ricardosierra /
phpspider
| Total Complexity | 383 |
| Total Lines | 2762 |
| Duplicated Lines | 0 % |
| Changes | 0 | ||
Complex classes like phpspider often do a lot of different things. To break such a class down, we need to identify a cohesive component within that class. A common approach to find such a component is to look for fields/methods that share the same prefixes, or suffixes.
Once you have determined the fields that belong together, you can apply the Extract Class refactoring. If the component makes sense as a sub-class, Extract Subclass is also a candidate, and is often faster.
While breaking up the class, it is a good idea to analyze how other classes use phpspider, and based on these observations, apply Extract Interface, too.
| 1 | <?php |
||
| 44 | class phpspider |
||
| 45 | { |
||
| 46 | /** |
||
| 47 | * 版本号 |
||
| 48 | * @var string |
||
| 49 | */ |
||
| 50 | const VERSION = '3.0.4'; |
||
| 51 | |||
| 52 | /** |
||
| 53 | * 爬虫爬取每个网页的时间间隔,0表示不延时, 单位: 毫秒 |
||
| 54 | */ |
||
| 55 | const INTERVAL = 0; |
||
| 56 | |||
| 57 | /** |
||
| 58 | * 爬虫爬取每个网页的超时时间, 单位: 秒 |
||
| 59 | */ |
||
| 60 | const TIMEOUT = 5; |
||
| 61 | |||
| 62 | /** |
||
| 63 | * 爬取失败次数, 不想失败重新爬取则设置为0 |
||
| 64 | */ |
||
| 65 | const MAX_TRY = 0; |
||
| 66 | |||
| 67 | /** |
||
| 68 | * 爬虫爬取网页所使用的浏览器类型: pc、ios、android |
||
| 69 | * 默认类型是PC |
||
| 70 | */ |
||
| 71 | const AGENT_PC = "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_11_6) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/52.0.2743.116 Safari/537.36"; |
||
| 72 | const AGENT_IOS = "Mozilla/5.0 (iPhone; CPU iPhone OS 9_3_3 like Mac OS X) AppleWebKit/601.1.46 (KHTML, like Gecko) Version/9.0 Mobile/13G34 Safari/601.1"; |
||
| 73 | const AGENT_ANDROID = "Mozilla/5.0 (Linux; U; Android 6.0.1;zh_cn; Le X820 Build/FEXCNFN5801507014S) AppleWebKit/537.36 (KHTML, like Gecko)Version/4.0 Chrome/49.0.0.0 Mobile Safari/537.36 EUI Browser/5.8.015S"; |
||
| 74 | |||
| 75 | /** |
||
| 76 | * pid文件的路径及名称 |
||
| 77 | * @var string |
||
| 78 | */ |
||
| 79 | //public static $pid_file = ''; |
||
| 80 | |||
| 81 | /** |
||
| 82 | * 日志目录, 默认在data根目录下 |
||
| 83 | * @var mixed |
||
| 84 | */ |
||
| 85 | //public static $log_file = ''; |
||
| 86 | |||
| 87 | /** |
||
| 88 | * 主任务进程ID |
||
| 89 | */ |
||
| 90 | //public static $master_pid = 0; |
||
| 91 | |||
| 92 | /** |
||
| 93 | * 所有任务进程ID |
||
| 94 | */ |
||
| 95 | //public static $taskpids = array(); |
||
| 96 | |||
| 97 | /** |
||
| 98 | * Daemonize. |
||
| 99 | * |
||
| 100 | * @var bool |
||
| 101 | */ |
||
| 102 | public static $daemonize = false; |
||
| 103 | |||
| 104 | /** |
||
| 105 | * 当前进程是否终止 |
||
| 106 | */ |
||
| 107 | public static $terminate = false; |
||
| 108 | |||
| 109 | /** |
||
| 110 | * 是否分布式 |
||
| 111 | */ |
||
| 112 | public static $multiserver = false; |
||
| 113 | |||
| 114 | /** |
||
| 115 | * 当前服务器ID |
||
| 116 | */ |
||
| 117 | public static $serverid = 1; |
||
| 118 | |||
| 119 | /** |
||
| 120 | * 主任务进程 |
||
| 121 | */ |
||
| 122 | public static $taskmaster = true; |
||
| 123 | |||
| 124 | /** |
||
| 125 | * 当前任务ID |
||
| 126 | */ |
||
| 127 | public static $taskid = 1; |
||
| 128 | |||
| 129 | /** |
||
| 130 | * 当前任务进程ID |
||
| 131 | */ |
||
| 132 | public static $taskpid = 1; |
||
| 133 | |||
| 134 | /** |
||
| 135 | * 并发任务数 |
||
| 136 | */ |
||
| 137 | public static $tasknum = 1; |
||
| 138 | |||
| 139 | /** |
||
| 140 | * 生成 |
||
| 141 | */ |
||
| 142 | public static $fork_task_complete = false; |
||
| 143 | |||
| 144 | /** |
||
| 145 | * 是否使用Redis |
||
| 146 | */ |
||
| 147 | public static $use_redis = false; |
||
| 148 | |||
| 149 | /** |
||
| 150 | * 是否保存爬虫运行状态 |
||
| 151 | */ |
||
| 152 | public static $save_running_state = false; |
||
| 153 | |||
| 154 | /** |
||
| 155 | * 配置 |
||
| 156 | */ |
||
| 157 | public static $configs = array(); |
||
| 158 | |||
| 159 | /** |
||
| 160 | * 要抓取的URL队列 |
||
| 161 | md5(url) => array( |
||
| 162 | 'url' => '', // 要爬取的URL |
||
| 163 | 'url_type' => '', // 要爬取的URL类型,scan_page、list_page、content_page |
||
| 164 | 'method' => 'get', // 默认为"GET"请求, 也支持"POST"请求 |
||
| 165 | 'headers' => array(), // 此url的Headers, 可以为空 |
||
| 166 | 'params' => array(), // 发送请求时需添加的参数, 可以为空 |
||
| 167 | 'context_data'=> '', // 此url附加的数据, 可以为空 |
||
| 168 | 'proxies' => false, // 是否使用代理 |
||
| 169 | 'try_num' => 0 // 抓取次数 |
||
| 170 | 'max_try' => 0 // 允许抓取失败次数 |
||
| 171 | ) |
||
| 172 | */ |
||
| 173 | public static $collect_queue = array(); |
||
| 174 | |||
| 175 | /** |
||
| 176 | * 要抓取的URL数组 |
||
| 177 | * md5($url) => time() |
||
| 178 | */ |
||
| 179 | public static $collect_urls = array(); |
||
| 180 | |||
| 181 | /** |
||
| 182 | * 要抓取的URL数量 |
||
| 183 | */ |
||
| 184 | public static $collect_urls_num = 0; |
||
| 185 | |||
| 186 | /** |
||
| 187 | * 已经抓取的URL数量 |
||
| 188 | */ |
||
| 189 | public static $collected_urls_num = 0; |
||
| 190 | |||
| 191 | /** |
||
| 192 | * 当前进程采集成功数 |
||
| 193 | */ |
||
| 194 | public static $collect_succ = 0; |
||
| 195 | |||
| 196 | /** |
||
| 197 | * 当前进程采集失败数 |
||
| 198 | */ |
||
| 199 | public static $collect_fail = 0; |
||
| 200 | |||
| 201 | /** |
||
| 202 | * 提取到的字段数 |
||
| 203 | */ |
||
| 204 | public static $fields_num = 0; |
||
| 205 | |||
| 206 | /** |
||
| 207 | * 采集深度 |
||
| 208 | */ |
||
| 209 | public static $depth_num = 0; |
||
| 210 | |||
| 211 | /** |
||
| 212 | * 爬虫开始时间 |
||
| 213 | */ |
||
| 214 | public static $time_start = 0; |
||
| 215 | |||
| 216 | /** |
||
| 217 | * 任务状态 |
||
| 218 | */ |
||
| 219 | public static $task_status = array(); |
||
| 220 | |||
| 221 | // 导出类型配置 |
||
| 222 | public static $export_type = ''; |
||
| 223 | public static $export_file = ''; |
||
| 224 | public static $export_conf = ''; |
||
| 225 | public static $export_table = ''; |
||
| 226 | |||
| 227 | // 数据库配置 |
||
| 228 | public static $db_config = array(); |
||
| 229 | // 队列配置 |
||
| 230 | public static $queue_config = array(); |
||
| 231 | |||
| 232 | // 运行面板参数长度 |
||
| 233 | public static $server_length = 10; |
||
| 234 | public static $tasknum_length = 8; |
||
| 235 | public static $taskid_length = 8; |
||
| 236 | public static $pid_length = 8; |
||
| 237 | public static $mem_length = 8; |
||
| 238 | public static $urls_length = 15; |
||
| 239 | public static $speed_length = 6; |
||
| 240 | |||
| 241 | /** |
||
| 242 | * 爬虫初始化时调用, 用来指定一些爬取前的操作 |
||
| 243 | * |
||
| 244 | * @var mixed |
||
| 245 | * @access public |
||
| 246 | */ |
||
| 247 | public $on_start = null; |
||
| 248 | |||
| 249 | /** |
||
| 250 | * 网页状态码回调 |
||
| 251 | * |
||
| 252 | * @var mixed |
||
| 253 | * @access public |
||
| 254 | */ |
||
| 255 | public $on_status_code = null; |
||
| 256 | |||
| 257 | /** |
||
| 258 | * 判断当前网页是否被反爬虫, 需要开发者实现 |
||
| 259 | * |
||
| 260 | * @var mixed |
||
| 261 | * @access public |
||
| 262 | */ |
||
| 263 | public $is_anti_spider = null; |
||
| 264 | |||
| 265 | /** |
||
| 266 | * 在一个网页下载完成之后调用, 主要用来对下载的网页进行处理 |
||
| 267 | * |
||
| 268 | * @var mixed |
||
| 269 | * @access public |
||
| 270 | */ |
||
| 271 | public $on_download_page = null; |
||
| 272 | |||
| 273 | /** |
||
| 274 | * 在一个attached_url对应的网页下载完成之后调用. 主要用来对下载的网页进行处理 |
||
| 275 | * |
||
| 276 | * @var mixed |
||
| 277 | * @access public |
||
| 278 | */ |
||
| 279 | public $on_download_attached_page = null; |
||
| 280 | |||
| 281 | /** |
||
| 282 | * 当前页面抽取到URL |
||
| 283 | * |
||
| 284 | * @var mixed |
||
| 285 | * @access public |
||
| 286 | */ |
||
| 287 | public $on_fetch_url = null; |
||
| 288 | |||
| 289 | /** |
||
| 290 | * URL属于入口页 |
||
| 291 | * 在爬取到入口url的内容之后, 添加新的url到待爬队列之前调用 |
||
| 292 | * 主要用来发现新的待爬url, 并且能给新发现的url附加数据 |
||
| 293 | * |
||
| 294 | * @var mixed |
||
| 295 | * @access public |
||
| 296 | */ |
||
| 297 | public $on_scan_page = null; |
||
| 298 | |||
| 299 | /** |
||
| 300 | * URL属于列表页 |
||
| 301 | * 在爬取到列表页url的内容之后, 添加新的url到待爬队列之前调用 |
||
| 302 | * 主要用来发现新的待爬url, 并且能给新发现的url附加数据 |
||
| 303 | * |
||
| 304 | * @var mixed |
||
| 305 | * @access public |
||
| 306 | */ |
||
| 307 | public $on_list_page = null; |
||
| 308 | |||
| 309 | /** |
||
| 310 | * URL属于内容页 |
||
| 311 | * 在爬取到内容页url的内容之后, 添加新的url到待爬队列之前调用 |
||
| 312 | * 主要用来发现新的待爬url, 并且能给新发现的url附加数据 |
||
| 313 | * |
||
| 314 | * @var mixed |
||
| 315 | * @access public |
||
| 316 | */ |
||
| 317 | public $on_content_page = null; |
||
| 318 | |||
| 319 | /** |
||
| 320 | * 在抽取到field内容之后调用, 对其中包含的img标签进行回调处理 |
||
| 321 | * |
||
| 322 | * @var mixed |
||
| 323 | * @access public |
||
| 324 | */ |
||
| 325 | public $on_handle_img = null; |
||
| 326 | |||
| 327 | /** |
||
| 328 | * 当一个field的内容被抽取到后进行的回调, 在此回调中可以对网页中抽取的内容作进一步处理 |
||
| 329 | * |
||
| 330 | * @var mixed |
||
| 331 | * @access public |
||
| 332 | */ |
||
| 333 | public $on_extract_field = null; |
||
| 334 | |||
| 335 | /** |
||
| 336 | * 在一个网页的所有field抽取完成之后, 可能需要对field进一步处理, 以发布到自己的网站 |
||
| 337 | * |
||
| 338 | * @var mixed |
||
| 339 | * @access public |
||
| 340 | */ |
||
| 341 | public $on_extract_page = null; |
||
| 342 | |||
| 343 | /** |
||
| 344 | * 如果抓取的页面是一个附件文件, 比如图片、视频、二进制文件、apk、ipad、exe |
||
| 345 | * 就不去分析他的内容提取field了, 提取field只针对HTML |
||
| 346 | * |
||
| 347 | * @var mixed |
||
| 348 | * @access public |
||
| 349 | */ |
||
| 350 | public $on_attachment_file = null; |
||
| 351 | |||
| 352 | function __construct($configs = array()) |
||
| 353 | { |
||
| 354 | // 产生时钟云，解决php7下面ctrl+c无法停止bug |
||
| 355 | declare(ticks = 1); |
||
| 356 | |||
| 357 | // 先打开以显示验证报错内容 |
||
| 358 | log::$log_show = true; |
||
| 359 | log::$log_file = isset($configs['log_file']) ? $configs['log_file'] : PATH_DATA.'/phpspider.log'; |
||
| 360 | log::$log_type = isset($configs['log_type']) ? $configs['log_type'] : false; |
||
| 361 | |||
| 362 | // 彩蛋 |
||
| 363 | $included_files = get_included_files(); |
||
| 364 | $content = file_get_contents($included_files[0]); |
||
| 365 | if (!preg_match("#/\* Do NOT delete this comment \*/#", $content) || !preg_match("#/\* 不要删除这段注释 \*/#", $content)) |
||
| 366 | { |
||
| 367 | $msg = "Unknown error..."; |
||
| 368 | log::error($msg); |
||
| 369 | exit; |
||
|
|
|||
| 370 | } |
||
| 371 | |||
| 372 | $configs['name'] = isset($configs['name']) ? $configs['name'] : 'phpspider'; |
||
| 373 | $configs['proxies'] = isset($configs['proxies']) ? $configs['proxies'] : ''; |
||
| 374 | $configs['user_agent'] = isset($configs['user_agent']) ? $configs['user_agent'] : self::AGENT_PC; |
||
| 375 | $configs['user_agents'] = isset($configs['user_agents']) ? $configs['user_agents'] : null; |
||
| 376 | $configs['client_ip'] = isset($configs['client_ip']) ? $configs['client_ip'] : null; |
||
| 377 | $configs['client_ips'] = isset($configs['client_ips']) ? $configs['client_ips'] : null; |
||
| 378 | $configs['interval'] = isset($configs['interval']) ? $configs['interval'] : self::INTERVAL; |
||
| 379 | $configs['timeout'] = isset($configs['timeout']) ? $configs['timeout'] : self::TIMEOUT; |
||
| 380 | $configs['max_try'] = isset($configs['max_try']) ? $configs['max_try'] : self::MAX_TRY; |
||
| 381 | $configs['max_depth'] = isset($configs['max_depth']) ? $configs['max_depth'] : 0; |
||
| 382 | $configs['max_fields'] = isset($configs['max_fields']) ? $configs['max_fields'] : 0; |
||
| 383 | $configs['export'] = isset($configs['export']) ? $configs['export'] : array(); |
||
| 384 | |||
| 385 | // csv、sql、db |
||
| 386 | self::$export_type = isset($configs['export']['type']) ? $configs['export']['type'] : ''; |
||
| 387 | self::$export_file = isset($configs['export']['file']) ? $configs['export']['file'] : ''; |
||
| 388 | self::$export_table = isset($configs['export']['table']) ? $configs['export']['table'] : ''; |
||
| 389 | self::$db_config = isset($configs['db_config']) ? $configs['db_config'] : array(); |
||
| 390 | self::$queue_config = isset($configs['queue_config']) ? $configs['queue_config'] : array(); |
||
| 391 | |||
| 392 | // 是否设置了并发任务数, 并且大于1, 而且不是windows环境 |
||
| 393 | if (isset($configs['tasknum']) && $configs['tasknum'] > 1 && !util::is_win()) |
||
| 394 | { |
||
| 395 | self::$tasknum = $configs['tasknum']; |
||
| 396 | } |
||
| 397 | |||
| 398 | // 是否设置了保留运行状态 |
||
| 399 | if (isset($configs['save_running_state'])) |
||
| 400 | { |
||
| 401 | self::$save_running_state = $configs['save_running_state']; |
||
| 402 | } |
||
| 403 | |||
| 404 | // 是否分布式 |
||
| 405 | if (isset($configs['multiserver'])) |
||
| 406 | { |
||
| 407 | self::$multiserver = $configs['multiserver']; |
||
| 408 | } |
||
| 409 | |||
| 410 | // 当前服务器ID |
||
| 411 | if (isset($configs['serverid'])) |
||
| 412 | { |
||
| 413 | self::$serverid = $configs['serverid']; |
||
| 414 | } |
||
| 415 | |||
| 416 | // 不同项目的采集以采集名称作为前缀区分 |
||
| 417 | if (isset($GLOBALS['config']['redis']['prefix'])) |
||
| 418 | { |
||
| 419 | $GLOBALS['config']['redis']['prefix'] = $GLOBALS['config']['redis']['prefix'].'-'.md5($configs['name']); |
||
| 420 | } |
||
| 421 | |||
| 422 | self::$configs = $configs; |
||
| 423 | } |
||
| 424 | |||
| 425 | public function add_scan_url($url, $options = array(), $allowed_repeat = true) |
||
| 426 | { |
||
| 427 | // 投递状态 |
||
| 428 | $status = false; |
||
| 429 | |||
| 430 | $link = $options; |
||
| 431 | $link['url'] = $url; |
||
| 432 | $link['url_type'] = 'scan_page'; |
||
| 433 | $link = $this->link_uncompress($link); |
||
| 434 | |||
| 435 | if ($this->is_list_page($url)) |
||
| 436 | { |
||
| 437 | $link['url_type'] = 'list_page'; |
||
| 438 | $status = $this->queue_lpush($link, $allowed_repeat); |
||
| 439 | } |
||
| 440 | elseif ($this->is_content_page($url)) |
||
| 441 | { |
||
| 442 | $link['url_type'] = 'content_page'; |
||
| 443 | $status = $this->queue_lpush($link, $allowed_repeat); |
||
| 444 | } |
||
| 445 | else |
||
| 446 | { |
||
| 447 | $status = $this->queue_lpush($link, $allowed_repeat); |
||
| 448 | } |
||
| 449 | |||
| 450 | if ($status) |
||
| 451 | { |
||
| 452 | if ($link['url_type'] == 'scan_page') |
||
| 453 | { |
||
| 454 | log::debug("Find scan page: {$url}"); |
||
| 455 | } |
||
| 456 | elseif ($link['url_type'] == 'list_page') |
||
| 457 | { |
||
| 458 | log::debug("Find list page: {$url}"); |
||
| 459 | } |
||
| 460 | elseif ($link['url_type'] == 'content_page') |
||
| 461 | { |
||
| 462 | log::debug("Find content page: {$url}"); |
||
| 463 | } |
||
| 464 | } |
||
| 465 | |||
| 466 | return $status; |
||
| 467 | } |
||
| 468 | |||
| 469 | /** |
||
| 470 | * 一般在 on_scan_page 和 on_list_page 回调函数中调用, 用来往待爬队列中添加url |
||
| 471 | * 两个进程同时调用这个方法, 传递相同url的时候, 就会出现url重复进入队列 |
||
| 472 | * |
||
| 473 | * @param mixed $url |
||
| 474 | * @param mixed $options |
||
| 475 | * @return void |
||
| 476 | * @author seatle <[email protected]> |
||
| 477 | * @created time :2016-09-18 10:17 |
||
| 478 | */ |
||
| 479 | public function add_url($url, $options = array(), $depth = 0) |
||
| 480 | { |
||
| 481 | // 投递状态 |
||
| 482 | $status = false; |
||
| 483 | |||
| 484 | $link = $options; |
||
| 485 | $link['url'] = $url; |
||
| 486 | $link['depth'] = $depth; |
||
| 487 | $link = $this->link_uncompress($link); |
||
| 488 | |||
| 489 | if ($this->is_list_page($url)) |
||
| 490 | { |
||
| 491 | $link['url_type'] = 'list_page'; |
||
| 492 | $status = $this->queue_lpush($link); |
||
| 493 | } |
||
| 494 | |||
| 495 | if ($this->is_content_page($url)) |
||
| 496 | { |
||
| 497 | $link['url_type'] = 'content_page'; |
||
| 498 | $status = $this->queue_lpush($link); |
||
| 499 | } |
||
| 500 | |||
| 501 | if ($status) |
||
| 502 | { |
||
| 503 | if ($link['url_type'] == 'scan_page') |
||
| 504 | { |
||
| 505 | log::debug("Find scan page: {$url}"); |
||
| 506 | } |
||
| 507 | elseif ($link['url_type'] == 'list_page') |
||
| 508 | { |
||
| 509 | log::debug("Find list page: {$url}"); |
||
| 510 | } |
||
| 511 | elseif ($link['url_type'] == 'content_page') |
||
| 512 | { |
||
| 513 | log::debug("Find content page: {$url}"); |
||
| 514 | } |
||
| 515 | } |
||
| 516 | |||
| 517 | return $status; |
||
| 518 | } |
||
| 519 | |||
| 520 | /** |
||
| 521 | * 是否入口页面 |
||
| 522 | * |
||
| 523 | * @param mixed $url |
||
| 524 | * @return void |
||
| 525 | * @author seatle <[email protected]> |
||
| 526 | * @created time :2016-10-12 19:06 |
||
| 527 | */ |
||
| 528 | public function is_scan_page($url) |
||
| 529 | { |
||
| 530 | $parse_url = parse_url($url); |
||
| 531 | if (empty($parse_url['host']) || !in_array($parse_url['host'], self::$configs['domains'])) |
||
| 532 | { |
||
| 533 | return false; |
||
| 534 | } |
||
| 535 | return true; |
||
| 536 | } |
||
| 537 | |||
| 538 | /** |
||
| 539 | * 是否列表页面 |
||
| 540 | * |
||
| 541 | * @param mixed $url |
||
| 542 | * @return void |
||
| 543 | * @author seatle <[email protected]> |
||
| 544 | * @created time :2016-10-12 19:06 |
||
| 545 | */ |
||
| 546 | public function is_list_page($url) |
||
| 547 | { |
||
| 548 | $result = false; |
||
| 549 | if (!empty(self::$configs['list_url_regexes'])) |
||
| 550 | { |
||
| 551 | foreach (self::$configs['list_url_regexes'] as $regex) |
||
| 552 | { |
||
| 553 | if (preg_match("#{$regex}#i", $url)) |
||
| 554 | { |
||
| 555 | $result = true; |
||
| 556 | break; |
||
| 557 | } |
||
| 558 | } |
||
| 559 | } |
||
| 560 | return $result; |
||
| 561 | } |
||
| 562 | |||
| 563 | /** |
||
| 564 | * 是否内容页面 |
||
| 565 | * |
||
| 566 | * @param mixed $url |
||
| 567 | * @return void |
||
| 568 | * @author seatle <[email protected]> |
||
| 569 | * @created time :2016-10-12 19:06 |
||
| 570 | */ |
||
| 571 | public function is_content_page($url) |
||
| 572 | { |
||
| 573 | $result = false; |
||
| 574 | if (!empty(self::$configs['content_url_regexes'])) |
||
| 575 | { |
||
| 576 | foreach (self::$configs['content_url_regexes'] as $regex) |
||
| 577 | { |
||
| 578 | if (preg_match("#{$regex}#i", $url)) |
||
| 579 | { |
||
| 580 | $result = true; |
||
| 581 | break; |
||
| 582 | } |
||
| 583 | } |
||
| 584 | } |
||
| 585 | return $result; |
||
| 586 | } |
||
| 587 | |||
| 588 | /** |
||
| 589 | * Parse command. |
||
| 590 | * php yourfile.php start | stop | status | kill |
||
| 591 | * |
||
| 592 | * @return void |
||
| 593 | */ |
||
| 594 | public function parse_command() |
||
| 595 | { |
||
| 596 | // 检查运行命令的参数 |
||
| 597 | global $argv; |
||
| 598 | $start_file = $argv[0]; |
||
| 599 | |||
| 600 | // 命令 |
||
| 601 | $command = isset($argv[1]) ? trim($argv[1]) : 'start'; |
||
| 602 | |||
| 603 | // 子命令, 目前只支持-d |
||
| 604 | $command2 = isset($argv[2]) ? $argv[2] : ''; |
||
| 605 | |||
| 606 | // 根据命令做相应处理 |
||
| 607 | switch($command) |
||
| 608 | { |
||
| 609 | // 启动 phpspider |
||
| 610 | case 'start': |
||
| 611 | if ($command2 === '-d') |
||
| 612 | { |
||
| 613 | self::$daemonize = true; |
||
| 614 | } |
||
| 615 | break; |
||
| 616 | case 'stop': |
||
| 617 | exec("ps aux | grep $start_file | grep -v grep | awk '{print $2}'", $info); |
||
| 618 | if (count($info) <= 1) |
||
| 619 | { |
||
| 620 | echo "PHPSpider[$start_file] not run\n"; |
||
| 621 | } |
||
| 622 | else |
||
| 623 | { |
||
| 624 | //echo "PHPSpider[$start_file] is stoping ...\n"; |
||
| 625 | echo "PHPSpider[$start_file] stop success"; |
||
| 626 | exec("ps aux | grep $start_file | grep -v grep | awk '{print $2}' |xargs kill -SIGINT", $info); |
||
| 627 | } |
||
| 628 | exit; |
||
| 629 | break; |
||
| 630 | case 'kill': |
||
| 631 | exec("ps aux | grep $start_file | grep -v grep | awk '{print $2}' |xargs kill -SIGKILL"); |
||
| 632 | break; |
||
| 633 | // 显示 phpspider 运行状态 |
||
| 634 | case 'status': |
||
| 635 | exit(0); |
||
| 636 | // 未知命令 |
||
| 637 | default : |
||
| 638 | exit("Usage: php yourfile.php {start|stop|status|kill}\n"); |
||
| 639 | } |
||
| 640 | } |
||
| 641 | |||
| 642 | /** |
||
| 643 | * Signal hander. |
||
| 644 | * |
||
| 645 | * @param int $signal |
||
| 646 | */ |
||
| 647 | public function signal_handler($signal) |
||
| 648 | { |
||
| 649 | switch ($signal) { |
||
| 650 | // Stop. |
||
| 651 | case SIGINT: |
||
| 652 | log::warn("Program stopping..."); |
||
| 653 | self::$terminate = true; |
||
| 654 | break; |
||
| 655 | // Show status. |
||
| 656 | case SIGUSR2: |
||
| 657 | echo "show status\n"; |
||
| 658 | break; |
||
| 659 | } |
||
| 660 | } |
||
| 661 | |||
| 662 | /** |
||
| 663 | * Install signal handler. |
||
| 664 | * |
||
| 665 | * @return void |
||
| 666 | */ |
||
| 667 | public function install_signal() |
||
| 668 | { |
||
| 669 | if (function_exists('pcntl_signal')) |
||
| 670 | { |
||
| 671 | // stop |
||
| 672 | pcntl_signal(SIGINT, array(__CLASS__, 'signal_handler'), false); |
||
| 673 | // status |
||
| 674 | pcntl_signal(SIGUSR2, array(__CLASS__, 'signal_handler'), false); |
||
| 675 | // ignore |
||
| 676 | pcntl_signal(SIGPIPE, SIG_IGN, false); |
||
| 677 | } |
||
| 678 | } |
||
| 679 | |||
| 680 | /** |
||
| 681 | * Run as deamon mode. |
||
| 682 | * |
||
| 683 | * @throws Exception |
||
| 684 | */ |
||
| 685 | protected static function daemonize() |
||
| 686 | { |
||
| 687 | if (!self::$daemonize) |
||
| 688 | { |
||
| 689 | return; |
||
| 690 | } |
||
| 691 | |||
| 692 | // fork前一定要关闭redis |
||
| 693 | queue::clear_link(); |
||
| 694 | |||
| 695 | umask(0); |
||
| 696 | $pid = pcntl_fork(); |
||
| 697 | if (-1 === $pid) |
||
| 698 | { |
||
| 699 | throw new Exception('fork fail'); |
||
| 700 | } |
||
| 701 | elseif ($pid > 0) |
||
| 702 | { |
||
| 703 | exit(0); |
||
| 704 | } |
||
| 705 | if (-1 === posix_setsid()) |
||
| 706 | { |
||
| 707 | throw new Exception("setsid fail"); |
||
| 708 | } |
||
| 709 | // Fork again avoid SVR4 system regain the control of terminal. |
||
| 710 | $pid = pcntl_fork(); |
||
| 711 | if (-1 === $pid) |
||
| 712 | { |
||
| 713 | throw new Exception("fork fail"); |
||
| 714 | } |
||
| 715 | elseif (0 !== $pid) |
||
| 716 | { |
||
| 717 | exit(0); |
||
| 718 | } |
||
| 719 | } |
||
| 720 | |||
| 721 | /** |
||
| 722 | * 检查是否终止当前进程 |
||
| 723 | * |
||
| 724 | * @return void |
||
| 725 | * @author seatle <[email protected]> |
||
| 726 | * @created time :2016-11-16 11:06 |
||
| 727 | */ |
||
| 728 | public function check_terminate() |
||
| 729 | { |
||
| 730 | if (!self::$terminate) |
||
| 731 | { |
||
| 732 | return false; |
||
| 733 | } |
||
| 734 | |||
| 735 | // 删除当前任务状态 |
||
| 736 | $this->del_task_status(self::$serverid, self::$taskid); |
||
| 737 | |||
| 738 | if (self::$taskmaster) |
||
| 739 | { |
||
| 740 | // 检查子进程是否都退出 |
||
| 741 | while (true) |
||
| 742 | { |
||
| 743 | $all_stop = true; |
||
| 744 | for ($i = 2; $i <= self::$tasknum; $i++) |
||
| 745 | { |
||
| 746 | // 只要一个还活着就说明没有完全退出 |
||
| 747 | $task_status = $this->get_task_status(self::$serverid, $i); |
||
| 748 | if ($task_status) |
||
| 749 | { |
||
| 750 | $all_stop = false; |
||
| 751 | } |
||
| 752 | } |
||
| 753 | if ($all_stop) |
||
| 754 | { |
||
| 755 | break; |
||
| 756 | } |
||
| 757 | else |
||
| 758 | { |
||
| 759 | log::warn("Task stop waiting..."); |
||
| 760 | } |
||
| 761 | sleep(1); |
||
| 762 | } |
||
| 763 | |||
| 764 | $this->del_server_list(self::$serverid); |
||
| 765 | |||
| 766 | // 显示最后结果 |
||
| 767 | log::$log_show = true; |
||
| 768 | |||
| 769 | $spider_time_run = util::time2second(intval(microtime(true) - self::$time_start)); |
||
| 770 | log::note("Spider finished in {$spider_time_run}"); |
||
| 771 | |||
| 772 | $get_collected_url_num = $this->get_collected_url_num(); |
||
| 773 | log::note("Total pages: {$get_collected_url_num} \n"); |
||
| 774 | } |
||
| 775 | exit(); |
||
| 776 | } |
||
| 777 | |||
| 778 | public function start() |
||
| 779 | { |
||
| 780 | $this->parse_command(); |
||
| 781 | |||
| 782 | // 爬虫开始时间 |
||
| 783 | self::$time_start = time(); |
||
| 784 | // 当前任务ID |
||
| 785 | self::$taskid = 1; |
||
| 786 | // 当前任务进程ID |
||
| 787 | self::$taskpid = function_exists('posix_getpid') ? posix_getpid() : 1; |
||
| 788 | self::$collect_succ = 0; |
||
| 789 | self::$collect_fail = 0; |
||
| 790 | |||
| 791 | //-------------------------------------------------------------------------------- |
||
| 792 | // 运行前验证 |
||
| 793 | //-------------------------------------------------------------------------------- |
||
| 794 | |||
| 795 | // 检查PHP版本 |
||
| 796 | if (version_compare(PHP_VERSION, '5.3.0', 'lt')) |
||
| 797 | { |
||
| 798 | log::error('PHP 5.3+ is required, currently installed version is: ' . phpversion()); |
||
| 799 | exit; |
||
| 800 | } |
||
| 801 | |||
| 802 | // 检查CURL扩展 |
||
| 803 | if(!function_exists('curl_init')) |
||
| 804 | { |
||
| 805 | log::error("The curl extension was not found"); |
||
| 806 | exit; |
||
| 807 | } |
||
| 808 | |||
| 809 | // 多任务需要pcntl扩展支持 |
||
| 810 | if (self::$tasknum > 1 && !function_exists('pcntl_fork')) |
||
| 811 | { |
||
| 812 | log::error("Multitasking needs pcntl, the pcntl extension was not found"); |
||
| 813 | exit; |
||
| 814 | } |
||
| 815 | |||
| 816 | // 守护进程需要pcntl扩展支持 |
||
| 817 | if (self::$daemonize && !function_exists('pcntl_fork')) |
||
| 818 | { |
||
| 819 | log::error("Daemonize needs pcntl, the pcntl extension was not found"); |
||
| 820 | exit; |
||
| 821 | } |
||
| 822 | |||
| 823 | // 集群、保存运行状态、多任务都需要Redis支持 |
||
| 824 | if (self::$multiserver || self::$save_running_state || self::$tasknum > 1) |
||
| 825 | { |
||
| 826 | self::$use_redis = true; |
||
| 827 | |||
| 828 | queue::set_connect('default', self::$queue_config); |
||
| 829 | if (!queue::init()) |
||
| 830 | { |
||
| 831 | if (self::$multiserver) |
||
| 832 | { |
||
| 833 | log::error("Multiserver needs Redis support, ".queue::$error); |
||
| 834 | exit; |
||
| 835 | } |
||
| 836 | |||
| 837 | if (self::$tasknum > 1) |
||
| 838 | { |
||
| 839 | log::error("Multitasking needs Redis support, ".queue::$error); |
||
| 840 | exit; |
||
| 841 | } |
||
| 842 | |||
| 843 | if (self::$save_running_state) |
||
| 844 | { |
||
| 845 | log::error("Spider kept running state needs Redis support, ".queue::$error); |
||
| 846 | exit; |
||
| 847 | } |
||
| 848 | } |
||
| 849 | } |
||
| 850 | |||
| 851 | // 检查导出 |
||
| 852 | $this->check_export(); |
||
| 853 | |||
| 854 | // 检查缓存 |
||
| 855 | $this->check_cache(); |
||
| 856 | |||
| 857 | // 检查 scan_urls |
||
| 858 | if (empty(self::$configs['scan_urls'])) |
||
| 859 | { |
||
| 860 | log::error("No scan url to start"); |
||
| 861 | exit; |
||
| 862 | } |
||
| 863 | |||
| 864 | foreach ( self::$configs['scan_urls'] as $url ) |
||
| 865 | { |
||
| 866 | // 只检查配置中的入口URL, 通过 add_scan_url 添加的不检查了. |
||
| 867 | if (!$this->is_scan_page($url)) |
||
| 868 | { |
||
| 869 | log::error("Domain of scan_urls (\"{$url}\") does not match the domains of the domain name"); |
||
| 870 | exit; |
||
| 871 | } |
||
| 872 | } |
||
| 873 | |||
| 874 | // windows 下没法显示面板, 强制显示日志 |
||
| 875 | if (util::is_win()) |
||
| 876 | { |
||
| 877 | self::$configs['name'] = iconv("UTF-8", "GB2312//IGNORE", self::$configs['name']); |
||
| 878 | log::$log_show = true; |
||
| 879 | } |
||
| 880 | else |
||
| 881 | { |
||
| 882 | log::$log_show = isset(self::$configs['log_show']) ? self::$configs['log_show'] : false; |
||
| 883 | } |
||
| 884 | |||
| 885 | if (self::$daemonize) |
||
| 886 | { |
||
| 887 | log::$log_show = true; |
||
| 888 | } |
||
| 889 | |||
| 890 | if (log::$log_show) |
||
| 891 | { |
||
| 892 | global $argv; |
||
| 893 | $start_file = $argv[0]; |
||
| 894 | |||
| 895 | $header = ""; |
||
| 896 | if (!util::is_win()) $header .= "\033[33m"; |
||
| 897 | $header .= "\n[ ".self::$configs['name']." Spider ] is started...\n\n"; |
||
| 898 | $header .= " * PHPSpider Version: ".self::VERSION."\n"; |
||
| 899 | $header .= " * Documentation: https://doc.phpspider.org\n"; |
||
| 900 | $header .= " * Task Number: ".self::$tasknum."\n\n"; |
||
| 901 | $header .= "Input \"php $start_file stop\" to quit. Start success.\n"; |
||
| 902 | if (!util::is_win()) $header .= "\033[0m"; |
||
| 903 | log::note($header); |
||
| 904 | } |
||
| 905 | |||
| 906 | // 如果是守护进程，恢复日志状态 |
||
| 907 | if (self::$daemonize) |
||
| 908 | { |
||
| 909 | log::$log_show = isset(self::$configs['log_show']) ? self::$configs['log_show'] : false; |
||
| 910 | } |
||
| 911 | |||
| 912 | // 多任务和分布式都要清掉, 当然分布式只清自己的 |
||
| 913 | $this->init_redis(); |
||
| 914 | |||
| 915 | //-------------------------------------------------------------------------------- |
||
| 916 | // 生成多任务 |
||
| 917 | //-------------------------------------------------------------------------------- |
||
| 918 | |||
| 919 | // 添加入口URL到队列 |
||
| 920 | foreach ( self::$configs['scan_urls'] as $url ) |
||
| 921 | { |
||
| 922 | // false 表示不允许重复 |
||
| 923 | $this->add_scan_url($url, null, false); |
||
| 924 | } |
||
| 925 | |||
| 926 | // 放这个位置, 可以添加入口页面 |
||
| 927 | if ($this->on_start) |
||
| 928 | { |
||
| 929 | call_user_func($this->on_start, $this); |
||
| 930 | } |
||
| 931 | |||
| 932 | if (!self::$daemonize) |
||
| 933 | { |
||
| 934 | if (!log::$log_show) |
||
| 935 | { |
||
| 936 | // 第一次先清屏 |
||
| 937 | $this->clear_echo(); |
||
| 938 | |||
| 939 | // 先显示一次面板, 然后下面再每次采集成功显示一次 |
||
| 940 | $this->display_ui(); |
||
| 941 | } |
||
| 942 | } |
||
| 943 | else |
||
| 944 | { |
||
| 945 | $this->daemonize(); |
||
| 946 | } |
||
| 947 | |||
| 948 | // 安装信号 |
||
| 949 | $this->install_signal(); |
||
| 950 | |||
| 951 | // 开始采集 |
||
| 952 | $this->do_collect_page(); |
||
| 953 | |||
| 954 | // 从服务器列表中删除当前服务器信息 |
||
| 955 | $this->del_server_list(self::$serverid); |
||
| 956 | } |
||
| 957 | |||
| 958 | /** |
||
| 959 | * 创建一个子进程 |
||
| 960 | * @param Worker $worker |
||
| 961 | * @throws Exception |
||
| 962 | */ |
||
| 963 | public function fork_one_task($taskid) |
||
| 964 | { |
||
| 965 | $pid = pcntl_fork(); |
||
| 966 | |||
| 967 | // 主进程记录子进程pid |
||
| 968 | if($pid > 0) |
||
| 969 | { |
||
| 970 | // 暂时没用 |
||
| 971 | //self::$taskpids[$taskid] = $pid; |
||
| 972 | } |
||
| 973 | // 子进程运行 |
||
| 974 | elseif(0 === $pid) |
||
| 975 | { |
||
| 976 | log::warn("Fork children task({$taskid}) successful..."); |
||
| 977 | |||
| 978 | // 初始化子进程参数 |
||
| 979 | self::$time_start = microtime(true); |
||
| 980 | self::$taskid = $taskid; |
||
| 981 | self::$taskmaster = false; |
||
| 982 | self::$taskpid = posix_getpid(); |
||
| 983 | self::$collect_succ = 0; |
||
| 984 | self::$collect_fail = 0; |
||
| 985 | |||
| 986 | $this->do_collect_page(); |
||
| 987 | |||
| 988 | // 这里用0表示正常退出 |
||
| 989 | exit(0); |
||
| 990 | } |
||
| 991 | else |
||
| 992 | { |
||
| 993 | log::error("Fork children task({$taskid}) fail..."); |
||
| 994 | exit; |
||
| 995 | } |
||
| 996 | } |
||
| 997 | |||
| 998 | public function do_collect_page() |
||
| 999 | { |
||
| 1000 | queue::set_connect('default', self::$queue_config); |
||
| 1001 | queue::init(); |
||
| 1002 | while( $queue_lsize = $this->queue_lsize() ) |
||
| 1003 | { |
||
| 1004 | // 如果是主任务 |
||
| 1005 | if (self::$taskmaster) |
||
| 1006 | { |
||
| 1007 | // 多任务下主任务未准备就绪 |
||
| 1008 | if (self::$tasknum > 1 && !self::$fork_task_complete) |
||
| 1009 | { |
||
| 1010 | // 主进程采集到两倍于任务数时, 生成子任务一起采集 |
||
| 1011 | if ( $queue_lsize > self::$tasknum*2 ) |
||
| 1012 | { |
||
| 1013 | self::$fork_task_complete = true; |
||
| 1014 | |||
| 1015 | // fork 子进程前一定要先干掉redis连接fd, 不然会存在进程互抢redis fd 问题 |
||
| 1016 | queue::clear_link(); |
||
| 1017 | // task进程从2开始, 1被master进程所使用 |
||
| 1018 | for ($i = 2; $i <= self::$tasknum; $i++) |
||
| 1019 | { |
||
| 1020 | $this->fork_one_task($i); |
||
| 1021 | } |
||
| 1022 | } |
||
| 1023 | } |
||
| 1024 | |||
| 1025 | // 抓取页面 |
||
| 1026 | $this->collect_page(); |
||
| 1027 | // 保存任务状态 |
||
| 1028 | $this->set_task_status(); |
||
| 1029 | |||
| 1030 | // 每采集成功一次页面, 就刷新一次面板 |
||
| 1031 | if (!log::$log_show && !self::$daemonize) |
||
| 1032 | { |
||
| 1033 | $this->display_ui(); |
||
| 1034 | } |
||
| 1035 | } |
||
| 1036 | // 如果是子任务 |
||
| 1037 | else |
||
| 1038 | { |
||
| 1039 | // 如果队列中的网页比任务数2倍多, 子任务可以采集, 否则等待... |
||
| 1040 | if ( $queue_lsize > self::$tasknum*2 ) |
||
| 1041 | { |
||
| 1042 | // 抓取页面 |
||
| 1043 | $this->collect_page(); |
||
| 1044 | // 保存任务状态 |
||
| 1045 | $this->set_task_status(); |
||
| 1046 | } |
||
| 1047 | else |
||
| 1048 | { |
||
| 1049 | log::warn("Task(".self::$taskid.") waiting..."); |
||
| 1050 | sleep(1); |
||
| 1051 | } |
||
| 1052 | } |
||
| 1053 | |||
| 1054 | // 检查进程是否收到关闭信号 |
||
| 1055 | $this->check_terminate(); |
||
| 1056 | } |
||
| 1057 | } |
||
| 1058 | |||
| 1059 | /** |
||
| 1060 | * 爬取页面 |
||
| 1061 | * |
||
| 1062 | * @param mixed $collect_url 要抓取的链接 |
||
| 1063 | * @return void |
||
| 1064 | * @author seatle <[email protected]> |
||
| 1065 | * @created time :2016-09-18 10:17 |
||
| 1066 | */ |
||
| 1067 | public function collect_page() |
||
| 1068 | { |
||
| 1069 | $get_collect_url_num = $this->get_collect_url_num(); |
||
| 1070 | log::info("Find pages: {$get_collect_url_num} "); |
||
| 1071 | |||
| 1072 | $queue_lsize = $this->queue_lsize(); |
||
| 1073 | log::info("Waiting for collect pages: {$queue_lsize} "); |
||
| 1074 | |||
| 1075 | $get_collected_url_num = $this->get_collected_url_num(); |
||
| 1076 | log::info("Collected pages: {$get_collected_url_num} "); |
||
| 1077 | |||
| 1078 | // 多任务的时候输出爬虫序号 |
||
| 1079 | if (self::$tasknum > 1) |
||
| 1080 | { |
||
| 1081 | log::info("Current task id: ".self::$taskid); |
||
| 1082 | } |
||
| 1083 | |||
| 1084 | // 先进先出 |
||
| 1085 | $link = $this->queue_rpop(); |
||
| 1086 | $link = $this->link_uncompress($link); |
||
| 1087 | $url = $link['url']; |
||
| 1088 | |||
| 1089 | // 标记为已爬取网页 |
||
| 1090 | $this->incr_collected_url_num($url); |
||
| 1091 | |||
| 1092 | // 爬取页面开始时间 |
||
| 1093 | $page_time_start = microtime(true); |
||
| 1094 | |||
| 1095 | requests::$input_encoding = null; |
||
| 1096 | $html = $this->request_url($url, $link); |
||
| 1097 | |||
| 1098 | if (!$html) |
||
| 1099 | { |
||
| 1100 | return false; |
||
| 1101 | } |
||
| 1102 | // 当前正在爬取的网页页面的对象 |
||
| 1103 | $page = array( |
||
| 1104 | 'url' => $url, |
||
| 1105 | 'raw' => $html, |
||
| 1106 | 'request' => array( |
||
| 1107 | 'url' => $url, |
||
| 1108 | 'method' => $link['method'], |
||
| 1109 | 'headers' => $link['headers'], |
||
| 1110 | 'params' => $link['params'], |
||
| 1111 | 'context_data' => $link['context_data'], |
||
| 1112 | 'try_num' => $link['try_num'], |
||
| 1113 | 'max_try' => $link['max_try'], |
||
| 1114 | 'depth' => $link['depth'], |
||
| 1115 | 'taskid' => self::$taskid, |
||
| 1116 | ), |
||
| 1117 | ); |
||
| 1118 | unset($html); |
||
| 1119 | |||
| 1120 | //-------------------------------------------------------------------------------- |
||
| 1121 | // 处理回调函数 |
||
| 1122 | //-------------------------------------------------------------------------------- |
||
| 1123 | |||
| 1124 | // 判断当前网页是否被反爬虫了, 需要开发者实现 |
||
| 1125 | if ($this->is_anti_spider) |
||
| 1126 | { |
||
| 1127 | $is_anti_spider = call_user_func($this->is_anti_spider, $url, $page['raw'], $this); |
||
| 1128 | // 如果在回调函数里面判断被反爬虫并且返回true |
||
| 1129 | if ($is_anti_spider) |
||
| 1130 | { |
||
| 1131 | return false; |
||
| 1132 | } |
||
| 1133 | } |
||
| 1134 | |||
| 1135 | // 在一个网页下载完成之后调用. 主要用来对下载的网页进行处理. |
||
| 1136 | // 比如下载了某个网页, 希望向网页的body中添加html标签 |
||
| 1137 | if ($this->on_download_page) |
||
| 1138 | { |
||
| 1139 | $return = call_user_func($this->on_download_page, $page, $this); |
||
| 1140 | // 针对那些老是忘记return的人 |
||
| 1141 | if (isset($return)) $page = $return; |
||
| 1142 | } |
||
| 1143 | |||
| 1144 | // 是否从当前页面分析提取URL |
||
| 1145 | // 回调函数如果返回false表示不需要再从此网页中发现待爬url |
||
| 1146 | $is_find_url = true; |
||
| 1147 | if ($link['url_type'] == 'scan_page') |
||
| 1148 | { |
||
| 1149 | if ($this->on_scan_page) |
||
| 1150 | { |
||
| 1151 | $return = call_user_func($this->on_scan_page, $page, $page['raw'], $this); |
||
| 1152 | if (isset($return)) $is_find_url = $return; |
||
| 1153 | } |
||
| 1154 | } |
||
| 1155 | elseif ($link['url_type'] == 'list_page') |
||
| 1156 | { |
||
| 1157 | if ($this->on_list_page) |
||
| 1158 | { |
||
| 1159 | $return = call_user_func($this->on_list_page, $page, $page['raw'], $this); |
||
| 1160 | if (isset($return)) $is_find_url = $return; |
||
| 1161 | } |
||
| 1162 | } |
||
| 1163 | elseif ($link['url_type'] == 'content_page') |
||
| 1164 | { |
||
| 1165 | if ($this->on_content_page) |
||
| 1166 | { |
||
| 1167 | $return = call_user_func($this->on_content_page, $page, $page['raw'], $this); |
||
| 1168 | if (isset($return)) $is_find_url = $return; |
||
| 1169 | } |
||
| 1170 | } |
||
| 1171 | |||
| 1172 | // on_scan_page、on_list_page、on_content_page 返回false表示不需要再从此网页中发现待爬url |
||
| 1173 | if ($is_find_url) |
||
| 1174 | { |
||
| 1175 | // 如果深度没有超过最大深度, 获取下一级URL |
||
| 1176 | if (self::$configs['max_depth'] == 0 || $link['depth'] < self::$configs['max_depth']) |
||
| 1177 | { |
||
| 1178 | // 分析提取HTML页面中的URL |
||
| 1179 | $this->get_urls($page['raw'], $url, $link['depth'] + 1); |
||
| 1180 | } |
||
| 1181 | } |
||
| 1182 | |||
| 1183 | // 如果是内容页, 分析提取HTML页面中的字段 |
||
| 1184 | // 列表页也可以提取数据的, source_type: urlcontext, 未实现 |
||
| 1185 | if ($link['url_type'] == 'content_page') |
||
| 1186 | { |
||
| 1187 | $this->get_html_fields($page['raw'], $url, $page); |
||
| 1188 | } |
||
| 1189 | |||
| 1190 | // 如果当前深度大于缓存的, 更新缓存 |
||
| 1191 | $this->incr_depth_num($link['depth']); |
||
| 1192 | |||
| 1193 | // 处理页面耗时时间 |
||
| 1194 | $time_run = round(microtime(true) - $page_time_start, 3); |
||
| 1195 | log::debug("Success process page {$url} in {$time_run} s"); |
||
| 1196 | |||
| 1197 | $spider_time_run = util::time2second(intval(microtime(true) - self::$time_start)); |
||
| 1198 | log::info("Spider running in {$spider_time_run}"); |
||
| 1199 | |||
| 1200 | // 爬虫爬取每个网页的时间间隔, 单位: 毫秒 |
||
| 1201 | if (!isset(self::$configs['interval'])) |
||
| 1202 | { |
||
| 1203 | // 默认睡眠100毫秒, 太快了会被认为是ddos |
||
| 1204 | self::$configs['interval'] = 100; |
||
| 1205 | } |
||
| 1206 | usleep(self::$configs['interval'] * 1000); |
||
| 1207 | } |
||
| 1208 | |||
| 1209 | /** |
||
| 1210 | * 下载网页, 得到网页内容 |
||
| 1211 | * |
||
| 1212 | * @param mixed $url |
||
| 1213 | * @param mixed $link |
||
| 1214 | * @return void |
||
| 1215 | * @author seatle <[email protected]> |
||
| 1216 | * @created time :2016-09-18 10:17 |
||
| 1217 | */ |
||
| 1218 | public function request_url($url, $link = array()) |
||
| 1219 | { |
||
| 1220 | $time_start = microtime(true); |
||
| 1221 | |||
| 1222 | //$url = "http://www.qiushibaike.com/article/117568316"; |
||
| 1223 | |||
| 1224 | // 设置了编码就不要让requests去判断了 |
||
| 1225 | if (isset(self::$configs['input_encoding'])) |
||
| 1226 | { |
||
| 1227 | requests::$input_encoding = self::$configs['input_encoding']; |
||
| 1228 | } |
||
| 1229 | // 得到的编码如果不是utf-8的要转成utf-8, 因为xpath只支持utf-8 |
||
| 1230 | requests::$output_encoding = 'utf-8'; |
||
| 1231 | requests::set_timeout(self::$configs['timeout']); |
||
| 1232 | requests::set_useragent(self::$configs['user_agent']); |
||
| 1233 | if (self::$configs['user_agents']) |
||
| 1234 | { |
||
| 1235 | requests::set_useragents(self::$configs['user_agents']); |
||
| 1236 | } |
||
| 1237 | if (self::$configs['client_ip']) |
||
| 1238 | { |
||
| 1239 | requests::set_client_ip(self::$configs['client_ip']); |
||
| 1240 | } |
||
| 1241 | if (self::$configs['client_ips']) |
||
| 1242 | { |
||
| 1243 | requests::set_client_ips(self::$configs['client_ips']); |
||
| 1244 | } |
||
| 1245 | |||
| 1246 | // 是否设置了代理 |
||
| 1247 | if (!empty($link['proxies'])) |
||
| 1248 | { |
||
| 1249 | requests::set_proxies($link['proxies']); |
||
| 1250 | // 自动切换IP |
||
| 1251 | requests::set_header('Proxy-Switch-Ip', 'yes'); |
||
| 1252 | } |
||
| 1253 | |||
| 1254 | // 如何设置了 HTTP Headers |
||
| 1255 | if (!empty($link['headers'])) |
||
| 1256 | { |
||
| 1257 | foreach ($link['headers'] as $k=>$v) |
||
| 1258 | { |
||
| 1259 | requests::set_header($k, $v); |
||
| 1260 | } |
||
| 1261 | } |
||
| 1262 | |||
| 1263 | $method = empty($link['method']) ? 'get' : strtolower($link['method']); |
||
| 1264 | $params = empty($link['params']) ? array() : $link['params']; |
||
| 1265 | $html = requests::$method($url, $params); |
||
| 1266 | // 此url附加的数据不为空, 比如内容页需要列表页一些数据, 拼接到后面去 |
||
| 1267 | if ($html && !empty($link['context_data'])) |
||
| 1268 | { |
||
| 1269 | $html .= $link['context_data']; |
||
| 1270 | } |
||
| 1271 | |||
| 1272 | $http_code = requests::$status_code; |
||
| 1273 | |||
| 1274 | if ($this->on_status_code) |
||
| 1275 | { |
||
| 1276 | $return = call_user_func($this->on_status_code, $http_code, $url, $html, $this); |
||
| 1277 | if (isset($return)) |
||
| 1278 | { |
||
| 1279 | $html = $return; |
||
| 1280 | } |
||
| 1281 | if (!$html) |
||
| 1282 | { |
||
| 1283 | return false; |
||
| 1284 | } |
||
| 1285 | } |
||
| 1286 | |||
| 1287 | if ($http_code != 200) |
||
| 1288 | { |
||
| 1289 | // 如果是301、302跳转, 抓取跳转后的网页内容 |
||
| 1290 | if ($http_code == 301 || $http_code == 302) |
||
| 1291 | { |
||
| 1292 | $info = requests::$info; |
||
| 1293 | if (isset($info['redirect_url'])) |
||
| 1294 | { |
||
| 1295 | $url = $info['redirect_url']; |
||
| 1296 | requests::$input_encoding = null; |
||
| 1297 | $html = $this->request_url($url, $link); |
||
| 1298 | if ($html && !empty($link['context_data'])) |
||
| 1299 | { |
||
| 1300 | $html .= $link['context_data']; |
||
| 1301 | } |
||
| 1302 | } |
||
| 1303 | else |
||
| 1304 | { |
||
| 1305 | return false; |
||
| 1306 | } |
||
| 1307 | } |
||
| 1308 | else |
||
| 1309 | { |
||
| 1310 | if ($http_code == 407) |
||
| 1311 | { |
||
| 1312 | // 扔到队列头部去, 继续采集 |
||
| 1313 | $this->queue_rpush($link); |
||
| 1314 | log::error("Failed to download page {$url}"); |
||
| 1315 | self::$collect_fail++; |
||
| 1316 | } |
||
| 1317 | elseif (in_array($http_code, array('0','502','503','429'))) |
||
| 1318 | { |
||
| 1319 | // 采集次数加一 |
||
| 1320 | $link['try_num']++; |
||
| 1321 | // 抓取次数 小于 允许抓取失败次数 |
||
| 1322 | if ( $link['try_num'] <= $link['max_try'] ) |
||
| 1323 | { |
||
| 1324 | // 扔到队列头部去, 继续采集 |
||
| 1325 | $this->queue_rpush($link); |
||
| 1326 | } |
||
| 1327 | log::error("Failed to download page {$url}, retry({$link['try_num']})"); |
||
| 1328 | } |
||
| 1329 | else |
||
| 1330 | { |
||
| 1331 | log::error("Failed to download page {$url}"); |
||
| 1332 | self::$collect_fail++; |
||
| 1333 | } |
||
| 1334 | log::error("HTTP CODE: {$http_code}"); |
||
| 1335 | return false; |
||
| 1336 | } |
||
| 1337 | } |
||
| 1338 | |||
| 1339 | // 爬取页面耗时时间 |
||
| 1340 | $time_run = round(microtime(true) - $time_start, 3); |
||
| 1341 | log::debug("Success download page {$url} in {$time_run} s"); |
||
| 1342 | self::$collect_succ++; |
||
| 1343 | |||
| 1344 | return $html; |
||
| 1345 | } |
||
| 1346 | |||
| 1347 | /** |
||
| 1348 | * 分析提取HTML页面中的URL |
||
| 1349 | * |
||
| 1350 | * @param mixed $html HTML内容 |
||
| 1351 | * @param mixed $collect_url 抓取的URL, 用来拼凑完整页面的URL |
||
| 1352 | * @return void |
||
| 1353 | * @author seatle <[email protected]> |
||
| 1354 | * @created time :2016-09-18 10:17 |
||
| 1355 | */ |
||
| 1356 | public function get_urls($html, $collect_url, $depth = 0) |
||
| 1357 | { |
||
| 1358 | //-------------------------------------------------------------------------------- |
||
| 1359 | // 正则匹配出页面中的URL |
||
| 1360 | //-------------------------------------------------------------------------------- |
||
| 1361 | $urls = selector::select($html, '//a/@href'); |
||
| 1362 | //preg_match_all("/<a.*href=[\"']{0,1}(.*)[\"']{0,1}[> \r\n\t]{1,}/isU", $html, $matchs); |
||
| 1363 | //$urls = array(); |
||
| 1364 | //if (!empty($matchs[1])) |
||
| 1365 | //{ |
||
| 1366 | //foreach ($matchs[1] as $url) |
||
| 1367 | //{ |
||
| 1368 | //$urls[] = str_replace(array("\"", "'",'&'), array("",'','&'), $url); |
||
| 1369 | //} |
||
| 1370 | //} |
||
| 1371 | |||
| 1372 | if (empty($urls)) |
||
| 1373 | { |
||
| 1374 | return false; |
||
| 1375 | } |
||
| 1376 | |||
| 1377 | // 如果页面上只有一个url，要把他转为数组，否则下面会报警告 |
||
| 1378 | if (!is_array($urls)) |
||
| 1379 | { |
||
| 1380 | $urls = array($urls); |
||
| 1381 | } |
||
| 1382 | |||
| 1383 | foreach ($urls as $key=>$url) |
||
| 1384 | { |
||
| 1385 | $urls[$key] = str_replace(array("\"", "'",'&'), array("",'','&'), $url); |
||
| 1386 | } |
||
| 1387 | |||
| 1388 | //-------------------------------------------------------------------------------- |
||
| 1389 | // 过滤和拼凑URL |
||
| 1390 | //-------------------------------------------------------------------------------- |
||
| 1391 | // 去除重复的RUL |
||
| 1392 | $urls = array_unique($urls); |
||
| 1393 | foreach ($urls as $k=>$url) |
||
| 1394 | { |
||
| 1395 | $url = trim($url); |
||
| 1396 | if (empty($url)) |
||
| 1397 | { |
||
| 1398 | continue; |
||
| 1399 | } |
||
| 1400 | |||
| 1401 | $val = $this->fill_url($url, $collect_url); |
||
| 1402 | if ($val) |
||
| 1403 | { |
||
| 1404 | $urls[$k] = $val; |
||
| 1405 | } |
||
| 1406 | else |
||
| 1407 | { |
||
| 1408 | unset($urls[$k]); |
||
| 1409 | } |
||
| 1410 | } |
||
| 1411 | |||
| 1412 | if (empty($urls)) |
||
| 1413 | { |
||
| 1414 | return false; |
||
| 1415 | } |
||
| 1416 | |||
| 1417 | //-------------------------------------------------------------------------------- |
||
| 1418 | // 把抓取到的URL放入队列 |
||
| 1419 | //-------------------------------------------------------------------------------- |
||
| 1420 | foreach ($urls as $url) |
||
| 1421 | { |
||
| 1422 | if ($this->on_fetch_url) |
||
| 1423 | { |
||
| 1424 | $return = call_user_func($this->on_fetch_url, $url, $this); |
||
| 1425 | $url = isset($return) ? $return : $url; |
||
| 1426 | unset($return); |
||
| 1427 | |||
| 1428 | // 如果 on_fetch_url 返回 false，此URL不入队列 |
||
| 1429 | if (!$url) |
||
| 1430 | { |
||
| 1431 | continue; |
||
| 1432 | } |
||
| 1433 | } |
||
| 1434 | |||
| 1435 | // 把当前页当做找到的url的Referer页 |
||
| 1436 | $options = array( |
||
| 1437 | 'headers' => array( |
||
| 1438 | 'Referer' => $collect_url, |
||
| 1439 | ) |
||
| 1440 | ); |
||
| 1441 | $this->add_url($url, $options, $depth); |
||
| 1442 | } |
||
| 1443 | } |
||
| 1444 | |||
| 1445 | /** |
||
| 1446 | * 获得完整的连接地址 |
||
| 1447 | * |
||
| 1448 | * @param mixed $url 要检查的URL |
||
| 1449 | * @param mixed $collect_url 从那个URL页面得到上面的URL |
||
| 1450 | * @return void |
||
| 1451 | * @author seatle <[email protected]> |
||
| 1452 | * @created time :2016-09-23 17:13 |
||
| 1453 | */ |
||
| 1454 | public function fill_url($url, $collect_url) |
||
| 1455 | { |
||
| 1456 | $url = trim($url); |
||
| 1457 | $collect_url = trim($collect_url); |
||
| 1458 | |||
| 1459 | // 排除JavaScript的连接 |
||
| 1460 | //if (strpos($url, "javascript:") !== false) |
||
| 1461 | if( preg_match("@^(javascript:|#|'|\")@i", $url) || $url == '') |
||
| 1462 | { |
||
| 1463 | return false; |
||
| 1464 | } |
||
| 1465 | // 排除没有被解析成功的语言标签 |
||
| 1466 | if(substr($url, 0, 3) == '<%=') |
||
| 1467 | { |
||
| 1468 | return false; |
||
| 1469 | } |
||
| 1470 | |||
| 1471 | $parse_url = @parse_url($collect_url); |
||
| 1472 | if (empty($parse_url['scheme']) || empty($parse_url['host'])) |
||
| 1473 | { |
||
| 1474 | return false; |
||
| 1475 | } |
||
| 1476 | // 过滤mailto、tel、sms、wechat、sinaweibo、weixin等协议 |
||
| 1477 | if (!in_array($parse_url['scheme'], array("http", "https"))) |
||
| 1478 | { |
||
| 1479 | return false; |
||
| 1480 | } |
||
| 1481 | $scheme = $parse_url['scheme']; |
||
| 1482 | $domain = $parse_url['host']; |
||
| 1483 | $path = empty($parse_url['path']) ? '' : $parse_url['path']; |
||
| 1484 | $base_url_path = $domain.$path; |
||
| 1485 | $base_url_path = preg_replace("/\/([^\/]*)\.(.*)$/","/",$base_url_path); |
||
| 1486 | $base_url_path = preg_replace("/\/$/",'',$base_url_path); |
||
| 1487 | |||
| 1488 | $i = $path_step = 0; |
||
| 1489 | $dstr = $pstr = ''; |
||
| 1490 | $pos = strpos($url,'#'); |
||
| 1491 | if($pos > 0) |
||
| 1492 | { |
||
| 1493 | // 去掉#和后面的字符串 |
||
| 1494 | $url = substr($url, 0, $pos); |
||
| 1495 | } |
||
| 1496 | |||
| 1497 | // 京东变态的都是 //www.jd.com/111.html |
||
| 1498 | if(substr($url, 0, 2) == '//') |
||
| 1499 | { |
||
| 1500 | $url = str_replace("//", "", $url); |
||
| 1501 | } |
||
| 1502 | // /1234.html |
||
| 1503 | elseif($url[0] == '/') |
||
| 1504 | { |
||
| 1505 | $url = $domain.$url; |
||
| 1506 | } |
||
| 1507 | // ./1234.html、../1234.html 这种类型的 |
||
| 1508 | elseif($url[0] == '.') |
||
| 1509 | { |
||
| 1510 | if(!isset($url[2])) |
||
| 1511 | { |
||
| 1512 | return false; |
||
| 1513 | } |
||
| 1514 | else |
||
| 1515 | { |
||
| 1516 | $urls = explode('/',$url); |
||
| 1517 | foreach($urls as $u) |
||
| 1518 | { |
||
| 1519 | if( $u == '..' ) |
||
| 1520 | { |
||
| 1521 | $path_step++; |
||
| 1522 | } |
||
| 1523 | // 遇到 ., 不知道为什么不直接写$u == '.', 貌似一样的 |
||
| 1524 | else if( $i < count($urls)-1 ) |
||
| 1525 | { |
||
| 1526 | $dstr .= $urls[$i].'/'; |
||
| 1527 | } |
||
| 1528 | else |
||
| 1529 | { |
||
| 1530 | $dstr .= $urls[$i]; |
||
| 1531 | } |
||
| 1532 | $i++; |
||
| 1533 | } |
||
| 1534 | $urls = explode('/',$base_url_path); |
||
| 1535 | if(count($urls) <= $path_step) |
||
| 1536 | { |
||
| 1537 | return false; |
||
| 1538 | } |
||
| 1539 | else |
||
| 1540 | { |
||
| 1541 | $pstr = ''; |
||
| 1542 | for($i=0;$i<count($urls)-$path_step;$i++){ $pstr .= $urls[$i].'/'; } |
||
| 1543 | $url = $pstr.$dstr; |
||
| 1544 | } |
||
| 1545 | } |
||
| 1546 | } |
||
| 1547 | else |
||
| 1548 | { |
||
| 1549 | if( strtolower(substr($url, 0, 7))=='http://' ) |
||
| 1550 | { |
||
| 1551 | $url = preg_replace('#^http://#i','',$url); |
||
| 1552 | $scheme = "http"; |
||
| 1553 | } |
||
| 1554 | else if( strtolower(substr($url, 0, 8))=='https://' ) |
||
| 1555 | { |
||
| 1556 | $url = preg_replace('#^https://#i','',$url); |
||
| 1557 | $scheme = "https"; |
||
| 1558 | } |
||
| 1559 | else |
||
| 1560 | { |
||
| 1561 | $url = $base_url_path.'/'.$url; |
||
| 1562 | } |
||
| 1563 | } |
||
| 1564 | // 两个 / 或以上的替换成一个 / |
||
| 1565 | $url = preg_replace('@/{1,}@i', '/', $url); |
||
| 1566 | $url = $scheme.'://'.$url; |
||
| 1567 | //echo $url;exit("\n"); |
||
| 1568 | |||
| 1569 | $parse_url = @parse_url($url); |
||
| 1570 | $domain = empty($parse_url['host']) ? $domain : $parse_url['host']; |
||
| 1571 | // 如果host不为空, 判断是不是要爬取的域名 |
||
| 1572 | if (!empty($parse_url['host'])) |
||
| 1573 | { |
||
| 1574 | //排除非域名下的url以提高爬取速度 |
||
| 1575 | if (!in_array($parse_url['host'], self::$configs['domains'])) |
||
| 1576 | { |
||
| 1577 | return false; |
||
| 1578 | } |
||
| 1579 | } |
||
| 1580 | |||
| 1581 | return $url; |
||
| 1582 | } |
||
| 1583 | |||
| 1584 | /** |
||
| 1585 | * 连接对象压缩 |
||
| 1586 | * |
||
| 1587 | * @return void |
||
| 1588 | * @author seatle <[email protected]> |
||
| 1589 | * @created time :2016-11-05 18:58 |
||
| 1590 | */ |
||
| 1591 | public function link_compress($link) |
||
| 1592 | { |
||
| 1593 | if (empty($link['url_type'])) |
||
| 1594 | { |
||
| 1595 | unset($link['url_type']); |
||
| 1596 | } |
||
| 1597 | |||
| 1598 | if (empty($link['method']) || strtolower($link['method']) == 'get') |
||
| 1599 | { |
||
| 1600 | unset($link['method']); |
||
| 1601 | } |
||
| 1602 | |||
| 1603 | if (empty($link['headers'])) |
||
| 1604 | { |
||
| 1605 | unset($link['headers']); |
||
| 1606 | } |
||
| 1607 | |||
| 1608 | if (empty($link['params'])) |
||
| 1609 | { |
||
| 1610 | unset($link['params']); |
||
| 1611 | } |
||
| 1612 | |||
| 1613 | if (empty($link['context_data'])) |
||
| 1614 | { |
||
| 1615 | unset($link['context_data']); |
||
| 1616 | } |
||
| 1617 | |||
| 1618 | if (empty($link['proxies'])) |
||
| 1619 | { |
||
| 1620 | unset($link['proxies']); |
||
| 1621 | } |
||
| 1622 | |||
| 1623 | if (empty($link['try_num'])) |
||
| 1624 | { |
||
| 1625 | unset($link['try_num']); |
||
| 1626 | } |
||
| 1627 | |||
| 1628 | if (empty($link['max_try'])) |
||
| 1629 | { |
||
| 1630 | unset($link['max_try']); |
||
| 1631 | } |
||
| 1632 | |||
| 1633 | if (empty($link['depth'])) |
||
| 1634 | { |
||
| 1635 | unset($link['depth']); |
||
| 1636 | } |
||
| 1637 | //$json = json_encode($link); |
||
| 1638 | //$json = gzdeflate($json); |
||
| 1639 | return $link; |
||
| 1640 | } |
||
| 1641 | |||
| 1642 | /** |
||
| 1643 | * 连接对象解压缩 |
||
| 1644 | * |
||
| 1645 | * @param mixed $link |
||
| 1646 | * @return void |
||
| 1647 | * @author seatle <[email protected]> |
||
| 1648 | * @created time :2016-11-05 18:58 |
||
| 1649 | */ |
||
| 1650 | public function link_uncompress($link) |
||
| 1651 | { |
||
| 1652 | $link = array( |
||
| 1653 | 'url' => isset($link['url']) ? $link['url'] : '', |
||
| 1654 | 'url_type' => isset($link['url_type']) ? $link['url_type'] : '', |
||
| 1655 | 'method' => isset($link['method']) ? $link['method'] : 'get', |
||
| 1656 | 'headers' => isset($link['headers']) ? $link['headers'] : array(), |
||
| 1657 | 'params' => isset($link['params']) ? $link['params'] : array(), |
||
| 1658 | 'context_data' => isset($link['context_data']) ? $link['context_data'] : '', |
||
| 1659 | 'proxies' => isset($link['proxies']) ? $link['proxies'] : self::$configs['proxies'], |
||
| 1660 | 'try_num' => isset($link['try_num']) ? $link['try_num'] : 0, |
||
| 1661 | 'max_try' => isset($link['max_try']) ? $link['max_try'] : self::$configs['max_try'], |
||
| 1662 | 'depth' => isset($link['depth']) ? $link['depth'] : 0, |
||
| 1663 | ); |
||
| 1664 | |||
| 1665 | return $link; |
||
| 1666 | } |
||
| 1667 | |||
| 1668 | /** |
||
| 1669 | * 分析提取HTML页面中的字段 |
||
| 1670 | * |
||
| 1671 | * @param mixed $html |
||
| 1672 | * @return void |
||
| 1673 | * @author seatle <[email protected]> |
||
| 1674 | * @created time :2016-09-18 10:17 |
||
| 1675 | */ |
||
| 1676 | public function get_html_fields($html, $url, $page) |
||
| 1677 | { |
||
| 1678 | $fields = $this->get_fields(self::$configs['fields'], $html, $url, $page); |
||
| 1679 | |||
| 1680 | if (!empty($fields)) |
||
| 1681 | { |
||
| 1682 | if ($this->on_extract_page) |
||
| 1683 | { |
||
| 1684 | $return = call_user_func($this->on_extract_page, $page, $fields); |
||
| 1685 | if (!isset($return)) |
||
| 1686 | { |
||
| 1687 | log::warn("on_extract_page return value can't be empty"); |
||
| 1688 | } |
||
| 1689 | elseif (!is_array($return)) |
||
| 1690 | { |
||
| 1691 | log::warn("on_extract_page return value must be an array"); |
||
| 1692 | } |
||
| 1693 | else |
||
| 1694 | { |
||
| 1695 | $fields = $return; |
||
| 1696 | } |
||
| 1697 | } |
||
| 1698 | |||
| 1699 | if (isset($fields) && is_array($fields)) |
||
| 1700 | { |
||
| 1701 | $fields_num = $this->incr_fields_num(); |
||
| 1702 | if (self::$configs['max_fields'] != 0 && $fields_num > self::$configs['max_fields']) |
||
| 1703 | { |
||
| 1704 | exit(0); |
||
| 1705 | } |
||
| 1706 | |||
| 1707 | if (version_compare(PHP_VERSION,'5.4.0','<')) |
||
| 1708 | { |
||
| 1709 | $fields_str = json_encode($fields); |
||
| 1710 | $fields_str = preg_replace_callback( "#\\\u([0-9a-f]{4})#i", function($matchs) { |
||
| 1711 | return iconv('UCS-2BE', 'UTF-8', pack('H4', $matchs[1])); |
||
| 1712 | }, $fields_str ); |
||
| 1713 | } |
||
| 1714 | else |
||
| 1715 | { |
||
| 1716 | $fields_str = json_encode($fields, JSON_UNESCAPED_UNICODE); |
||
| 1717 | } |
||
| 1718 | |||
| 1719 | if (util::is_win()) |
||
| 1720 | { |
||
| 1721 | $fields_str = mb_convert_encoding($fields_str, 'gb2312', 'utf-8'); |
||
| 1722 | } |
||
| 1723 | log::info("Result[{$fields_num}]: ".$fields_str); |
||
| 1724 | |||
| 1725 | // 如果设置了导出选项 |
||
| 1726 | if (!empty(self::$configs['export'])) |
||
| 1727 | { |
||
| 1728 | self::$export_type = isset(self::$configs['export']['type']) ? self::$configs['export']['type'] : ''; |
||
| 1729 | if (self::$export_type == 'csv') |
||
| 1730 | { |
||
| 1731 | util::put_file(self::$export_file, util::format_csv($fields)."\n", FILE_APPEND); |
||
| 1732 | } |
||
| 1733 | elseif (self::$export_type == 'sql') |
||
| 1734 | { |
||
| 1735 | $sql = db::insert(self::$export_table, $fields, true); |
||
| 1736 | util::put_file(self::$export_file, $sql.";\n", FILE_APPEND); |
||
| 1737 | } |
||
| 1738 | elseif (self::$export_type == 'db') |
||
| 1739 | { |
||
| 1740 | db::insert(self::$export_table, $fields); |
||
| 1741 | } |
||
| 1742 | } |
||
| 1743 | } |
||
| 1744 | } |
||
| 1745 | } |
||
| 1746 | |||
| 1747 | /** |
||
| 1748 | * 根据配置提取HTML代码块中的字段 |
||
| 1749 | * |
||
| 1750 | * @param mixed $confs |
||
| 1751 | * @param mixed $html |
||
| 1752 | * @param mixed $page |
||
| 1753 | * @return void |
||
| 1754 | * @author seatle <[email protected]> |
||
| 1755 | * @created time :2016-09-23 17:13 |
||
| 1756 | */ |
||
| 1757 | public function get_fields($confs, $html, $url, $page) |
||
| 1914 | } |
||
| 1915 | |||
| 1916 | /** |
||
| 1917 | * 验证导出 |
||
| 1918 | * |
||
| 1919 | * @return void |
||
| 1920 | * @author seatle <[email protected]> |
||
| 1921 | * @created time :2016-10-02 23:37 |
||
| 1922 | */ |
||
| 1923 | public function check_export() |
||
| 1924 | { |
||
| 1925 | // 如果设置了导出选项 |
||
| 1926 | if (!empty(self::$configs['export'])) |
||
| 1927 | { |
||
| 1928 | if (self::$export_type == 'csv') |
||
| 1929 | { |
||
| 1930 | if (empty(self::$export_file)) |
||
| 1931 | { |
||
| 1932 | log::error("Export data into CSV files need to Set the file path."); |
||
| 1933 | exit; |
||
| 1934 | } |
||
| 1935 | } |
||
| 1936 | elseif (self::$export_type == 'sql') |
||
| 1937 | { |
||
| 1938 | if (empty(self::$export_file)) |
||
| 1939 | { |
||
| 1940 | log::error("Export data into SQL files need to Set the file path."); |
||
| 1941 | exit; |
||
| 1942 | } |
||
| 1943 | } |
||
| 1944 | elseif (self::$export_type == 'db') |
||
| 1945 | { |
||
| 1946 | if (!function_exists('mysqli_connect')) |
||
| 1947 | { |
||
| 1948 | log::error("Export data to a database need Mysql support, Error: Unable to load mysqli extension."); |
||
| 1949 | exit; |
||
| 1950 | } |
||
| 1951 | |||
| 1952 | if (empty(self::$db_config)) |
||
| 1953 | { |
||
| 1954 | log::error("Export data to a database need Mysql support, Error: You not set a config array for connect."); |
||
| 1955 | exit; |
||
| 1956 | } |
||
| 1957 | |||
| 1958 | $config = self::$db_config; |
||
| 1959 | @mysqli_connect($config['host'], $config['user'], $config['pass'], $config['name'], $config['port']); |
||
| 1960 | if(mysqli_connect_errno()) |
||
| 1961 | { |
||
| 1962 | log::error("Export data to a database need Mysql support, Error: ".mysqli_connect_error()); |
||
| 1963 | exit; |
||
| 1964 | } |
||
| 1965 | |||
| 1966 | db::set_connect('default', $config); |
||
| 1967 | db::_init(); |
||
| 1968 | |||
| 1969 | if (!db::table_exists(self::$export_table)) |
||
| 1970 | { |
||
| 1971 | log::error("Table ".self::$export_table." does not exist"); |
||
| 1972 | exit; |
||
| 1973 | } |
||
| 1974 | } |
||
| 1975 | } |
||
| 1976 | } |
||
| 1977 | |||
| 1978 | public function check_cache() |
||
| 1979 | { |
||
| 1980 | if (!self::$use_redis || self::$save_running_state) |
||
| 1981 | { |
||
| 1982 | return false; |
||
| 1983 | } |
||
| 1984 | |||
| 1985 | //if (queue::exists("collect_queue")) |
||
| 1986 | $keys = queue::keys("*"); |
||
| 1987 | $count = count($keys); |
||
| 1988 | if ($count != 0) |
||
| 1989 | { |
||
| 1990 | // After this operation, 4,318 kB of additional disk space will be used. |
||
| 1991 | // Do you want to continue? [Y/n] |
||
| 1992 | //$msg = "发现Redis中有采集数据, 是否继续执行, 不继续则清空Redis数据重新采集\n"; |
||
| 1993 | $msg = "Found that the data of Redis, no continue will empty Redis data start again\n"; |
||
| 1994 | $msg .= "Do you want to continue? [Y/n]"; |
||
| 1995 | fwrite(STDOUT, $msg); |
||
| 1996 | $arg = strtolower(trim(fgets(STDIN))); |
||
| 1997 | $arg = empty($arg) || !in_array($arg, array('y','n')) ? 'y' : $arg; |
||
| 1998 | if ($arg == 'n') |
||
| 1999 | { |
||
| 2000 | foreach ($keys as $key) |
||
| 2001 | { |
||
| 2002 | $key = str_replace($GLOBALS['config']['redis']['prefix'].":", "", $key); |
||
| 2003 | queue::del($key); |
||
| 2004 | } |
||
| 2005 | } |
||
| 2006 | } |
||
| 2007 | } |
||
| 2008 | |||
| 2009 | public function init_redis() |
||
| 2010 | { |
||
| 2011 | if (!self::$use_redis) |
||
| 2012 | { |
||
| 2013 | return false; |
||
| 2014 | } |
||
| 2015 | |||
| 2016 | // 添加当前服务器到服务器列表 |
||
| 2017 | $this->add_server_list(self::$serverid, self::$tasknum); |
||
| 2018 | |||
| 2019 | // 删除当前服务器的任务状态 |
||
| 2020 | // 对于被强制退出的进程有用 |
||
| 2021 | for ($i = 1; $i <= self::$tasknum; $i++) |
||
| 2022 | { |
||
| 2023 | $this->del_task_status(self::$serverid, $i); |
||
| 2024 | } |
||
| 2025 | } |
||
| 2026 | |||
| 2027 | /** |
||
| 2028 | * 设置任务状态, 主进程和子进程每成功采集一个页面后调用 |
||
| 2029 | * |
||
| 2030 | * @return void |
||
| 2031 | * @author seatle <[email protected]> |
||
| 2032 | * @created time :2016-10-30 23:56 |
||
| 2033 | */ |
||
| 2034 | public function set_task_status() |
||
| 2035 | { |
||
| 2036 | // 每采集成功一个页面, 生成当前进程状态到文件, 供主进程使用 |
||
| 2037 | $mem = round(memory_get_usage(true)/(1024*1024),2); |
||
| 2038 | $use_time = microtime(true) - self::$time_start; |
||
| 2039 | $speed = round((self::$collect_succ + self::$collect_fail) / $use_time, 2); |
||
| 2040 | $status = array( |
||
| 2041 | 'id' => self::$taskid, |
||
| 2042 | 'pid' => self::$taskpid, |
||
| 2043 | 'mem' => $mem, |
||
| 2044 | 'collect_succ' => self::$collect_succ, |
||
| 2045 | 'collect_fail' => self::$collect_fail, |
||
| 2046 | 'speed' => $speed, |
||
| 2047 | ); |
||
| 2048 | $task_status = json_encode($status); |
||
| 2049 | |||
| 2050 | if (self::$use_redis) |
||
| 2051 | { |
||
| 2052 | $key = "server-".self::$serverid."-task_status-".self::$taskid; |
||
| 2053 | queue::set($key, $task_status); |
||
| 2054 | } |
||
| 2055 | else |
||
| 2056 | { |
||
| 2057 | self::$task_status = array($task_status); |
||
| 2058 | } |
||
| 2059 | } |
||
| 2060 | |||
| 2061 | /** |
||
| 2062 | * 删除任务状态 |
||
| 2063 | * |
||
| 2064 | * @return void |
||
| 2065 | * @author seatle <[email protected]> |
||
| 2066 | * @created time :2016-11-16 11:06 |
||
| 2067 | */ |
||
| 2068 | public function del_task_status($serverid, $taskid) |
||
| 2069 | { |
||
| 2070 | if (!self::$use_redis) |
||
| 2071 | { |
||
| 2072 | return false; |
||
| 2073 | } |
||
| 2074 | $key = "server-{$serverid}-task_status-{$taskid}"; |
||
| 2075 | queue::del($key); |
||
| 2076 | } |
||
| 2077 | |||
| 2078 | /** |
||
| 2079 | * 获得任务状态, 主进程才会调用 |
||
| 2080 | * |
||
| 2081 | * @return void |
||
| 2082 | * @author seatle <[email protected]> |
||
| 2083 | * @created time :2016-10-30 23:56 |
||
| 2084 | */ |
||
| 2085 | public function get_task_status($serverid, $taskid) |
||
| 2095 | } |
||
| 2096 | |||
| 2097 | /** |
||
| 2098 | * 获得任务状态, 主进程才会调用 |
||
| 2099 | * |
||
| 2100 | * @return void |
||
| 2101 | * @author seatle <[email protected]> |
||
| 2102 | * @created time :2016-10-30 23:56 |
||
| 2103 | */ |
||
| 2104 | public function get_task_status_list($serverid = 1, $tasknum) |
||
| 2120 | } |
||
| 2121 | |||
| 2122 | /** |
||
| 2123 | * 添加当前服务器信息到服务器列表 |
||
| 2124 | * |
||
| 2125 | * @return void |
||
| 2126 | * @author seatle <[email protected]> |
||
| 2127 | * @created time :2016-11-16 11:06 |
||
| 2128 | */ |
||
| 2129 | public function add_server_list($serverid, $tasknum) |
||
| 2130 | { |
||
| 2131 | if (!self::$use_redis) |
||
| 2132 | { |
||
| 2133 | return false; |
||
| 2134 | } |
||
| 2135 | |||
| 2136 | // 更新服务器列表 |
||
| 2137 | $server_list_json = queue::get("server_list"); |
||
| 2138 | $server_list = array(); |
||
| 2139 | if (!$server_list_json) |
||
| 2140 | { |
||
| 2141 | $server_list[$serverid] = array( |
||
| 2142 | 'serverid' => $serverid, |
||
| 2143 | 'tasknum' => $tasknum, |
||
| 2144 | 'time' => time(), |
||
| 2145 | ); |
||
| 2146 | } |
||
| 2147 | else |
||
| 2148 | { |
||
| 2149 | $server_list = json_decode($server_list_json, true); |
||
| 2150 | $server_list[$serverid] = array( |
||
| 2151 | 'serverid' => $serverid, |
||
| 2152 | 'tasknum' => $tasknum, |
||
| 2153 | 'time' => time(), |
||
| 2154 | ); |
||
| 2155 | ksort($server_list); |
||
| 2156 | } |
||
| 2157 | queue::set("server_list", json_encode($server_list)); |
||
| 2158 | } |
||
| 2159 | |||
| 2160 | /** |
||
| 2161 | * 从服务器列表中删除当前服务器信息 |
||
| 2162 | * |
||
| 2163 | * @return void |
||
| 2164 | * @author seatle <[email protected]> |
||
| 2165 | * @created time :2016-11-16 11:06 |
||
| 2166 | */ |
||
| 2167 | public function del_server_list($serverid) |
||
| 2168 | { |
||
| 2169 | if (!self::$use_redis) |
||
| 2170 | { |
||
| 2171 | return false; |
||
| 2172 | } |
||
| 2173 | |||
| 2174 | $server_list_json = queue::get("server_list"); |
||
| 2175 | $server_list = array(); |
||
| 2176 | if ($server_list_json) |
||
| 2177 | { |
||
| 2178 | $server_list = json_decode($server_list_json, true); |
||
| 2179 | if (isset($server_list[$serverid])) |
||
| 2180 | { |
||
| 2181 | unset($server_list[$serverid]); |
||
| 2182 | } |
||
| 2183 | |||
| 2184 | // 删除完当前的任务列表如果还存在，就更新一下Redis |
||
| 2185 | if (!empty($server_list)) |
||
| 2186 | { |
||
| 2187 | ksort($server_list); |
||
| 2188 | queue::set("server_list", json_encode($server_list)); |
||
| 2189 | } |
||
| 2190 | } |
||
| 2191 | } |
||
| 2192 | |||
| 2193 | /** |
||
| 2194 | * 获取等待爬取页面数量 |
||
| 2195 | * |
||
| 2196 | * @param mixed $url |
||
| 2197 | * @return void |
||
| 2198 | * @author seatle <[email protected]> |
||
| 2199 | * @created time :2016-09-23 17:13 |
||
| 2200 | */ |
||
| 2201 | public function get_collect_url_num() |
||
| 2202 | { |
||
| 2203 | if (self::$use_redis) |
||
| 2204 | { |
||
| 2205 | $count = queue::get("collect_urls_num"); |
||
| 2206 | } |
||
| 2207 | else |
||
| 2208 | { |
||
| 2209 | $count = self::$collect_urls_num; |
||
| 2210 | } |
||
| 2211 | return $count; |
||
| 2212 | } |
||
| 2213 | |||
| 2214 | /** |
||
| 2215 | * 获取已经爬取页面数量 |
||
| 2216 | * |
||
| 2217 | * @param mixed $url |
||
| 2218 | * @return void |
||
| 2219 | * @author seatle <[email protected]> |
||
| 2220 | * @created time :2016-09-23 17:13 |
||
| 2221 | */ |
||
| 2222 | public function get_collected_url_num() |
||
| 2223 | { |
||
| 2224 | if (self::$use_redis) |
||
| 2225 | { |
||
| 2226 | $count = queue::get("collected_urls_num"); |
||
| 2227 | } |
||
| 2228 | else |
||
| 2229 | { |
||
| 2230 | $count = self::$collected_urls_num; |
||
| 2231 | } |
||
| 2232 | return $count; |
||
| 2233 | } |
||
| 2234 | |||
| 2235 | /** |
||
| 2236 | * 已采集页面数量加一 |
||
| 2237 | * |
||
| 2238 | * @param mixed $url |
||
| 2239 | * @return void |
||
| 2240 | * @author seatle <[email protected]> |
||
| 2241 | * @created time :2016-09-23 17:13 |
||
| 2242 | */ |
||
| 2243 | public function incr_collected_url_num($url) |
||
| 2244 | { |
||
| 2245 | if (self::$use_redis) |
||
| 2246 | { |
||
| 2247 | queue::incr("collected_urls_num"); |
||
| 2248 | } |
||
| 2249 | else |
||
| 2250 | { |
||
| 2251 | self::$collected_urls_num++; |
||
| 2252 | } |
||
| 2253 | } |
||
| 2254 | |||
| 2255 | /** |
||
| 2256 | * 从队列左边插入 |
||
| 2257 | * |
||
| 2258 | * @return void |
||
| 2259 | * @author seatle <[email protected]> |
||
| 2260 | * @created time :2016-09-23 17:13 |
||
| 2261 | */ |
||
| 2262 | public function queue_lpush($link = array(), $allowed_repeat = false) |
||
| 2263 | { |
||
| 2264 | if (empty($link) || empty($link['url'])) |
||
| 2265 | { |
||
| 2266 | return false; |
||
| 2267 | } |
||
| 2268 | |||
| 2269 | $url = $link['url']; |
||
| 2270 | $link = $this->link_compress($link); |
||
| 2271 | |||
| 2272 | $status = false; |
||
| 2273 | if (self::$use_redis) |
||
| 2274 | { |
||
| 2275 | $key = "collect_urls-".md5($url); |
||
| 2276 | $lock = "lock-".$key; |
||
| 2277 | // 加锁: 一个进程一个进程轮流处理 |
||
| 2278 | if (queue::lock($lock)) |
||
| 2279 | { |
||
| 2280 | $exists = queue::exists($key); |
||
| 2281 | // 不存在或者当然URL可重复入 |
||
| 2282 | if (!$exists || $allowed_repeat) |
||
| 2283 | { |
||
| 2284 | // 待爬取网页记录数加一 |
||
| 2285 | queue::incr("collect_urls_num"); |
||
| 2286 | // 先标记为待爬取网页 |
||
| 2287 | queue::set($key, time()); |
||
| 2288 | // 入队列 |
||
| 2289 | $link = json_encode($link); |
||
| 2290 | queue::lpush("collect_queue", $link); |
||
| 2291 | $status = true; |
||
| 2292 | } |
||
| 2293 | // 解锁 |
||
| 2294 | queue::unlock($lock); |
||
| 2295 | } |
||
| 2296 | } |
||
| 2297 | else |
||
| 2298 | { |
||
| 2299 | $key = md5($url); |
||
| 2300 | if (!array_key_exists($key, self::$collect_urls)) |
||
| 2301 | { |
||
| 2302 | self::$collect_urls_num++; |
||
| 2303 | self::$collect_urls[$key] = time(); |
||
| 2304 | array_push(self::$collect_queue, $link); |
||
| 2305 | $status = true; |
||
| 2306 | } |
||
| 2307 | } |
||
| 2308 | return $status; |
||
| 2309 | } |
||
| 2310 | |||
| 2311 | /** |
||
| 2312 | * 从队列右边插入 |
||
| 2313 | * |
||
| 2314 | * @return void |
||
| 2315 | * @author seatle <[email protected]> |
||
| 2316 | * @created time :2016-09-23 17:13 |
||
| 2317 | */ |
||
| 2318 | public function queue_rpush($link = array(), $allowed_repeat = false) |
||
| 2319 | { |
||
| 2320 | if (empty($link) || empty($link['url'])) |
||
| 2321 | { |
||
| 2322 | return false; |
||
| 2323 | } |
||
| 2324 | |||
| 2325 | $url = $link['url']; |
||
| 2326 | |||
| 2327 | $status = false; |
||
| 2328 | if (self::$use_redis) |
||
| 2329 | { |
||
| 2330 | $key = "collect_urls-".md5($url); |
||
| 2331 | $lock = "lock-".$key; |
||
| 2332 | // 加锁: 一个进程一个进程轮流处理 |
||
| 2333 | if (queue::lock($lock)) |
||
| 2334 | { |
||
| 2335 | $exists = queue::exists($key); |
||
| 2336 | // 不存在或者当然URL可重复入 |
||
| 2337 | if (!$exists || $allowed_repeat) |
||
| 2338 | { |
||
| 2339 | // 待爬取网页记录数加一 |
||
| 2340 | queue::incr("collect_urls_num"); |
||
| 2341 | // 先标记为待爬取网页 |
||
| 2342 | queue::set($key, time()); |
||
| 2343 | // 入队列 |
||
| 2344 | $link = json_encode($link); |
||
| 2345 | queue::rpush("collect_queue", $link); |
||
| 2346 | $status = true; |
||
| 2347 | } |
||
| 2348 | // 解锁 |
||
| 2349 | queue::unlock($lock); |
||
| 2350 | } |
||
| 2351 | } |
||
| 2352 | else |
||
| 2353 | { |
||
| 2354 | $key = md5($url); |
||
| 2355 | if (!array_key_exists($key, self::$collect_urls)) |
||
| 2356 | { |
||
| 2357 | self::$collect_urls_num++; |
||
| 2358 | self::$collect_urls[$key] = time(); |
||
| 2359 | array_unshift(self::$collect_queue, $link); |
||
| 2360 | $status = true; |
||
| 2361 | } |
||
| 2362 | } |
||
| 2363 | return $status; |
||
| 2364 | } |
||
| 2365 | |||
| 2366 | /** |
||
| 2367 | * 从队列左边取出 |
||
| 2368 | * 后进先出 |
||
| 2369 | * 可以避免采集内容页有分页的时候采集失败数据拼凑不全 |
||
| 2370 | * 还可以按顺序采集列表页 |
||
| 2371 | * |
||
| 2372 | * @return void |
||
| 2373 | * @author seatle <[email protected]> |
||
| 2374 | * @created time :2016-09-23 17:13 |
||
| 2375 | */ |
||
| 2376 | public function queue_lpop() |
||
| 2377 | { |
||
| 2378 | if (self::$use_redis) |
||
| 2379 | { |
||
| 2380 | $link = queue::lpop("collect_queue"); |
||
| 2381 | $link = json_decode($link, true); |
||
| 2382 | } |
||
| 2383 | else |
||
| 2384 | { |
||
| 2385 | $link = array_pop(self::$collect_queue); |
||
| 2386 | } |
||
| 2387 | return $link; |
||
| 2388 | } |
||
| 2389 | |||
| 2390 | /** |
||
| 2391 | * 从队列右边取出 |
||
| 2392 | * |
||
| 2393 | * @return void |
||
| 2394 | * @author seatle <[email protected]> |
||
| 2395 | * @created time :2016-09-23 17:13 |
||
| 2396 | */ |
||
| 2397 | public function queue_rpop() |
||
| 2398 | { |
||
| 2399 | if (self::$use_redis) |
||
| 2400 | { |
||
| 2401 | $link = queue::rpop("collect_queue"); |
||
| 2402 | $link = json_decode($link, true); |
||
| 2403 | } |
||
| 2404 | else |
||
| 2405 | { |
||
| 2406 | $link = array_shift(self::$collect_queue); |
||
| 2407 | } |
||
| 2408 | return $link; |
||
| 2409 | } |
||
| 2410 | |||
| 2411 | /** |
||
| 2412 | * 队列长度 |
||
| 2413 | * |
||
| 2414 | * @return void |
||
| 2415 | * @author seatle <[email protected]> |
||
| 2416 | * @created time :2016-09-23 17:13 |
||
| 2417 | */ |
||
| 2418 | public function queue_lsize() |
||
| 2419 | { |
||
| 2420 | if (self::$use_redis) |
||
| 2421 | { |
||
| 2422 | $lsize = queue::lsize("collect_queue"); |
||
| 2423 | } |
||
| 2424 | else |
||
| 2425 | { |
||
| 2426 | $lsize = count(self::$collect_queue); |
||
| 2427 | } |
||
| 2428 | return $lsize; |
||
| 2429 | } |
||
| 2430 | |||
| 2431 | /** |
||
| 2432 | * 采集深度加一 |
||
| 2433 | * |
||
| 2434 | * @return void |
||
| 2435 | * @author seatle <[email protected]> |
||
| 2436 | * @created time :2016-09-23 17:13 |
||
| 2437 | */ |
||
| 2438 | public function incr_depth_num($depth) |
||
| 2439 | { |
||
| 2440 | if (self::$use_redis) |
||
| 2441 | { |
||
| 2442 | $lock = "lock-depth_num"; |
||
| 2443 | // 锁2秒 |
||
| 2444 | if (queue::lock($lock, time(), 2)) |
||
| 2445 | { |
||
| 2446 | if (queue::get("depth_num") < $depth) |
||
| 2447 | { |
||
| 2448 | queue::set("depth_num", $depth); |
||
| 2449 | } |
||
| 2450 | |||
| 2451 | queue::unlock($lock); |
||
| 2452 | } |
||
| 2453 | } |
||
| 2454 | else |
||
| 2455 | { |
||
| 2456 | if (self::$depth_num < $depth) |
||
| 2457 | { |
||
| 2458 | self::$depth_num = $depth; |
||
| 2459 | } |
||
| 2460 | } |
||
| 2461 | } |
||
| 2462 | |||
| 2463 | /** |
||
| 2464 | * 获得采集深度 |
||
| 2465 | * |
||
| 2466 | * @return void |
||
| 2467 | * @author seatle <[email protected]> |
||
| 2468 | * @created time :2016-09-23 17:13 |
||
| 2469 | */ |
||
| 2470 | public function get_depth_num() |
||
| 2471 | { |
||
| 2472 | if (self::$use_redis) |
||
| 2473 | { |
||
| 2474 | $depth_num = queue::get("depth_num"); |
||
| 2475 | return $depth_num ? $depth_num : 0; |
||
| 2476 | } |
||
| 2477 | else |
||
| 2478 | { |
||
| 2479 | return self::$depth_num; |
||
| 2480 | } |
||
| 2481 | } |
||
| 2482 | |||
| 2483 | /** |
||
| 2484 | * 提取到的field数目加一 |
||
| 2485 | * |
||
| 2486 | * @return void |
||
| 2487 | * @author seatle <[email protected]> |
||
| 2488 | * @created time :2016-09-23 17:13 |
||
| 2489 | */ |
||
| 2490 | public function incr_fields_num() |
||
| 2491 | { |
||
| 2492 | if (self::$use_redis) |
||
| 2493 | { |
||
| 2494 | $fields_num = queue::incr("fields_num"); |
||
| 2495 | } |
||
| 2496 | else |
||
| 2497 | { |
||
| 2498 | self::$fields_num++; |
||
| 2499 | $fields_num = self::$fields_num; |
||
| 2500 | } |
||
| 2501 | return $fields_num; |
||
| 2502 | } |
||
| 2503 | |||
| 2504 | /** |
||
| 2505 | * 提取到的field数目 |
||
| 2506 | * |
||
| 2507 | * @return void |
||
| 2508 | * @author seatle <[email protected]> |
||
| 2509 | * @created time :2016-09-23 17:13 |
||
| 2510 | */ |
||
| 2511 | public function get_fields_num() |
||
| 2512 | { |
||
| 2513 | if (self::$use_redis) |
||
| 2514 | { |
||
| 2515 | $fields_num = queue::get("fields_num"); |
||
| 2516 | } |
||
| 2517 | else |
||
| 2518 | { |
||
| 2519 | $fields_num = self::$fields_num; |
||
| 2520 | } |
||
| 2521 | return $fields_num ? $fields_num : 0; |
||
| 2522 | } |
||
| 2523 | |||
| 2524 | /** |
||
| 2525 | * 采用xpath分析提取字段 |
||
| 2526 | * |
||
| 2527 | * @param mixed $html |
||
| 2528 | * @param mixed $selector |
||
| 2529 | * @return void |
||
| 2530 | * @author seatle <[email protected]> |
||
| 2531 | * @created time :2016-09-18 10:17 |
||
| 2532 | */ |
||
| 2533 | public function get_fields_xpath($html, $selector, $fieldname) |
||
| 2541 | } |
||
| 2542 | |||
| 2543 | /** |
||
| 2544 | * 采用正则分析提取字段 |
||
| 2545 | * |
||
| 2546 | * @param mixed $html |
||
| 2547 | * @param mixed $selector |
||
| 2548 | * @return void |
||
| 2549 | * @author seatle <[email protected]> |
||
| 2550 | * @created time :2016-09-18 10:17 |
||
| 2551 | */ |
||
| 2552 | public function get_fields_regex($html, $selector, $fieldname) |
||
| 2553 | { |
||
| 2554 | $result = selector::select($html, $selector, 'regex'); |
||
| 2555 | if (selector::$error) |
||
| 2556 | { |
||
| 2557 | log::error("Field(\"{$fieldname}\") ".selector::$error."\n"); |
||
| 2558 | } |
||
| 2559 | return $result; |
||
| 2560 | } |
||
| 2561 | |||
| 2562 | /** |
||
| 2563 | * 采用CSS选择器提取字段 |
||
| 2564 | * |
||
| 2565 | * @param mixed $html |
||
| 2566 | * @param mixed $selector |
||
| 2567 | * @param mixed $fieldname |
||
| 2568 | * @return void |
||
| 2569 | * @author seatle <[email protected]> |
||
| 2570 | * @created time :2016-09-18 10:17 |
||
| 2571 | */ |
||
| 2572 | public function get_fields_css($html, $selector, $fieldname) |
||
| 2573 | { |
||
| 2574 | $result = selector::select($html, $selector, 'css'); |
||
| 2575 | if (selector::$error) |
||
| 2576 | { |
||
| 2577 | log::error("Field(\"{$fieldname}\") ".selector::$error."\n"); |
||
| 2578 | } |
||
| 2579 | return $result; |
||
| 2580 | } |
||
| 2581 | |||
| 2582 | /** |
||
| 2583 | * 清空shell输出内容 |
||
| 2584 | * |
||
| 2585 | * @return void |
||
| 2586 | * @author seatle <[email protected]> |
||
| 2587 | * @created time :2016-11-16 11:06 |
||
| 2588 | */ |
||
| 2589 | public function clear_echo() |
||
| 2590 | { |
||
| 2591 | $arr = array(27, 91, 72, 27, 91, 50, 74); |
||
| 2592 | foreach ($arr as $a) |
||
| 2593 | { |
||
| 2594 | print chr($a); |
||
| 2595 | } |
||
| 2596 | //array_map(create_function('$a', 'print chr($a);'), array(27, 91, 72, 27, 91, 50, 74)); |
||
| 2597 | } |
||
| 2598 | |||
| 2599 | /** |
||
| 2600 | * 替换shell输出内容 |
||
| 2601 | * |
||
| 2602 | * @param mixed $message |
||
| 2603 | * @param mixed $force_clear_lines |
||
| 2604 | * @return void |
||
| 2605 | * @author seatle <[email protected]> |
||
| 2606 | * @created time :2016-11-16 11:06 |
||
| 2607 | */ |
||
| 2608 | public function replace_echo($message, $force_clear_lines = NULL) |
||
| 2609 | { |
||
| 2610 | static $last_lines = 0; |
||
| 2611 | |||
| 2612 | if(!is_null($force_clear_lines)) |
||
| 2613 | { |
||
| 2614 | $last_lines = $force_clear_lines; |
||
| 2615 | } |
||
| 2616 | |||
| 2617 | // 获取终端宽度 |
||
| 2618 | $toss = $status = null; |
||
| 2619 | $term_width = exec('tput cols', $toss, $status); |
||
| 2620 | if($status || empty($term_width)) |
||
| 2621 | { |
||
| 2622 | $term_width = 64; // Arbitrary fall-back term width. |
||
| 2623 | } |
||
| 2624 | |||
| 2625 | $line_count = 0; |
||
| 2626 | foreach(explode("\n", $message) as $line) |
||
| 2627 | { |
||
| 2628 | $line_count += count(str_split($line, $term_width)); |
||
| 2629 | } |
||
| 2630 | |||
| 2631 | // Erasure MAGIC: Clear as many lines as the last output had. |
||
| 2632 | for($i = 0; $i < $last_lines; $i++) |
||
| 2633 | { |
||
| 2634 | // Return to the beginning of the line |
||
| 2635 | echo "\r"; |
||
| 2636 | // Erase to the end of the line |
||
| 2637 | echo "\033[K"; |
||
| 2638 | // Move cursor Up a line |
||
| 2639 | echo "\033[1A"; |
||
| 2640 | // Return to the beginning of the line |
||
| 2641 | echo "\r"; |
||
| 2642 | // Erase to the end of the line |
||
| 2643 | echo "\033[K"; |
||
| 2644 | // Return to the beginning of the line |
||
| 2645 | echo "\r"; |
||
| 2646 | // Can be consolodated into |
||
| 2647 | // echo "\r\033[K\033[1A\r\033[K\r"; |
||
| 2648 | } |
||
| 2649 | |||
| 2650 | $last_lines = $line_count; |
||
| 2651 | |||
| 2652 | echo $message."\n"; |
||
| 2653 | } |
||
| 2654 | |||
| 2655 | /** |
||
| 2656 | * 展示启动界面, Windows 不会到这里来 |
||
| 2657 | * @return void |
||
| 2658 | */ |
||
| 2659 | public function display_ui() |
||
| 2660 | { |
||
| 2661 | $loadavg = sys_getloadavg(); |
||
| 2662 | foreach ($loadavg as $k=>$v) |
||
| 2663 | { |
||
| 2664 | $loadavg[$k] = round($v, 2); |
||
| 2665 | } |
||
| 2666 | $display_str = "\033[1A\n\033[K-----------------------------\033[47;30m PHPSPIDER \033[0m-----------------------------\n\033[0m"; |
||
| 2667 | //$display_str = "-----------------------------\033[47;30m PHPSPIDER \033[0m-----------------------------\n\033[0m"; |
||
| 2668 | $run_time_str = util::time2second(time()-self::$time_start, false); |
||
| 2669 | $display_str .= 'PHPSpider version:' . self::VERSION . " PHP version:" . PHP_VERSION . "\n"; |
||
| 2670 | $display_str .= 'start time:'. date('Y-m-d H:i:s', self::$time_start).' run ' . $run_time_str . " \n"; |
||
| 2671 | |||
| 2672 | $display_str .= 'spider name: ' . self::$configs['name'] . "\n"; |
||
| 2673 | if (self::$multiserver) |
||
| 2674 | { |
||
| 2675 | $display_str .= 'server id: ' . self::$serverid."\n"; |
||
| 2676 | } |
||
| 2677 | $display_str .= 'task number: ' . self::$tasknum . "\n"; |
||
| 2678 | $display_str .= 'load average: ' . implode(", ", $loadavg) . "\n"; |
||
| 2679 | $display_str .= "document: https://doc.phpspider.org\n"; |
||
| 2680 | |||
| 2681 | $display_str .= $this->display_task_ui(); |
||
| 2682 | |||
| 2683 | if (self::$multiserver) |
||
| 2684 | { |
||
| 2685 | $display_str .= $this->display_server_ui(); |
||
| 2686 | } |
||
| 2687 | |||
| 2688 | $display_str .= $this->display_collect_ui(); |
||
| 2689 | |||
| 2690 | // 清屏 |
||
| 2691 | //$this->clear_echo(); |
||
| 2692 | // 返回到第一行,第一列 |
||
| 2693 | //echo "\033[0;0H"; |
||
| 2694 | $display_str .= "---------------------------------------------------------------------\n"; |
||
| 2695 | $display_str .= "Press Ctrl-C to quit. Start success."; |
||
| 2696 | if (self::$terminate) |
||
| 2697 | { |
||
| 2698 | $display_str .= "\n\033[33mWait for the process exits...\033[0m"; |
||
| 2699 | } |
||
| 2700 | //echo $display_str; |
||
| 2701 | $this->replace_echo($display_str); |
||
| 2702 | } |
||
| 2703 | |||
| 2704 | public function display_task_ui() |
||
| 2735 | } |
||
| 2736 | |||
| 2737 | public function display_server_ui() |
||
| 2738 | { |
||
| 2739 | $display_str = "-------------------------------\033[47;30m SERVER \033[0m------------------------------\n"; |
||
| 2781 | } |
||
| 2782 | |||
| 2783 | public function display_collect_ui() |
||
| 2784 | { |
||
| 2785 | $display_str = "---------------------------\033[47;30m COLLECT STATUS \033[0m--------------------------\n"; |
||
| 2786 | |||
| 2787 | $display_str .= "\033[47;30mfind pages\033[0m". str_pad('', 16-strlen('find pages')). |
||
| 2788 | "\033[47;30mqueue\033[0m". str_pad('', 14-strlen('queue')). |
||
| 2789 | "\033[47;30mcollected\033[0m". str_pad('', 15-strlen('collected')). |
||
| 2790 | "\033[47;30mfields\033[0m". str_pad('', 15-strlen('fields')). |
||
| 2791 | "\033[47;30mdepth\033[0m". str_pad('', 12-strlen('depth')). |
||
| 2792 | "\n"; |
||
| 2793 | |||
| 2794 | $collect = $this->get_collect_url_num(); |
||
| 2795 | $collected = $this->get_collected_url_num(); |
||
| 2796 | $queue = $this->queue_lsize(); |
||
| 2797 | $fields = $this->get_fields_num(); |
||
| 2798 | $depth = $this->get_depth_num(); |
||
| 2799 | $display_str .= str_pad($collect, 16); |
||
| 2806 | } |
||
| 2807 | |||
| 2808 | /** |
||
| 2809 | * 判断是否附件文件 |
||
| 2810 | * |
||
| 2811 | * @return void |
||
| 2812 | * @author seatle <[email protected]> |
||
| 2813 | * @created time :2016-09-23 17:13 |
||
| 2814 | */ |
||
| 2815 | //public function is_attachment_file($url) |
||
| 2816 | //{ |
||
| 2817 | //$mime_types = $GLOBALS['config']['mimetype']; |
||
| 2818 | //$mime_types_flip = array_flip($mime_types); |
||
| 2819 | |||
| 2820 | //$pathinfo = pathinfo($url); |
||
| 2821 | //$fileext = isset($pathinfo['extension']) ? $pathinfo['extension'] : ''; |
||
| 2822 | |||
| 2823 | //$fileinfo = array(); |
||
| 2824 | //// 存在文件后缀并且是配置里面的后缀 |
||
| 2825 | //if (!empty($fileext) && isset($mime_types_flip[$fileext])) |
||
| 2826 | //{ |
||
| 2827 | //stream_context_set_default( |
||
| 2871 |
Loading history...
In general, usage of exit should be done with care and only when running in a scripting context like a CLI script.