| 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 |
|
| 12 |
namespace Symfony\Component\Mime\Header; |
| 13 |
|
| 14 |
use Symfony\Component\Mime\Encoder\QpMimeHeaderEncoder; |
| 15 |
|
| 16 |
/** |
| 17 |
* An abstract base MIME Header. |
| 18 |
* |
| 19 |
* @author Chris Corbyn |
| 20 |
*/ |
| 21 |
abstract class AbstractHeader implements HeaderInterface |
| 22 |
{ |
| 23 |
public const PHRASE_PATTERN = '(?:(?:(?:(?:(?:(?:(?:[ \t]*(?:\r\n))?[ \t])?(\((?:(?:(?:[ \t]*(?:\r\n))?[ \t])|(?:(?:[\x01-\x08\x0B\x0C\x0E-\x19\x7F]|[\x21-\x27\x2A-\x5B\x5D-\x7E])|(?:\\[\x00-\x08\x0B\x0C\x0E-\x7F])|(?1)))*(?:(?:[ \t]*(?:\r\n))?[ \t])?\)))*(?:(?:(?:(?:[ \t]*(?:\r\n))?[ \t])?(\((?:(?:(?:[ \t]*(?:\r\n))?[ \t])|(?:(?:[\x01-\x08\x0B\x0C\x0E-\x19\x7F]|[\x21-\x27\x2A-\x5B\x5D-\x7E])|(?:\\[\x00-\x08\x0B\x0C\x0E-\x7F])|(?1)))*(?:(?:[ \t]*(?:\r\n))?[ \t])?\)))|(?:(?:[ \t]*(?:\r\n))?[ \t])))?[a-zA-Z0-9!#\$%&\'\*\+\-\/=\?\^_`\{\}\|~]+(?:(?:(?:(?:[ \t]*(?:\r\n))?[ \t])?(\((?:(?:(?:[ \t]*(?:\r\n))?[ \t])|(?:(?:[\x01-\x08\x0B\x0C\x0E-\x19\x7F]|[\x21-\x27\x2A-\x5B\x5D-\x7E])|(?:\\[\x00-\x08\x0B\x0C\x0E-\x7F])|(?1)))*(?:(?:[ \t]*(?:\r\n))?[ \t])?\)))*(?:(?:(?:(?:[ \t]*(?:\r\n))?[ \t])?(\((?:(?:(?:[ \t]*(?:\r\n))?[ \t])|(?:(?:[\x01-\x08\x0B\x0C\x0E-\x19\x7F]|[\x21-\x27\x2A-\x5B\x5D-\x7E])|(?:\\[\x00-\x08\x0B\x0C\x0E-\x7F])|(?1)))*(?:(?:[ \t]*(?:\r\n))?[ \t])?\)))|(?:(?:[ \t]*(?:\r\n))?[ \t])))?)|(?:(?:(?:(?:(?:[ \t]*(?:\r\n))?[ \t])?(\((?:(?:(?:[ \t]*(?:\r\n))?[ \t])|(?:(?:[\x01-\x08\x0B\x0C\x0E-\x19\x7F]|[\x21-\x27\x2A-\x5B\x5D-\x7E])|(?:\\[\x00-\x08\x0B\x0C\x0E-\x7F])|(?1)))*(?:(?:[ \t]*(?:\r\n))?[ \t])?\)))*(?:(?:(?:(?:[ \t]*(?:\r\n))?[ \t])?(\((?:(?:(?:[ \t]*(?:\r\n))?[ \t])|(?:(?:[\x01-\x08\x0B\x0C\x0E-\x19\x7F]|[\x21-\x27\x2A-\x5B\x5D-\x7E])|(?:\\[\x00-\x08\x0B\x0C\x0E-\x7F])|(?1)))*(?:(?:[ \t]*(?:\r\n))?[ \t])?\)))|(?:(?:[ \t]*(?:\r\n))?[ \t])))?"((?:(?:[ \t]*(?:\r\n))?[ \t])?(?:(?:[\x01-\x08\x0B\x0C\x0E-\x19\x7F]|[\x21\x23-\x5B\x5D-\x7E])|(?:\\[\x00-\x08\x0B\x0C\x0E-\x7F])))*(?:(?:[ \t]*(?:\r\n))?[ \t])?"(?:(?:(?:(?:[ \t]*(?:\r\n))?[ \t])?(\((?:(?:(?:[ \t]*(?:\r\n))?[ \t])|(?:(?:[\x01-\x08\x0B\x0C\x0E-\x19\x7F]|[\x21-\x27\x2A-\x5B\x5D-\x7E])|(?:\\[\x00-\x08\x0B\x0C\x0E-\x7F])|(?1)))*(?:(?:[ \t]*(?:\r\n))?[ \t])?\)))*(?:(?:(?:(?:[ \t]*(?:\r\n))?[ \t])?(\((?:(?:(?:[ \t]*(?:\r\n))?[ \t])|(?:(?:[\x01-\x08\x0B\x0C\x0E-\x19\x7F]|[\x21-\x27\x2A-\x5B\x5D-\x7E])|(?:\\[\x00-\x08\x0B\x0C\x0E-\x7F])|(?1)))*(?:(?:[ \t]*(?:\r\n))?[ \t])?\)))|(?:(?:[ \t]*(?:\r\n))?[ \t])))?))+?)'; |
| 24 |
|
| 25 |
private static $encoder; |
| 26 |
|
| 27 |
private $name; |
| 28 |
private $lineLength = 76; |
| 29 |
private $lang; |
| 30 |
private $charset = 'utf-8'; |
| 31 |
|
| 32 |
public function __construct(string $name) |
| 33 |
{ |
| 34 |
$this->name = $name; |
| 35 |
} |
| 36 |
|
| 37 |
public function setCharset(string $charset) |
| 38 |
{ |
| 39 |
$this->charset = $charset; |
| 40 |
} |
| 41 |
|
| 42 |
public function getCharset(): ?string |
| 43 |
{ |
| 44 |
return $this->charset; |
| 45 |
} |
| 46 |
|
| 47 |
/** |
| 48 |
* Set the language used in this Header. |
| 49 |
* |
| 50 |
* For example, for US English, 'en-us'. |
| 51 |
*/ |
| 52 |
public function setLanguage(string $lang) |
| 53 |
{ |
| 54 |
$this->lang = $lang; |
| 55 |
} |
| 56 |
|
| 57 |
public function getLanguage(): ?string |
| 58 |
{ |
| 59 |
return $this->lang; |
| 60 |
} |
| 61 |
|
| 62 |
public function getName(): string |
| 63 |
{ |
| 64 |
return $this->name; |
| 65 |
} |
| 66 |
|
| 67 |
public function setMaxLineLength(int $lineLength) |
| 68 |
{ |
| 69 |
$this->lineLength = $lineLength; |
| 70 |
} |
| 71 |
|
| 72 |
public function getMaxLineLength(): int |
| 73 |
{ |
| 74 |
return $this->lineLength; |
| 75 |
} |
| 76 |
|
| 77 |
public function toString(): string |
| 78 |
{ |
| 79 |
return $this->tokensToString($this->toTokens()); |
| 80 |
} |
| 81 |
|
| 82 |
/** |
| 83 |
* Produces a compliant, formatted RFC 2822 'phrase' based on the string given. |
| 84 |
* |
| 85 |
* @param string $string as displayed |
| 86 |
* @param bool $shorten the first line to make remove for header name |
| 87 |
*/ |
| 88 |
protected function createPhrase(HeaderInterface $header, string $string, string $charset, bool $shorten = false): string |
| 89 |
{ |
| 90 |
// Treat token as exactly what was given |
| 91 |
$phraseStr = $string; |
| 92 |
|
| 93 |
// If it's not valid |
| 94 |
if (!preg_match('/^'.self::PHRASE_PATTERN.'$/D', $phraseStr)) { |
| 95 |
// .. but it is just ascii text, try escaping some characters |
| 96 |
// and make it a quoted-string |
| 97 |
if (preg_match('/^[\x00-\x08\x0B\x0C\x0E-\x7F]*$/D', $phraseStr)) { |
| 98 |
foreach (['\\', '"'] as $char) { |
| 99 |
$phraseStr = str_replace($char, '\\'.$char, $phraseStr); |
| 100 |
} |
| 101 |
$phraseStr = '"'.$phraseStr.'"'; |
| 102 |
} else { |
| 103 |
// ... otherwise it needs encoding |
| 104 |
// Determine space remaining on line if first line |
| 105 |
if ($shorten) { |
| 106 |
$usedLength = \strlen($header->getName().': '); |
| 107 |
} else { |
| 108 |
$usedLength = 0; |
| 109 |
} |
| 110 |
$phraseStr = $this->encodeWords($header, $string, $usedLength); |
| 111 |
} |
| 112 |
} elseif (str_contains($phraseStr, '(')) { |
| 113 |
foreach (['\\', '"'] as $char) { |
| 114 |
$phraseStr = str_replace($char, '\\'.$char, $phraseStr); |
| 115 |
} |
| 116 |
$phraseStr = '"'.$phraseStr.'"'; |
| 117 |
} |
| 118 |
|
| 119 |
return $phraseStr; |
| 120 |
} |
| 121 |
|
| 122 |
/** |
| 123 |
* Encode needed word tokens within a string of input. |
| 124 |
*/ |
| 125 |
protected function encodeWords(HeaderInterface $header, string $input, int $usedLength = -1): string |
| 126 |
{ |
| 127 |
$value = ''; |
| 128 |
$tokens = $this->getEncodableWordTokens($input); |
| 129 |
foreach ($tokens as $token) { |
| 130 |
// See RFC 2822, Sect 2.2 (really 2.2 ??) |
| 131 |
if ($this->tokenNeedsEncoding($token)) { |
| 132 |
// Don't encode starting WSP |
| 133 |
$firstChar = substr($token, 0, 1); |
| 134 |
switch ($firstChar) { |
| 135 |
case ' ': |
| 136 |
case "\t": |
| 137 |
$value .= $firstChar; |
| 138 |
$token = substr($token, 1); |
| 139 |
} |
| 140 |
|
| 141 |
if (-1 == $usedLength) { |
| 142 |
$usedLength = \strlen($header->getName().': ') + \strlen($value); |
| 143 |
} |
| 144 |
$value .= $this->getTokenAsEncodedWord($token, $usedLength); |
| 145 |
} else { |
| 146 |
$value .= $token; |
| 147 |
} |
| 148 |
} |
| 149 |
|
| 150 |
return $value; |
| 151 |
} |
| 152 |
|
| 153 |
protected function tokenNeedsEncoding(string $token): bool |
| 154 |
{ |
| 155 |
return (bool) preg_match('~[\x00-\x08\x10-\x19\x7F-\xFF\r\n]~', $token); |
| 156 |
} |
| 157 |
|
| 158 |
/** |
| 159 |
* Splits a string into tokens in blocks of words which can be encoded quickly. |
| 160 |
* |
| 161 |
* @return string[] |
| 162 |
*/ |
| 163 |
protected function getEncodableWordTokens(string $string): array |
| 164 |
{ |
| 165 |
$tokens = []; |
| 166 |
$encodedToken = ''; |
| 167 |
// Split at all whitespace boundaries |
| 168 |
foreach (preg_split('~(?=[\t ])~', $string) as $token) { |
| 169 |
if ($this->tokenNeedsEncoding($token)) { |
| 170 |
$encodedToken .= $token; |
| 171 |
} else { |
| 172 |
if ('' !== $encodedToken) { |
| 173 |
$tokens[] = $encodedToken; |
| 174 |
$encodedToken = ''; |
| 175 |
} |
| 176 |
$tokens[] = $token; |
| 177 |
} |
| 178 |
} |
| 179 |
if ('' !== $encodedToken) { |
| 180 |
$tokens[] = $encodedToken; |
| 181 |
} |
| 182 |
|
| 183 |
foreach ($tokens as $i => $token) { |
| 184 |
// whitespace(s) between 2 encoded tokens |
| 185 |
if ( |
| 186 |
0 < $i |
| 187 |
&& isset($tokens[$i + 1]) |
| 188 |
&& preg_match('~^[\t ]+$~', $token) |
| 189 |
&& $this->tokenNeedsEncoding($tokens[$i - 1]) |
| 190 |
&& $this->tokenNeedsEncoding($tokens[$i + 1]) |
| 191 |
) { |
| 192 |
$tokens[$i - 1] .= $token.$tokens[$i + 1]; |
| 193 |
array_splice($tokens, $i, 2); |
| 194 |
} |
| 195 |
} |
| 196 |
|
| 197 |
return $tokens; |
| 198 |
} |
| 199 |
|
| 200 |
/** |
| 201 |
* Get a token as an encoded word for safe insertion into headers. |
| 202 |
*/ |
| 203 |
protected function getTokenAsEncodedWord(string $token, int $firstLineOffset = 0): string |
| 204 |
{ |
| 205 |
if (null === self::$encoder) { |
| 206 |
self::$encoder = new QpMimeHeaderEncoder(); |
| 207 |
} |
| 208 |
|
| 209 |
// Adjust $firstLineOffset to account for space needed for syntax |
| 210 |
$charsetDecl = $this->charset; |
| 211 |
if (null !== $this->lang) { |
| 212 |
$charsetDecl .= '*'.$this->lang; |
| 213 |
} |
| 214 |
$encodingWrapperLength = \strlen('=?'.$charsetDecl.'?'.self::$encoder->getName().'??='); |
| 215 |
|
| 216 |
if ($firstLineOffset >= 75) { |
| 217 |
// Does this logic need to be here? |
| 218 |
$firstLineOffset = 0; |
| 219 |
} |
| 220 |
|
| 221 |
$encodedTextLines = explode("\r\n", |
| 222 |
self::$encoder->encodeString($token, $this->charset, $firstLineOffset, 75 - $encodingWrapperLength) |
| 223 |
); |
| 224 |
|
| 225 |
if ('iso-2022-jp' !== strtolower($this->charset)) { |
| 226 |
// special encoding for iso-2022-jp using mb_encode_mimeheader |
| 227 |
foreach ($encodedTextLines as $lineNum => $line) { |
| 228 |
$encodedTextLines[$lineNum] = '=?'.$charsetDecl.'?'.self::$encoder->getName().'?'.$line.'?='; |
| 229 |
} |
| 230 |
} |
| 231 |
|
| 232 |
return implode("\r\n ", $encodedTextLines); |
| 233 |
} |
| 234 |
|
| 235 |
/** |
| 236 |
* Generates tokens from the given string which include CRLF as individual tokens. |
| 237 |
* |
| 238 |
* @return string[] |
| 239 |
*/ |
| 240 |
protected function generateTokenLines(string $token): array |
| 241 |
{ |
| 242 |
return preg_split('~(\r\n)~', $token, -1, \PREG_SPLIT_DELIM_CAPTURE); |
| 243 |
} |
| 244 |
|
| 245 |
/** |
| 246 |
* Generate a list of all tokens in the final header. |
| 247 |
*/ |
| 248 |
protected function toTokens(?string $string = null): array |
| 249 |
{ |
| 250 |
if (null === $string) { |
| 251 |
$string = $this->getBodyAsString(); |
| 252 |
} |
| 253 |
|
| 254 |
$tokens = []; |
| 255 |
// Generate atoms; split at all invisible boundaries followed by WSP |
| 256 |
foreach (preg_split('~(?=[ \t])~', $string) as $token) { |
| 257 |
$newTokens = $this->generateTokenLines($token); |
| 258 |
foreach ($newTokens as $newToken) { |
| 259 |
$tokens[] = $newToken; |
| 260 |
} |
| 261 |
} |
| 262 |
|
| 263 |
return $tokens; |
| 264 |
} |
| 265 |
|
| 266 |
/** |
| 267 |
* Takes an array of tokens which appear in the header and turns them into |
| 268 |
* an RFC 2822 compliant string, adding FWSP where needed. |
| 269 |
* |
| 270 |
* @param string[] $tokens |
| 271 |
*/ |
| 272 |
private function tokensToString(array $tokens): string |
| 273 |
{ |
| 274 |
$lineCount = 0; |
| 275 |
$headerLines = []; |
| 276 |
$headerLines[] = $this->name.': '; |
| 277 |
$currentLine = &$headerLines[$lineCount++]; |
| 278 |
|
| 279 |
// Build all tokens back into compliant header |
| 280 |
foreach ($tokens as $i => $token) { |
| 281 |
// Line longer than specified maximum or token was just a new line |
| 282 |
if (("\r\n" === $token) || |
| 283 |
($i > 0 && \strlen($currentLine.$token) > $this->lineLength) |
| 284 |
&& '' !== $currentLine) { |
| 285 |
$headerLines[] = ''; |
| 286 |
$currentLine = &$headerLines[$lineCount++]; |
| 287 |
} |
| 288 |
|
| 289 |
// Append token to the line |
| 290 |
if ("\r\n" !== $token) { |
| 291 |
$currentLine .= $token; |
| 292 |
} |
| 293 |
} |
| 294 |
|
| 295 |
// Implode with FWS (RFC 2822, 2.2.3) |
| 296 |
return implode("\r\n", $headerLines); |
| 297 |
} |
| 298 |
} |
| 299 |
|
