]);	1 ✔
59	}
60
61
62	/**
63	* Callback for: <tag attr="...">.
64	* @param string[] $matches
65	*/
66	public function patternTag(Texy\LineParser $parser, array $matches): ?string
67	{
68	[, $mEnd, $mTag, $mAttr, $mEmpty] = $matches;	1 ✔
69	// [1] => /
70	// [2] => tag
71	// [3] => attributes
72	// [4] => /
73
74	$isStart = $mEnd !== '/';	1 ✔
75	$isEmpty = $mEmpty === '/';	1 ✔
76	if (!$isEmpty && str_ends_with($mAttr, '/')) { // uvizlo v $mAttr?	1 ✔
77	$mAttr = substr($mAttr, 0, -1);	×
78	$isEmpty = true;	×
79	}
80
81	// error - can't close empty element
82	if ($isEmpty && !$isStart) {	1 ✔
83	return null;	×
84	}
85
86	// error - end element with atttrs
87	$mAttr = trim(strtr($mAttr, "\n", ' '));	1 ✔
88	if ($mAttr && !$isStart) {	1 ✔
89	return null;	1 ✔
90	}
91
92	$el = new HtmlElement($mTag);	1 ✔
93	if ($isStart) {	1 ✔
94	$el->attrs = $this->parseAttributes($mAttr);	1 ✔
95	}
96
97	$res = $this->texy->invokeAroundHandlers('htmlTag', $parser, [$el, $isStart, $isEmpty]);	1 ✔
98
99	if ($res instanceof HtmlElement) {	1 ✔
100	return $this->texy->protect($isStart ? $res->startTag() : $res->endTag(), $res->getContentType());	1 ✔
101	}
102
103	return $res;	1 ✔
104	}
105
106
107	/**
108	* Finish invocation.
109	*/
110	private function solveTag(
111	Texy\HandlerInvocation $invocation,
112	HtmlElement $el,
113	bool $isStart,
114	?bool $forceEmpty = null,
115	): ?HtmlElement
116	{
117	$texy = $this->texy;	1 ✔
118
119	// tag & attibutes
120	$allowedTags = $texy->allowedTags; // speed-up	1 ✔
121	if (!$allowedTags) {	1 ✔
122	return null; // all tags are disabled	1 ✔
123	}
124
125	// convert case
126	$name = $el->getName();	1 ✔
127	assert($name !== null);
128	$lower = strtolower($name);	1 ✔
129	if (isset($texy->getDTD()[$lower]) \|\| $name === strtoupper($name)) {	1 ✔
130	// complete UPPER convert to lower
131	$name = $lower;	1 ✔
132	$el->setName($name);	1 ✔
133	}
134
135	if (is_array($allowedTags)) {	1 ✔
136	if (!isset($allowedTags[$name])) {	1 ✔
137	return null;	1 ✔
138	}
139	} else { // allowedTags === Texy\Texy::ALL
140	if ($forceEmpty) {	1 ✔
141	$el->setName($name, empty: true);	1 ✔
142	}
143	}
144
145	// end tag? we are finished
146	if (!$isStart) {	1 ✔
147	return $el;	1 ✔
148	}
149
150	$this->applyAttrs($el->attrs, is_array($allowedTags) ? $allowedTags[$name] : $texy::ALL);	1 ✔
151	$this->applyClasses($el->attrs, $texy->getAllowedProps()[0]);	1 ✔
152	$this->applyStyles($el->attrs, $texy->getAllowedProps()[1]);	1 ✔
153	if (!$this->validateAttrs($el, $texy)) {	1 ✔
154	return null;	1 ✔
155	}
156
157	$el->validateAttrs($texy->getDTD());	1 ✔
158
159	return $el;	1 ✔
160	}
161
162
163	/**
164	* Finish invocation.
165	*/
166	private function solveComment(Texy\HandlerInvocation $invocation, string $content): string
167	{
168	if (!$this->passComment) {	1 ✔
169	return '';	×
170	}
171
172	// sanitize comment
173	$content = Texy\Regexp::replace($content, '#-{2,}#', ' - ');	1 ✔
174	$content = trim($content, '-');	1 ✔
175
176	return $this->texy->protect('<!--' . $content . '-->', Texy\Texy::CONTENT_MARKUP);	1 ✔
177	}
178
179
180	/**
181	* @param array<string, array<string\|int\|bool>\|string\|int\|bool\|null> $attrs
182	* @param bool\|string[] $allowedAttrs
183	*/
184	private function applyAttrs(array &$attrs, bool\|array $allowedAttrs): void
185	{
186	if (!$allowedAttrs) {	1 ✔
187	$attrs = [];	1 ✔
188
189	} elseif (is_array($allowedAttrs)) {	1 ✔
190	// skip disabled
191	$allowedAttrs = array_flip($allowedAttrs);	1 ✔
192	foreach ($attrs as $key => $foo) {	1 ✔
193	if (!isset($allowedAttrs[$key])) {	1 ✔
194	unset($attrs[$key]);	×
195	}
196	}
197	}
198	}	1 ✔
199
200
201	/**
202	* @param array<string, string\|int\|bool\|array<string\|int\|bool>\|null> $attrs
203	* @param array<string, int>\|bool $allowedClasses
204	*/
205	private function applyClasses(array &$attrs, bool\|array $allowedClasses): void
206	{
207	if (!isset($attrs['class'])) {	1 ✔
208	} elseif (is_array($allowedClasses)) {	1 ✔
209	$attrs['class'] = is_string($attrs['class']) ? explode(' ', $attrs['class']) : (array) $attrs['class'];	×
210	foreach ($attrs['class'] as $key => $value) {	×
211	if (!isset($allowedClasses[$value])) {	×
212	unset($attrs['class'][$key]); // id & class are case-sensitive	×
213	}
214	}
215	} elseif ($allowedClasses !== Texy\Texy::ALL) {	1 ✔
216	$attrs['class'] = null;	×
217	}
218
219	if (!isset($attrs['id'])) {	1 ✔
220	} elseif (is_array($allowedClasses)) {	1 ✔
221	if (!is_string($attrs['id']) \|\| !isset($allowedClasses['#' . $attrs['id']])) {	×
222	$attrs['id'] = null;	×
223	}
224
225	} elseif ($allowedClasses !== Texy\Texy::ALL) {	1 ✔
226	$attrs['id'] = null;	×
227	}
228	}	1 ✔
229
230
231	/**
232	* @param array<string, string\|int\|bool\|array<string\|int\|bool>\|null> $attrs
233	* @param array<string, int>\|bool $allowedStyles
234	*/
235	private function applyStyles(array &$attrs, bool\|array $allowedStyles): void
236	{
237	if (!isset($attrs['style'])) {	1 ✔
238	} elseif (is_array($allowedStyles)) {	1 ✔
239	if (is_string($attrs['style'])) {	×
240	$parts = explode(';', $attrs['style']);	×
241	$attrs['style'] = [];	×
242	foreach ($parts as $value) {	×
243	if (count($pair = explode(':', $value, 2)) === 2) {	×
244	$attrs['style'][trim($pair[0])] = trim($pair[1]);	×
245	}
246	}
247	} else {
248	$attrs['style'] = (array) $attrs['style'];	×
249	}
250
251	foreach ($attrs['style'] as $key => $value) {	×
252	if (!isset($allowedStyles[strtolower((string) $key)])) { // CSS is case-insensitive	×
253	unset($attrs['style'][$key]);	×
254	}
255	}
256	} elseif ($allowedStyles !== Texy\Texy::ALL) {	1 ✔
257	$attrs['style'] = null;	×
258	}
259	}	1 ✔
260
261
262	private function validateAttrs(HtmlElement $el, Texy\Texy $texy): bool
263	{
264	foreach (['src', 'href', 'name', 'id'] as $attr) {	1 ✔
265	if (isset($el->attrs[$attr])) {	1 ✔
266	$el->attrs[$attr] = is_string($el->attrs[$attr])	1 ✔
267	? trim($el->attrs[$attr])	1 ✔
268	: '';	×
269	if ($el->attrs[$attr] === '') {	1 ✔
270	unset($el->attrs[$attr]);	×
271	}
272	}
273	}
274
275	$name = $el->getName();	1 ✔
276	if ($name === 'img') {	1 ✔
277	if (!isset($el->attrs['src'])) {	1 ✔
278	return false;	×
279	}
280
281	assert(is_string($el->attrs['src']));
282	if (!$texy->checkURL($el->attrs['src'], $texy::FILTER_IMAGE)) {	1 ✔
283	return false;	×
284	}
285
286	$texy->summary['images'][] = $el->attrs['src'];	1 ✔
287
288	} elseif ($name === 'a') {	1 ✔
289	if (!isset($el->attrs['href']) && !isset($el->attrs['name']) && !isset($el->attrs['id'])) {	1 ✔
290	return false;	1 ✔
291	}
292
293	if (isset($el->attrs['href'])) {	1 ✔
294	assert(is_string($el->attrs['href']));
295	if ($texy->linkModule->forceNoFollow && str_contains($el->attrs['href'], '//')) {	1 ✔
296	settype($el->attrs['rel'], 'array');	1 ✔
297	$el->attrs['rel'][] = 'nofollow';	1 ✔
298	}
299
300	if (!$texy->checkURL($el->attrs['href'], $texy::FILTER_ANCHOR)) {	1 ✔
301	return false;	1 ✔
302	}
303
304	$texy->summary['links'][] = $el->attrs['href'];	1 ✔
305	}
306
307	} elseif (preg_match('#^h[1-6]#i', $name ?? '')) {	1 ✔
308	$texy->headingModule->TOC[] = [	1 ✔
309	'el' => $el,	1 ✔
310	'level' => (int) substr($name, 1),	1 ✔
311	'type' => 'html',	1 ✔
312	];
313	}
314
315	return true;	1 ✔
316	}
317
318
319	/** @return array<string, string\|bool> */
320	private function parseAttributes(string $attrs): array
321	{
322	$matches = $res = [];	1 ✔
323	preg_match_all(	1 ✔
324	'#([a-z0-9\_:-]+)\s(?:=\s(\'[^\']\'\|"[^"]"\|[^\'"\s]+))?()#isu',	1 ✔
325	$attrs,
326	$matches,
327	PREG_SET_ORDER,	1 ✔
328	);
329
330	foreach ($matches as $m) {	1 ✔
331	$key = strtolower($m[1]);	1 ✔
332	$value = $m[2];	1 ✔
333	if ($value === '') {	1 ✔
334	$res[$key] = true;	1 ✔
335	} elseif ($value[0] === '\'' \|\| $value[0] === '"') {	1 ✔
336	$res[$key] = Texy\Helpers::unescapeHtml(substr($value, 1, -1));	1 ✔
337	} else {
338	$res[$key] = Texy\Helpers::unescapeHtml($value);	1 ✔
339	}
340	}
341
342	return $res;	1 ✔
343	}
344	}

dg / texy / 38492668891

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous