1<?php
2
3/*
4 * This file is part of the Symfony package.
5 *
6 * (c) Fabien Potencier <fabien@symfony.com>
7 *
8 * For the full copyright and license information, please view the LICENSE
9 * file that was distributed with this source code.
10 */
11
12namespace Symfony\Component\Mime;
13
14/**
15 * @author Fabien Potencier <fabien@symfony.com>
16 * @author Xavier De Cock <xdecock@gmail.com>
17 *
18 * @internal
19 */
20final class CharacterStream
21{
22    /** Pre-computed for optimization */
23    private const UTF8_LENGTH_MAP = [
24        "\x00" => 1, "\x01" => 1, "\x02" => 1, "\x03" => 1, "\x04" => 1, "\x05" => 1, "\x06" => 1, "\x07" => 1,
25        "\x08" => 1, "\x09" => 1, "\x0a" => 1, "\x0b" => 1, "\x0c" => 1, "\x0d" => 1, "\x0e" => 1, "\x0f" => 1,
26        "\x10" => 1, "\x11" => 1, "\x12" => 1, "\x13" => 1, "\x14" => 1, "\x15" => 1, "\x16" => 1, "\x17" => 1,
27        "\x18" => 1, "\x19" => 1, "\x1a" => 1, "\x1b" => 1, "\x1c" => 1, "\x1d" => 1, "\x1e" => 1, "\x1f" => 1,
28        "\x20" => 1, "\x21" => 1, "\x22" => 1, "\x23" => 1, "\x24" => 1, "\x25" => 1, "\x26" => 1, "\x27" => 1,
29        "\x28" => 1, "\x29" => 1, "\x2a" => 1, "\x2b" => 1, "\x2c" => 1, "\x2d" => 1, "\x2e" => 1, "\x2f" => 1,
30        "\x30" => 1, "\x31" => 1, "\x32" => 1, "\x33" => 1, "\x34" => 1, "\x35" => 1, "\x36" => 1, "\x37" => 1,
31        "\x38" => 1, "\x39" => 1, "\x3a" => 1, "\x3b" => 1, "\x3c" => 1, "\x3d" => 1, "\x3e" => 1, "\x3f" => 1,
32        "\x40" => 1, "\x41" => 1, "\x42" => 1, "\x43" => 1, "\x44" => 1, "\x45" => 1, "\x46" => 1, "\x47" => 1,
33        "\x48" => 1, "\x49" => 1, "\x4a" => 1, "\x4b" => 1, "\x4c" => 1, "\x4d" => 1, "\x4e" => 1, "\x4f" => 1,
34        "\x50" => 1, "\x51" => 1, "\x52" => 1, "\x53" => 1, "\x54" => 1, "\x55" => 1, "\x56" => 1, "\x57" => 1,
35        "\x58" => 1, "\x59" => 1, "\x5a" => 1, "\x5b" => 1, "\x5c" => 1, "\x5d" => 1, "\x5e" => 1, "\x5f" => 1,
36        "\x60" => 1, "\x61" => 1, "\x62" => 1, "\x63" => 1, "\x64" => 1, "\x65" => 1, "\x66" => 1, "\x67" => 1,
37        "\x68" => 1, "\x69" => 1, "\x6a" => 1, "\x6b" => 1, "\x6c" => 1, "\x6d" => 1, "\x6e" => 1, "\x6f" => 1,
38        "\x70" => 1, "\x71" => 1, "\x72" => 1, "\x73" => 1, "\x74" => 1, "\x75" => 1, "\x76" => 1, "\x77" => 1,
39        "\x78" => 1, "\x79" => 1, "\x7a" => 1, "\x7b" => 1, "\x7c" => 1, "\x7d" => 1, "\x7e" => 1, "\x7f" => 1,
40        "\x80" => 0, "\x81" => 0, "\x82" => 0, "\x83" => 0, "\x84" => 0, "\x85" => 0, "\x86" => 0, "\x87" => 0,
41        "\x88" => 0, "\x89" => 0, "\x8a" => 0, "\x8b" => 0, "\x8c" => 0, "\x8d" => 0, "\x8e" => 0, "\x8f" => 0,
42        "\x90" => 0, "\x91" => 0, "\x92" => 0, "\x93" => 0, "\x94" => 0, "\x95" => 0, "\x96" => 0, "\x97" => 0,
43        "\x98" => 0, "\x99" => 0, "\x9a" => 0, "\x9b" => 0, "\x9c" => 0, "\x9d" => 0, "\x9e" => 0, "\x9f" => 0,
44        "\xa0" => 0, "\xa1" => 0, "\xa2" => 0, "\xa3" => 0, "\xa4" => 0, "\xa5" => 0, "\xa6" => 0, "\xa7" => 0,
45        "\xa8" => 0, "\xa9" => 0, "\xaa" => 0, "\xab" => 0, "\xac" => 0, "\xad" => 0, "\xae" => 0, "\xaf" => 0,
46        "\xb0" => 0, "\xb1" => 0, "\xb2" => 0, "\xb3" => 0, "\xb4" => 0, "\xb5" => 0, "\xb6" => 0, "\xb7" => 0,
47        "\xb8" => 0, "\xb9" => 0, "\xba" => 0, "\xbb" => 0, "\xbc" => 0, "\xbd" => 0, "\xbe" => 0, "\xbf" => 0,
48        "\xc0" => 2, "\xc1" => 2, "\xc2" => 2, "\xc3" => 2, "\xc4" => 2, "\xc5" => 2, "\xc6" => 2, "\xc7" => 2,
49        "\xc8" => 2, "\xc9" => 2, "\xca" => 2, "\xcb" => 2, "\xcc" => 2, "\xcd" => 2, "\xce" => 2, "\xcf" => 2,
50        "\xd0" => 2, "\xd1" => 2, "\xd2" => 2, "\xd3" => 2, "\xd4" => 2, "\xd5" => 2, "\xd6" => 2, "\xd7" => 2,
51        "\xd8" => 2, "\xd9" => 2, "\xda" => 2, "\xdb" => 2, "\xdc" => 2, "\xdd" => 2, "\xde" => 2, "\xdf" => 2,
52        "\xe0" => 3, "\xe1" => 3, "\xe2" => 3, "\xe3" => 3, "\xe4" => 3, "\xe5" => 3, "\xe6" => 3, "\xe7" => 3,
53        "\xe8" => 3, "\xe9" => 3, "\xea" => 3, "\xeb" => 3, "\xec" => 3, "\xed" => 3, "\xee" => 3, "\xef" => 3,
54        "\xf0" => 4, "\xf1" => 4, "\xf2" => 4, "\xf3" => 4, "\xf4" => 4, "\xf5" => 4, "\xf6" => 4, "\xf7" => 4,
55        "\xf8" => 5, "\xf9" => 5, "\xfa" => 5, "\xfb" => 5, "\xfc" => 6, "\xfd" => 6, "\xfe" => 0, "\xff" => 0,
56    ];
57
58    private $data = '';
59    private $dataSize = 0;
60    private $map = [];
61    private $charCount = 0;
62    private $currentPos = 0;
63    private $fixedWidth = 0;
64
65    /**
66     * @param resource|string $input
67     */
68    public function __construct($input, ?string $charset = 'utf-8')
69    {
70        $charset = strtolower(trim($charset)) ?: 'utf-8';
71        if ('utf-8' === $charset || 'utf8' === $charset) {
72            $this->fixedWidth = 0;
73            $this->map = ['p' => [], 'i' => []];
74        } else {
75            switch ($charset) {
76                // 16 bits
77                case 'ucs2':
78                case 'ucs-2':
79                case 'utf16':
80                case 'utf-16':
81                    $this->fixedWidth = 2;
82                    break;
83
84                // 32 bits
85                case 'ucs4':
86                case 'ucs-4':
87                case 'utf32':
88                case 'utf-32':
89                    $this->fixedWidth = 4;
90                break;
91
92                // 7-8 bit charsets: (us-)?ascii, (iso|iec)-?8859-?[0-9]+, windows-?125[0-9], cp-?[0-9]+, ansi, macintosh,
93                //                   koi-?7, koi-?8-?.+, mik, (cork|t1), v?iscii
94                // and fallback
95                default:
96                    $this->fixedWidth = 1;
97            }
98        }
99        if (\is_resource($input)) {
100            $blocks = 512;
101            if (stream_get_meta_data($input)['seekable'] ?? false) {
102                rewind($input);
103            }
104            while (false !== $read = fread($input, $blocks)) {
105                $this->write($read);
106            }
107        } else {
108            $this->write($input);
109        }
110    }
111
112    public function read(int $length): ?string
113    {
114        if ($this->currentPos >= $this->charCount) {
115            return null;
116        }
117        $length = ($this->currentPos + $length > $this->charCount) ? $this->charCount - $this->currentPos : $length;
118        if ($this->fixedWidth > 0) {
119            $len = $length * $this->fixedWidth;
120            $ret = substr($this->data, $this->currentPos * $this->fixedWidth, $len);
121            $this->currentPos += $length;
122        } else {
123            $end = $this->currentPos + $length;
124            $end = $end > $this->charCount ? $this->charCount : $end;
125            $ret = '';
126            $start = 0;
127            if ($this->currentPos > 0) {
128                $start = $this->map['p'][$this->currentPos - 1];
129            }
130            $to = $start;
131            for (; $this->currentPos < $end; ++$this->currentPos) {
132                if (isset($this->map['i'][$this->currentPos])) {
133                    $ret .= substr($this->data, $start, $to - $start).'?';
134                    $start = $this->map['p'][$this->currentPos];
135                } else {
136                    $to = $this->map['p'][$this->currentPos];
137                }
138            }
139            $ret .= substr($this->data, $start, $to - $start);
140        }
141
142        return $ret;
143    }
144
145    public function readBytes(int $length): ?array
146    {
147        if (null !== $read = $this->read($length)) {
148            return array_map('ord', str_split($read, 1));
149        }
150
151        return null;
152    }
153
154    public function setPointer(int $charOffset): void
155    {
156        if ($this->charCount < $charOffset) {
157            $charOffset = $this->charCount;
158        }
159        $this->currentPos = $charOffset;
160    }
161
162    public function write(string $chars): void
163    {
164        $ignored = '';
165        $this->data .= $chars;
166        if ($this->fixedWidth > 0) {
167            $strlen = \strlen($chars);
168            $ignoredL = $strlen % $this->fixedWidth;
169            $ignored = $ignoredL ? substr($chars, -$ignoredL) : '';
170            $this->charCount += ($strlen - $ignoredL) / $this->fixedWidth;
171        } else {
172            $this->charCount += $this->getUtf8CharPositions($chars, $this->dataSize, $ignored);
173        }
174        $this->dataSize = \strlen($this->data) - \strlen($ignored);
175    }
176
177    private function getUtf8CharPositions(string $string, int $startOffset, string &$ignoredChars): int
178    {
179        $strlen = \strlen($string);
180        $charPos = \count($this->map['p']);
181        $foundChars = 0;
182        $invalid = false;
183        for ($i = 0; $i < $strlen; ++$i) {
184            $char = $string[$i];
185            $size = self::UTF8_LENGTH_MAP[$char];
186            if (0 == $size) {
187                /* char is invalid, we must wait for a resync */
188                $invalid = true;
189                continue;
190            }
191
192            if ($invalid) {
193                /* We mark the chars as invalid and start a new char */
194                $this->map['p'][$charPos + $foundChars] = $startOffset + $i;
195                $this->map['i'][$charPos + $foundChars] = true;
196                ++$foundChars;
197                $invalid = false;
198            }
199            if (($i + $size) > $strlen) {
200                $ignoredChars = substr($string, $i);
201                break;
202            }
203            for ($j = 1; $j < $size; ++$j) {
204                $char = $string[$i + $j];
205                if ($char > "\x7F" && $char < "\xC0") {
206                    // Valid - continue parsing
207                } else {
208                    /* char is invalid, we must wait for a resync */
209                    $invalid = true;
210                    continue 2;
211                }
212            }
213            /* Ok we got a complete char here */
214            $this->map['p'][$charPos + $foundChars] = $startOffset + $i + $size;
215            $i += $j - 1;
216            ++$foundChars;
217        }
218
219        return $foundChars;
220    }
221}
222