|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
declare(strict_types=1); |
|
4
|
|
|
|
|
5
|
|
|
/* |
|
6
|
|
|
* This file is part of the box project. |
|
7
|
|
|
* |
|
8
|
|
|
* (c) Kevin Herrera <[email protected]> |
|
9
|
|
|
* Théo Fidry <[email protected]> |
|
10
|
|
|
* |
|
11
|
|
|
* This source file is subject to the MIT license that is bundled |
|
12
|
|
|
* with this source code in the file LICENSE. |
|
13
|
|
|
*/ |
|
14
|
|
|
|
|
15
|
|
|
namespace KevinGH\Box\Compactor; |
|
16
|
|
|
|
|
17
|
|
|
use KevinGH\Box\Annotation\CompactedFormatter; |
|
18
|
|
|
use KevinGH\Box\Annotation\DocblockAnnotationParser; |
|
19
|
|
|
use phpDocumentor\Reflection\DocBlockFactory; |
|
20
|
|
|
use PhpToken; |
|
21
|
|
|
use RuntimeException; |
|
22
|
|
|
use Webmozart\Assert\Assert; |
|
23
|
|
|
use function array_pop; |
|
24
|
|
|
use function array_slice; |
|
25
|
|
|
use function array_splice; |
|
26
|
|
|
use function count; |
|
27
|
|
|
use function is_int; |
|
28
|
|
|
use function ltrim; |
|
29
|
|
|
use function preg_replace; |
|
30
|
|
|
use function str_repeat; |
|
31
|
|
|
use const T_COMMENT; |
|
32
|
|
|
use const T_DOC_COMMENT; |
|
33
|
|
|
use const T_WHITESPACE; |
|
34
|
|
|
|
|
35
|
|
|
/** |
|
36
|
|
|
* A PHP source code compactor copied from Composer. |
|
37
|
|
|
* |
|
38
|
|
|
* @see https://github.com/composer/composer/blob/a8df30c09be550bffc37ba540fb7c7f0383c3944/src/Composer/Compiler.php#L214 |
|
39
|
|
|
* |
|
40
|
|
|
* @author Kevin Herrera <[email protected]> |
|
41
|
|
|
* @author Fabien Potencier <[email protected]> |
|
42
|
|
|
* @author Jordi Boggiano <[email protected]> |
|
43
|
|
|
* @author Théo Fidry <[email protected]> |
|
44
|
|
|
* @author Juliette Reinders Folmer <[email protected]> |
|
45
|
|
|
* @author Alessandro Chitolina <[email protected]> |
|
46
|
|
|
* |
|
47
|
|
|
* @private |
|
48
|
|
|
*/ |
|
49
|
|
|
final class Php extends FileExtensionCompactor |
|
50
|
|
|
{ |
|
51
|
|
|
/** |
|
52
|
|
|
* @param list<string> $ignoredAnnotations |
|
|
|
|
|
|
53
|
|
|
*/ |
|
54
|
|
|
public static function create(array $ignoredAnnotations): self |
|
55
|
|
|
{ |
|
56
|
|
|
return new self( |
|
57
|
|
|
new DocblockAnnotationParser( |
|
58
|
|
|
DocBlockFactory::createInstance(), |
|
59
|
|
|
new CompactedFormatter(), |
|
60
|
|
|
$ignoredAnnotations, |
|
61
|
|
|
), |
|
62
|
|
|
); |
|
63
|
|
|
} |
|
64
|
|
|
|
|
65
|
|
|
public function __construct( |
|
66
|
|
|
private readonly ?DocblockAnnotationParser $annotationParser, |
|
67
|
|
|
array $extensions = ['php'], |
|
68
|
|
|
) { |
|
69
|
|
|
parent::__construct($extensions); |
|
70
|
|
|
} |
|
71
|
|
|
|
|
72
|
|
|
protected function compactContent(string $contents): string |
|
73
|
|
|
{ |
|
74
|
|
|
$output = ''; |
|
75
|
|
|
$tokens = PhpToken::tokenize($contents); |
|
76
|
|
|
$tokenCount = count($tokens); |
|
77
|
|
|
|
|
78
|
|
|
for ($index = 0; $index < $tokenCount; ++$index) { |
|
79
|
|
|
$token = $tokens[$index]; |
|
80
|
|
|
$tokenText = $token->text; |
|
81
|
|
|
|
|
82
|
|
|
if ($token->is([T_COMMENT, T_DOC_COMMENT])) { |
|
83
|
|
|
if (str_starts_with($tokenText, '#[')) { |
|
84
|
|
|
// This is, in all likelihood, the start of a PHP >= 8.0 attribute. |
|
85
|
|
|
// Note: $tokens may be updated by reference as well! |
|
86
|
|
|
$retokenized = $this->retokenizeAttribute($tokens, $index); |
|
87
|
|
|
|
|
88
|
|
|
if (null !== $retokenized) { |
|
89
|
|
|
array_splice($tokens, $index, 1, $retokenized); |
|
90
|
|
|
$tokenCount = count($tokens); |
|
91
|
|
|
} |
|
92
|
|
|
|
|
93
|
|
|
$attributeCloser = self::findAttributeCloser($tokens, $index); |
|
94
|
|
|
|
|
95
|
|
|
if (is_int($attributeCloser)) { |
|
96
|
|
|
$output .= '#['; |
|
97
|
|
|
} else { |
|
98
|
|
|
// Turns out this was not an attribute. Treat it as a plain comment. |
|
99
|
|
|
$output .= str_repeat("\n", mb_substr_count($tokenText, "\n")); |
|
100
|
|
|
} |
|
101
|
|
|
} elseif (str_contains($tokenText, '@')) { |
|
102
|
|
|
try { |
|
103
|
|
|
$output .= $this->compactAnnotations($tokenText); |
|
104
|
|
|
} catch (RuntimeException) { |
|
105
|
|
|
$output .= $tokenText; |
|
106
|
|
|
} |
|
107
|
|
|
} else { |
|
108
|
|
|
$output .= str_repeat("\n", mb_substr_count($tokenText, "\n")); |
|
109
|
|
|
} |
|
110
|
|
|
} elseif ($token->is(T_WHITESPACE)) { |
|
111
|
|
|
$whitespace = $tokenText; |
|
112
|
|
|
$previousIndex = ($index - 1); |
|
113
|
|
|
|
|
114
|
|
|
// Handle whitespace potentially being split into two tokens after attribute retokenization. |
|
115
|
|
|
$nextToken = $tokens[$index + 1] ?? null; |
|
116
|
|
|
|
|
117
|
|
|
if (null !== $nextToken |
|
118
|
|
|
&& $nextToken->is(T_WHITESPACE) |
|
119
|
|
|
) { |
|
120
|
|
|
$whitespace .= $nextToken->text; |
|
121
|
|
|
++$index; |
|
122
|
|
|
} |
|
123
|
|
|
|
|
124
|
|
|
// reduce wide spaces |
|
125
|
|
|
$whitespace = preg_replace('{[ \t]+}', ' ', $whitespace); |
|
126
|
|
|
|
|
127
|
|
|
// normalize newlines to \n |
|
128
|
|
|
$whitespace = preg_replace('{(?:\r\n|\r|\n)}', "\n", $whitespace); |
|
129
|
|
|
|
|
130
|
|
|
// If the new line was split off from the whitespace token due to it being included in |
|
131
|
|
|
// the previous (comment) token (PHP < 8), remove leading spaces. |
|
132
|
|
|
|
|
133
|
|
|
$previousToken = $tokens[$previousIndex]; |
|
134
|
|
|
|
|
135
|
|
|
if ($previousToken->is(T_COMMENT) |
|
136
|
|
|
&& str_contains($previousToken->text, "\n") |
|
137
|
|
|
) { |
|
138
|
|
|
$whitespace = ltrim($whitespace, ' '); |
|
139
|
|
|
} |
|
140
|
|
|
|
|
141
|
|
|
// trim leading spaces |
|
142
|
|
|
$whitespace = preg_replace('{\n +}', "\n", $whitespace); |
|
143
|
|
|
|
|
144
|
|
|
$output .= $whitespace; |
|
145
|
|
|
} else { |
|
146
|
|
|
$output .= $tokenText; |
|
147
|
|
|
} |
|
148
|
|
|
} |
|
149
|
|
|
|
|
150
|
|
|
return $output; |
|
151
|
|
|
} |
|
152
|
|
|
|
|
153
|
|
|
private function compactAnnotations(string $docblock): string |
|
154
|
|
|
{ |
|
155
|
|
|
if (null === $this->annotationParser) { |
|
156
|
|
|
return $docblock; |
|
157
|
|
|
} |
|
158
|
|
|
|
|
159
|
|
|
$breaksNbr = mb_substr_count($docblock, "\n"); |
|
160
|
|
|
|
|
161
|
|
|
$annotations = $this->annotationParser->parse($docblock); |
|
162
|
|
|
|
|
163
|
|
|
if ([] === $annotations) { |
|
164
|
|
|
return str_repeat("\n", $breaksNbr); |
|
165
|
|
|
} |
|
166
|
|
|
|
|
167
|
|
|
$compactedDocblock = '/**'; |
|
168
|
|
|
|
|
169
|
|
|
foreach ($annotations as $annotation) { |
|
170
|
|
|
$compactedDocblock .= "\n".$annotation; |
|
171
|
|
|
} |
|
172
|
|
|
|
|
173
|
|
|
$breaksNbr -= count($annotations); |
|
174
|
|
|
|
|
175
|
|
|
if ($breaksNbr > 0) { |
|
176
|
|
|
$compactedDocblock .= str_repeat("\n", $breaksNbr - 1); |
|
177
|
|
|
$compactedDocblock .= "\n*/"; |
|
178
|
|
|
} else { |
|
179
|
|
|
// A space is required here to avoid having /***/ |
|
180
|
|
|
$compactedDocblock .= ' */'; |
|
181
|
|
|
} |
|
182
|
|
|
|
|
183
|
|
|
return $compactedDocblock; |
|
184
|
|
|
} |
|
185
|
|
|
|
|
186
|
|
|
/** |
|
187
|
|
|
* @param list<PhpToken> $tokens |
|
188
|
|
|
*/ |
|
189
|
|
|
private static function findAttributeCloser(array $tokens, int $opener): ?int |
|
190
|
|
|
{ |
|
191
|
|
|
$tokenCount = count($tokens); |
|
192
|
|
|
$brackets = [$opener]; |
|
193
|
|
|
$closer = null; |
|
194
|
|
|
|
|
195
|
|
|
for ($i = ($opener + 1); $i < $tokenCount; ++$i) { |
|
196
|
|
|
$tokenText = $tokens[$i]->text; |
|
197
|
|
|
|
|
198
|
|
|
// Allow for short arrays within attributes. |
|
199
|
|
|
if ('[' === $tokenText) { |
|
200
|
|
|
$brackets[] = $i; |
|
201
|
|
|
|
|
202
|
|
|
continue; |
|
203
|
|
|
} |
|
204
|
|
|
|
|
205
|
|
|
if (']' === $tokenText) { |
|
206
|
|
|
array_pop($brackets); |
|
207
|
|
|
|
|
208
|
|
|
if (0 === count($brackets)) { |
|
209
|
|
|
$closer = $i; |
|
210
|
|
|
break; |
|
211
|
|
|
} |
|
212
|
|
|
} |
|
213
|
|
|
} |
|
214
|
|
|
|
|
215
|
|
|
return $closer; |
|
216
|
|
|
} |
|
217
|
|
|
|
|
218
|
|
|
/** |
|
219
|
|
|
* @param non-empty-list<PhpToken> $tokens |
|
|
|
|
|
|
220
|
|
|
*/ |
|
221
|
|
|
private function retokenizeAttribute(array &$tokens, int $opener): ?array |
|
222
|
|
|
{ |
|
223
|
|
|
Assert::keyExists($tokens, $opener); |
|
224
|
|
|
|
|
225
|
|
|
$token = $tokens[$opener]; |
|
226
|
|
|
$attributeBody = mb_substr($token->text, 2); |
|
227
|
|
|
$subTokens = PhpToken::tokenize('<?php '.$attributeBody); |
|
228
|
|
|
|
|
229
|
|
|
// Replace the PHP open tag with the attribute opener as a simple token. |
|
230
|
|
|
array_splice($subTokens, 0, 1, ['#[']); |
|
231
|
|
|
|
|
232
|
|
|
$closer = self::findAttributeCloser($subTokens, 0); |
|
233
|
|
|
|
|
234
|
|
|
// Multi-line attribute or attribute containing something which looks like a PHP close tag. |
|
235
|
|
|
// Retokenize the rest of the file after the attribute opener. |
|
236
|
|
|
if (null === $closer) { |
|
237
|
|
|
foreach (array_slice($tokens, $opener + 1) as $token) { |
|
238
|
|
|
$attributeBody .= $token->text; |
|
239
|
|
|
} |
|
240
|
|
|
|
|
241
|
|
|
$subTokens = PhpToken::tokenize('<?php '.$attributeBody); |
|
242
|
|
|
array_splice($subTokens, 0, 1, ['#[']); |
|
243
|
|
|
|
|
244
|
|
|
$closer = self::findAttributeCloser($subTokens, 0); |
|
245
|
|
|
|
|
246
|
|
|
if (null !== $closer) { |
|
247
|
|
|
array_splice( |
|
248
|
|
|
$tokens, |
|
249
|
|
|
$opener + 1, |
|
250
|
|
|
count($tokens), |
|
251
|
|
|
array_slice($subTokens, $closer + 1), |
|
252
|
|
|
); |
|
253
|
|
|
|
|
254
|
|
|
$subTokens = array_slice($subTokens, 0, $closer + 1); |
|
255
|
|
|
} |
|
256
|
|
|
} |
|
257
|
|
|
|
|
258
|
|
|
if (null === $closer) { |
|
259
|
|
|
return null; |
|
260
|
|
|
} |
|
261
|
|
|
|
|
262
|
|
|
return $subTokens; |
|
263
|
|
|
} |
|
264
|
|
|
} |
|
265
|
|
|
|
The issue could also be caused by a filter entry in the build configuration. If the path has been excluded in your configuration, e.g.
excluded_paths: ["lib/*"], you can move it to the dependency path list as follows:For further information see https://scrutinizer-ci.com/docs/tools/php/php-scrutinizer/#list-dependency-paths