1 | <?php |
---|
2 | /** |
---|
3 | * HTTP request representation specific for the WAF. |
---|
4 | * |
---|
5 | * @package automattic/jetpack-waf |
---|
6 | */ |
---|
7 | |
---|
8 | namespace Automattic\Jetpack\Waf; |
---|
9 | |
---|
10 | require_once __DIR__ . '/functions.php'; |
---|
11 | |
---|
12 | /** |
---|
13 | * Request representation. |
---|
14 | * |
---|
15 | * @template RequestFile as array{ name: string, filename: string } |
---|
16 | */ |
---|
17 | class Waf_Request { |
---|
18 | /** |
---|
19 | * The request URL, broken into three pieces: the host, the filename, and the query string |
---|
20 | * |
---|
21 | * @example for `https://wordpress.com/index.php?myvar=red` |
---|
22 | * $this->url = [ 'https://wordpress.com', '/index.php', '?myvar=red' ] |
---|
23 | * @var array{ 0: string, 1: string, 2: string }|null |
---|
24 | */ |
---|
25 | protected $url = null; |
---|
26 | |
---|
27 | /** |
---|
28 | * Trusted proxies. |
---|
29 | * |
---|
30 | * @var array List of trusted proxy IP addresses. |
---|
31 | */ |
---|
32 | private $trusted_proxies = array(); |
---|
33 | |
---|
34 | /** |
---|
35 | * Trusted headers. |
---|
36 | * |
---|
37 | * @var array List of headers to trust from the trusted proxies. |
---|
38 | */ |
---|
39 | private $trusted_headers = array(); |
---|
40 | |
---|
41 | /** |
---|
42 | * Sets the list of IP addresses for the proxies to trust. Trusted headers will only be accepted as the |
---|
43 | * user IP address from these IP adresses. |
---|
44 | * |
---|
45 | * Popular choices include: |
---|
46 | * - 192.168.0.1 |
---|
47 | * - 10.0.0.1 |
---|
48 | * |
---|
49 | * @param array $proxies List of proxy IP addresses. |
---|
50 | * @return void |
---|
51 | */ |
---|
52 | public function set_trusted_proxies( $proxies ) { |
---|
53 | $this->trusted_proxies = (array) $proxies; |
---|
54 | } |
---|
55 | |
---|
56 | /** |
---|
57 | * Sets the list of headers to be trusted from the proxies. These headers will only be taken into account |
---|
58 | * if the request comes from a trusted proxy as configured with set_trusted_proxies(). |
---|
59 | * |
---|
60 | * Popular choices include: |
---|
61 | * - HTTP_CLIENT_IP |
---|
62 | * - HTTP_X_FORWARDED_FOR |
---|
63 | * - HTTP_X_FORWARDED |
---|
64 | * - HTTP_X_CLUSTER_CLIENT_IP |
---|
65 | * - HTTP_FORWARDED_FOR |
---|
66 | * - HTTP_FORWARDED |
---|
67 | * |
---|
68 | * @param array $headers List of HTTP header strings. |
---|
69 | * @return void |
---|
70 | */ |
---|
71 | public function set_trusted_headers( $headers ) { |
---|
72 | $this->trusted_headers = (array) $headers; |
---|
73 | } |
---|
74 | |
---|
75 | /** |
---|
76 | * Determines the users real IP address based on the settings passed to set_trusted_proxies() and |
---|
77 | * set_trusted_headers() before. On CLI, this will be null. |
---|
78 | * |
---|
79 | * @return string|null |
---|
80 | */ |
---|
81 | public function get_real_user_ip_address() { |
---|
82 | $remote_addr = ! empty( $_SERVER['REMOTE_ADDR'] ) ? wp_unslash( $_SERVER['REMOTE_ADDR'] ) : null; // phpcs:ignore WordPress.Security.ValidatedSanitizedInput.InputNotSanitized |
---|
83 | |
---|
84 | if ( in_array( $remote_addr, $this->trusted_proxies, true ) ) { |
---|
85 | $ip_by_header = $this->get_ip_by_header( array_merge( $this->trusted_headers, array( 'REMOTE_ADDR' ) ) ); |
---|
86 | if ( ! empty( $ip_by_header ) ) { |
---|
87 | return $ip_by_header; |
---|
88 | } |
---|
89 | } |
---|
90 | |
---|
91 | return $remote_addr; |
---|
92 | } |
---|
93 | |
---|
94 | /** |
---|
95 | * Iterates through a given list of HTTP headers and attempts to get the IP address from the header that |
---|
96 | * a proxy sends along. Make sure you trust the IP address before calling this method. |
---|
97 | * |
---|
98 | * @param array $headers The list of headers to check. |
---|
99 | * @return string|null |
---|
100 | */ |
---|
101 | private function get_ip_by_header( $headers ) { |
---|
102 | foreach ( $headers as $key ) { |
---|
103 | if ( isset( $_SERVER[ $key ] ) ) { |
---|
104 | foreach ( explode( ',', wp_unslash( $_SERVER[ $key ] ) ) as $ip ) { // phpcs:ignore WordPress.Security.ValidatedSanitizedInput.InputNotSanitized -- filter_var is applied below. |
---|
105 | $ip = trim( $ip ); |
---|
106 | |
---|
107 | if ( filter_var( $ip, FILTER_VALIDATE_IP ) !== false ) { |
---|
108 | return $ip; |
---|
109 | } |
---|
110 | } |
---|
111 | } |
---|
112 | } |
---|
113 | |
---|
114 | return null; |
---|
115 | } |
---|
116 | |
---|
117 | /** |
---|
118 | * Returns the headers that were sent with this request |
---|
119 | * |
---|
120 | * @return array{ 0: string, 1: scalar }[] |
---|
121 | */ |
---|
122 | public function get_headers() { |
---|
123 | $value = array(); |
---|
124 | $has_content_type = false; |
---|
125 | $has_content_length = false; |
---|
126 | foreach ( $_SERVER as $k => $v ) { |
---|
127 | $k = strtolower( $k ); |
---|
128 | if ( 'http_' === substr( $k, 0, 5 ) ) { |
---|
129 | $value[] = array( $this->normalize_header_name( substr( $k, 5 ) ), $v ); |
---|
130 | } elseif ( 'content_type' === $k && '' !== $v ) { |
---|
131 | $has_content_type = true; |
---|
132 | $value[] = array( 'content-type', $v ); |
---|
133 | } elseif ( 'content_length' === $k && '' !== $v ) { |
---|
134 | $has_content_length = true; |
---|
135 | $value[] = array( 'content-length', $v ); |
---|
136 | } |
---|
137 | } |
---|
138 | if ( ! $has_content_type ) { |
---|
139 | // default Content-Type per RFC 7231 section 3.1.5.5. |
---|
140 | $value[] = array( 'content-type', 'application/octet-stream' ); |
---|
141 | } |
---|
142 | if ( ! $has_content_length ) { |
---|
143 | $value[] = array( 'content-length', '0' ); |
---|
144 | } |
---|
145 | |
---|
146 | return $value; |
---|
147 | } |
---|
148 | |
---|
149 | /** |
---|
150 | * Returns the value of a specific header that was sent with this request |
---|
151 | * |
---|
152 | * @param string $name The name of the header to retrieve. |
---|
153 | * @return string |
---|
154 | */ |
---|
155 | public function get_header( $name ) { |
---|
156 | $name = $this->normalize_header_name( $name ); |
---|
157 | foreach ( $this->get_headers() as list( $header_name, $header_value ) ) { |
---|
158 | if ( $header_name === $name ) { |
---|
159 | return $header_value; |
---|
160 | } |
---|
161 | } |
---|
162 | return ''; |
---|
163 | } |
---|
164 | |
---|
165 | /** |
---|
166 | * Change a header name to all-lowercase and replace spaces and underscores with dashes. |
---|
167 | * |
---|
168 | * @param string $name The header name to normalize. |
---|
169 | * @return string |
---|
170 | */ |
---|
171 | public function normalize_header_name( $name ) { |
---|
172 | return str_replace( array( ' ', '_' ), '-', strtolower( $name ) ); |
---|
173 | } |
---|
174 | |
---|
175 | /** |
---|
176 | * Get the method for this request (GET, POST, etc). |
---|
177 | * |
---|
178 | * @return string |
---|
179 | */ |
---|
180 | public function get_method() { |
---|
181 | return isset( $_SERVER['REQUEST_METHOD'] ) |
---|
182 | ? filter_var( wp_unslash( $_SERVER['REQUEST_METHOD'] ), FILTER_DEFAULT ) |
---|
183 | : ''; |
---|
184 | } |
---|
185 | |
---|
186 | /** |
---|
187 | * Get the protocol for this request (HTTP, HTTPS, etc) |
---|
188 | * |
---|
189 | * @return string |
---|
190 | */ |
---|
191 | public function get_protocol() { |
---|
192 | return isset( $_SERVER['SERVER_PROTOCOL'] ) |
---|
193 | ? filter_var( wp_unslash( $_SERVER['SERVER_PROTOCOL'] ), FILTER_DEFAULT ) |
---|
194 | : ''; |
---|
195 | } |
---|
196 | |
---|
197 | /** |
---|
198 | * Returns the URL parts for this request. |
---|
199 | * |
---|
200 | * @see $this->url |
---|
201 | * @return array{ 0: string, 1: string, 2: string } |
---|
202 | */ |
---|
203 | protected function get_url() { |
---|
204 | if ( null !== $this->url ) { |
---|
205 | return $this->url; |
---|
206 | } |
---|
207 | |
---|
208 | $uri = isset( $_SERVER['REQUEST_URI'] ) ? filter_var( wp_unslash( $_SERVER['REQUEST_URI'] ), FILTER_DEFAULT ) : '/'; |
---|
209 | if ( false !== strpos( $uri, '?' ) ) { |
---|
210 | // remove the query string (we'll pull it from elsewhere later) |
---|
211 | $uri = urldecode( substr( $uri, 0, strpos( $uri, '?' ) ) ); |
---|
212 | } else { |
---|
213 | $uri = urldecode( $uri ); |
---|
214 | } |
---|
215 | $query_string = isset( $_SERVER['QUERY_STRING'] ) ? '?' . filter_var( wp_unslash( $_SERVER['QUERY_STRING'] ), FILTER_DEFAULT ) : ''; |
---|
216 | if ( 1 === preg_match( '/^https?:\/\//', $uri ) ) { |
---|
217 | // sometimes $_SERVER[REQUEST_URI] already includes the full domain name |
---|
218 | $uri_host = substr( $uri, 0, strpos( $uri, '/', 8 ) ); |
---|
219 | $uri_path = substr( $uri, strlen( $uri_host ) ); |
---|
220 | $this->url = array( $uri_host, $uri_path, $query_string ); |
---|
221 | } else { |
---|
222 | // otherwise build the URI manually |
---|
223 | $uri_scheme = ( ! empty( $_SERVER['HTTPS'] ) && 'off' !== $_SERVER['HTTPS'] ) |
---|
224 | ? 'https' |
---|
225 | : 'http'; |
---|
226 | $uri_host = isset( $_SERVER['HTTP_HOST'] ) |
---|
227 | ? filter_var( wp_unslash( $_SERVER['HTTP_HOST'] ), FILTER_DEFAULT ) |
---|
228 | : ( |
---|
229 | isset( $_SERVER['SERVER_NAME'] ) |
---|
230 | ? filter_var( wp_unslash( $_SERVER['SERVER_NAME'] ), FILTER_DEFAULT ) |
---|
231 | : '' |
---|
232 | ); |
---|
233 | $uri_port = isset( $_SERVER['SERVER_PORT'] ) |
---|
234 | ? filter_var( wp_unslash( $_SERVER['SERVER_PORT'] ), FILTER_SANITIZE_NUMBER_INT ) |
---|
235 | : ''; |
---|
236 | // we only need to include the port if it's non-standard |
---|
237 | if ( $uri_port && ( 'http' === $uri_scheme && '80' !== $uri_port || 'https' === $uri_scheme && '443' !== $uri_port ) ) { |
---|
238 | $uri_port = ':' . $uri_port; |
---|
239 | } else { |
---|
240 | $uri_port = ''; |
---|
241 | } |
---|
242 | $this->url = array( |
---|
243 | $uri_scheme . '://' . $uri_host . $uri_port, |
---|
244 | $uri, |
---|
245 | $query_string, |
---|
246 | ); |
---|
247 | } |
---|
248 | return $this->url; |
---|
249 | } |
---|
250 | |
---|
251 | /** |
---|
252 | * Get the requested URI |
---|
253 | * |
---|
254 | * @param boolean $include_host If true, the scheme and domain will be included in the returned string (i.e. 'https://wordpress.com/index.php). |
---|
255 | * If false, only the requested URI path will be returned (i.e. '/index.php'). |
---|
256 | * @return string |
---|
257 | */ |
---|
258 | public function get_uri( $include_host = false ) { |
---|
259 | list( $host, $file, $query ) = $this->get_url(); |
---|
260 | |
---|
261 | return ( $include_host ? $host : '' ) . $file . $query; |
---|
262 | } |
---|
263 | |
---|
264 | /** |
---|
265 | * Return the filename part of the request |
---|
266 | * |
---|
267 | * @example for 'https://wordpress.com/some/page?id=5', return '/some/page' |
---|
268 | * @return string |
---|
269 | */ |
---|
270 | public function get_filename() { |
---|
271 | return $this->get_url()[1]; |
---|
272 | } |
---|
273 | |
---|
274 | /** |
---|
275 | * Return the basename part of the request |
---|
276 | * |
---|
277 | * @example for 'https://wordpress.com/some/page.php?id=5', return 'page.php' |
---|
278 | * @return string |
---|
279 | */ |
---|
280 | public function get_basename() { |
---|
281 | // Get the filename part of the request |
---|
282 | $filename = $this->get_filename(); |
---|
283 | // Normalize slashes |
---|
284 | $filename = str_replace( '\\', '/', $filename ); |
---|
285 | // Remove trailing slashes |
---|
286 | $filename = rtrim( $filename, '/' ); |
---|
287 | // Return the basename |
---|
288 | $offset = strrpos( $filename, '/' ); |
---|
289 | return $offset !== false ? substr( $filename, $offset + 1 ) : $filename; |
---|
290 | } |
---|
291 | |
---|
292 | /** |
---|
293 | * Return the query string. If present, it will be prefixed with '?'. Otherwise, it will be an empty string. |
---|
294 | * |
---|
295 | * @return string |
---|
296 | */ |
---|
297 | public function get_query_string() { |
---|
298 | return $this->get_url()[2]; |
---|
299 | } |
---|
300 | |
---|
301 | /** |
---|
302 | * Returns the request body. |
---|
303 | * |
---|
304 | * @return string |
---|
305 | */ |
---|
306 | public function get_body() { |
---|
307 | $body = file_get_contents( 'php://input' ); |
---|
308 | return false === $body ? '' : $body; |
---|
309 | } |
---|
310 | |
---|
311 | /** |
---|
312 | * Returns the cookies |
---|
313 | * |
---|
314 | * @return array<string, string> |
---|
315 | */ |
---|
316 | public function get_cookies() { |
---|
317 | return flatten_array( $_COOKIE ); |
---|
318 | } |
---|
319 | |
---|
320 | /** |
---|
321 | * Returns the GET variables |
---|
322 | * |
---|
323 | * @return array<string, mixed|array> |
---|
324 | */ |
---|
325 | public function get_get_vars() { |
---|
326 | return flatten_array( $_GET ); |
---|
327 | } |
---|
328 | |
---|
329 | /** |
---|
330 | * Returns the POST variables |
---|
331 | * |
---|
332 | * @return array<string, mixed|array> |
---|
333 | */ |
---|
334 | public function get_post_vars() { |
---|
335 | // Attempt to decode JSON requests. |
---|
336 | if ( strpos( $this->get_header( 'content-type' ), 'application/json' ) !== false ) { |
---|
337 | $decoded_json = json_decode( $this->get_body(), true ) ?? array(); |
---|
338 | return flatten_array( $decoded_json, 'json', true ); |
---|
339 | } |
---|
340 | |
---|
341 | return flatten_array( $_POST ); |
---|
342 | } |
---|
343 | |
---|
344 | /** |
---|
345 | * Returns the files that were uploaded with this request (i.e. what's in the $_FILES superglobal) |
---|
346 | * |
---|
347 | * @return RequestFile[] |
---|
348 | */ |
---|
349 | public function get_files() { |
---|
350 | $files = array(); |
---|
351 | foreach ( $_FILES as $field_name => $arr ) { |
---|
352 | // flatten the values in case we were given inputs with brackets |
---|
353 | foreach ( flatten_array( $arr ) as list( $arr_key, $arr_value ) ) { |
---|
354 | if ( $arr_key === 'name' ) { |
---|
355 | // if this file was a simple (non-nested) name and unique, then just add it. |
---|
356 | $files[] = array( |
---|
357 | 'name' => $field_name, |
---|
358 | 'filename' => $arr_value, |
---|
359 | ); |
---|
360 | } elseif ( 'name[' === substr( $arr_key, 0, 5 ) ) { |
---|
361 | // otherwise this was a file with a nested name and/or multiple files with the same name |
---|
362 | $files[] = array( |
---|
363 | 'name' => $field_name . substr( $arr_key, 4 ), |
---|
364 | 'filename' => $arr_value, |
---|
365 | ); |
---|
366 | } |
---|
367 | } |
---|
368 | } |
---|
369 | return $files; |
---|
370 | } |
---|
371 | } |
---|