|
1
|
|
|
<?php |
|
2
|
|
|
// +---------------------------------------------------------------------- |
|
3
|
|
|
// | PHPSpider [ A PHP Framework For Crawler ] |
|
4
|
|
|
// +---------------------------------------------------------------------- |
|
5
|
|
|
// | Copyright (c) 2006-2014 https://doc.phpspider.org All rights reserved. |
|
6
|
|
|
// +---------------------------------------------------------------------- |
|
7
|
|
|
// | Licensed ( http://www.apache.org/licenses/LICENSE-2.0 ) |
|
8
|
|
|
// +---------------------------------------------------------------------- |
|
9
|
|
|
// | Author: Seatle Yang <[email protected]> |
|
10
|
|
|
// +---------------------------------------------------------------------- |
|
11
|
|
|
|
|
12
|
|
|
//---------------------------------- |
|
13
|
|
|
// PHPSpider核心类文件 |
|
14
|
|
|
// *********** |
|
15
|
|
|
// 泛域名抓取优化版 BY KEN [email protected] |
|
16
|
|
|
// *********** |
|
17
|
|
|
// * 泛域名设置：domain = array('*') |
|
18
|
|
|
// * 增加子域名数量限制 $max_sub_num = 100 |
|
19
|
|
|
//---------------------------------- |
|
20
|
|
|
|
|
21
|
|
|
namespace phpspider\core; |
|
22
|
|
|
|
|
23
|
|
|
require_once __DIR__.'/constants.php'; |
|
24
|
|
|
|
|
25
|
|
|
use Exception; |
|
26
|
|
|
use phpspider\core\db; |
|
27
|
|
|
use phpspider\core\log; |
|
28
|
|
|
use phpspider\core\queue; |
|
29
|
|
|
use phpspider\core\requests; |
|
30
|
|
|
use phpspider\core\selector; |
|
31
|
|
|
use phpspider\core\util; |
|
32
|
|
|
|
|
33
|
|
|
// 启动的时候生成data目录 |
|
34
|
|
|
util::path_exists(PATH_DATA); |
|
35
|
|
|
util::path_exists(PATH_DATA.'/lock'); |
|
36
|
|
|
util::path_exists(PATH_DATA.'/log'); |
|
37
|
|
|
util::path_exists(PATH_DATA.'/cache'); |
|
38
|
|
|
util::path_exists(PATH_DATA.'/status'); |
|
39
|
|
|
|
|
40
|
|
|
class phpspider |
|
41
|
|
|
{ |
|
42
|
|
|
/** |
|
43
|
|
|
* 版本号 |
|
44
|
|
|
* @var string |
|
45
|
|
|
*/ |
|
46
|
|
|
const VERSION = '2.1.5'; |
|
47
|
|
|
|
|
48
|
|
|
/** |
|
49
|
|
|
* 爬虫爬取每个网页的时间间隔,0表示不延时, 单位: 毫秒 |
|
50
|
|
|
*/ |
|
51
|
|
|
const INTERVAL = 100; |
|
52
|
|
|
|
|
53
|
|
|
/** |
|
54
|
|
|
* 爬虫爬取每个网页的超时时间, 单位: 秒 |
|
55
|
|
|
*/ |
|
56
|
|
|
const TIMEOUT = 5; |
|
57
|
|
|
|
|
58
|
|
|
/** |
|
59
|
|
|
* 爬取失败次数, 不想失败重新爬取则设置为0 |
|
60
|
|
|
*/ |
|
61
|
|
|
const MAX_TRY = 0; |
|
62
|
|
|
|
|
63
|
|
|
/** |
|
64
|
|
|
* 爬虫爬取网页所使用的浏览器类型: pc/Mac、ios、android |
|
65
|
|
|
* 默认类型是PC |
|
66
|
|
|
*/ |
|
67
|
|
|
const AGENT_PC = 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_11_6) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.139 Safari/537.36'; |
|
68
|
|
|
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'; |
|
69
|
|
|
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'; |
|
70
|
|
|
|
|
71
|
|
|
/** |
|
72
|
|
|
* pid文件的路径及名称 |
|
73
|
|
|
* @var string |
|
74
|
|
|
*/ |
|
75
|
|
|
//public static $pid_file = ''; |
|
76
|
|
|
|
|
77
|
|
|
/** |
|
78
|
|
|
* 日志目录, 默认在data根目录下 |
|
79
|
|
|
* @var mixed |
|
80
|
|
|
*/ |
|
81
|
|
|
//public static $log_file = ''; |
|
82
|
|
|
|
|
83
|
|
|
/** |
|
84
|
|
|
* 主任务进程ID |
|
85
|
|
|
*/ |
|
86
|
|
|
//public static $master_pid = 0; |
|
87
|
|
|
|
|
88
|
|
|
/** |
|
89
|
|
|
* 所有任务进程ID |
|
90
|
|
|
*/ |
|
91
|
|
|
//public static $taskpids = array(); |
|
92
|
|
|
|
|
93
|
|
|
/** |
|
94
|
|
|
* Daemonize. |
|
95
|
|
|
* |
|
96
|
|
|
* @var bool |
|
97
|
|
|
*/ |
|
98
|
|
|
public static $daemonize = false; |
|
99
|
|
|
|
|
100
|
|
|
/** |
|
101
|
|
|
* 当前进程是否终止 |
|
102
|
|
|
*/ |
|
103
|
|
|
public static $terminate = false; |
|
104
|
|
|
|
|
105
|
|
|
/** |
|
106
|
|
|
* 是否分布式 |
|
107
|
|
|
*/ |
|
108
|
|
|
public static $multiserver = false; |
|
109
|
|
|
|
|
110
|
|
|
/** |
|
111
|
|
|
* 当前服务器ID |
|
112
|
|
|
*/ |
|
113
|
|
|
public static $serverid = 1; |
|
114
|
|
|
|
|
115
|
|
|
/** |
|
116
|
|
|
* 主任务进程 |
|
117
|
|
|
*/ |
|
118
|
|
|
public static $taskmaster = true; |
|
119
|
|
|
|
|
120
|
|
|
/** |
|
121
|
|
|
* 当前任务ID |
|
122
|
|
|
*/ |
|
123
|
|
|
public static $taskid = 1; |
|
124
|
|
|
|
|
125
|
|
|
/** |
|
126
|
|
|
* 当前任务进程ID |
|
127
|
|
|
*/ |
|
128
|
|
|
public static $taskpid = 1; |
|
129
|
|
|
|
|
130
|
|
|
/** |
|
131
|
|
|
* 并发任务数 |
|
132
|
|
|
*/ |
|
133
|
|
|
public static $tasknum = 1; |
|
134
|
|
|
|
|
135
|
|
|
/** |
|
136
|
|
|
* 生成 |
|
137
|
|
|
*/ |
|
138
|
|
|
public static $fork_task_complete = false; |
|
139
|
|
|
|
|
140
|
|
|
/** |
|
141
|
|
|
* 是否使用Redis |
|
142
|
|
|
*/ |
|
143
|
|
|
public static $use_redis = false; |
|
144
|
|
|
|
|
145
|
|
|
/** |
|
146
|
|
|
* 是否保存爬虫运行状态 |
|
147
|
|
|
*/ |
|
148
|
|
|
public static $save_running_state = false; |
|
149
|
|
|
|
|
150
|
|
|
/** |
|
151
|
|
|
* 配置 |
|
152
|
|
|
*/ |
|
153
|
|
|
public static $configs = array(); |
|
154
|
|
|
|
|
155
|
|
|
/** |
|
156
|
|
|
* 要抓取的URL队列 |
|
157
|
|
|
md5(url) => array( |
|
158
|
|
|
'url' => '', // 要爬取的URL |
|
159
|
|
|
'url_type' => '', // 要爬取的URL类型,scan_page、list_page、content_page |
|
160
|
|
|
'method' => 'get', // 默认为"GET"请求, 也支持"POST"请求 |
|
161
|
|
|
'headers' => array(), // 此url的Headers, 可以为空 |
|
162
|
|
|
'params' => array(), // 发送请求时需添加的参数, 可以为空 |
|
163
|
|
|
'context_data'=> '', // 此url附加的数据, 可以为空 |
|
164
|
|
|
'proxy' => false, // 是否使用代理 |
|
165
|
|
|
'try_num' => 0 // 抓取次数 |
|
166
|
|
|
'max_try' => 0 // 允许抓取失败次数 |
|
167
|
|
|
) |
|
168
|
|
|
*/ |
|
169
|
|
|
public static $collect_queue = array(); |
|
170
|
|
|
|
|
171
|
|
|
/** |
|
172
|
|
|
* 要抓取的URL数组 |
|
173
|
|
|
* md5($url) => time() |
|
174
|
|
|
*/ |
|
175
|
|
|
public static $collect_urls = array(); |
|
176
|
|
|
|
|
177
|
|
|
/** |
|
178
|
|
|
* 要抓取的URL数量 |
|
179
|
|
|
*/ |
|
180
|
|
|
public static $collect_urls_num = 0; |
|
181
|
|
|
|
|
182
|
|
|
/** |
|
183
|
|
|
* 已经抓取的URL数量 |
|
184
|
|
|
*/ |
|
185
|
|
|
public static $collected_urls_num = 0; |
|
186
|
|
|
|
|
187
|
|
|
/** |
|
188
|
|
|
* 当前进程采集成功数 |
|
189
|
|
|
*/ |
|
190
|
|
|
public static $collect_succ = 0; |
|
191
|
|
|
|
|
192
|
|
|
/** |
|
193
|
|
|
* 当前进程采集失败数 |
|
194
|
|
|
*/ |
|
195
|
|
|
public static $collect_fail = 0; |
|
196
|
|
|
|
|
197
|
|
|
/** |
|
198
|
|
|
* 提取到的字段数 |
|
199
|
|
|
*/ |
|
200
|
|
|
public static $fields_num = 0; |
|
201
|
|
|
|
|
202
|
|
|
/** |
|
203
|
|
|
* 【KEN】提取到的页面数按域名计数容器 结构为 domain => number |
|
204
|
|
|
*/ |
|
205
|
|
|
public static $pages_num = array(); |
|
206
|
|
|
|
|
207
|
|
|
/** |
|
208
|
|
|
* 【KEN】单域名允许抓取的最大页面数,0为不限制 |
|
209
|
|
|
*/ |
|
210
|
|
|
public static $max_pages = 0; |
|
211
|
|
|
|
|
212
|
|
|
/** |
|
213
|
|
|
* 【KEN】花费的抓取时长计数容器 结构为 domain => number |
|
214
|
|
|
*/ |
|
215
|
|
|
public static $duration = array(); |
|
216
|
|
|
|
|
217
|
|
|
/** |
|
218
|
|
|
* 【KEN】单域名允许抓取的最大时长，单位秒,0为不限制 |
|
219
|
|
|
*/ |
|
220
|
|
|
public static $max_duration = 0; |
|
221
|
|
|
|
|
222
|
|
|
/** |
|
223
|
|
|
* 【KEN】单域名最大子域名发现数量 防止掉进蜘蛛池，推荐值：3000（多数大型网站上限） |
|
224
|
|
|
*/ |
|
225
|
|
|
public static $max_sub_num = 3000; //建议值 3000 |
|
226
|
|
|
|
|
227
|
|
|
/** |
|
228
|
|
|
* 【KEN】子进程未获取任务，超时退出前，等待计时器 |
|
229
|
|
|
*/ |
|
230
|
|
|
|
|
231
|
|
|
public static $stand_by_time = 0; |
|
232
|
|
|
|
|
233
|
|
|
/** |
|
234
|
|
|
* 【KEN】子进程未获取任务，超时退出前，最大等待时长/秒，全部任务束后，子进程将会等待的时间，以便有缓冲时间，获得新的任务 |
|
235
|
|
|
*/ |
|
236
|
|
|
public static $max_stand_by_time = 60; //建议值 60 |
|
237
|
|
|
|
|
238
|
|
|
/** |
|
239
|
|
|
* 【KEN】每个主机并发上限，降低对方网站流量压力和减少被阻挡概率，建议值 6 ，须与 queue_order = rand 一起使用 |
|
240
|
|
|
*/ |
|
241
|
|
|
public static $max_task_per_host = 0; //0值和非0值会使用不同类型的队列缓存库，从0改为非0值或从非0值改为0需清空队列缓存库再运行，否则任务无法添加 |
|
242
|
|
|
public static $task_per_host_counter = array(); //计数容器 |
|
243
|
|
|
|
|
244
|
|
|
/** |
|
245
|
|
|
* 采集深度 |
|
246
|
|
|
*/ |
|
247
|
|
|
public static $depth_num = 0; |
|
248
|
|
|
|
|
249
|
|
|
/** |
|
250
|
|
|
* 爬虫开始时间 |
|
251
|
|
|
*/ |
|
252
|
|
|
public static $time_start = 0; |
|
253
|
|
|
|
|
254
|
|
|
/** |
|
255
|
|
|
* 任务状态 |
|
256
|
|
|
*/ |
|
257
|
|
|
public static $task_status = array(); |
|
258
|
|
|
|
|
259
|
|
|
// 导出类型配置 |
|
260
|
|
|
public static $export_type = ''; |
|
261
|
|
|
public static $export_file = ''; |
|
262
|
|
|
public static $export_conf = ''; |
|
263
|
|
|
public static $export_table = ''; |
|
264
|
|
|
|
|
265
|
|
|
// 数据库配置 |
|
266
|
|
|
public static $db_config = array(); |
|
267
|
|
|
// 队列配置 |
|
268
|
|
|
public static $queue_config = array(); |
|
269
|
|
|
|
|
270
|
|
|
// 运行面板参数长度 |
|
271
|
|
|
public static $server_length = 10; |
|
272
|
|
|
public static $tasknum_length = 8; |
|
273
|
|
|
public static $taskid_length = 8; |
|
274
|
|
|
public static $pid_length = 8; |
|
275
|
|
|
public static $mem_length = 8; |
|
276
|
|
|
public static $urls_length = 15; |
|
277
|
|
|
public static $speed_length = 6; |
|
278
|
|
|
|
|
279
|
|
|
/** |
|
280
|
|
|
* 爬虫初始化时调用, 用来指定一些爬取前的操作 |
|
281
|
|
|
* |
|
282
|
|
|
* @var mixed |
|
283
|
|
|
* @access public |
|
284
|
|
|
*/ |
|
285
|
|
|
public $on_start = null; |
|
286
|
|
|
|
|
287
|
|
|
/** |
|
288
|
|
|
* URL采集前调用 |
|
289
|
|
|
* 比如有时需要根据某个特定的URL，来决定这次的请求是否使用代理 / 或使用哪个代理 |
|
290
|
|
|
* |
|
291
|
|
|
* @var mixed |
|
292
|
|
|
* @access public |
|
293
|
|
|
*/ |
|
294
|
|
|
public $on_before_download_page = null; |
|
295
|
|
|
|
|
296
|
|
|
/** |
|
297
|
|
|
* 网页状态码回调 |
|
298
|
|
|
* |
|
299
|
|
|
* @var mixed |
|
300
|
|
|
* @access public |
|
301
|
|
|
*/ |
|
302
|
|
|
public $on_status_code = null; |
|
303
|
|
|
|
|
304
|
|
|
/** |
|
305
|
|
|
* 判断当前网页是否被反爬虫, 需要开发者实现 |
|
306
|
|
|
* |
|
307
|
|
|
* @var mixed |
|
308
|
|
|
* @access public |
|
309
|
|
|
*/ |
|
310
|
|
|
public $is_anti_spider = null; |
|
311
|
|
|
|
|
312
|
|
|
/** |
|
313
|
|
|
* 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 |
|
314
|
|
|
* |
|
315
|
|
|
* @var mixed |
|
316
|
|
|
* @access public |
|
317
|
|
|
*/ |
|
318
|
|
|
public $on_download_page = null; |
|
319
|
|
|
|
|
320
|
|
|
/** |
|
321
|
|
|
* 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 |
|
322
|
|
|
* |
|
323
|
|
|
* @var mixed |
|
324
|
|
|
* @access public |
|
325
|
|
|
*/ |
|
326
|
|
|
public $on_download_attached_page = null; |
|
327
|
|
|
|
|
328
|
|
|
/** |
|
329
|
|
|
* 当前页面抽取到URL |
|
330
|
|
|
* |
|
331
|
|
|
* @var mixed |
|
332
|
|
|
* @access public |
|
333
|
|
|
*/ |
|
334
|
|
|
public $on_fetch_url = null; |
|
335
|
|
|
|
|
336
|
|
|
/** |
|
337
|
|
|
* URL属于入口页 |
|
338
|
|
|
* 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 |
|
339
|
|
|
* 主要用来发现新的待爬url, 并且能给新发现的url附加数据 |
|
340
|
|
|
* |
|
341
|
|
|
* @var mixed |
|
342
|
|
|
* @access public |
|
343
|
|
|
*/ |
|
344
|
|
|
public $on_scan_page = null; |
|
345
|
|
|
|
|
346
|
|
|
/** |
|
347
|
|
|
* URL属于列表页 |
|
348
|
|
|
* 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 |
|
349
|
|
|
* 主要用来发现新的待爬url, 并且能给新发现的url附加数据 |
|
350
|
|
|
* |
|
351
|
|
|
* @var mixed |
|
352
|
|
|
* @access public |
|
353
|
|
|
*/ |
|
354
|
|
|
public $on_list_page = null; |
|
355
|
|
|
|
|
356
|
|
|
/** |
|
357
|
|
|
* URL属于内容页 |
|
358
|
|
|
* 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 |
|
359
|
|
|
* 主要用来发现新的待爬url, 并且能给新发现的url附加数据 |
|
360
|
|
|
* |
|
361
|
|
|
* @var mixed |
|
362
|
|
|
* @access public |
|
363
|
|
|
*/ |
|
364
|
|
|
public $on_content_page = null; |
|
365
|
|
|
|
|
366
|
|
|
/** |
|
367
|
|
|
* 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 |
|
368
|
|
|
* |
|
369
|
|
|
* @var mixed |
|
370
|
|
|
* @access public |
|
371
|
|
|
*/ |
|
372
|
|
|
public $on_handle_img = null; |
|
373
|
|
|
|
|
374
|
|
|
/** |
|
375
|
|
|
* 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 |
|
376
|
|
|
* |
|
377
|
|
|
* @var mixed |
|
378
|
|
|
* @access public |
|
379
|
|
|
*/ |
|
380
|
|
|
public $on_extract_field = null; |
|
381
|
|
|
|
|
382
|
|
|
/** |
|
383
|
|
|
* 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 |
|
384
|
|
|
* |
|
385
|
|
|
* @var mixed |
|
386
|
|
|
* @access public |
|
387
|
|
|
*/ |
|
388
|
|
|
public $on_extract_page = null; |
|
389
|
|
|
|
|
390
|
|
|
/** |
|
391
|
|
|
* 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 |
|
392
|
|
|
* 就不去分析他的内容提取field了, 提取field只针对HTML |
|
393
|
|
|
* |
|
394
|
|
|
* @var mixed |
|
395
|
|
|
* @access public |
|
396
|
|
|
*/ |
|
397
|
|
|
public $on_attachment_file = null; |
|
398
|
|
|
|
|
399
|
|
|
public function __construct($configs = array()) |
|
400
|
|
|
{ |
|
401
|
|
|
// 产生时钟云，解决php7下面ctrl+c无法停止bug |
|
402
|
|
|
declare(ticks = 1); |
|
403
|
|
|
|
|
404
|
|
|
// 先打开以显示验证报错内容 |
|
405
|
|
|
log::$log_show = true; |
|
406
|
|
|
log::$log_file = isset($configs['log_file']) ? $configs['log_file'] : PATH_DATA.'/phpspider.log'; |
|
407
|
|
|
log::$log_type = isset($configs['log_type']) ? $configs['log_type'] : false; |
|
408
|
|
|
|
|
409
|
|
|
// 彩蛋 |
|
410
|
|
|
$included_files = get_included_files(); |
|
411
|
|
|
$content = file_get_contents($included_files[0]); |
|
412
|
|
|
if (!preg_match("#/\* Do NOT delete this comment \*/#", $content) || !preg_match("#/\* 不要删除这段注释 \*/#", $content)) |
|
413
|
|
|
{ |
|
414
|
|
|
$msg = "Unknown error..."; |
|
415
|
|
|
log::error($msg); |
|
416
|
|
|
exit; |
|
|
|
|
|
|
417
|
|
|
} |
|
418
|
|
|
|
|
419
|
|
|
$configs['name'] = isset($configs['name']) ? $configs['name'] : 'phpspider'; |
|
420
|
|
|
$configs['proxy'] = isset($configs['proxy']) ? $configs['proxy'] : false; |
|
421
|
|
|
$configs['user_agent'] = isset($configs['user_agent']) ? $configs['user_agent'] : self::AGENT_PC; |
|
422
|
|
|
$configs['client_ip'] = isset($configs['client_ip']) ? $configs['client_ip'] : array(); |
|
423
|
|
|
$configs['interval'] = isset($configs['interval']) ? $configs['interval'] : self::INTERVAL; |
|
424
|
|
|
$configs['timeout'] = isset($configs['timeout']) ? $configs['timeout'] : self::TIMEOUT; |
|
425
|
|
|
$configs['max_try'] = isset($configs['max_try']) ? $configs['max_try'] : self::MAX_TRY; |
|
426
|
|
|
$configs['max_depth'] = isset($configs['max_depth']) ? $configs['max_depth'] : 0; |
|
427
|
|
|
$configs['max_fields'] = isset($configs['max_fields']) ? $configs['max_fields'] : 0; |
|
428
|
|
|
$configs['export'] = isset($configs['export']) ? $configs['export'] : array(); |
|
429
|
|
|
//新增参数 BY KEN <[email protected]> |
|
430
|
|
|
$configs['max_pages'] = isset($configs['max_pages']) ? $configs['max_pages'] : self::$max_pages; |
|
431
|
|
|
$configs['max_duration'] = isset($configs['max_duration']) ? $configs['max_duration'] : self::$max_duration; |
|
432
|
|
|
$configs['max_sub_num'] = isset($configs['max_sub_num']) ? $configs['max_sub_num'] : self::$max_sub_num; |
|
433
|
|
|
$configs['max_stand_by_time'] = isset($configs['max_stand_by_time']) ? $configs['max_stand_by_time'] : self::$max_stand_by_time; |
|
434
|
|
|
$configs['max_task_per_host'] = isset($configs['max_task_per_host']) ? $configs['max_task_per_host'] : self::$max_task_per_host; |
|
435
|
|
|
//启用 host并发上限时，队列参数强制为随机 |
|
436
|
|
|
if ($configs['max_task_per_host'] > 0) |
|
437
|
|
|
{ |
|
438
|
|
|
$configs['queue_order'] = 'rand'; |
|
439
|
|
|
} |
|
440
|
|
|
else |
|
441
|
|
|
{ |
|
442
|
|
|
$configs['queue_order'] = isset($configs['queue_order']) ? $configs['queue_order'] : 'list'; |
|
443
|
|
|
} |
|
444
|
|
|
|
|
445
|
|
|
// csv、sql、db |
|
446
|
|
|
self::$export_type = isset($configs['export']['type']) ? $configs['export']['type'] : ''; |
|
447
|
|
|
self::$export_file = isset($configs['export']['file']) ? $configs['export']['file'] : ''; |
|
448
|
|
|
self::$export_table = isset($configs['export']['table']) ? $configs['export']['table'] : ''; |
|
449
|
|
|
self::$db_config = isset($configs['db_config']) ? $configs['db_config'] : array(); |
|
450
|
|
|
self::$queue_config = isset($configs['queue_config']) ? $configs['queue_config'] : array(); |
|
451
|
|
|
|
|
452
|
|
|
// 是否设置了并发任务数, 并且大于1, 而且不是windows环境 |
|
453
|
|
|
if (isset($configs['tasknum']) && $configs['tasknum'] > 1 && !util::is_win()) |
|
454
|
|
|
{ |
|
455
|
|
|
self::$tasknum = $configs['tasknum']; |
|
456
|
|
|
} |
|
457
|
|
|
|
|
458
|
|
|
// 是否设置了保留运行状态 |
|
459
|
|
|
if (isset($configs['save_running_state'])) |
|
460
|
|
|
{ |
|
461
|
|
|
self::$save_running_state = $configs['save_running_state']; |
|
462
|
|
|
} |
|
463
|
|
|
|
|
464
|
|
|
// 是否分布式 |
|
465
|
|
|
if (isset($configs['multiserver'])) |
|
466
|
|
|
{ |
|
467
|
|
|
self::$multiserver = $configs['multiserver']; |
|
468
|
|
|
} |
|
469
|
|
|
|
|
470
|
|
|
// 当前服务器ID |
|
471
|
|
|
if (isset($configs['serverid'])) |
|
472
|
|
|
{ |
|
473
|
|
|
self::$serverid = $configs['serverid']; |
|
474
|
|
|
} |
|
475
|
|
|
|
|
476
|
|
|
// 不同项目的采集以采集名称作为前缀区分 缩短 spider name md5长度到4位，减少内存占用 |
|
477
|
|
|
if (isset(self::$queue_config['prefix'])) |
|
478
|
|
|
{ |
|
479
|
|
|
self::$queue_config['prefix'] = self::$queue_config['prefix'].'-'.substr(md5($configs['name']), 0, 4); |
|
480
|
|
|
} |
|
481
|
|
|
|
|
482
|
|
|
self::$configs = $configs; |
|
483
|
|
|
} |
|
484
|
|
|
|
|
485
|
|
|
public function get_config($name) |
|
486
|
|
|
{ |
|
487
|
|
|
return empty(self::$configs[$name]) ? array() : self::$configs[$name]; |
|
488
|
|
|
} |
|
489
|
|
|
|
|
490
|
|
|
public function add_scan_url($url, $options = array(), $allowed_repeat = true) |
|
491
|
|
|
{ |
|
492
|
|
|
// 投递状态 |
|
493
|
|
|
$status = false; |
|
494
|
|
|
//限制最大子域名数量 |
|
495
|
|
|
if ( ! empty(self::$configs['max_sub_num'])) |
|
496
|
|
|
{ |
|
497
|
|
|
//抓取到的子域名超过指定数量，就丢掉此域名 |
|
498
|
|
|
$sub_domain_count = $this->sub_domain_count($url); |
|
499
|
|
|
if ($sub_domain_count > self::$configs['max_sub_num']) |
|
500
|
|
|
{ |
|
501
|
|
|
log::debug('Task('.self::$taskid.') subdomin = '.$sub_domain_count.' more than '.self::$configs['max_sub_num'].",add_scan_url $url [Skip]"); |
|
502
|
|
|
return $status; |
|
503
|
|
|
} |
|
504
|
|
|
} |
|
505
|
|
|
|
|
506
|
|
|
$link = $options; |
|
507
|
|
|
$link['url'] = $url; |
|
508
|
|
|
$link['url_type'] = 'scan_page'; |
|
509
|
|
|
$link = $this->link_uncompress($link); |
|
|
|
|
|
|
510
|
|
|
|
|
511
|
|
|
if ($this->is_content_page($url)) |
|
|
|
|
|
|
512
|
|
|
{ |
|
513
|
|
|
$link['url_type'] = 'content_page'; |
|
514
|
|
|
$status = $this->queue_lpush($link, $allowed_repeat); |
|
|
|
|
|
|
515
|
|
|
} |
|
516
|
|
|
elseif ($this->is_list_page($url)) |
|
|
|
|
|
|
517
|
|
|
{ |
|
518
|
|
|
$link['url_type'] = 'list_page'; |
|
519
|
|
|
$status = $this->queue_lpush($link, $allowed_repeat); |
|
|
|
|
|
|
520
|
|
|
} |
|
521
|
|
|
else |
|
522
|
|
|
{ |
|
523
|
|
|
$status = $this->queue_lpush($link, $allowed_repeat); |
|
|
|
|
|
|
524
|
|
|
} |
|
525
|
|
|
|
|
526
|
|
|
if ($status) |
|
|
|
|
|
|
527
|
|
|
{ |
|
528
|
|
|
if ($link['url_type'] == 'scan_page') |
|
529
|
|
|
{ |
|
530
|
|
|
log::debug("Find scan page: {$url}"); |
|
531
|
|
|
} |
|
532
|
|
|
elseif ($link['url_type'] == 'content_page') |
|
533
|
|
|
{ |
|
534
|
|
|
log::debug("Find content page: {$url}"); |
|
535
|
|
|
} |
|
536
|
|
|
elseif ($link['url_type'] == 'list_page') |
|
537
|
|
|
{ |
|
538
|
|
|
log::debug("Find list page: {$url}"); |
|
539
|
|
|
} |
|
540
|
|
|
} |
|
541
|
|
|
|
|
542
|
|
|
return $status; |
|
543
|
|
|
} |
|
544
|
|
|
|
|
545
|
|
|
/** |
|
546
|
|
|
* 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 |
|
547
|
|
|
* 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 |
|
548
|
|
|
* |
|
549
|
|
|
* @param mixed $url |
|
550
|
|
|
* @param mixed $options |
|
551
|
|
|
* @return void |
|
552
|
|
|
* @author seatle <[email protected]> |
|
553
|
|
|
* @created time :2016-09-18 10:17 |
|
554
|
|
|
*/ |
|
555
|
|
|
public function add_url($url, $options = array(), $depth = 0) |
|
556
|
|
|
{ |
|
557
|
|
|
// 投递状态 |
|
558
|
|
|
$status = false; |
|
559
|
|
|
//限制最大子域名数量 |
|
560
|
|
|
if ( ! empty(self::$configs['max_sub_num'])) |
|
561
|
|
|
{ |
|
562
|
|
|
//抓取超过 max_sub_num 子域名的，就丢掉 |
|
563
|
|
|
$sub_domain_count = $this->sub_domain_count($url); |
|
564
|
|
|
if ($sub_domain_count > self::$configs['max_sub_num']) |
|
565
|
|
|
{ |
|
566
|
|
|
log::debug('Task('.self::$taskid.') subdomin = '.$sub_domain_count.' more than '.self::$configs['max_sub_num'].",add_url $url [Skip]"); |
|
567
|
|
|
//echo '[on_download_page] ' . $domain . "'s subdomin > 1000 ,Skip!\n"; |
|
568
|
|
|
return $status; |
|
|
|
|
|
|
569
|
|
|
} |
|
570
|
|
|
} |
|
571
|
|
|
$link = $options; |
|
572
|
|
|
$link['url'] = $url; |
|
573
|
|
|
$link['depth'] = $depth; |
|
574
|
|
|
$link = $this->link_uncompress($link); |
|
|
|
|
|
|
575
|
|
|
|
|
576
|
|
|
if ($this->is_content_page($url)) |
|
|
|
|
|
|
577
|
|
|
{ |
|
578
|
|
|
$link['url_type'] = 'content_page'; |
|
579
|
|
|
$status = $this->queue_lpush($link); |
|
|
|
|
|
|
580
|
|
|
} |
|
581
|
|
|
elseif ($this->is_list_page($url)) |
|
|
|
|
|
|
582
|
|
|
{ |
|
583
|
|
|
$link['url_type'] = 'list_page'; |
|
584
|
|
|
$status = $this->queue_lpush($link); |
|
|
|
|
|
|
585
|
|
|
} |
|
586
|
|
|
|
|
587
|
|
|
if ($status) |
|
|
|
|
|
|
588
|
|
|
{ |
|
589
|
|
|
if ($link['url_type'] == 'scan_page') |
|
590
|
|
|
{ |
|
591
|
|
|
log::debug("Find scan page: {$url}"); |
|
592
|
|
|
} |
|
593
|
|
|
elseif ($link['url_type'] == 'content_page') |
|
594
|
|
|
{ |
|
595
|
|
|
log::debug("Find content page: {$url}"); |
|
596
|
|
|
} |
|
597
|
|
|
elseif ($link['url_type'] == 'list_page') |
|
598
|
|
|
{ |
|
599
|
|
|
log::debug("Find list page: {$url}"); |
|
600
|
|
|
} |
|
601
|
|
|
} |
|
602
|
|
|
|
|
603
|
|
|
return $status; |
|
|
|
|
|
|
604
|
|
|
} |
|
605
|
|
|
|
|
606
|
|
|
/** |
|
607
|
|
|
* 是否入口页面 |
|
608
|
|
|
* |
|
609
|
|
|
* @param mixed $url |
|
610
|
|
|
* @return void |
|
611
|
|
|
* @author seatle <[email protected]> |
|
612
|
|
|
* @created time :2016-10-12 19:06 |
|
613
|
|
|
*/ |
|
614
|
|
|
public function is_scan_page($url) |
|
615
|
|
|
{ |
|
616
|
|
|
$parse_url = parse_url($url); |
|
617
|
|
|
//2018-1-3 通配所有域名 |
|
618
|
|
|
if ( ! empty($parse_url['host']) and self::$configs['domains'][0] == '*') |
|
619
|
|
|
{ |
|
620
|
|
|
return true; |
|
|
|
|
|
|
621
|
|
|
} |
|
622
|
|
|
//限定域名 |
|
623
|
|
|
if (empty($parse_url['host']) || ! in_array($parse_url['host'], self::$configs['domains'])) |
|
624
|
|
|
{ |
|
625
|
|
|
return false; |
|
|
|
|
|
|
626
|
|
|
} |
|
627
|
|
|
return true; |
|
|
|
|
|
|
628
|
|
|
} |
|
629
|
|
|
|
|
630
|
|
|
/** |
|
631
|
|
|
* 是否列表页面 |
|
632
|
|
|
* |
|
633
|
|
|
* @param mixed $url |
|
634
|
|
|
* @return void |
|
635
|
|
|
* @author seatle <[email protected]> |
|
636
|
|
|
* @created time :2016-10-12 19:06 |
|
637
|
|
|
*/ |
|
638
|
|
|
public function is_list_page($url) |
|
639
|
|
|
{ |
|
640
|
|
|
$result = false; |
|
641
|
|
|
//过滤下载类型文件 20180209 |
|
642
|
|
|
if (preg_match('/\.(zip|7z|cab|rar|iso|gho|jar|ace|tar|gz|bz2|z|xml|pdf|doc|txt|rtf|snd|xls|xlsx|docx|apk|ipa|flv|midi|mps|pls|pps|ppa|pwz|mp3|mp4|mpeg|mpe|asf|asx|mpg|3gp|mov|m4v|mkv|vob|vod|mod|ogg|rm|rmvb|wmv|avi|dat|exe|wps|js|css|bmp|jpg|png|gif|ico|tiff|jpeg|svg|webp|mpa|mdb|bin)$/iu', $url)) |
|
643
|
|
|
{ |
|
644
|
|
|
return false; |
|
|
|
|
|
|
645
|
|
|
} |
|
646
|
|
|
|
|
647
|
|
|
//增加 要排除的列表页特征正则 BY KEN <[email protected]> |
|
648
|
|
|
if ( ! empty(self::$configs['list_url_regexes_remove'])) |
|
649
|
|
|
{ |
|
650
|
|
|
foreach (self::$configs['list_url_regexes_remove'] as $regex) |
|
651
|
|
|
{ |
|
652
|
|
|
if (preg_match("#{$regex}#i", $url)) |
|
653
|
|
|
{ |
|
654
|
|
|
return false; |
|
|
|
|
|
|
655
|
|
|
} |
|
656
|
|
|
} |
|
657
|
|
|
} |
|
658
|
|
|
|
|
659
|
|
|
//增加无列表页选项，即所有页面都要抓取内容，包含列表页 |
|
660
|
|
|
if (empty(self::$configs['list_url_regexes']) or self::$configs['list_url_regexes'][0] == 'x') |
|
661
|
|
|
{ |
|
662
|
|
|
return false; |
|
|
|
|
|
|
663
|
|
|
} |
|
664
|
|
|
|
|
665
|
|
|
//增加泛列表页，即所有页面都是列表页，只抓取链接，不抓取内容 |
|
666
|
|
|
if (self::$configs['list_url_regexes'][0] == '*') |
|
667
|
|
|
{ |
|
668
|
|
|
return true; |
|
|
|
|
|
|
669
|
|
|
} |
|
670
|
|
|
|
|
671
|
|
|
if ( ! empty(self::$configs['list_url_regexes'])) |
|
672
|
|
|
{ |
|
673
|
|
|
foreach (self::$configs['list_url_regexes'] as $regex) |
|
674
|
|
|
{ |
|
675
|
|
|
if (preg_match("#{$regex}#i", $url)) |
|
676
|
|
|
{ |
|
677
|
|
|
$result = true; |
|
678
|
|
|
break; |
|
679
|
|
|
} |
|
680
|
|
|
} |
|
681
|
|
|
} |
|
682
|
|
|
return $result; |
|
|
|
|
|
|
683
|
|
|
} |
|
684
|
|
|
|
|
685
|
|
|
/** |
|
686
|
|
|
* 是否内容页面 |
|
687
|
|
|
* |
|
688
|
|
|
* @param mixed $url |
|
689
|
|
|
* @return void |
|
690
|
|
|
* @author seatle <[email protected]> |
|
691
|
|
|
* @created time :2016-10-12 19:06 |
|
692
|
|
|
*/ |
|
693
|
|
|
public function is_content_page($url) |
|
694
|
|
|
{ |
|
695
|
|
|
$result = false; |
|
696
|
|
|
//过滤下载类型文件 20180209 |
|
697
|
|
|
if (preg_match('/\.(zip|7z|cab|rar|iso|gho|jar|ace|tar|gz|bz2|z|xml|pdf|doc|txt|rtf|snd|xls|xlsx|docx|apk|ipa|flv|midi|mps|pls|pps|ppa|pwz|mp3|mp4|mpeg|mpe|asf|asx|mpg|3gp|mov|m4v|mkv|vob|vod|mod|ogg|rm|rmvb|wmv|avi|dat|exe|wps|js|css|bmp|jpg|png|gif|ico|tiff|jpeg|svg|webp|mpa|mdb|bin)$/iu', $url)) |
|
698
|
|
|
{ |
|
699
|
|
|
return false; |
|
|
|
|
|
|
700
|
|
|
} |
|
701
|
|
|
|
|
702
|
|
|
//增加 要排除的内容页特征正则 BY KEN <[email protected]> |
|
703
|
|
|
if ( ! empty(self::$configs['content_url_regexes_remove'])) |
|
704
|
|
|
{ |
|
705
|
|
|
foreach (self::$configs['content_url_regexes_remove'] as $regex) |
|
706
|
|
|
{ |
|
707
|
|
|
if (preg_match("#{$regex}#i", $url)) |
|
708
|
|
|
{ |
|
709
|
|
|
return false; |
|
|
|
|
|
|
710
|
|
|
} |
|
711
|
|
|
} |
|
712
|
|
|
} |
|
713
|
|
|
|
|
714
|
|
|
//增加泛内容模式，即所有页面都要提取内容 |
|
715
|
|
|
if (empty(self::$configs['content_url_regexes']) or self::$configs['content_url_regexes'][0] == '*') |
|
716
|
|
|
{ |
|
717
|
|
|
return true; |
|
|
|
|
|
|
718
|
|
|
} |
|
719
|
|
|
//无内容，泛列表模式，即所有页面都不提取内容 |
|
720
|
|
|
if (self::$configs['content_url_regexes'][0] == 'x') |
|
721
|
|
|
{ |
|
722
|
|
|
return false; |
|
|
|
|
|
|
723
|
|
|
} |
|
724
|
|
|
|
|
725
|
|
|
if ( ! empty(self::$configs['content_url_regexes'])) |
|
726
|
|
|
{ |
|
727
|
|
|
foreach (self::$configs['content_url_regexes'] as $regex) |
|
728
|
|
|
{ |
|
729
|
|
|
if (preg_match("#{$regex}#i", $url)) |
|
730
|
|
|
{ |
|
731
|
|
|
$result = true; |
|
732
|
|
|
break; |
|
733
|
|
|
} |
|
734
|
|
|
} |
|
735
|
|
|
} |
|
736
|
|
|
return $result; |
|
|
|
|
|
|
737
|
|
|
} |
|
738
|
|
|
|
|
739
|
|
|
/** |
|
740
|
|
|
* Parse command. |
|
741
|
|
|
* php yourfile.php start | stop | status | kill |
|
742
|
|
|
* |
|
743
|
|
|
* @return void |
|
744
|
|
|
*/ |
|
745
|
|
|
public function parse_command() |
|
746
|
|
|
{ |
|
747
|
|
|
// 检查运行命令的参数 |
|
748
|
|
|
global $argv; |
|
749
|
|
|
$start_file = $argv[0]; |
|
750
|
|
|
|
|
751
|
|
|
// 命令 |
|
752
|
|
|
$command = isset($argv[1]) ? trim($argv[1]) : 'start'; |
|
753
|
|
|
|
|
754
|
|
|
// 子命令, 目前只支持-d |
|
755
|
|
|
$command2 = isset($argv[2]) ? $argv[2] : ''; |
|
756
|
|
|
|
|
757
|
|
|
// 根据命令做相应处理 |
|
758
|
|
|
switch($command) |
|
759
|
|
|
{ |
|
760
|
|
|
// 启动 phpspider |
|
761
|
|
|
case 'start': |
|
762
|
|
|
if ($command2 === '-d') |
|
763
|
|
|
{ |
|
764
|
|
|
self::$daemonize = true; |
|
765
|
|
|
} |
|
766
|
|
|
break; |
|
767
|
|
|
case 'stop': |
|
768
|
|
|
exec("ps aux | grep $start_file | grep -v grep | awk '{print $2}'", $info); |
|
769
|
|
|
if (count($info) <= 1) |
|
770
|
|
|
{ |
|
771
|
|
|
echo "PHPSpider[$start_file] not run\n"; |
|
772
|
|
|
} |
|
773
|
|
|
else |
|
774
|
|
|
{ |
|
775
|
|
|
//echo "PHPSpider[$start_file] is stoping ...\n"; |
|
776
|
|
|
echo "PHPSpider[$start_file] stop success"; |
|
777
|
|
|
exec("ps aux | grep $start_file | grep -v grep | awk '{print $2}' |xargs kill -SIGINT", $info); |
|
778
|
|
|
} |
|
779
|
|
|
exit; |
|
|
|
|
|
|
780
|
|
|
break; |
|
781
|
|
|
case 'kill': |
|
782
|
|
|
exec("ps aux | grep $start_file | grep -v grep | awk '{print $2}' |xargs kill -SIGKILL"); |
|
783
|
|
|
break; |
|
784
|
|
|
// 显示 phpspider 运行状态 |
|
785
|
|
|
case 'status': |
|
786
|
|
|
exit(0); |
|
|
|
|
|
|
787
|
|
|
// 未知命令 |
|
788
|
|
|
default : |
|
789
|
|
|
exit("Usage: php yourfile.php {start|stop|status|kill}\n"); |
|
|
|
|
|
|
790
|
|
|
} |
|
791
|
|
|
} |
|
792
|
|
|
|
|
793
|
|
|
/** |
|
794
|
|
|
* Signal hander. |
|
795
|
|
|
* |
|
796
|
|
|
* @param int $signal |
|
797
|
|
|
*/ |
|
798
|
|
|
public function signal_handler($signal) |
|
799
|
|
|
{ |
|
800
|
|
|
switch ($signal) |
|
801
|
|
|
{ |
|
802
|
|
|
// Stop. |
|
803
|
|
|
case SIGINT: |
|
804
|
|
|
log::warn('Program stopping...'); |
|
805
|
|
|
self::$terminate = true; |
|
806
|
|
|
break; |
|
807
|
|
|
// Show status. |
|
808
|
|
|
case SIGUSR2: |
|
809
|
|
|
echo "show status\n"; |
|
810
|
|
|
break; |
|
811
|
|
|
} |
|
812
|
|
|
} |
|
813
|
|
|
|
|
814
|
|
|
/** |
|
815
|
|
|
* Install signal handler. |
|
816
|
|
|
* |
|
817
|
|
|
* @return void |
|
818
|
|
|
*/ |
|
819
|
|
|
public function install_signal() |
|
820
|
|
|
{ |
|
821
|
|
|
if (function_exists('pcntl_signal')) |
|
822
|
|
|
{ |
|
823
|
|
|
// stop |
|
824
|
|
|
// static调用方式 |
|
825
|
|
|
//pcntl_signal(SIGINT, array(__CLASS__, 'signal_handler'), false); |
|
826
|
|
|
pcntl_signal(SIGINT, array(&$this, 'signal_handler'), false); |
|
827
|
|
|
// status |
|
828
|
|
|
pcntl_signal(SIGUSR2, array(&$this, 'signal_handler'), false); |
|
829
|
|
|
// ignore |
|
830
|
|
|
pcntl_signal(SIGPIPE, SIG_IGN, false); |
|
831
|
|
|
} |
|
832
|
|
|
} |
|
833
|
|
|
|
|
834
|
|
|
/** |
|
835
|
|
|
* Run as deamon mode. |
|
836
|
|
|
* |
|
837
|
|
|
* @throws Exception |
|
838
|
|
|
*/ |
|
839
|
|
|
protected static function daemonize() |
|
840
|
|
|
{ |
|
841
|
|
|
if (!self::$daemonize) |
|
842
|
|
|
{ |
|
843
|
|
|
return; |
|
844
|
|
|
} |
|
845
|
|
|
|
|
846
|
|
|
// fork前一定要关闭redis |
|
847
|
|
|
queue::clear_link(); |
|
848
|
|
|
|
|
849
|
|
|
umask(0); |
|
850
|
|
|
$pid = pcntl_fork(); |
|
851
|
|
|
if (-1 === $pid) |
|
852
|
|
|
{ |
|
853
|
|
|
throw new Exception('fork fail'); |
|
854
|
|
|
} |
|
855
|
|
|
elseif ($pid > 0) |
|
856
|
|
|
{ |
|
857
|
|
|
exit(0); |
|
|
|
|
|
|
858
|
|
|
} |
|
859
|
|
|
if (-1 === posix_setsid()) |
|
860
|
|
|
{ |
|
861
|
|
|
throw new Exception('setsid fail'); |
|
862
|
|
|
} |
|
863
|
|
|
// Fork again avoid SVR4 system regain the control of terminal. |
|
864
|
|
|
$pid = pcntl_fork(); |
|
865
|
|
|
if (-1 === $pid) |
|
866
|
|
|
{ |
|
867
|
|
|
throw new Exception('fork fail'); |
|
868
|
|
|
} |
|
869
|
|
|
elseif (0 !== $pid) |
|
870
|
|
|
{ |
|
871
|
|
|
exit(0); |
|
|
|
|
|
|
872
|
|
|
} |
|
873
|
|
|
} |
|
874
|
|
|
|
|
875
|
|
|
/** |
|
876
|
|
|
* 检查是否终止当前进程 |
|
877
|
|
|
* |
|
878
|
|
|
* @return void |
|
879
|
|
|
* @author seatle <[email protected]> |
|
880
|
|
|
* @created time :2016-11-16 11:06 |
|
881
|
|
|
*/ |
|
882
|
|
|
public function check_terminate() |
|
883
|
|
|
{ |
|
884
|
|
|
if (!self::$terminate) |
|
885
|
|
|
{ |
|
886
|
|
|
return false; |
|
|
|
|
|
|
887
|
|
|
} |
|
888
|
|
|
|
|
889
|
|
|
// 删除当前任务状态 |
|
890
|
|
|
$this->del_task_status(self::$serverid, self::$taskid); |
|
891
|
|
|
|
|
892
|
|
|
if (self::$taskmaster) |
|
893
|
|
|
{ |
|
894
|
|
|
// 检查子进程是否都退出 |
|
895
|
|
|
while (true) |
|
896
|
|
|
{ |
|
897
|
|
|
$all_stop = true; |
|
898
|
|
|
for ($i = 2; $i <= self::$tasknum; $i++) |
|
899
|
|
|
{ |
|
900
|
|
|
// 只要一个还活着就说明没有完全退出 |
|
901
|
|
|
$task_status = $this->get_task_status(self::$serverid, $i); |
|
|
|
|
|
|
902
|
|
|
if ($task_status) |
|
903
|
|
|
{ |
|
904
|
|
|
$all_stop = false; |
|
905
|
|
|
} |
|
906
|
|
|
} |
|
907
|
|
|
if ($all_stop) |
|
908
|
|
|
{ |
|
909
|
|
|
break; |
|
910
|
|
|
} |
|
911
|
|
|
else |
|
912
|
|
|
{ |
|
913
|
|
|
log::warn('Task stop waiting...'); |
|
914
|
|
|
} |
|
915
|
|
|
sleep(1); |
|
916
|
|
|
} |
|
917
|
|
|
|
|
918
|
|
|
$this->del_server_list(self::$serverid); |
|
919
|
|
|
|
|
920
|
|
|
// 显示最后结果 |
|
921
|
|
|
log::$log_show = true; |
|
922
|
|
|
|
|
923
|
|
|
$spider_time_run = util::time2second(intval(microtime(true) - self::$time_start)); |
|
924
|
|
|
log::note("Spider finished in {$spider_time_run}"); |
|
925
|
|
|
|
|
926
|
|
|
$get_collected_url_num = $this->get_collected_url_num(); |
|
|
|
|
|
|
927
|
|
|
log::note("Total pages: {$get_collected_url_num} \n"); |
|
928
|
|
|
} |
|
929
|
|
|
exit(); |
|
|
|
|
|
|
930
|
|
|
} |
|
931
|
|
|
|
|
932
|
|
|
public function start() |
|
933
|
|
|
{ |
|
934
|
|
|
$this->parse_command(); |
|
935
|
|
|
|
|
936
|
|
|
// 爬虫开始时间 |
|
937
|
|
|
self::$time_start = time(); |
|
938
|
|
|
// 当前任务ID |
|
939
|
|
|
self::$taskid = 1; |
|
940
|
|
|
// 当前任务进程ID |
|
941
|
|
|
self::$taskpid = function_exists('posix_getpid') ? posix_getpid() : 1; |
|
942
|
|
|
self::$collect_succ = 0; |
|
943
|
|
|
self::$collect_fail = 0; |
|
944
|
|
|
|
|
945
|
|
|
//-------------------------------------------------------------------------------- |
|
946
|
|
|
// 运行前验证 |
|
947
|
|
|
//-------------------------------------------------------------------------------- |
|
948
|
|
|
|
|
949
|
|
|
// 检查PHP版本 |
|
950
|
|
|
if (version_compare(PHP_VERSION, '5.3.0', 'lt')) |
|
951
|
|
|
{ |
|
952
|
|
|
log::error('PHP 5.3+ is required, currently installed version is: ' . phpversion()); |
|
953
|
|
|
exit; |
|
|
|
|
|
|
954
|
|
|
} |
|
955
|
|
|
|
|
956
|
|
|
// 检查CURL扩展 |
|
957
|
|
|
if(!function_exists('curl_init')) |
|
958
|
|
|
{ |
|
959
|
|
|
log::error('The curl extension was not found'); |
|
960
|
|
|
exit; |
|
|
|
|
|
|
961
|
|
|
} |
|
962
|
|
|
|
|
963
|
|
|
// 多任务需要pcntl扩展支持 |
|
964
|
|
|
if (self::$tasknum > 1 && !function_exists('pcntl_fork')) |
|
965
|
|
|
{ |
|
966
|
|
|
log::error('Multitasking needs pcntl, the pcntl extension was not found'); |
|
967
|
|
|
exit; |
|
|
|
|
|
|
968
|
|
|
} |
|
969
|
|
|
|
|
970
|
|
|
// 守护进程需要pcntl扩展支持 |
|
971
|
|
|
if (self::$daemonize && !function_exists('pcntl_fork')) |
|
972
|
|
|
{ |
|
973
|
|
|
log::error('Daemonize needs pcntl, the pcntl extension was not found'); |
|
974
|
|
|
exit; |
|
|
|
|
|
|
975
|
|
|
} |
|
976
|
|
|
|
|
977
|
|
|
// 集群、保存运行状态、多任务都需要Redis支持 |
|
978
|
|
|
if ( self::$multiserver || self::$save_running_state || self::$tasknum > 1 ) |
|
979
|
|
|
{ |
|
980
|
|
|
self::$use_redis = true; |
|
981
|
|
|
|
|
982
|
|
|
queue::set_connect('default', self::$queue_config); |
|
983
|
|
|
if (!queue::init()) |
|
984
|
|
|
{ |
|
985
|
|
|
if ( self::$multiserver ) |
|
986
|
|
|
{ |
|
987
|
|
|
log::error('Multiserver needs Redis support, '.queue::$error); |
|
988
|
|
|
exit; |
|
|
|
|
|
|
989
|
|
|
} |
|
990
|
|
|
|
|
991
|
|
|
if ( self::$tasknum > 1 ) |
|
992
|
|
|
{ |
|
993
|
|
|
log::error('Multitasking needs Redis support, '.queue::$error); |
|
994
|
|
|
exit; |
|
|
|
|
|
|
995
|
|
|
} |
|
996
|
|
|
|
|
997
|
|
|
if ( self::$save_running_state ) |
|
998
|
|
|
{ |
|
999
|
|
|
log::error('Spider kept running state needs Redis support, '.queue::$error); |
|
1000
|
|
|
exit; |
|
|
|
|
|
|
1001
|
|
|
} |
|
1002
|
|
|
} |
|
1003
|
|
|
} |
|
1004
|
|
|
|
|
1005
|
|
|
// 检查导出 |
|
1006
|
|
|
$this->check_export(); |
|
1007
|
|
|
|
|
1008
|
|
|
// 检查缓存 |
|
1009
|
|
|
$this->check_cache(); |
|
1010
|
|
|
|
|
1011
|
|
|
// 检查 scan_urls |
|
1012
|
|
|
if (empty(self::$configs['scan_urls'])) |
|
1013
|
|
|
{ |
|
1014
|
|
|
log::error('No scan url to start'); |
|
1015
|
|
|
exit; |
|
|
|
|
|
|
1016
|
|
|
} |
|
1017
|
|
|
|
|
1018
|
|
|
foreach ( self::$configs['scan_urls'] as $url ) |
|
1019
|
|
|
{ |
|
1020
|
|
|
// 只检查配置中的入口URL, 通过 add_scan_url 添加的不检查了. |
|
1021
|
|
|
if (!$this->is_scan_page($url)) |
|
1022
|
|
|
{ |
|
1023
|
|
|
log::error("Domain of scan_urls (\"{$url}\") does not match the domains of the domain name"); |
|
1024
|
|
|
exit; |
|
|
|
|
|
|
1025
|
|
|
} |
|
1026
|
|
|
} |
|
1027
|
|
|
|
|
1028
|
|
|
// windows 下没法显示面板, 强制显示日志 |
|
1029
|
|
|
if (util::is_win()) |
|
1030
|
|
|
{ |
|
1031
|
|
|
self::$configs['name'] = iconv('UTF-8', 'GB2312//IGNORE', self::$configs['name']); |
|
1032
|
|
|
log::$log_show = true; |
|
1033
|
|
|
} |
|
1034
|
|
|
// 守护进程下也显示日志 |
|
1035
|
|
|
elseif (self::$daemonize) |
|
1036
|
|
|
{ |
|
1037
|
|
|
log::$log_show = true; |
|
1038
|
|
|
} |
|
1039
|
|
|
else |
|
1040
|
|
|
{ |
|
1041
|
|
|
log::$log_show = isset(self::$configs['log_show']) ? self::$configs['log_show'] : false; |
|
1042
|
|
|
} |
|
1043
|
|
|
|
|
1044
|
|
|
if (log::$log_show) |
|
1045
|
|
|
{ |
|
1046
|
|
|
global $argv; |
|
1047
|
|
|
$start_file = $argv[0]; |
|
1048
|
|
|
|
|
1049
|
|
|
$header = ''; |
|
1050
|
|
|
if ( ! util::is_win()) |
|
1051
|
|
|
{ |
|
1052
|
|
|
$header .= "\033[33m"; |
|
1053
|
|
|
} |
|
1054
|
|
|
|
|
1055
|
|
|
$header .= "\n[ ".self::$configs['name']." Spider ] is started...\n\n"; |
|
1056
|
|
|
$header .= ' * PHPSpider Version: '.self::VERSION."\n"; |
|
1057
|
|
|
$header .= " * Documentation: https://doc.phpspider.org\n"; |
|
1058
|
|
|
$header .= ' * Task Number: '.self::$tasknum."\n\n"; |
|
1059
|
|
|
$header .= "Input \"php $start_file stop\" to quit. Start success.\n"; |
|
1060
|
|
|
if ( ! util::is_win()) |
|
1061
|
|
|
{ |
|
1062
|
|
|
$header .= "\033[0m"; |
|
1063
|
|
|
} |
|
1064
|
|
|
|
|
1065
|
|
|
log::note($header); |
|
1066
|
|
|
} |
|
1067
|
|
|
|
|
1068
|
|
|
// 如果是守护进程，恢复日志状态 |
|
1069
|
|
|
//if (self::$daemonize) |
|
1070
|
|
|
//{ |
|
1071
|
|
|
//log::$log_show = isset(self::$configs['log_show']) ? self::$configs['log_show'] : false; |
|
1072
|
|
|
//} |
|
1073
|
|
|
|
|
1074
|
|
|
// 多任务和分布式都要清掉, 当然分布式只清自己的 |
|
1075
|
|
|
$this->init_redis(); |
|
1076
|
|
|
|
|
1077
|
|
|
//-------------------------------------------------------------------------------- |
|
1078
|
|
|
// 生成多任务 |
|
1079
|
|
|
//-------------------------------------------------------------------------------- |
|
1080
|
|
|
|
|
1081
|
|
|
// 添加入口URL到队列 |
|
1082
|
|
|
foreach ( self::$configs['scan_urls'] as $url ) |
|
1083
|
|
|
{ |
|
1084
|
|
|
// false 表示不允许重复 |
|
1085
|
|
|
$this->add_scan_url($url, null, false); |
|
1086
|
|
|
} |
|
1087
|
|
|
|
|
1088
|
|
|
// 放这个位置, 可以添加入口页面 |
|
1089
|
|
|
if ($this->on_start) |
|
1090
|
|
|
{ |
|
1091
|
|
|
call_user_func($this->on_start, $this); |
|
1092
|
|
|
} |
|
1093
|
|
|
|
|
1094
|
|
|
if (!self::$daemonize) |
|
1095
|
|
|
{ |
|
1096
|
|
|
if (!log::$log_show) |
|
1097
|
|
|
{ |
|
1098
|
|
|
// 第一次先清屏 |
|
1099
|
|
|
$this->clear_echo(); |
|
1100
|
|
|
|
|
1101
|
|
|
// 先显示一次面板, 然后下面再每次采集成功显示一次 |
|
1102
|
|
|
$this->display_ui(); |
|
1103
|
|
|
} |
|
1104
|
|
|
} |
|
1105
|
|
|
else |
|
1106
|
|
|
{ |
|
1107
|
|
|
$this->daemonize(); |
|
1108
|
|
|
} |
|
1109
|
|
|
|
|
1110
|
|
|
// 安装信号 |
|
1111
|
|
|
$this->install_signal(); |
|
1112
|
|
|
|
|
1113
|
|
|
// 开始采集 |
|
1114
|
|
|
$this->do_collect_page(); |
|
1115
|
|
|
|
|
1116
|
|
|
// 从服务器列表中删除当前服务器信息 |
|
1117
|
|
|
$this->del_server_list(self::$serverid); |
|
1118
|
|
|
} |
|
1119
|
|
|
|
|
1120
|
|
|
/** |
|
1121
|
|
|
* 创建一个子进程 |
|
1122
|
|
|
* @param Worker $worker |
|
|
|
|
|
|
1123
|
|
|
* @throws Exception |
|
1124
|
|
|
*/ |
|
1125
|
|
|
public function fork_one_task($taskid) |
|
1126
|
|
|
{ |
|
1127
|
|
|
$pid = pcntl_fork(); |
|
1128
|
|
|
|
|
1129
|
|
|
// 主进程记录子进程pid |
|
1130
|
|
|
if($pid > 0) |
|
1131
|
|
|
{ |
|
1132
|
|
|
// 暂时没用 |
|
1133
|
|
|
//self::$taskpids[$taskid] = $pid; |
|
1134
|
|
|
} |
|
1135
|
|
|
// 子进程运行 |
|
1136
|
|
|
elseif (0 === $pid) |
|
1137
|
|
|
{ |
|
1138
|
|
|
log::warn("Fork children task({$taskid}) successful..."); |
|
1139
|
|
|
|
|
1140
|
|
|
// 初始化子进程参数 |
|
1141
|
|
|
self::$time_start = microtime(true); |
|
1142
|
|
|
self::$taskid = $taskid; |
|
1143
|
|
|
self::$taskmaster = false; |
|
1144
|
|
|
self::$taskpid = posix_getpid(); |
|
1145
|
|
|
self::$collect_succ = 0; |
|
1146
|
|
|
self::$collect_fail = 0; |
|
1147
|
|
|
|
|
1148
|
|
|
queue::set_connect('default', self::$queue_config); |
|
1149
|
|
|
queue::init(); |
|
1150
|
|
|
|
|
1151
|
|
|
//退出前计时，等待1分钟，如果获取不到新任务，再退出 |
|
1152
|
|
|
self::$stand_by_time = 0; |
|
1153
|
|
|
while (self::$stand_by_time < self::$configs['max_stand_by_time']) |
|
1154
|
|
|
{ |
|
1155
|
|
|
$this->do_collect_page(); |
|
1156
|
|
|
log::warn('Task('.self::$taskid.') Stand By '.self::$stand_by_time.'/'.self::$configs['max_stand_by_time'].' s'); |
|
1157
|
|
|
self::$stand_by_time++; |
|
1158
|
|
|
sleep(1); |
|
1159
|
|
|
} |
|
1160
|
|
|
$queue_lsize = $this->queue_lsize(); |
|
|
|
|
|
|
1161
|
|
|
log::warn('Task('.self::$taskid.') exit : queue_lsize = '.$queue_lsize); |
|
|
|
|
|
|
1162
|
|
|
$this->del_task_status(self::$serverid, $taskid); |
|
1163
|
|
|
|
|
1164
|
|
|
// 这里用0表示子进程正常退出 |
|
1165
|
|
|
exit(0); |
|
|
|
|
|
|
1166
|
|
|
} |
|
1167
|
|
|
else |
|
1168
|
|
|
{ |
|
1169
|
|
|
log::error("Fork children task({$taskid}) fail..."); |
|
1170
|
|
|
exit; |
|
|
|
|
|
|
1171
|
|
|
} |
|
1172
|
|
|
} |
|
1173
|
|
|
|
|
1174
|
|
|
public function do_collect_page() |
|
1175
|
|
|
{ |
|
1176
|
|
|
while( $queue_lsize = $this->queue_lsize() ) |
|
|
|
|
|
|
1177
|
|
|
{ |
|
1178
|
|
|
// 如果是主任务 |
|
1179
|
|
|
if (self::$taskmaster) |
|
1180
|
|
|
{ |
|
1181
|
|
|
// 多任务下主任务未准备就绪 |
|
1182
|
|
|
if (self::$tasknum > 1 && !self::$fork_task_complete) |
|
1183
|
|
|
{ |
|
1184
|
|
|
// 主进程采集到多于任务数2个时, 生成子任务一起采集 |
|
1185
|
|
|
if ($queue_lsize > self::$tasknum + 2) |
|
1186
|
|
|
{ |
|
1187
|
|
|
self::$fork_task_complete = true; |
|
1188
|
|
|
|
|
1189
|
|
|
// fork 子进程前一定要先干掉redis连接fd, 不然会存在进程互抢redis fd 问题 |
|
1190
|
|
|
queue::clear_link(); |
|
1191
|
|
|
// task进程从2开始, 1被master进程所使用 |
|
1192
|
|
|
for ($i = 2; $i <= self::$tasknum; $i++) |
|
1193
|
|
|
{ |
|
1194
|
|
|
$this->fork_one_task($i); |
|
1195
|
|
|
} |
|
1196
|
|
|
} |
|
1197
|
|
|
} |
|
1198
|
|
|
//在主进程中，保存当前配置到缓存，以使子进程可实时读取动态修改后的配置 20180209 |
|
1199
|
|
|
if (self::$use_redis and ! empty(self::$configs)) |
|
1200
|
|
|
{ |
|
1201
|
|
|
queue::set('configs_'.self::$configs['name'], json_encode(self::$configs)); |
|
1202
|
|
|
} |
|
1203
|
|
|
// 抓取页面 |
|
1204
|
|
|
$this->collect_page(); |
|
1205
|
|
|
// 保存任务状态 |
|
1206
|
|
|
$this->set_task_status(); |
|
1207
|
|
|
|
|
1208
|
|
|
// 每采集成功一次页面, 就刷新一次面板 |
|
1209
|
|
|
if (!log::$log_show && !self::$daemonize) |
|
1210
|
|
|
{ |
|
1211
|
|
|
$this->display_ui(); |
|
1212
|
|
|
} |
|
1213
|
|
|
} |
|
1214
|
|
|
// 如果是子任务 |
|
1215
|
|
|
else |
|
1216
|
|
|
{ |
|
1217
|
|
|
// 主进程采集到多于任务数2个时, 子任务可以采集, 否则等待... |
|
1218
|
|
|
if ($queue_lsize > self::$taskid + 2) |
|
1219
|
|
|
{ |
|
1220
|
|
|
//在子进程中，从内存中实时读取当前最新配置，用于适应主进程常驻内存模式，无限循环后的配置变动 20180209 |
|
1221
|
|
|
if (self::$use_redis and ! empty(self::$configs)) |
|
1222
|
|
|
{ |
|
1223
|
|
|
if ($configs_active = queue::get('configs_'.self::$configs['name'])) |
|
|
|
|
|
|
1224
|
|
|
{ |
|
1225
|
|
|
self::$configs = json_decode($configs_active, true); |
|
|
|
|
|
|
1226
|
|
|
} |
|
1227
|
|
|
} |
|
1228
|
|
|
// 抓取页面 |
|
1229
|
|
|
$this->collect_page(); |
|
1230
|
|
|
// 保存任务状态 |
|
1231
|
|
|
$this->set_task_status(); |
|
1232
|
|
|
} |
|
1233
|
|
|
else |
|
1234
|
|
|
{ |
|
1235
|
|
|
log::warn('Task('.self::$taskid.') waiting...reason: queue_lsize = '.$queue_lsize.' < tasknum = '.self::$tasknum); |
|
|
|
|
|
|
1236
|
|
|
sleep(1); |
|
1237
|
|
|
} |
|
1238
|
|
|
} |
|
1239
|
|
|
|
|
1240
|
|
|
// 检查进程是否收到关闭信号 |
|
1241
|
|
|
$this->check_terminate(); |
|
1242
|
|
|
} |
|
1243
|
|
|
} |
|
1244
|
|
|
|
|
1245
|
|
|
/** |
|
1246
|
|
|
* 爬取页面 |
|
1247
|
|
|
* |
|
1248
|
|
|
* @param mixed $collect_url 要抓取的链接 |
|
1249
|
|
|
* @return void |
|
1250
|
|
|
* @author seatle <[email protected]> |
|
1251
|
|
|
* @created time :2016-09-18 10:17 |
|
1252
|
|
|
*/ |
|
1253
|
|
|
public function collect_page() |
|
1254
|
|
|
{ |
|
1255
|
|
|
//减少非必要 queue_lsize 查询 20180214 |
|
1256
|
|
|
if (isset(self::$configs['log_type']) and strstr(self::$configs['log_type'], 'info')) |
|
1257
|
|
|
{ |
|
1258
|
|
|
$get_collect_url_num = $this->get_collect_url_num(); |
|
|
|
|
|
|
1259
|
|
|
log::info('task id: '.self::$taskid." Find pages: {$get_collect_url_num} "); |
|
1260
|
|
|
|
|
1261
|
|
|
$queue_lsize = $this->queue_lsize(); |
|
|
|
|
|
|
1262
|
|
|
log::info('task id: '.self::$taskid." Waiting for collect pages: {$queue_lsize} "); |
|
1263
|
|
|
|
|
1264
|
|
|
$get_collected_url_num = $this->get_collected_url_num(); |
|
|
|
|
|
|
1265
|
|
|
log::info('task id: '.self::$taskid." Collected pages: {$get_collected_url_num} "); |
|
1266
|
|
|
|
|
1267
|
|
|
// 多任务的时候输出爬虫序号 |
|
1268
|
|
|
if (self::$tasknum > 1) |
|
1269
|
|
|
{ |
|
1270
|
|
|
log::info('Current task id: '.self::$taskid); |
|
1271
|
|
|
} |
|
1272
|
|
|
} |
|
1273
|
|
|
//顺序提取任务，先进先出(当配置 queue_order = rand ，先进先出无效，都为随机提取任务) |
|
1274
|
|
|
$link = $this->queue_rpop(); |
|
|
|
|
|
|
1275
|
|
|
|
|
1276
|
|
|
if (empty($link)) |
|
1277
|
|
|
{ |
|
1278
|
|
|
log::warn('Task('.self::$taskid.') Get Task link Fail...Stand By...'); |
|
1279
|
|
|
return false; |
|
|
|
|
|
|
1280
|
|
|
} |
|
1281
|
|
|
$link = $this->link_uncompress($link); |
|
|
|
|
|
|
1282
|
|
|
if (empty($link['url'])) |
|
1283
|
|
|
{ |
|
1284
|
|
|
log::warn('Task('.self::$taskid.') Get Task url Fail...Stand By...'); |
|
1285
|
|
|
return false; |
|
|
|
|
|
|
1286
|
|
|
} |
|
1287
|
|
|
self::$stand_by_time = 0; //接到任务，则超时退出计时重置 |
|
1288
|
|
|
|
|
1289
|
|
|
$url = $link['url']; |
|
1290
|
|
|
|
|
1291
|
|
|
//限制单域名最大url数量 20180213 |
|
1292
|
|
|
if (isset(self::$configs['max_pages']) and self::$configs['max_pages'] > 0) |
|
1293
|
|
|
{ |
|
1294
|
|
|
$domain_pages_num = $this->incr_pages_num($url); |
|
|
|
|
|
|
1295
|
|
|
if ($domain_pages_num > self::$configs['max_pages']) |
|
1296
|
|
|
{ |
|
1297
|
|
|
log::debug('Task('.self::$taskid.') pages = '.$domain_pages_num.' more than '.self::$configs['max_pages'].", $url [Skip]"); |
|
|
|
|
|
|
1298
|
|
|
return false; |
|
|
|
|
|
|
1299
|
|
|
} |
|
1300
|
|
|
} |
|
1301
|
|
|
|
|
1302
|
|
|
//限制单域名最大花费时长 20180213 |
|
1303
|
|
|
if (isset(self::$configs['max_duration']) and self::$configs['max_duration'] > 0) |
|
1304
|
|
|
{ |
|
1305
|
|
|
$domain_duration = $this->get_duration_num($url); |
|
|
|
|
|
|
1306
|
|
|
if ($domain_duration > self::$configs['max_duration']) |
|
1307
|
|
|
{ |
|
1308
|
|
|
log::debug('Task('.self::$taskid.') duration = '.$domain_duration.' more than '.self::$configs['max_duration'].", $url [Skip]"); |
|
|
|
|
|
|
1309
|
|
|
return false; |
|
|
|
|
|
|
1310
|
|
|
} |
|
1311
|
|
|
} |
|
1312
|
|
|
|
|
1313
|
|
|
//当前 host 并发检测 2018-5 BY KEN <[email protected]> |
|
1314
|
|
|
if (self::$configs['max_task_per_host'] > 0) |
|
1315
|
|
|
{ |
|
1316
|
|
|
$task_per_host = $this->get_task_per_host_num($url); |
|
1317
|
|
|
if ($task_per_host < self::$configs['max_task_per_host']) |
|
1318
|
|
|
{ |
|
1319
|
|
|
$task_per_host = $this->incr_task_per_host($url); |
|
|
|
|
|
|
1320
|
|
|
} |
|
1321
|
|
|
else |
|
1322
|
|
|
{ |
|
1323
|
|
|
log::warn('Task('.self::$taskid.') task_per_host = '.$task_per_host.' > '.self::$configs['max_task_per_host'].' ; URL: '.$url.' will be retry later...'); |
|
1324
|
|
|
$this->queue_lpush($link); //放回队列 |
|
1325
|
|
|
usleep(100000); |
|
1326
|
|
|
return false; |
|
|
|
|
|
|
1327
|
|
|
} |
|
1328
|
|
|
} |
|
1329
|
|
|
|
|
1330
|
|
|
// 已采集页面数量 +1 |
|
1331
|
|
|
$this->incr_collected_url_num($url); |
|
1332
|
|
|
|
|
1333
|
|
|
// 爬取页面开始时间 |
|
1334
|
|
|
$page_time_start = microtime(true); |
|
1335
|
|
|
|
|
1336
|
|
|
// 下载页面前执行 |
|
1337
|
|
|
// 比如有时需要根据某个特定的URL，来决定这次的请求是否使用代理 / 或使用哪个代理 |
|
1338
|
|
|
if ($this->on_before_download_page) |
|
1339
|
|
|
{ |
|
1340
|
|
|
$return = call_user_func($this->on_before_download_page, $url, $link, $this); |
|
1341
|
|
|
if (isset($return)) $link = $return; |
|
1342
|
|
|
} |
|
1343
|
|
|
|
|
1344
|
|
|
requests::$input_encoding = null; |
|
1345
|
|
|
$html = $this->request_url($url, $link); |
|
|
|
|
|
|
1346
|
|
|
|
|
1347
|
|
|
//记录速度较慢域名花费抓取时间 20180213 |
|
1348
|
|
|
$time_run = round(microtime(true) - $page_time_start); |
|
1349
|
|
|
if ($time_run > 1) |
|
1350
|
|
|
{ |
|
1351
|
|
|
$this->incr_duration_num($url, $time_run); |
|
1352
|
|
|
} |
|
1353
|
|
|
|
|
1354
|
|
|
// 爬完页面开始处理时间 |
|
1355
|
|
|
$page_time_start = microtime(true); |
|
1356
|
|
|
|
|
1357
|
|
|
if (!$html) |
|
|
|
|
|
|
1358
|
|
|
{ |
|
1359
|
|
|
return false; |
|
|
|
|
|
|
1360
|
|
|
} |
|
1361
|
|
|
// 当前正在爬取的网页页面的对象 |
|
1362
|
|
|
$page = array( |
|
1363
|
|
|
'url' => $url, |
|
1364
|
|
|
'raw' => $html, |
|
1365
|
|
|
'request' => array( |
|
1366
|
|
|
'url' => $url, |
|
1367
|
|
|
'method' => $link['method'], |
|
1368
|
|
|
'headers' => $link['headers'], |
|
1369
|
|
|
'params' => $link['params'], |
|
1370
|
|
|
'context_data' => $link['context_data'], |
|
1371
|
|
|
'try_num' => $link['try_num'], |
|
1372
|
|
|
'max_try' => $link['max_try'], |
|
1373
|
|
|
'depth' => $link['depth'], |
|
1374
|
|
|
'taskid' => self::$taskid, |
|
1375
|
|
|
), |
|
1376
|
|
|
); |
|
1377
|
|
|
//printf("memory usage: %.2f M\n", memory_get_usage() / 1024 / 1024 ); |
|
1378
|
|
|
unset($html); |
|
1379
|
|
|
|
|
1380
|
|
|
//-------------------------------------------------------------------------------- |
|
1381
|
|
|
// 处理回调函数 |
|
1382
|
|
|
//-------------------------------------------------------------------------------- |
|
1383
|
|
|
|
|
1384
|
|
|
// 判断当前网页是否被反爬虫了, 需要开发者实现 |
|
1385
|
|
|
if ($this->is_anti_spider) |
|
1386
|
|
|
{ |
|
1387
|
|
|
$is_anti_spider = call_user_func($this->is_anti_spider, $url, $page['raw'], $this); |
|
1388
|
|
|
// 如果在回调函数里面判断被反爬虫并且返回true |
|
1389
|
|
|
if ($is_anti_spider) |
|
1390
|
|
|
{ |
|
1391
|
|
|
return false; |
|
1392
|
|
|
} |
|
1393
|
|
|
} |
|
1394
|
|
|
|
|
1395
|
|
|
// 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. |
|
1396
|
|
|
// 比如下载了某个网页, 希望向网页的body中添加html标签 |
|
1397
|
|
|
if ($this->on_download_page) |
|
1398
|
|
|
{ |
|
1399
|
|
|
$return = call_user_func($this->on_download_page, $page, $this); |
|
1400
|
|
|
// 针对那些老是忘记return的人 |
|
1401
|
|
|
if (isset($return)) |
|
1402
|
|
|
{ |
|
1403
|
|
|
$page = $return; |
|
1404
|
|
|
} |
|
1405
|
|
|
unset($return); |
|
1406
|
|
|
} |
|
1407
|
|
|
|
|
1408
|
|
|
// 是否从当前页面分析提取URL |
|
1409
|
|
|
// 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 |
|
1410
|
|
|
$is_find_url = true; |
|
1411
|
|
|
if ($link['url_type'] == 'scan_page') |
|
1412
|
|
|
{ |
|
1413
|
|
|
if ($this->on_scan_page) |
|
1414
|
|
|
{ |
|
1415
|
|
|
$return = call_user_func($this->on_scan_page, $page, $page['raw'], $this); |
|
1416
|
|
|
if (isset($return)) |
|
1417
|
|
|
{ |
|
1418
|
|
|
$is_find_url = $return; |
|
1419
|
|
|
} |
|
1420
|
|
|
|
|
1421
|
|
|
unset($return); |
|
1422
|
|
|
} |
|
1423
|
|
|
} |
|
1424
|
|
|
elseif ($link['url_type'] == 'content_page') |
|
1425
|
|
|
{ |
|
1426
|
|
|
if ($this->on_content_page) |
|
1427
|
|
|
{ |
|
1428
|
|
|
$return = call_user_func($this->on_content_page, $page, $page['raw'], $this); |
|
1429
|
|
|
if (isset($return)) |
|
1430
|
|
|
{ |
|
1431
|
|
|
$is_find_url = $return; |
|
1432
|
|
|
} |
|
1433
|
|
|
unset($return); |
|
1434
|
|
|
} |
|
1435
|
|
|
} |
|
1436
|
|
|
elseif ($link['url_type'] == 'list_page') |
|
1437
|
|
|
{ |
|
1438
|
|
|
if ($this->on_list_page) |
|
1439
|
|
|
{ |
|
1440
|
|
|
$return = call_user_func($this->on_list_page, $page, $page['raw'], $this); |
|
1441
|
|
|
if (isset($return)) |
|
1442
|
|
|
{ |
|
1443
|
|
|
$is_find_url = $return; |
|
1444
|
|
|
} |
|
1445
|
|
|
unset($return); |
|
1446
|
|
|
} |
|
1447
|
|
|
} |
|
1448
|
|
|
|
|
1449
|
|
|
// 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 |
|
1450
|
|
|
if ($is_find_url) |
|
1451
|
|
|
{ |
|
1452
|
|
|
// 如果深度没有超过最大深度, 获取下一级URL |
|
1453
|
|
|
if (self::$configs['max_depth'] == 0 || $link['depth'] < self::$configs['max_depth']) |
|
1454
|
|
|
{ |
|
1455
|
|
|
// 分析提取HTML页面中的URL |
|
1456
|
|
|
$this->get_urls($page['raw'], $url, $link['depth'] + 1); |
|
1457
|
|
|
} |
|
1458
|
|
|
} |
|
1459
|
|
|
|
|
1460
|
|
|
// 如果是内容页, 分析提取HTML页面中的字段 |
|
1461
|
|
|
// 列表页也可以提取数据的, source_type: urlcontext, 未实现 |
|
1462
|
|
|
if ($link['url_type'] == 'content_page') |
|
1463
|
|
|
{ |
|
1464
|
|
|
$this->get_html_fields($page['raw'], $url, $page); |
|
1465
|
|
|
} |
|
1466
|
|
|
|
|
1467
|
|
|
// 如果当前深度大于缓存的, 更新缓存 |
|
1468
|
|
|
$this->incr_depth_num($link['depth']); |
|
1469
|
|
|
|
|
1470
|
|
|
// 处理页面耗时时间 |
|
1471
|
|
|
$time_run = round(microtime(true) - $page_time_start, 3); |
|
1472
|
|
|
log::debug('task id: '.self::$taskid." Success process page {$url} in {$time_run} s"); |
|
1473
|
|
|
|
|
1474
|
|
|
$spider_time_run = util::time2second(intval(microtime(true) - self::$time_start)); |
|
1475
|
|
|
log::info('task id: '.self::$taskid." Spider running in {$spider_time_run}"); |
|
1476
|
|
|
|
|
1477
|
|
|
// 爬虫爬取每个网页的时间间隔, 单位: 毫秒 |
|
1478
|
|
|
if (!isset(self::$configs['interval'])) |
|
1479
|
|
|
{ |
|
1480
|
|
|
// 默认睡眠100毫秒, 太快了会被认为是ddos |
|
1481
|
|
|
self::$configs['interval'] = 100; |
|
1482
|
|
|
} |
|
1483
|
|
|
usleep(self::$configs['interval'] * 1000); |
|
1484
|
|
|
} |
|
1485
|
|
|
|
|
1486
|
|
|
/** |
|
1487
|
|
|
* 下载网页, 得到网页内容 |
|
1488
|
|
|
* |
|
1489
|
|
|
* @param mixed $url |
|
1490
|
|
|
* @param mixed $link |
|
1491
|
|
|
* @return void |
|
1492
|
|
|
* @author seatle <[email protected]> |
|
1493
|
|
|
* @created time :2016-09-18 10:17 |
|
1494
|
|
|
*/ |
|
1495
|
|
|
public function request_url($url, $link = array()) |
|
1496
|
|
|
{ |
|
1497
|
|
|
$time_start = microtime(true); |
|
1498
|
|
|
|
|
1499
|
|
|
//$url = "http://www.qiushibaike.com/article/117568316"; |
|
1500
|
|
|
|
|
1501
|
|
|
// 设置了编码就不要让requests去判断了 |
|
1502
|
|
|
if (isset(self::$configs['input_encoding'])) |
|
1503
|
|
|
{ |
|
1504
|
|
|
requests::$input_encoding = self::$configs['input_encoding']; |
|
1505
|
|
|
} |
|
1506
|
|
|
// 得到的编码如果不是utf-8的要转成utf-8, 因为xpath只支持utf-8 |
|
1507
|
|
|
requests::$output_encoding = 'utf-8'; |
|
1508
|
|
|
requests::set_timeout(self::$configs['timeout']); |
|
1509
|
|
|
requests::set_useragent(self::$configs['user_agent']); |
|
1510
|
|
|
|
|
1511
|
|
|
// 先删除伪造IP |
|
1512
|
|
|
requests::del_client_ip(); |
|
1513
|
|
|
// 是否设置了伪造IP |
|
1514
|
|
|
if (self::$configs['client_ip']) |
|
1515
|
|
|
{ |
|
1516
|
|
|
requests::set_client_ip(self::$configs['client_ip']); |
|
1517
|
|
|
} |
|
1518
|
|
|
|
|
1519
|
|
|
// 先删除代理，免得前一个URL的代理被带过来了 |
|
1520
|
|
|
requests::del_proxy(); |
|
1521
|
|
|
// 是否设置了代理 |
|
1522
|
|
|
if ($link['proxy']) |
|
1523
|
|
|
{ |
|
1524
|
|
|
requests::set_proxy($link['proxy']); |
|
1525
|
|
|
} |
|
1526
|
|
|
|
|
1527
|
|
|
// 如何设置了 HTTP Headers |
|
1528
|
|
|
if (!empty($link['headers'])) |
|
1529
|
|
|
{ |
|
1530
|
|
|
foreach ($link['headers'] as $k=>$v) |
|
1531
|
|
|
{ |
|
1532
|
|
|
requests::set_header($k, $v); |
|
1533
|
|
|
} |
|
1534
|
|
|
} |
|
1535
|
|
|
//限制 http 请求模式为 get 或 post |
|
1536
|
|
|
$method = trim(strtolower($link['method'])); |
|
1537
|
|
|
$method = ($method == 'post') ? 'post' : 'get'; |
|
1538
|
|
|
$params = empty($link['params']) ? array() : $link['params']; |
|
1539
|
|
|
$html = requests::$method($url, $params); |
|
1540
|
|
|
// 此url附加的数据不为空, 比如内容页需要列表页一些数据, 拼接到后面去 |
|
1541
|
|
|
if ($html && !empty($link['context_data'])) |
|
1542
|
|
|
{ |
|
1543
|
|
|
$html .= $link['context_data']; |
|
1544
|
|
|
} |
|
1545
|
|
|
|
|
1546
|
|
|
$http_code = requests::$status_code; |
|
1547
|
|
|
|
|
1548
|
|
|
//请求完成 host 的并发计数减 1 2018-5 BY KEN <[email protected]> |
|
1549
|
|
|
if (self::$configs['max_task_per_host'] > 0) |
|
1550
|
|
|
{ |
|
1551
|
|
|
$this->incr_task_per_host($url, 'decr'); |
|
1552
|
|
|
} |
|
1553
|
|
|
|
|
1554
|
|
|
if ($this->on_status_code) |
|
1555
|
|
|
{ |
|
1556
|
|
|
$return = call_user_func($this->on_status_code, $http_code, $url, $html, $this); |
|
1557
|
|
|
if (isset($return)) |
|
1558
|
|
|
{ |
|
1559
|
|
|
$html = $return; |
|
1560
|
|
|
} |
|
1561
|
|
|
unset($return); |
|
1562
|
|
|
if ( ! $html) |
|
1563
|
|
|
{ |
|
1564
|
|
|
return false; |
|
|
|
|
|
|
1565
|
|
|
} |
|
1566
|
|
|
} |
|
1567
|
|
|
|
|
1568
|
|
|
if ($http_code != 200) |
|
1569
|
|
|
{ |
|
1570
|
|
|
// 如果是301、302跳转, 抓取跳转后的网页内容 |
|
1571
|
|
|
if ($http_code == 301 || $http_code == 302) |
|
1572
|
|
|
{ |
|
1573
|
|
|
$info = requests::$info; |
|
1574
|
|
|
//if (isset($info['redirect_url'])) |
|
1575
|
|
|
if (!empty($info['redirect_url'])) |
|
1576
|
|
|
{ |
|
1577
|
|
|
$url = $info['redirect_url']; |
|
1578
|
|
|
requests::$input_encoding = null; |
|
1579
|
|
|
$method = empty($link['method']) ? 'get' : strtolower($link['method']); |
|
1580
|
|
|
$params = empty($link['params']) ? array() : $link['params']; |
|
1581
|
|
|
$html = requests::$method($url, $params); |
|
1582
|
|
|
// 有跳转的就直接获取就好，不要调用自己，容易进入死循环 |
|
1583
|
|
|
//$html = $this->request_url($url, $link); |
|
1584
|
|
|
if ($html && !empty($link['context_data'])) |
|
1585
|
|
|
{ |
|
1586
|
|
|
$html .= $link['context_data']; |
|
1587
|
|
|
} |
|
1588
|
|
|
} |
|
1589
|
|
|
else |
|
1590
|
|
|
{ |
|
1591
|
|
|
return false; |
|
|
|
|
|
|
1592
|
|
|
} |
|
1593
|
|
|
} |
|
1594
|
|
|
else |
|
1595
|
|
|
{ |
|
1596
|
|
|
if ( ! empty(self::$configs['max_try']) and $http_code == 407) |
|
1597
|
|
|
{ |
|
1598
|
|
|
// 扔到队列头部去, 继续采集 |
|
1599
|
|
|
$this->queue_rpush($link); |
|
1600
|
|
|
log::error("Failed to download page {$url}"); |
|
1601
|
|
|
self::$collect_fail++; |
|
1602
|
|
|
} |
|
1603
|
|
|
elseif ( ! empty(self::$configs['max_try']) and in_array($http_code, array('0', '502', '503', '429'))) |
|
1604
|
|
|
{ |
|
1605
|
|
|
// 采集次数加一 |
|
1606
|
|
|
$link['try_num']++; |
|
1607
|
|
|
// 抓取次数 小于 允许抓取失败次数 |
|
1608
|
|
|
if ( $link['try_num'] <= $link['max_try'] ) |
|
1609
|
|
|
{ |
|
1610
|
|
|
// 扔到队列头部去, 继续采集 |
|
1611
|
|
|
$this->queue_rpush($link); |
|
1612
|
|
|
} |
|
1613
|
|
|
log::error("Failed to download page {$url}, retry({$link['try_num']})"); |
|
1614
|
|
|
} |
|
1615
|
|
|
else |
|
1616
|
|
|
{ |
|
1617
|
|
|
log::error("Failed to download page {$url}"); |
|
1618
|
|
|
self::$collect_fail++; |
|
1619
|
|
|
} |
|
1620
|
|
|
log::error("HTTP CODE: {$http_code}"); |
|
1621
|
|
|
return false; |
|
|
|
|
|
|
1622
|
|
|
} |
|
1623
|
|
|
} |
|
1624
|
|
|
|
|
1625
|
|
|
// 爬取页面耗时时间 |
|
1626
|
|
|
$time_run = round(microtime(true) - $time_start, 3); |
|
1627
|
|
|
log::debug("Success download page {$url} in {$time_run} s"); |
|
1628
|
|
|
self::$collect_succ++; |
|
1629
|
|
|
|
|
1630
|
|
|
return $html; |
|
1631
|
|
|
} |
|
1632
|
|
|
|
|
1633
|
|
|
/** |
|
1634
|
|
|
* 分析提取HTML页面中的URL |
|
1635
|
|
|
* |
|
1636
|
|
|
* @param mixed $html HTML内容 |
|
1637
|
|
|
* @param mixed $collect_url 抓取的URL, 用来拼凑完整页面的URL |
|
1638
|
|
|
* @return void |
|
1639
|
|
|
* @author seatle <[email protected]> |
|
1640
|
|
|
* @created time :2016-09-18 10:17 |
|
1641
|
|
|
*/ |
|
1642
|
|
|
public function get_urls($html, $collect_url, $depth = 0) |
|
1643
|
|
|
{ |
|
1644
|
|
|
//-------------------------------------------------------------------------------- |
|
1645
|
|
|
// 正则匹配出页面中的URL |
|
1646
|
|
|
//-------------------------------------------------------------------------------- |
|
1647
|
|
|
$urls = selector::select($html, '//a/@href'); |
|
1648
|
|
|
//preg_match_all("/<a.*href=[\"']{0,1}(.*)[\"']{0,1}[> \r\n\t]{1,}/isU", $html, $matchs); |
|
1649
|
|
|
//$urls = array(); |
|
1650
|
|
|
//if (!empty($matchs[1])) |
|
1651
|
|
|
//{ |
|
1652
|
|
|
//foreach ($matchs[1] as $url) |
|
1653
|
|
|
//{ |
|
1654
|
|
|
//$urls[] = str_replace(array("\"", "'",'&'), array("",'','&'), $url); |
|
1655
|
|
|
//} |
|
1656
|
|
|
//} |
|
1657
|
|
|
|
|
1658
|
|
|
if (empty($urls)) |
|
1659
|
|
|
{ |
|
1660
|
|
|
return false; |
|
|
|
|
|
|
1661
|
|
|
} |
|
1662
|
|
|
|
|
1663
|
|
|
// 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 |
|
1664
|
|
|
if (!is_array($urls)) |
|
1665
|
|
|
{ |
|
1666
|
|
|
$urls = array($urls); |
|
1667
|
|
|
} |
|
1668
|
|
|
|
|
1669
|
|
|
foreach ($urls as $key=>$url) |
|
1670
|
|
|
{ |
|
1671
|
|
|
//限制最大子域名数量 |
|
1672
|
|
|
if ( ! empty(self::$configs['max_sub_num'])) |
|
1673
|
|
|
{ |
|
1674
|
|
|
//抓取子域名超过超过指定值，就丢掉 |
|
1675
|
|
|
$sub_domain_count = $this->sub_domain_count($url); |
|
1676
|
|
|
if ($sub_domain_count > self::$configs['max_sub_num']) |
|
1677
|
|
|
{ |
|
1678
|
|
|
unset($urls[$key]); |
|
1679
|
|
|
log::debug('Task('.self::$taskid.') subdomin = '.$sub_domain_count.' more than '.self::$configs['max_sub_num'].",get_urls $url [Skip]"); |
|
1680
|
|
|
continue; |
|
1681
|
|
|
} |
|
1682
|
|
|
} |
|
1683
|
|
|
$urls[$key] = str_replace(array('"', "'", '&'), array('', '', '&'), $url); |
|
1684
|
|
|
} |
|
1685
|
|
|
|
|
1686
|
|
|
//-------------------------------------------------------------------------------- |
|
1687
|
|
|
// 过滤和拼凑URL |
|
1688
|
|
|
//-------------------------------------------------------------------------------- |
|
1689
|
|
|
// 去除重复的URL |
|
1690
|
|
|
$urls = array_unique($urls); |
|
1691
|
|
|
foreach ($urls as $k=>$url) |
|
1692
|
|
|
{ |
|
1693
|
|
|
$url = trim($url); |
|
1694
|
|
|
if (empty($url)) |
|
1695
|
|
|
{ |
|
1696
|
|
|
continue; |
|
1697
|
|
|
} |
|
1698
|
|
|
|
|
1699
|
|
|
$val = $this->fill_url($url, $collect_url); |
|
|
|
|
|
|
1700
|
|
|
|
|
1701
|
|
|
//限制单域名最大url数量 20180213 |
|
1702
|
|
|
if ($val and isset(self::$configs['max_pages']) and self::$configs['max_pages'] > 0) |
|
1703
|
|
|
{ |
|
1704
|
|
|
$domain_pages_num = $this->incr_pages_num($val); |
|
|
|
|
|
|
1705
|
|
|
if ($domain_pages_num > self::$configs['max_pages']) |
|
1706
|
|
|
{ |
|
1707
|
|
|
continue; |
|
1708
|
|
|
} |
|
1709
|
|
|
} |
|
1710
|
|
|
|
|
1711
|
|
|
if ($val) |
|
1712
|
|
|
{ |
|
1713
|
|
|
$urls[$k] = $val; |
|
1714
|
|
|
} |
|
1715
|
|
|
else |
|
1716
|
|
|
{ |
|
1717
|
|
|
unset($urls[$k]); |
|
1718
|
|
|
} |
|
1719
|
|
|
} |
|
1720
|
|
|
|
|
1721
|
|
|
if (empty($urls)) |
|
1722
|
|
|
{ |
|
1723
|
|
|
return false; |
|
|
|
|
|
|
1724
|
|
|
} |
|
1725
|
|
|
|
|
1726
|
|
|
//-------------------------------------------------------------------------------- |
|
1727
|
|
|
// 把抓取到的URL放入队列 |
|
1728
|
|
|
//-------------------------------------------------------------------------------- |
|
1729
|
|
|
foreach ($urls as $url) |
|
1730
|
|
|
{ |
|
1731
|
|
|
if ($this->on_fetch_url) |
|
1732
|
|
|
{ |
|
1733
|
|
|
$return = call_user_func($this->on_fetch_url, $url, $this); |
|
1734
|
|
|
$url = isset($return) ? $return : $url; |
|
1735
|
|
|
unset($return); |
|
1736
|
|
|
|
|
1737
|
|
|
// 如果 on_fetch_url 返回 false，此URL不入队列 |
|
1738
|
|
|
if (!$url) |
|
1739
|
|
|
{ |
|
1740
|
|
|
continue; |
|
1741
|
|
|
} |
|
1742
|
|
|
} |
|
1743
|
|
|
|
|
1744
|
|
|
// 把当前页当做找到的url的Referer页 |
|
1745
|
|
|
$options = array( |
|
1746
|
|
|
'headers' => array( |
|
1747
|
|
|
'Referer' => $collect_url, |
|
1748
|
|
|
) |
|
1749
|
|
|
); |
|
1750
|
|
|
$this->add_url($url, $options, $depth); |
|
1751
|
|
|
} |
|
1752
|
|
|
} |
|
1753
|
|
|
|
|
1754
|
|
|
/** |
|
1755
|
|
|
* 获得完整的连接地址 |
|
1756
|
|
|
* |
|
1757
|
|
|
* @param mixed $url 要检查的URL |
|
1758
|
|
|
* @param mixed $collect_url 从那个URL页面得到上面的URL |
|
1759
|
|
|
* @return void |
|
1760
|
|
|
* @author seatle <[email protected]> |
|
1761
|
|
|
* @created time :2016-09-23 17:13 |
|
1762
|
|
|
*/ |
|
1763
|
|
|
public function fill_url($url, $collect_url) |
|
1764
|
|
|
{ |
|
1765
|
|
|
$url = trim($url); |
|
1766
|
|
|
$collect_url = trim($collect_url); |
|
1767
|
|
|
|
|
1768
|
|
|
// 排除JavaScript的连接 |
|
1769
|
|
|
//if (strpos($url, "javascript:") !== false) |
|
1770
|
|
|
if (preg_match("@^(mailto|javascript:|#|'|\")@i", $url) || $url == '') |
|
1771
|
|
|
{ |
|
1772
|
|
|
return false; |
|
|
|
|
|
|
1773
|
|
|
} |
|
1774
|
|
|
// 排除没有被解析成功的语言标签 |
|
1775
|
|
|
if (substr($url, 0, 3) == '<%=' or substr($url, 0, 1) == '{' or substr($url, 0, 2) == ' {') |
|
1776
|
|
|
// if(substr($url, 0, 3) == '<%=') |
|
1777
|
|
|
{ |
|
1778
|
|
|
return false; |
|
|
|
|
|
|
1779
|
|
|
} |
|
1780
|
|
|
|
|
1781
|
|
|
$parse_url = @parse_url($collect_url); |
|
1782
|
|
|
if (empty($parse_url['scheme']) || empty($parse_url['host'])) |
|
1783
|
|
|
{ |
|
1784
|
|
|
return false; |
|
|
|
|
|
|
1785
|
|
|
} |
|
1786
|
|
|
// 过滤mailto、tel、sms、wechat、sinaweibo、weixin等协议 |
|
1787
|
|
|
if ( ! in_array($parse_url['scheme'], array('http', 'https'))) |
|
1788
|
|
|
{ |
|
1789
|
|
|
return false; |
|
|
|
|
|
|
1790
|
|
|
} |
|
1791
|
|
|
$scheme = $parse_url['scheme']; |
|
1792
|
|
|
$domain = $parse_url['host']; |
|
1793
|
|
|
$path = empty($parse_url['path']) ? '' : $parse_url['path']; |
|
1794
|
|
|
$base_url_path = $domain.$path; |
|
1795
|
|
|
$base_url_path = preg_replace("/\/([^\/]*)\.(.*)$/", '/', $base_url_path); |
|
1796
|
|
|
$base_url_path = preg_replace("/\/$/", '', $base_url_path); |
|
1797
|
|
|
$i = $path_step = 0; |
|
1798
|
|
|
$dstr = $pstr = ''; |
|
|
|
|
|
|
1799
|
|
|
$pos = strpos($url, '#'); |
|
1800
|
|
|
if ($pos > 0) |
|
1801
|
|
|
{ |
|
1802
|
|
|
// 去掉#和后面的字符串 |
|
1803
|
|
|
$url = substr($url, 0, $pos); |
|
1804
|
|
|
} |
|
1805
|
|
|
|
|
1806
|
|
|
// 修正url格式为 //www.jd.com/111.html 为正确的http |
|
1807
|
|
|
if (substr($url, 0, 2) == '//') |
|
1808
|
|
|
{ |
|
1809
|
|
|
$url = preg_replace('/^\/\//iu', '', $url); |
|
1810
|
|
|
} |
|
1811
|
|
|
// /1234.html |
|
1812
|
|
|
elseif($url[0] == '/') |
|
1813
|
|
|
{ |
|
1814
|
|
|
$url = $domain.$url; |
|
1815
|
|
|
} |
|
1816
|
|
|
// ./1234.html、../1234.html 这种类型的 |
|
1817
|
|
|
elseif($url[0] == '.') |
|
1818
|
|
|
{ |
|
1819
|
|
|
if(!isset($url[2])) |
|
1820
|
|
|
{ |
|
1821
|
|
|
return false; |
|
|
|
|
|
|
1822
|
|
|
} |
|
1823
|
|
|
else |
|
1824
|
|
|
{ |
|
1825
|
|
|
$urls = explode('/',$url); |
|
1826
|
|
|
foreach($urls as $u) |
|
1827
|
|
|
{ |
|
1828
|
|
|
if( $u == '..' ) |
|
1829
|
|
|
{ |
|
1830
|
|
|
$path_step++; |
|
1831
|
|
|
} |
|
1832
|
|
|
// 遇到 ., 不知道为什么不直接写$u == '.', 貌似一样的 |
|
1833
|
|
|
else if( $i < count($urls)-1 ) |
|
1834
|
|
|
{ |
|
1835
|
|
|
$dstr .= $urls[$i].'/'; |
|
1836
|
|
|
} |
|
1837
|
|
|
else |
|
1838
|
|
|
{ |
|
1839
|
|
|
$dstr .= $urls[$i]; |
|
1840
|
|
|
} |
|
1841
|
|
|
$i++; |
|
1842
|
|
|
} |
|
1843
|
|
|
$urls = explode('/',$base_url_path); |
|
1844
|
|
|
if(count($urls) <= $path_step) |
|
1845
|
|
|
{ |
|
1846
|
|
|
return false; |
|
|
|
|
|
|
1847
|
|
|
} |
|
1848
|
|
|
else |
|
1849
|
|
|
{ |
|
1850
|
|
|
$pstr = ''; |
|
1851
|
|
|
for($i=0;$i<count($urls)-$path_step;$i++){ $pstr .= $urls[$i].'/'; } |
|
1852
|
|
|
$url = $pstr.$dstr; |
|
1853
|
|
|
} |
|
1854
|
|
|
} |
|
1855
|
|
|
} |
|
1856
|
|
|
else |
|
1857
|
|
|
{ |
|
1858
|
|
|
if( strtolower(substr($url, 0, 7))=='http://' ) |
|
1859
|
|
|
{ |
|
1860
|
|
|
$url = preg_replace('#^http://#i', '', $url); |
|
1861
|
|
|
$scheme = 'http'; |
|
1862
|
|
|
} |
|
1863
|
|
|
else if( strtolower(substr($url, 0, 8))=='https://' ) |
|
1864
|
|
|
{ |
|
1865
|
|
|
$url = preg_replace('#^https://#i','',$url); |
|
1866
|
|
|
$scheme = "https"; |
|
1867
|
|
|
} |
|
1868
|
|
|
// 相对路径，像 1111.html 这种 |
|
1869
|
|
|
else |
|
1870
|
|
|
{ |
|
1871
|
|
|
$arr = explode("/", $base_url_path); |
|
1872
|
|
|
// 去掉空值 |
|
1873
|
|
|
$arr = array_filter($arr); |
|
1874
|
|
|
$base_url_path = implode("/", $arr); |
|
1875
|
|
|
$url = $base_url_path.'/'.$url; |
|
1876
|
|
|
} |
|
1877
|
|
|
} |
|
1878
|
|
|
// 两个 / 或以上的替换成一个 / |
|
1879
|
|
|
$url = preg_replace('/\/{1,}/i', '/', $url); |
|
1880
|
|
|
$url = $scheme.'://'.$url; |
|
1881
|
|
|
|
|
1882
|
|
|
$parse_url = @parse_url($url); |
|
1883
|
|
|
$domain = empty($parse_url['host']) ? $domain : $parse_url['host']; |
|
|
|
|
|
|
1884
|
|
|
// 如果host不为空, 判断是不是要爬取的域名 |
|
1885
|
|
|
if ( ! empty($parse_url['host'])) |
|
1886
|
|
|
{ |
|
1887
|
|
|
//2018-1-3 通配所有域名 |
|
1888
|
|
|
if (empty(self::$configs['domains']) or self::$configs['domains'][0] == '*') |
|
1889
|
|
|
{ |
|
1890
|
|
|
return $url; |
|
|
|
|
|
|
1891
|
|
|
} |
|
1892
|
|
|
//排除非域名下的url以提高爬取速度 |
|
1893
|
|
|
if (!in_array($parse_url['host'], self::$configs['domains'])) |
|
1894
|
|
|
{ |
|
1895
|
|
|
return false; |
|
|
|
|
|
|
1896
|
|
|
} |
|
1897
|
|
|
} |
|
1898
|
|
|
|
|
1899
|
|
|
return $url; |
|
|
|
|
|
|
1900
|
|
|
} |
|
1901
|
|
|
|
|
1902
|
|
|
/** |
|
1903
|
|
|
* 连接对象压缩 |
|
1904
|
|
|
* |
|
1905
|
|
|
* @return void |
|
1906
|
|
|
* @author seatle <[email protected]> |
|
1907
|
|
|
* @created time :2016-11-05 18:58 |
|
1908
|
|
|
*/ |
|
1909
|
|
|
public function link_compress($link) |
|
1910
|
|
|
{ |
|
1911
|
|
|
if (empty($link['url_type'])) |
|
1912
|
|
|
{ |
|
1913
|
|
|
unset($link['url_type']); |
|
1914
|
|
|
} |
|
1915
|
|
|
|
|
1916
|
|
|
if (empty($link['method']) || strtolower($link['method']) == 'get') |
|
1917
|
|
|
{ |
|
1918
|
|
|
unset($link['method']); |
|
1919
|
|
|
} |
|
1920
|
|
|
|
|
1921
|
|
|
if (empty($link['headers'])) |
|
1922
|
|
|
{ |
|
1923
|
|
|
unset($link['headers']); |
|
1924
|
|
|
} |
|
1925
|
|
|
|
|
1926
|
|
|
if (empty($link['params'])) |
|
1927
|
|
|
{ |
|
1928
|
|
|
unset($link['params']); |
|
1929
|
|
|
} |
|
1930
|
|
|
|
|
1931
|
|
|
if (empty($link['context_data'])) |
|
1932
|
|
|
{ |
|
1933
|
|
|
unset($link['context_data']); |
|
1934
|
|
|
} |
|
1935
|
|
|
|
|
1936
|
|
|
if (empty($link['proxy'])) |
|
1937
|
|
|
{ |
|
1938
|
|
|
unset($link['proxy']); |
|
1939
|
|
|
} |
|
1940
|
|
|
|
|
1941
|
|
|
if (empty($link['try_num'])) |
|
1942
|
|
|
{ |
|
1943
|
|
|
unset($link['try_num']); |
|
1944
|
|
|
} |
|
1945
|
|
|
|
|
1946
|
|
|
if (empty($link['max_try'])) |
|
1947
|
|
|
{ |
|
1948
|
|
|
unset($link['max_try']); |
|
1949
|
|
|
} |
|
1950
|
|
|
|
|
1951
|
|
|
if (empty($link['depth'])) |
|
1952
|
|
|
{ |
|
1953
|
|
|
unset($link['depth']); |
|
1954
|
|
|
} |
|
1955
|
|
|
//$json = json_encode($link); |
|
1956
|
|
|
//$json = gzdeflate($json); |
|
1957
|
|
|
return $link; |
|
1958
|
|
|
} |
|
1959
|
|
|
|
|
1960
|
|
|
/** |
|
1961
|
|
|
* 连接对象解压缩 |
|
1962
|
|
|
* |
|
1963
|
|
|
* @param mixed $link |
|
1964
|
|
|
* @return void |
|
1965
|
|
|
* @author seatle <[email protected]> |
|
1966
|
|
|
* @created time :2016-11-05 18:58 |
|
1967
|
|
|
*/ |
|
1968
|
|
|
public function link_uncompress($link) |
|
1969
|
|
|
{ |
|
1970
|
|
|
$link = array( |
|
1971
|
|
|
'url' => isset($link['url']) ? $link['url'] : '', |
|
1972
|
|
|
'url_type' => isset($link['url_type']) ? $link['url_type'] : '', |
|
1973
|
|
|
'method' => isset($link['method']) ? $link['method'] : 'get', |
|
1974
|
|
|
'headers' => isset($link['headers']) ? $link['headers'] : array(), |
|
1975
|
|
|
'params' => isset($link['params']) ? $link['params'] : array(), |
|
1976
|
|
|
'context_data' => isset($link['context_data']) ? $link['context_data'] : '', |
|
1977
|
|
|
'proxy' => isset($link['proxy']) ? $link['proxy'] : self::$configs['proxy'], |
|
1978
|
|
|
'try_num' => isset($link['try_num']) ? $link['try_num'] : 0, |
|
1979
|
|
|
'max_try' => isset($link['max_try']) ? $link['max_try'] : self::$configs['max_try'], |
|
1980
|
|
|
'depth' => isset($link['depth']) ? $link['depth'] : 0, |
|
1981
|
|
|
); |
|
1982
|
|
|
|
|
1983
|
|
|
return $link; |
|
1984
|
|
|
} |
|
1985
|
|
|
|
|
1986
|
|
|
/** |
|
1987
|
|
|
* 分析提取HTML页面中的字段 |
|
1988
|
|
|
* |
|
1989
|
|
|
* @param mixed $html |
|
1990
|
|
|
* @return void |
|
1991
|
|
|
* @author seatle <[email protected]> |
|
1992
|
|
|
* @created time :2016-09-18 10:17 |
|
1993
|
|
|
*/ |
|
1994
|
|
|
public function get_html_fields($html, $url, $page) |
|
1995
|
|
|
{ |
|
1996
|
|
|
$fields = $this->get_fields(self::$configs['fields'], $html, $url, $page); |
|
|
|
|
|
|
1997
|
|
|
|
|
1998
|
|
|
if (!empty($fields)) |
|
1999
|
|
|
{ |
|
2000
|
|
|
if ($this->on_extract_page) |
|
2001
|
|
|
{ |
|
2002
|
|
|
$return = call_user_func($this->on_extract_page, $page, $fields); |
|
2003
|
|
|
if (!isset($return)) |
|
2004
|
|
|
{ |
|
2005
|
|
|
log::warn("on_extract_page return value can't be empty"); |
|
2006
|
|
|
} |
|
2007
|
|
|
// 返回false，跳过当前页面，内容不入库 |
|
2008
|
|
|
elseif ($return === false) |
|
2009
|
|
|
{ |
|
2010
|
|
|
return false; |
|
2011
|
|
|
} |
|
2012
|
|
|
elseif (!is_array($return)) |
|
2013
|
|
|
{ |
|
2014
|
|
|
log::warn('on_extract_page return value must be an array'); |
|
2015
|
|
|
} |
|
2016
|
|
|
else |
|
2017
|
|
|
{ |
|
2018
|
|
|
$fields = $return; |
|
2019
|
|
|
} |
|
2020
|
|
|
} |
|
2021
|
|
|
|
|
2022
|
|
|
if (isset($fields) && is_array($fields)) |
|
2023
|
|
|
{ |
|
2024
|
|
|
$fields_num = $this->incr_fields_num(); |
|
2025
|
|
|
if (self::$configs['max_fields'] != 0 && $fields_num > self::$configs['max_fields']) |
|
2026
|
|
|
{ |
|
2027
|
|
|
exit(0); |
|
|
|
|
|
|
2028
|
|
|
} |
|
2029
|
|
|
|
|
2030
|
|
|
if (version_compare(PHP_VERSION,'5.4.0','<')) |
|
2031
|
|
|
{ |
|
2032
|
|
|
$fields_str = json_encode($fields); |
|
2033
|
|
|
$fields_str = preg_replace_callback("#\\\u([0-9a-f]{4})#i", function ($matchs) |
|
2034
|
|
|
{ |
|
2035
|
|
|
return @iconv('UCS-2BE', 'UTF-8', pack('H4', $matchs[1])); |
|
2036
|
|
|
}, $fields_str); |
|
2037
|
|
|
} |
|
2038
|
|
|
else |
|
2039
|
|
|
{ |
|
2040
|
|
|
$fields_str = json_encode($fields, JSON_UNESCAPED_UNICODE); |
|
2041
|
|
|
} |
|
2042
|
|
|
|
|
2043
|
|
|
if (util::is_win()) |
|
2044
|
|
|
{ |
|
2045
|
|
|
$fields_str = mb_convert_encoding($fields_str, 'gb2312', 'utf-8'); |
|
2046
|
|
|
} |
|
2047
|
|
|
log::info("Result[{$fields_num}]: ".$fields_str); |
|
2048
|
|
|
|
|
2049
|
|
|
// 如果设置了导出选项 |
|
2050
|
|
|
if (!empty(self::$configs['export'])) |
|
2051
|
|
|
{ |
|
2052
|
|
|
self::$export_type = isset(self::$configs['export']['type']) ? self::$configs['export']['type'] : ''; |
|
2053
|
|
|
if (self::$export_type == 'csv') |
|
2054
|
|
|
{ |
|
2055
|
|
|
util::put_file(self::$export_file, util::format_csv($fields)."\n", FILE_APPEND); |
|
2056
|
|
|
} |
|
2057
|
|
|
elseif (self::$export_type == 'sql') |
|
2058
|
|
|
{ |
|
2059
|
|
|
$sql = db::insert(self::$export_table, $fields, true); |
|
2060
|
|
|
util::put_file(self::$export_file, $sql.";\n", FILE_APPEND); |
|
2061
|
|
|
} |
|
2062
|
|
|
elseif (self::$export_type == 'db') |
|
2063
|
|
|
{ |
|
2064
|
|
|
db::insert(self::$export_table, $fields); |
|
2065
|
|
|
} |
|
2066
|
|
|
} |
|
2067
|
|
|
} |
|
2068
|
|
|
} |
|
2069
|
|
|
} |
|
2070
|
|
|
|
|
2071
|
|
|
/** |
|
2072
|
|
|
* 根据配置提取HTML代码块中的字段 |
|
2073
|
|
|
* |
|
2074
|
|
|
* @param mixed $confs |
|
2075
|
|
|
* @param mixed $html |
|
2076
|
|
|
* @param mixed $page |
|
2077
|
|
|
* @return void |
|
2078
|
|
|
* @author seatle <[email protected]> |
|
2079
|
|
|
* @created time :2016-09-23 17:13 |
|
2080
|
|
|
*/ |
|
2081
|
|
|
public function get_fields($confs, $html, $url, $page) |
|
2082
|
|
|
{ |
|
2083
|
|
|
$fields = array(); |
|
2084
|
|
|
foreach ($confs as $conf) |
|
2085
|
|
|
{ |
|
2086
|
|
|
// 当前field抽取到的内容是否是有多项 |
|
2087
|
|
|
$repeated = isset($conf['repeated']) && $conf['repeated'] ? true : false; |
|
2088
|
|
|
// 当前field抽取到的内容是否必须有值 |
|
2089
|
|
|
$required = isset($conf['required']) && $conf['required'] ? true : false; |
|
2090
|
|
|
|
|
2091
|
|
|
if (empty($conf['name'])) |
|
2092
|
|
|
{ |
|
2093
|
|
|
log::error("The field name is null, please check your \"fields\" and add the name of the field\n"); |
|
2094
|
|
|
exit; |
|
|
|
|
|
|
2095
|
|
|
} |
|
2096
|
|
|
|
|
2097
|
|
|
$values = NULL; |
|
2098
|
|
|
// 如果定义抽取规则 |
|
2099
|
|
|
if (!empty($conf['selector'])) |
|
2100
|
|
|
{ |
|
2101
|
|
|
// 如果这个field是上一个field的附带连接 |
|
2102
|
|
|
if (isset($conf['source_type']) && $conf['source_type']=='attached_url') |
|
2103
|
|
|
{ |
|
2104
|
|
|
// 取出上个field的内容作为连接, 内容分页是不进队列直接下载网页的 |
|
2105
|
|
|
if (!empty($fields[$conf['attached_url']])) |
|
2106
|
|
|
{ |
|
2107
|
|
|
$collect_url = $this->fill_url($fields[$conf['attached_url']], $url); |
|
|
|
|
|
|
2108
|
|
|
log::debug("Find attached content page: {$collect_url}"); |
|
2109
|
|
|
$link['url'] = $collect_url; |
|
2110
|
|
|
$link = $this->link_uncompress($link); |
|
|
|
|
|
|
2111
|
|
|
requests::$input_encoding = null; |
|
2112
|
|
|
$html = $this->request_url($collect_url, $link); |
|
|
|
|
|
|
2113
|
|
|
// 在一个attached_url对应的网页下载完成之后调用. 主要用来对下载的网页进行处理. |
|
2114
|
|
|
if ($this->on_download_attached_page) |
|
2115
|
|
|
{ |
|
2116
|
|
|
$return = call_user_func($this->on_download_attached_page, $html, $this); |
|
2117
|
|
|
if (isset($return)) |
|
2118
|
|
|
{ |
|
2119
|
|
|
$html = $return; |
|
2120
|
|
|
} |
|
2121
|
|
|
} |
|
2122
|
|
|
|
|
2123
|
|
|
// 请求获取完分页数据后把连接删除了 |
|
2124
|
|
|
unset($fields[$conf['attached_url']]); |
|
2125
|
|
|
} |
|
2126
|
|
|
} |
|
2127
|
|
|
|
|
2128
|
|
|
// 没有设置抽取规则的类型 或者 设置为 xpath |
|
2129
|
|
|
if (!isset($conf['selector_type']) || $conf['selector_type']=='xpath') |
|
2130
|
|
|
{ |
|
2131
|
|
|
// 如果找不到，返回的是false |
|
2132
|
|
|
$values = $this->get_fields_xpath($html, $conf['selector'], $conf['name']); |
|
|
|
|
|
|
2133
|
|
|
} |
|
2134
|
|
|
elseif ($conf['selector_type']=='css') |
|
2135
|
|
|
{ |
|
2136
|
|
|
$values = $this->get_fields_css($html, $conf['selector'], $conf['name']); |
|
|
|
|
|
|
2137
|
|
|
} |
|
2138
|
|
|
elseif ($conf['selector_type']=='regex') |
|
2139
|
|
|
{ |
|
2140
|
|
|
$values = $this->get_fields_regex($html, $conf['selector'], $conf['name']); |
|
|
|
|
|
|
2141
|
|
|
} |
|
2142
|
|
|
|
|
2143
|
|
|
// field不为空而且存在子配置 |
|
2144
|
|
|
if (isset($values) && !empty($conf['children'])) |
|
2145
|
|
|
{ |
|
2146
|
|
|
// 如果提取到的结果是字符串，就转为数组，方便下面统一foreach |
|
2147
|
|
|
if (!is_array($values)) |
|
2148
|
|
|
{ |
|
2149
|
|
|
$values = array($values); |
|
2150
|
|
|
} |
|
2151
|
|
|
$child_values = array(); |
|
2152
|
|
|
// 父项抽取到的html作为子项的提取内容 |
|
2153
|
|
|
foreach ($values as $child_html) |
|
2154
|
|
|
{ |
|
2155
|
|
|
// 递归调用本方法, 所以多少子项目都支持 |
|
2156
|
|
|
$child_value = $this->get_fields($conf['children'], $child_html, $url, $page); |
|
|
|
|
|
|
2157
|
|
|
if (!empty($child_value)) |
|
2158
|
|
|
{ |
|
2159
|
|
|
$child_values[] = $child_value; |
|
2160
|
|
|
} |
|
2161
|
|
|
} |
|
2162
|
|
|
// 有子项就存子项的数组, 没有就存HTML代码块 |
|
2163
|
|
|
if (!empty($child_values)) |
|
2164
|
|
|
{ |
|
2165
|
|
|
$values = $child_values; |
|
2166
|
|
|
} |
|
2167
|
|
|
} |
|
2168
|
|
|
} |
|
2169
|
|
|
|
|
2170
|
|
|
if (!isset($values)) |
|
2171
|
|
|
{ |
|
2172
|
|
|
// 如果值为空而且值设置为必须项, 跳出foreach循环 |
|
2173
|
|
|
if ($required) |
|
2174
|
|
|
{ |
|
2175
|
|
|
log::warn("Selector {$conf['name']}[{$conf['selector']}] not found, It's a must"); |
|
2176
|
|
|
// 清空整个 fields，当前页面就等于略过了 |
|
2177
|
|
|
$fields = array(); |
|
2178
|
|
|
break; |
|
2179
|
|
|
} |
|
2180
|
|
|
// 避免内容分页时attached_url拼接时候string + array了 |
|
2181
|
|
|
$fields[$conf['name']] = ''; |
|
2182
|
|
|
//$fields[$conf['name']] = array(); |
|
2183
|
|
|
} |
|
2184
|
|
|
else |
|
2185
|
|
|
{ |
|
2186
|
|
|
if (is_array($values)) |
|
2187
|
|
|
{ |
|
2188
|
|
|
if ($repeated) |
|
2189
|
|
|
{ |
|
2190
|
|
|
$fields[$conf['name']] = $values; |
|
2191
|
|
|
} |
|
2192
|
|
|
else |
|
2193
|
|
|
{ |
|
2194
|
|
|
$fields[$conf['name']] = $values[0]; |
|
2195
|
|
|
} |
|
2196
|
|
|
} |
|
2197
|
|
|
else |
|
2198
|
|
|
{ |
|
2199
|
|
|
$fields[$conf['name']] = $values; |
|
2200
|
|
|
} |
|
2201
|
|
|
// 不重复抽取则只取第一个元素 |
|
2202
|
|
|
//$fields[$conf['name']] = $repeated ? $values : $values[0]; |
|
2203
|
|
|
} |
|
2204
|
|
|
} |
|
2205
|
|
|
|
|
2206
|
|
|
if (!empty($fields)) |
|
2207
|
|
|
{ |
|
2208
|
|
|
foreach ($fields as $fieldname => $data) |
|
2209
|
|
|
{ |
|
2210
|
|
|
$pattern = "/<img\s+.*?src=[\"']{0,1}(.*)[\"']{0,1}[> \r\n\t]{1,}/isu"; |
|
2211
|
|
|
/*$pattern = "/<img.*?src=[\'|\"](.*?(?:[\.gif|\.jpg|\.jpeg|\.png]))[\'|\"].*?[\/]?>/i"; */ |
|
2212
|
|
|
// 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 |
|
2213
|
|
|
if ($this->on_handle_img && preg_match($pattern, $data)) |
|
2214
|
|
|
{ |
|
2215
|
|
|
$return = call_user_func($this->on_handle_img, $fieldname, $data); |
|
2216
|
|
|
if (!isset($return)) |
|
2217
|
|
|
{ |
|
2218
|
|
|
log::warn("on_handle_img return value can't be empty\n"); |
|
2219
|
|
|
} |
|
2220
|
|
|
else |
|
2221
|
|
|
{ |
|
2222
|
|
|
// 有数据才会执行 on_handle_img 方法, 所以这里不要被替换没了 |
|
2223
|
|
|
$data = $return; |
|
2224
|
|
|
} |
|
2225
|
|
|
} |
|
2226
|
|
|
|
|
2227
|
|
|
// 当一个field的内容被抽取到后进行的回调, 在此回调中可以对网页中抽取的内容作进一步处理 |
|
2228
|
|
|
if ($this->on_extract_field) |
|
2229
|
|
|
{ |
|
2230
|
|
|
$return = call_user_func($this->on_extract_field, $fieldname, $data, $page); |
|
2231
|
|
|
if (!isset($return)) |
|
2232
|
|
|
{ |
|
2233
|
|
|
log::warn("on_extract_field return value can't be empty\n"); |
|
2234
|
|
|
} |
|
2235
|
|
|
else |
|
2236
|
|
|
{ |
|
2237
|
|
|
// 有数据才会执行 on_extract_field 方法, 所以这里不要被替换没了 |
|
2238
|
|
|
$fields[$fieldname] = $return; |
|
2239
|
|
|
} |
|
2240
|
|
|
} |
|
2241
|
|
|
} |
|
2242
|
|
|
} |
|
2243
|
|
|
|
|
2244
|
|
|
return $fields; |
|
2245
|
|
|
} |
|
2246
|
|
|
|
|
2247
|
|
|
/** |
|
2248
|
|
|
* 验证导出 |
|
2249
|
|
|
* |
|
2250
|
|
|
* @return void |
|
2251
|
|
|
* @author seatle <[email protected]> |
|
2252
|
|
|
* @created time :2016-10-02 23:37 |
|
2253
|
|
|
*/ |
|
2254
|
|
|
public function check_export() |
|
2255
|
|
|
{ |
|
2256
|
|
|
// 如果设置了导出选项 |
|
2257
|
|
|
if (!empty(self::$configs['export'])) |
|
2258
|
|
|
{ |
|
2259
|
|
|
if (self::$export_type == 'csv') |
|
2260
|
|
|
{ |
|
2261
|
|
|
if (empty(self::$export_file)) |
|
2262
|
|
|
{ |
|
2263
|
|
|
log::error('Export data into CSV files need to Set the file path.'); |
|
2264
|
|
|
exit; |
|
|
|
|
|
|
2265
|
|
|
} |
|
2266
|
|
|
} |
|
2267
|
|
|
elseif (self::$export_type == 'sql') |
|
2268
|
|
|
{ |
|
2269
|
|
|
if (empty(self::$export_file)) |
|
2270
|
|
|
{ |
|
2271
|
|
|
log::error('Export data into SQL files need to Set the file path.'); |
|
2272
|
|
|
exit; |
|
|
|
|
|
|
2273
|
|
|
} |
|
2274
|
|
|
} |
|
2275
|
|
|
elseif (self::$export_type == 'db') |
|
2276
|
|
|
{ |
|
2277
|
|
|
if (!function_exists('mysqli_connect')) |
|
2278
|
|
|
{ |
|
2279
|
|
|
log::error('Export data to a database need Mysql support, unable to load mysqli extension.'); |
|
2280
|
|
|
exit; |
|
|
|
|
|
|
2281
|
|
|
} |
|
2282
|
|
|
|
|
2283
|
|
|
if (empty(self::$db_config)) |
|
2284
|
|
|
{ |
|
2285
|
|
|
log::error('Export data to a database need Mysql support, you have not set a config array for connect.'); |
|
2286
|
|
|
exit; |
|
|
|
|
|
|
2287
|
|
|
} |
|
2288
|
|
|
|
|
2289
|
|
|
$config = self::$db_config; |
|
2290
|
|
|
@mysqli_connect($config['host'], $config['user'], $config['pass'], $config['name'], $config['port']); |
|
|
|
|
|
|
2291
|
|
|
if(mysqli_connect_errno()) |
|
2292
|
|
|
{ |
|
2293
|
|
|
log::error('Export data to a database need Mysql support, '.mysqli_connect_error()); |
|
2294
|
|
|
exit; |
|
|
|
|
|
|
2295
|
|
|
} |
|
2296
|
|
|
|
|
2297
|
|
|
db::set_connect('default', $config); |
|
2298
|
|
|
db::_init(); |
|
2299
|
|
|
|
|
2300
|
|
|
if (!db::table_exists(self::$export_table)) |
|
2301
|
|
|
{ |
|
2302
|
|
|
log::error('Table '.self::$export_table.' does not exist'); |
|
2303
|
|
|
exit; |
|
|
|
|
|
|
2304
|
|
|
} |
|
2305
|
|
|
} |
|
2306
|
|
|
} |
|
2307
|
|
|
} |
|
2308
|
|
|
|
|
2309
|
|
|
public function check_cache() |
|
2310
|
|
|
{ |
|
2311
|
|
|
if ( !self::$use_redis || self::$save_running_state) |
|
2312
|
|
|
{ |
|
2313
|
|
|
return false; |
|
2314
|
|
|
} |
|
2315
|
|
|
|
|
2316
|
|
|
// 这个位置要改 |
|
2317
|
|
|
//$keys = queue::keys("*"); |
|
2318
|
|
|
//$count = count($keys); |
|
2319
|
|
|
// 直接检查db，清空的时候整个db清空，所以注意db不要跟其他项目混用 |
|
2320
|
|
|
$count = queue::dbsize(); |
|
|
|
|
|
|
2321
|
|
|
if ( $count > 0 ) |
|
2322
|
|
|
{ |
|
2323
|
|
|
// After this operation, 4,318 kB of additional disk space will be used. |
|
2324
|
|
|
// Do you want to continue? [Y/n] |
|
2325
|
|
|
//$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"; |
|
2326
|
|
|
$msg = "Found that the data of Redis, no continue will empty Redis data start again\n"; |
|
2327
|
|
|
$msg .= 'Do you want to continue? [Y/n]'; |
|
2328
|
|
|
fwrite(STDOUT, $msg); |
|
2329
|
|
|
$arg = strtolower(trim(fgets(STDIN))); |
|
2330
|
|
|
$arg = empty($arg) || !in_array($arg, array('Y', 'N', 'y','n')) ? 'y' : strtolower($arg); |
|
2331
|
|
|
if ($arg == 'n') |
|
2332
|
|
|
{ |
|
2333
|
|
|
log::warn('Clear redis data...'); |
|
2334
|
|
|
queue::flushdb(); |
|
2335
|
|
|
// 下面这种性能太差了 |
|
2336
|
|
|
//foreach ($keys as $key) |
|
2337
|
|
|
//{ |
|
2338
|
|
|
//$key = str_replace(self::$queue_config['prefix'].':', '', $key); |
|
2339
|
|
|
//queue::del($key); |
|
2340
|
|
|
//} |
|
2341
|
|
|
} |
|
2342
|
|
|
} |
|
2343
|
|
|
} |
|
2344
|
|
|
|
|
2345
|
|
|
public function init_redis() |
|
2346
|
|
|
{ |
|
2347
|
|
|
if (!self::$use_redis) |
|
2348
|
|
|
{ |
|
2349
|
|
|
return false; |
|
2350
|
|
|
} |
|
2351
|
|
|
|
|
2352
|
|
|
// 添加当前服务器到服务器列表 |
|
2353
|
|
|
$this->add_server_list(self::$serverid, self::$tasknum); |
|
2354
|
|
|
|
|
2355
|
|
|
// 删除当前服务器的任务状态 |
|
2356
|
|
|
// 对于被强制退出的进程有用 |
|
2357
|
|
|
for ($i = 1; $i <= self::$tasknum; $i++) |
|
2358
|
|
|
{ |
|
2359
|
|
|
$this->del_task_status(self::$serverid, $i); |
|
2360
|
|
|
} |
|
2361
|
|
|
} |
|
2362
|
|
|
|
|
2363
|
|
|
/** |
|
2364
|
|
|
* 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 |
|
2365
|
|
|
* |
|
2366
|
|
|
* @return void |
|
2367
|
|
|
* @author seatle <[email protected]> |
|
2368
|
|
|
* @created time :2016-10-30 23:56 |
|
2369
|
|
|
*/ |
|
2370
|
|
|
public function set_task_status() |
|
2371
|
|
|
{ |
|
2372
|
|
|
// 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 |
|
2373
|
|
|
$mem = round(memory_get_usage(true)/(1024*1024),2); |
|
2374
|
|
|
$use_time = microtime(true) - self::$time_start; |
|
2375
|
|
|
$speed = round((self::$collect_succ + self::$collect_fail) / $use_time, 2); |
|
2376
|
|
|
$status = array( |
|
2377
|
|
|
'id' => self::$taskid, |
|
2378
|
|
|
'pid' => self::$taskpid, |
|
2379
|
|
|
'mem' => $mem, |
|
2380
|
|
|
'collect_succ' => self::$collect_succ, |
|
2381
|
|
|
'collect_fail' => self::$collect_fail, |
|
2382
|
|
|
'speed' => $speed, |
|
2383
|
|
|
); |
|
2384
|
|
|
$task_status = json_encode($status); |
|
2385
|
|
|
|
|
2386
|
|
|
if (self::$use_redis) |
|
2387
|
|
|
{ |
|
2388
|
|
|
$key = 'server-'.self::$serverid.'-task_status-'.self::$taskid; |
|
2389
|
|
|
queue::set($key, $task_status); |
|
2390
|
|
|
} |
|
2391
|
|
|
else |
|
2392
|
|
|
{ |
|
2393
|
|
|
self::$task_status = array($task_status); |
|
2394
|
|
|
} |
|
2395
|
|
|
} |
|
2396
|
|
|
|
|
2397
|
|
|
/** |
|
2398
|
|
|
* 删除任务状态 |
|
2399
|
|
|
* |
|
2400
|
|
|
* @return void |
|
2401
|
|
|
* @author seatle <[email protected]> |
|
2402
|
|
|
* @created time :2016-11-16 11:06 |
|
2403
|
|
|
*/ |
|
2404
|
|
|
public function del_task_status($serverid, $taskid) |
|
2405
|
|
|
{ |
|
2406
|
|
|
if (!self::$use_redis) |
|
2407
|
|
|
{ |
|
2408
|
|
|
return false; |
|
|
|
|
|
|
2409
|
|
|
} |
|
2410
|
|
|
$key = "server-{$serverid}-task_status-{$taskid}"; |
|
2411
|
|
|
queue::del($key); |
|
2412
|
|
|
} |
|
2413
|
|
|
|
|
2414
|
|
|
/** |
|
2415
|
|
|
* 获得任务状态, 主进程才会调用 |
|
2416
|
|
|
* |
|
2417
|
|
|
* @return void |
|
2418
|
|
|
* @author seatle <[email protected]> |
|
2419
|
|
|
* @created time :2016-10-30 23:56 |
|
2420
|
|
|
*/ |
|
2421
|
|
|
public function get_task_status($serverid, $taskid) |
|
2422
|
|
|
{ |
|
2423
|
|
|
if (!self::$use_redis) |
|
2424
|
|
|
{ |
|
2425
|
|
|
return false; |
|
|
|
|
|
|
2426
|
|
|
} |
|
2427
|
|
|
|
|
2428
|
|
|
$key = "server-{$serverid}-task_status-{$taskid}"; |
|
2429
|
|
|
$task_status = queue::get($key); |
|
|
|
|
|
|
2430
|
|
|
return $task_status; |
|
2431
|
|
|
} |
|
2432
|
|
|
|
|
2433
|
|
|
/** |
|
2434
|
|
|
* 获得任务状态, 主进程才会调用 |
|
2435
|
|
|
* |
|
2436
|
|
|
* @return void |
|
2437
|
|
|
* @author seatle <[email protected]> |
|
2438
|
|
|
* @created time :2016-10-30 23:56 |
|
2439
|
|
|
*/ |
|
2440
|
|
|
public function get_task_status_list($serverid = 1, $tasknum) |
|
2441
|
|
|
{ |
|
2442
|
|
|
$task_status = array(); |
|
2443
|
|
|
if (self::$use_redis) |
|
2444
|
|
|
{ |
|
2445
|
|
|
for ($i = 1; $i <= $tasknum; $i++) |
|
2446
|
|
|
{ |
|
2447
|
|
|
$key = "server-{$serverid}-task_status-".$i; |
|
2448
|
|
|
$task_status[] = queue::get($key); |
|
|
|
|
|
|
2449
|
|
|
} |
|
2450
|
|
|
} |
|
2451
|
|
|
else |
|
2452
|
|
|
{ |
|
2453
|
|
|
$task_status = self::$task_status; |
|
2454
|
|
|
} |
|
2455
|
|
|
return $task_status; |
|
|
|
|
|
|
2456
|
|
|
} |
|
2457
|
|
|
|
|
2458
|
|
|
/** |
|
2459
|
|
|
* 添加当前服务器信息到服务器列表 |
|
2460
|
|
|
* |
|
2461
|
|
|
* @return void |
|
2462
|
|
|
* @author seatle <[email protected]> |
|
2463
|
|
|
* @created time :2016-11-16 11:06 |
|
2464
|
|
|
*/ |
|
2465
|
|
|
public function add_server_list($serverid, $tasknum) |
|
2466
|
|
|
{ |
|
2467
|
|
|
if (!self::$use_redis) |
|
2468
|
|
|
{ |
|
2469
|
|
|
return false; |
|
|
|
|
|
|
2470
|
|
|
} |
|
2471
|
|
|
|
|
2472
|
|
|
// 更新服务器列表 |
|
2473
|
|
|
$server_list_json = queue::get('server_list'); |
|
|
|
|
|
|
2474
|
|
|
$server_list = array(); |
|
2475
|
|
|
if ( ! $server_list_json) |
|
|
|
|
|
|
2476
|
|
|
{ |
|
2477
|
|
|
$server_list[$serverid] = array( |
|
2478
|
|
|
'serverid' => $serverid, |
|
2479
|
|
|
'tasknum' => $tasknum, |
|
2480
|
|
|
'time' => time(), |
|
2481
|
|
|
); |
|
2482
|
|
|
} |
|
2483
|
|
|
else |
|
2484
|
|
|
{ |
|
2485
|
|
|
$server_list = json_decode($server_list_json, true); |
|
2486
|
|
|
$server_list[$serverid] = array( |
|
2487
|
|
|
'serverid' => $serverid, |
|
2488
|
|
|
'tasknum' => $tasknum, |
|
2489
|
|
|
'time' => time(), |
|
2490
|
|
|
); |
|
2491
|
|
|
ksort($server_list); |
|
2492
|
|
|
} |
|
2493
|
|
|
queue::set('server_list', json_encode($server_list)); |
|
2494
|
|
|
} |
|
2495
|
|
|
|
|
2496
|
|
|
/** |
|
2497
|
|
|
* 从服务器列表中删除当前服务器信息 |
|
2498
|
|
|
* |
|
2499
|
|
|
* @return void |
|
2500
|
|
|
* @author seatle <[email protected]> |
|
2501
|
|
|
* @created time :2016-11-16 11:06 |
|
2502
|
|
|
*/ |
|
2503
|
|
|
public function del_server_list($serverid) |
|
2504
|
|
|
{ |
|
2505
|
|
|
if (!self::$use_redis) |
|
2506
|
|
|
{ |
|
2507
|
|
|
return false; |
|
|
|
|
|
|
2508
|
|
|
} |
|
2509
|
|
|
|
|
2510
|
|
|
$server_list_json = queue::get('server_list'); |
|
|
|
|
|
|
2511
|
|
|
$server_list = array(); |
|
|
|
|
|
|
2512
|
|
|
if ($server_list_json) |
|
|
|
|
|
|
2513
|
|
|
{ |
|
2514
|
|
|
$server_list = json_decode($server_list_json, true); |
|
2515
|
|
|
if (isset($server_list[$serverid])) |
|
2516
|
|
|
{ |
|
2517
|
|
|
unset($server_list[$serverid]); |
|
2518
|
|
|
} |
|
2519
|
|
|
|
|
2520
|
|
|
// 删除完当前的任务列表如果还存在，就更新一下Redis |
|
2521
|
|
|
if (!empty($server_list)) |
|
2522
|
|
|
{ |
|
2523
|
|
|
ksort($server_list); |
|
2524
|
|
|
queue::set('server_list', json_encode($server_list)); |
|
2525
|
|
|
} |
|
2526
|
|
|
} |
|
2527
|
|
|
} |
|
2528
|
|
|
|
|
2529
|
|
|
/** |
|
2530
|
|
|
* 获取等待爬取页面数量 |
|
2531
|
|
|
* |
|
2532
|
|
|
* @param mixed $url |
|
2533
|
|
|
* @return void |
|
2534
|
|
|
* @author seatle <[email protected]> |
|
2535
|
|
|
* @created time :2016-09-23 17:13 |
|
2536
|
|
|
*/ |
|
2537
|
|
|
public function get_collect_url_num() |
|
2538
|
|
|
{ |
|
2539
|
|
|
if (self::$use_redis) |
|
2540
|
|
|
{ |
|
2541
|
|
|
$count = queue::get('collect_urls_num'); |
|
|
|
|
|
|
2542
|
|
|
} |
|
2543
|
|
|
else |
|
2544
|
|
|
{ |
|
2545
|
|
|
$count = self::$collect_urls_num; |
|
2546
|
|
|
} |
|
2547
|
|
|
return $count; |
|
|
|
|
|
|
2548
|
|
|
} |
|
2549
|
|
|
|
|
2550
|
|
|
/** |
|
2551
|
|
|
* 获取已经爬取页面数量 |
|
2552
|
|
|
* |
|
2553
|
|
|
* @param mixed $url |
|
2554
|
|
|
* @return void |
|
2555
|
|
|
* @author seatle <[email protected]> |
|
2556
|
|
|
* @created time :2016-09-23 17:13 |
|
2557
|
|
|
*/ |
|
2558
|
|
|
public function get_collected_url_num() |
|
2559
|
|
|
{ |
|
2560
|
|
|
if (self::$use_redis) |
|
2561
|
|
|
{ |
|
2562
|
|
|
$count = queue::get('collected_urls_num'); |
|
|
|
|
|
|
2563
|
|
|
} |
|
2564
|
|
|
else |
|
2565
|
|
|
{ |
|
2566
|
|
|
$count = self::$collected_urls_num; |
|
2567
|
|
|
} |
|
2568
|
|
|
return $count; |
|
|
|
|
|
|
2569
|
|
|
} |
|
2570
|
|
|
|
|
2571
|
|
|
/** |
|
2572
|
|
|
* 已采集页面数量加一 |
|
2573
|
|
|
* |
|
2574
|
|
|
* @param mixed $url |
|
2575
|
|
|
* @return void |
|
2576
|
|
|
* @author seatle <[email protected]> |
|
2577
|
|
|
* @created time :2016-09-23 17:13 |
|
2578
|
|
|
*/ |
|
2579
|
|
|
public function incr_collected_url_num($url) |
|
2580
|
|
|
{ |
|
2581
|
|
|
if (self::$use_redis) |
|
2582
|
|
|
{ |
|
2583
|
|
|
queue::incr('collected_urls_num'); |
|
2584
|
|
|
} |
|
2585
|
|
|
else |
|
2586
|
|
|
{ |
|
2587
|
|
|
self::$collected_urls_num++; |
|
2588
|
|
|
} |
|
2589
|
|
|
} |
|
2590
|
|
|
|
|
2591
|
|
|
/** |
|
2592
|
|
|
* 从队列左边插入 |
|
2593
|
|
|
* |
|
2594
|
|
|
* @return void |
|
2595
|
|
|
* @author seatle <[email protected]> |
|
2596
|
|
|
* @created time :2016-09-23 17:13 |
|
2597
|
|
|
*/ |
|
2598
|
|
|
public function queue_lpush($link = array(), $allowed_repeat = false) |
|
2599
|
|
|
{ |
|
2600
|
|
|
if (empty($link) || empty($link['url'])) |
|
2601
|
|
|
{ |
|
2602
|
|
|
return false; |
|
|
|
|
|
|
2603
|
|
|
} |
|
2604
|
|
|
|
|
2605
|
|
|
$url = $link['url']; |
|
2606
|
|
|
$link = $this->link_compress($link); |
|
|
|
|
|
|
2607
|
|
|
|
|
2608
|
|
|
$status = false; |
|
2609
|
|
|
if (self::$use_redis) |
|
2610
|
|
|
{ |
|
2611
|
|
|
$key = 'collect_urls-'.md5($url); |
|
2612
|
|
|
$lock = 'lock-'.$key; |
|
2613
|
|
|
// 加锁: 一个进程一个进程轮流处理 |
|
2614
|
|
|
if (queue::lock($lock)) |
|
|
|
|
|
|
2615
|
|
|
{ |
|
2616
|
|
|
$exists = queue::exists($key); |
|
|
|
|
|
|
2617
|
|
|
// 不存在或者当然URL可重复入 |
|
2618
|
|
|
if (!$exists || $allowed_repeat) |
|
|
|
|
|
|
2619
|
|
|
{ |
|
2620
|
|
|
// 待爬取网页记录数加一 |
|
2621
|
|
|
queue::incr('collect_urls_num'); |
|
2622
|
|
|
// 先标记为待爬取网页 |
|
2623
|
|
|
queue::set($key, time()); |
|
2624
|
|
|
// 入队列 |
|
2625
|
|
|
$link = json_encode($link); |
|
2626
|
|
|
//根据采集设置为顺序采集还是随机采集，使用列表或集合对象 2018-5 BY KEN <[email protected]> |
|
2627
|
|
|
if (self::$configs['queue_order'] == 'rand') |
|
2628
|
|
|
{ |
|
2629
|
|
|
queue::sadd('collect_queue', $link); |
|
2630
|
|
|
} |
|
2631
|
|
|
else |
|
2632
|
|
|
{ |
|
2633
|
|
|
queue::lpush('collect_queue', $link); |
|
2634
|
|
|
} |
|
2635
|
|
|
$status = true; |
|
2636
|
|
|
} |
|
2637
|
|
|
// 解锁 |
|
2638
|
|
|
queue::unlock($lock); |
|
2639
|
|
|
} |
|
2640
|
|
|
} |
|
2641
|
|
|
else |
|
2642
|
|
|
{ |
|
2643
|
|
|
$key = md5($url); |
|
2644
|
|
|
if (!array_key_exists($key, self::$collect_urls)) |
|
2645
|
|
|
{ |
|
2646
|
|
|
self::$collect_urls_num++; |
|
2647
|
|
|
self::$collect_urls[$key] = time(); |
|
2648
|
|
|
array_push(self::$collect_queue, $link); |
|
2649
|
|
|
$status = true; |
|
2650
|
|
|
} |
|
2651
|
|
|
} |
|
2652
|
|
|
return $status; |
|
|
|
|
|
|
2653
|
|
|
} |
|
2654
|
|
|
|
|
2655
|
|
|
/** |
|
2656
|
|
|
* 从队列右边插入 |
|
2657
|
|
|
* |
|
2658
|
|
|
* @return void |
|
2659
|
|
|
* @author seatle <[email protected]> |
|
2660
|
|
|
* @created time :2016-09-23 17:13 |
|
2661
|
|
|
*/ |
|
2662
|
|
|
public function queue_rpush($link = array(), $allowed_repeat = false) |
|
2663
|
|
|
{ |
|
2664
|
|
|
if (empty($link) || empty($link['url'])) |
|
2665
|
|
|
{ |
|
2666
|
|
|
return false; |
|
|
|
|
|
|
2667
|
|
|
} |
|
2668
|
|
|
|
|
2669
|
|
|
$url = $link['url']; |
|
2670
|
|
|
|
|
2671
|
|
|
$status = false; |
|
2672
|
|
|
if (self::$use_redis) |
|
2673
|
|
|
{ |
|
2674
|
|
|
$key = 'collect_urls-'.md5($url); |
|
2675
|
|
|
$lock = 'lock-'.$key; |
|
2676
|
|
|
// 加锁: 一个进程一个进程轮流处理 |
|
2677
|
|
|
if (queue::lock($lock)) |
|
|
|
|
|
|
2678
|
|
|
{ |
|
2679
|
|
|
$exists = queue::exists($key); |
|
|
|
|
|
|
2680
|
|
|
// 不存在或者当然URL可重复入 |
|
2681
|
|
|
if ( ! $exists || $allowed_repeat) |
|
|
|
|
|
|
2682
|
|
|
{ |
|
2683
|
|
|
// 待爬取网页记录数加一 |
|
2684
|
|
|
queue::incr('collect_urls_num'); |
|
2685
|
|
|
// 先标记为待爬取网页 |
|
2686
|
|
|
queue::set($key, time()); |
|
2687
|
|
|
// 入队列 |
|
2688
|
|
|
$link = json_encode($link); |
|
2689
|
|
|
//根据采集设置为顺序采集还是随机采集，使用列表或集合对象 2018-5 BY KEN <[email protected]> |
|
2690
|
|
|
if (self::$configs['queue_order'] == 'rand') |
|
2691
|
|
|
{ |
|
2692
|
|
|
queue::sadd('collect_queue', $link); //无序集合 |
|
2693
|
|
|
} |
|
2694
|
|
|
else |
|
2695
|
|
|
{ |
|
2696
|
|
|
queue::rpush('collect_queue', $link); //有序列表 |
|
2697
|
|
|
} |
|
2698
|
|
|
$status = true; |
|
2699
|
|
|
} |
|
2700
|
|
|
// 解锁 |
|
2701
|
|
|
queue::unlock($lock); |
|
2702
|
|
|
} |
|
2703
|
|
|
} |
|
2704
|
|
|
else |
|
2705
|
|
|
{ |
|
2706
|
|
|
$key = md5($url); |
|
2707
|
|
|
if (!array_key_exists($key, self::$collect_urls)) |
|
2708
|
|
|
{ |
|
2709
|
|
|
self::$collect_urls_num++; |
|
2710
|
|
|
self::$collect_urls[$key] = time(); |
|
2711
|
|
|
array_unshift(self::$collect_queue, $link); |
|
2712
|
|
|
$status = true; |
|
2713
|
|
|
} |
|
2714
|
|
|
} |
|
2715
|
|
|
return $status; |
|
|
|
|
|
|
2716
|
|
|
} |
|
2717
|
|
|
|
|
2718
|
|
|
/** |
|
2719
|
|
|
* 从队列左边取出 |
|
2720
|
|
|
* 后进先出 |
|
2721
|
|
|
* 可以避免采集内容页有分页的时候采集失败数据拼凑不全 |
|
2722
|
|
|
* 还可以按顺序采集列表页 |
|
2723
|
|
|
* |
|
2724
|
|
|
* @return void |
|
2725
|
|
|
* @author seatle <[email protected]> |
|
2726
|
|
|
* @created time :2016-09-23 17:13 |
|
2727
|
|
|
*/ |
|
2728
|
|
|
public function queue_lpop() |
|
2729
|
|
|
{ |
|
2730
|
|
|
if (self::$use_redis) |
|
2731
|
|
|
{ |
|
2732
|
|
|
//根据采集设置为顺序采集还是随机采集，使用列表或集合对象 |
|
2733
|
|
|
if (self::$configs['queue_order'] == 'rand') |
|
2734
|
|
|
{ |
|
2735
|
|
|
$link = queue::spop('collect_queue'); |
|
|
|
|
|
|
2736
|
|
|
} |
|
2737
|
|
|
else |
|
2738
|
|
|
{ |
|
2739
|
|
|
$link = queue::lpop('collect_queue'); |
|
|
|
|
|
|
2740
|
|
|
} |
|
2741
|
|
|
$link = json_decode($link, true); |
|
|
|
|
|
|
2742
|
|
|
} |
|
2743
|
|
|
else |
|
2744
|
|
|
{ |
|
2745
|
|
|
$link = array_pop(self::$collect_queue); |
|
2746
|
|
|
} |
|
2747
|
|
|
return $link; |
|
2748
|
|
|
} |
|
2749
|
|
|
|
|
2750
|
|
|
/** |
|
2751
|
|
|
* 从队列右边取出 |
|
2752
|
|
|
* |
|
2753
|
|
|
* @return void |
|
2754
|
|
|
* @author seatle <[email protected]> |
|
2755
|
|
|
* @created time :2016-09-23 17:13 |
|
2756
|
|
|
*/ |
|
2757
|
|
|
public function queue_rpop() |
|
2758
|
|
|
{ |
|
2759
|
|
|
if (self::$use_redis) |
|
2760
|
|
|
{ |
|
2761
|
|
|
//根据采集设置为顺序采集还是随机采集，使用列表或集合对象 |
|
2762
|
|
|
if (self::$configs['queue_order'] == 'rand') |
|
2763
|
|
|
{ |
|
2764
|
|
|
$link = queue::spop('collect_queue'); |
|
|
|
|
|
|
2765
|
|
|
} |
|
2766
|
|
|
else |
|
2767
|
|
|
{ |
|
2768
|
|
|
$link = queue::rpop('collect_queue'); |
|
|
|
|
|
|
2769
|
|
|
} |
|
2770
|
|
|
$link = json_decode($link, true); |
|
|
|
|
|
|
2771
|
|
|
} |
|
2772
|
|
|
else |
|
2773
|
|
|
{ |
|
2774
|
|
|
$link = array_shift(self::$collect_queue); |
|
2775
|
|
|
} |
|
2776
|
|
|
return $link; |
|
2777
|
|
|
} |
|
2778
|
|
|
|
|
2779
|
|
|
/** |
|
2780
|
|
|
* 队列长度 |
|
2781
|
|
|
* |
|
2782
|
|
|
* @return void |
|
2783
|
|
|
* @author seatle <[email protected]> |
|
2784
|
|
|
* @created time :2016-09-23 17:13 |
|
2785
|
|
|
*/ |
|
2786
|
|
|
public function queue_lsize() |
|
2787
|
|
|
{ |
|
2788
|
|
|
if (self::$use_redis) |
|
2789
|
|
|
{ |
|
2790
|
|
|
//根据采集设置为顺序采集还是随机采集，使用列表或集合对象 |
|
2791
|
|
|
if (self::$configs['queue_order'] == 'rand') |
|
2792
|
|
|
{ |
|
2793
|
|
|
$lsize = queue::scard('collect_queue'); |
|
|
|
|
|
|
2794
|
|
|
} |
|
2795
|
|
|
else |
|
2796
|
|
|
{ |
|
2797
|
|
|
$lsize = queue::lsize('collect_queue'); |
|
|
|
|
|
|
2798
|
|
|
} |
|
2799
|
|
|
} |
|
2800
|
|
|
else |
|
2801
|
|
|
{ |
|
2802
|
|
|
$lsize = count(self::$collect_queue); |
|
2803
|
|
|
} |
|
2804
|
|
|
return $lsize; |
|
|
|
|
|
|
2805
|
|
|
} |
|
2806
|
|
|
|
|
2807
|
|
|
/** |
|
2808
|
|
|
* 采集深度加一 |
|
2809
|
|
|
* |
|
2810
|
|
|
* @return void |
|
2811
|
|
|
* @author seatle <[email protected]> |
|
2812
|
|
|
* @created time :2016-09-23 17:13 |
|
2813
|
|
|
*/ |
|
2814
|
|
|
public function incr_depth_num($depth) |
|
2815
|
|
|
{ |
|
2816
|
|
|
if (self::$use_redis) |
|
2817
|
|
|
{ |
|
2818
|
|
|
$lock = 'lock-depth_num'; |
|
2819
|
|
|
// 锁2秒 |
|
2820
|
|
|
if (queue::lock($lock, time(), 2)) |
|
|
|
|
|
|
2821
|
|
|
{ |
|
2822
|
|
|
if (queue::get('depth_num') < $depth) |
|
|
|
|
|
|
2823
|
|
|
{ |
|
2824
|
|
|
queue::set('depth_num', $depth); |
|
2825
|
|
|
} |
|
2826
|
|
|
|
|
2827
|
|
|
queue::unlock($lock); |
|
2828
|
|
|
} |
|
2829
|
|
|
} |
|
2830
|
|
|
else |
|
2831
|
|
|
{ |
|
2832
|
|
|
if (self::$depth_num < $depth) |
|
2833
|
|
|
{ |
|
2834
|
|
|
self::$depth_num = $depth; |
|
2835
|
|
|
} |
|
2836
|
|
|
} |
|
2837
|
|
|
} |
|
2838
|
|
|
|
|
2839
|
|
|
/** |
|
2840
|
|
|
* 获得采集深度 |
|
2841
|
|
|
* |
|
2842
|
|
|
* @return void |
|
2843
|
|
|
* @author seatle <[email protected]> |
|
2844
|
|
|
* @created time :2016-09-23 17:13 |
|
2845
|
|
|
*/ |
|
2846
|
|
|
public function get_depth_num() |
|
2847
|
|
|
{ |
|
2848
|
|
|
if (self::$use_redis) |
|
2849
|
|
|
{ |
|
2850
|
|
|
$depth_num = queue::get('depth_num'); |
|
|
|
|
|
|
2851
|
|
|
return $depth_num ? $depth_num : 0; |
|
|
|
|
|
|
2852
|
|
|
} |
|
2853
|
|
|
else |
|
2854
|
|
|
{ |
|
2855
|
|
|
return self::$depth_num; |
|
|
|
|
|
|
2856
|
|
|
} |
|
2857
|
|
|
} |
|
2858
|
|
|
|
|
2859
|
|
|
/** |
|
2860
|
|
|
* 提取到的field数目加一 |
|
2861
|
|
|
* |
|
2862
|
|
|
* @return void |
|
2863
|
|
|
* @author seatle <[email protected]> |
|
2864
|
|
|
* @created time :2016-09-23 17:13 |
|
2865
|
|
|
*/ |
|
2866
|
|
|
public function incr_fields_num() |
|
2867
|
|
|
{ |
|
2868
|
|
|
if (self::$use_redis) |
|
2869
|
|
|
{ |
|
2870
|
|
|
$fields_num = queue::incr('fields_num'); |
|
|
|
|
|
|
2871
|
|
|
} |
|
2872
|
|
|
else |
|
2873
|
|
|
{ |
|
2874
|
|
|
self::$fields_num++; |
|
2875
|
|
|
$fields_num = self::$fields_num; |
|
2876
|
|
|
} |
|
2877
|
|
|
return $fields_num; |
|
|
|
|
|
|
2878
|
|
|
} |
|
2879
|
|
|
|
|
2880
|
|
|
/** |
|
2881
|
|
|
* 提取到的field数目 |
|
2882
|
|
|
* |
|
2883
|
|
|
* @return void |
|
2884
|
|
|
* @author seatle <[email protected]> |
|
2885
|
|
|
* @created time :2016-09-23 17:13 |
|
2886
|
|
|
*/ |
|
2887
|
|
|
public function get_fields_num() |
|
2888
|
|
|
{ |
|
2889
|
|
|
if (self::$use_redis) |
|
2890
|
|
|
{ |
|
2891
|
|
|
$fields_num = queue::get('fields_num'); |
|
|
|
|
|
|
2892
|
|
|
} |
|
2893
|
|
|
else |
|
2894
|
|
|
{ |
|
2895
|
|
|
$fields_num = self::$fields_num; |
|
2896
|
|
|
} |
|
2897
|
|
|
return $fields_num ? $fields_num : 0; |
|
|
|
|
|
|
2898
|
|
|
} |
|
2899
|
|
|
|
|
2900
|
|
|
/** |
|
2901
|
|
|
* 提取到的pages数目加一，用于限制单域名采集页数上限 |
|
2902
|
|
|
* |
|
2903
|
|
|
* @return void |
|
2904
|
|
|
* @author KEN <[email protected]> |
|
2905
|
|
|
* @created time :2018-05 |
|
2906
|
|
|
*/ |
|
2907
|
|
|
public function incr_pages_num($url = '') |
|
2908
|
|
|
{ |
|
2909
|
|
|
if ( ! empty($url)) |
|
2910
|
|
|
{ |
|
2911
|
|
|
$domain = $this->getRootDomain($url, 'host'); |
|
2912
|
|
|
} |
|
2913
|
|
|
if (empty($domain)) |
|
2914
|
|
|
{ |
|
2915
|
|
|
$domain = 'all'; |
|
2916
|
|
|
} |
|
2917
|
|
|
if (self::$use_redis) |
|
2918
|
|
|
{ |
|
2919
|
|
|
$pages_num[$domain] = queue::incr('pages_num:'.$domain); |
|
|
|
|
|
|
2920
|
|
|
} |
|
2921
|
|
|
else |
|
2922
|
|
|
{ |
|
2923
|
|
|
if (empty(self::$pages_num[$domain])) |
|
2924
|
|
|
{ |
|
2925
|
|
|
self::$pages_num[$domain] = 1; |
|
2926
|
|
|
} |
|
2927
|
|
|
else |
|
2928
|
|
|
{ |
|
2929
|
|
|
self::$pages_num[$domain]++; |
|
2930
|
|
|
} |
|
2931
|
|
|
$pages_num[$domain] = self::$pages_num[$domain]; |
|
2932
|
|
|
} |
|
2933
|
|
|
return $pages_num[$domain]; |
|
2934
|
|
|
} |
|
2935
|
|
|
|
|
2936
|
|
|
/** |
|
2937
|
|
|
* 超过1秒的慢速采集时间计数，用于限制单域名总采集时间上限 |
|
2938
|
|
|
* |
|
2939
|
|
|
* @return void |
|
2940
|
|
|
* @author KEN <[email protected]> |
|
2941
|
|
|
* @created time :2018-05 |
|
2942
|
|
|
*/ |
|
2943
|
|
|
public function incr_duration_num($url = '', $time_run = 1) |
|
2944
|
|
|
{ |
|
2945
|
|
|
if ( ! empty($url)) |
|
2946
|
|
|
{ |
|
2947
|
|
|
$domain = $this->getRootDomain($url); |
|
2948
|
|
|
} |
|
2949
|
|
|
if (empty($domain)) |
|
2950
|
|
|
{ |
|
2951
|
|
|
$domain = 'all'; |
|
2952
|
|
|
} |
|
2953
|
|
|
if (self::$use_redis) |
|
2954
|
|
|
{ |
|
2955
|
|
|
$duration[$domain] = queue::incr('duration:'.$domain, $time_run); |
|
|
|
|
|
|
2956
|
|
|
} |
|
2957
|
|
|
else |
|
2958
|
|
|
{ |
|
2959
|
|
|
if (empty(self::$duration[$domain])) |
|
2960
|
|
|
{ |
|
2961
|
|
|
self::$duration[$domain] = $time_run; |
|
2962
|
|
|
} |
|
2963
|
|
|
else |
|
2964
|
|
|
{ |
|
2965
|
|
|
self::$duration[$domain] += $time_run; |
|
2966
|
|
|
} |
|
2967
|
|
|
$duration[$domain] = self::$duration[$domain]; |
|
2968
|
|
|
} |
|
2969
|
|
|
return $duration[$domain]; |
|
2970
|
|
|
} |
|
2971
|
|
|
|
|
2972
|
|
|
/** |
|
2973
|
|
|
* 读取单域名总慢速采集（响应超过1秒）的时间 |
|
2974
|
|
|
* |
|
2975
|
|
|
* @return void |
|
2976
|
|
|
* @author KEN <[email protected]> |
|
2977
|
|
|
* @created time :2018-04 |
|
2978
|
|
|
*/ |
|
2979
|
|
|
public function get_duration_num($url = '') |
|
2980
|
|
|
{ |
|
2981
|
|
|
if ( ! empty($url)) |
|
2982
|
|
|
{ |
|
2983
|
|
|
$domain = $this->getRootDomain($url); |
|
2984
|
|
|
} |
|
2985
|
|
|
if (empty($domain)) |
|
2986
|
|
|
{ |
|
2987
|
|
|
$domain = 'all'; |
|
2988
|
|
|
} |
|
2989
|
|
|
if (self::$use_redis) |
|
2990
|
|
|
{ |
|
2991
|
|
|
$duration[$domain] = queue::get('duration:'.$domain); |
|
|
|
|
|
|
2992
|
|
|
} |
|
2993
|
|
|
else |
|
2994
|
|
|
{ |
|
2995
|
|
|
$duration[$domain] = ! empty(self::$duration[$domain]) ? self::$duration[$domain] : 0; |
|
2996
|
|
|
} |
|
2997
|
|
|
return $duration[$domain] ? $duration[$domain] : 0; |
|
|
|
|
|
|
2998
|
|
|
} |
|
2999
|
|
|
|
|
3000
|
|
|
/** |
|
3001
|
|
|
* 单 host 当前并发计数 |
|
3002
|
|
|
* @return int |
|
3003
|
|
|
* @author KEN <[email protected]> |
|
3004
|
|
|
* @created time :2018-05-28 16:40 |
|
3005
|
|
|
*/ |
|
3006
|
|
|
public function incr_task_per_host($url = '', $type = 'incr') |
|
3007
|
|
|
{ |
|
3008
|
|
|
if (empty($url)) |
|
3009
|
|
|
{ |
|
3010
|
|
|
return false; |
|
|
|
|
|
|
3011
|
|
|
} |
|
3012
|
|
|
$domain = $this->getRootDomain($url, 'host'); |
|
3013
|
|
|
if (empty($domain)) |
|
3014
|
|
|
{ |
|
3015
|
|
|
return false; |
|
|
|
|
|
|
3016
|
|
|
} |
|
3017
|
|
|
if (self::$use_redis) |
|
3018
|
|
|
{ |
|
3019
|
|
|
if ($type == 'decr') |
|
3020
|
|
|
{ |
|
3021
|
|
|
$task_per_host_counter[$domain] = queue::decr('task_per_host:'.$domain); |
|
|
|
|
|
|
3022
|
|
|
} |
|
3023
|
|
|
else |
|
3024
|
|
|
{ |
|
3025
|
|
|
$task_per_host_counter[$domain] = queue::incr('task_per_host:'.$domain); |
|
|
|
|
|
|
3026
|
|
|
} |
|
3027
|
|
|
} |
|
3028
|
|
|
else |
|
3029
|
|
|
{ |
|
3030
|
|
|
|
|
3031
|
|
|
if (empty(self::$task_per_host_counter[$domain])) |
|
3032
|
|
|
{ |
|
3033
|
|
|
self::$task_per_host_counter[$domain] = 1; |
|
3034
|
|
|
} |
|
3035
|
|
|
else |
|
3036
|
|
|
{ |
|
3037
|
|
|
if ($type == 'decr') |
|
3038
|
|
|
{ |
|
3039
|
|
|
self::$task_per_host_counter[$domain]--; |
|
3040
|
|
|
} |
|
3041
|
|
|
else |
|
3042
|
|
|
{ |
|
3043
|
|
|
self::$task_per_host_counter[$domain]++; |
|
3044
|
|
|
} |
|
3045
|
|
|
} |
|
3046
|
|
|
$task_per_host_counter[$domain] = self::$task_per_host_counter[$domain]; |
|
3047
|
|
|
} |
|
3048
|
|
|
return $task_per_host_counter[$domain]; |
|
3049
|
|
|
} |
|
3050
|
|
|
|
|
3051
|
|
|
//获取url所属 host 当前并发数量 KEN <[email protected]> |
|
3052
|
|
|
public function get_task_per_host_num($url) |
|
3053
|
|
|
{ |
|
3054
|
|
|
if (empty($url)) |
|
3055
|
|
|
{ |
|
3056
|
|
|
return 0; |
|
3057
|
|
|
} |
|
3058
|
|
|
$domain = $this->getRootDomain($url, 'host'); |
|
3059
|
|
|
if (empty($domain)) |
|
3060
|
|
|
{ |
|
3061
|
|
|
return 0; |
|
3062
|
|
|
} |
|
3063
|
|
|
if (self::$use_redis) |
|
3064
|
|
|
{ |
|
3065
|
|
|
$count = queue::get('task_per_host:'.$domain); |
|
|
|
|
|
|
3066
|
|
|
} |
|
3067
|
|
|
else |
|
3068
|
|
|
{ |
|
3069
|
|
|
$count = self::$task_per_host_counter[$domain]; |
|
3070
|
|
|
} |
|
3071
|
|
|
return $count; |
|
3072
|
|
|
} |
|
3073
|
|
|
|
|
3074
|
|
|
/** |
|
3075
|
|
|
* 采用xpath分析提取字段 |
|
3076
|
|
|
* |
|
3077
|
|
|
* @param mixed $html |
|
3078
|
|
|
* @param mixed $selector |
|
3079
|
|
|
* @return void |
|
3080
|
|
|
* @author seatle <[email protected]> |
|
3081
|
|
|
* @created time :2016-09-18 10:17 |
|
3082
|
|
|
*/ |
|
3083
|
|
|
public function get_fields_xpath($html, $selector, $fieldname) |
|
3084
|
|
|
{ |
|
3085
|
|
|
$result = selector::select($html, $selector); |
|
3086
|
|
|
if (selector::$error) |
|
3087
|
|
|
{ |
|
3088
|
|
|
log::error("Field(\"{$fieldname}\") ".selector::$error."\n"); |
|
3089
|
|
|
} |
|
3090
|
|
|
return $result; |
|
|
|
|
|
|
3091
|
|
|
} |
|
3092
|
|
|
|
|
3093
|
|
|
/** |
|
3094
|
|
|
* 采用正则分析提取字段 |
|
3095
|
|
|
* |
|
3096
|
|
|
* @param mixed $html |
|
3097
|
|
|
* @param mixed $selector |
|
3098
|
|
|
* @return void |
|
3099
|
|
|
* @author seatle <[email protected]> |
|
3100
|
|
|
* @created time :2016-09-18 10:17 |
|
3101
|
|
|
*/ |
|
3102
|
|
|
public function get_fields_regex($html, $selector, $fieldname) |
|
3103
|
|
|
{ |
|
3104
|
|
|
$result = selector::select($html, $selector, 'regex'); |
|
3105
|
|
|
if (selector::$error) |
|
3106
|
|
|
{ |
|
3107
|
|
|
log::error("Field(\"{$fieldname}\") ".selector::$error."\n"); |
|
3108
|
|
|
} |
|
3109
|
|
|
return $result; |
|
|
|
|
|
|
3110
|
|
|
} |
|
3111
|
|
|
|
|
3112
|
|
|
/** |
|
3113
|
|
|
* 采用CSS选择器提取字段 |
|
3114
|
|
|
* |
|
3115
|
|
|
* @param mixed $html |
|
3116
|
|
|
* @param mixed $selector |
|
3117
|
|
|
* @param mixed $fieldname |
|
3118
|
|
|
* @return void |
|
3119
|
|
|
* @author seatle <[email protected]> |
|
3120
|
|
|
* @created time :2016-09-18 10:17 |
|
3121
|
|
|
*/ |
|
3122
|
|
|
public function get_fields_css($html, $selector, $fieldname) |
|
3123
|
|
|
{ |
|
3124
|
|
|
$result = selector::select($html, $selector, 'css'); |
|
3125
|
|
|
if (selector::$error) |
|
3126
|
|
|
{ |
|
3127
|
|
|
log::error("Field(\"{$fieldname}\") ".selector::$error."\n"); |
|
3128
|
|
|
} |
|
3129
|
|
|
return $result; |
|
|
|
|
|
|
3130
|
|
|
} |
|
3131
|
|
|
|
|
3132
|
|
|
/** |
|
3133
|
|
|
* 清空shell输出内容 |
|
3134
|
|
|
* |
|
3135
|
|
|
* @return void |
|
3136
|
|
|
* @author seatle <[email protected]> |
|
3137
|
|
|
* @created time :2016-11-16 11:06 |
|
3138
|
|
|
*/ |
|
3139
|
|
|
public function clear_echo() |
|
3140
|
|
|
{ |
|
3141
|
|
|
$arr = array(27, 91, 72, 27, 91, 50, 74); |
|
3142
|
|
|
foreach ($arr as $a) |
|
3143
|
|
|
{ |
|
3144
|
|
|
print chr($a); |
|
3145
|
|
|
} |
|
3146
|
|
|
//array_map(create_function('$a', 'print chr($a);'), array(27, 91, 72, 27, 91, 50, 74)); |
|
3147
|
|
|
} |
|
3148
|
|
|
|
|
3149
|
|
|
/** |
|
3150
|
|
|
* 替换shell输出内容 |
|
3151
|
|
|
* |
|
3152
|
|
|
* @param mixed $message |
|
3153
|
|
|
* @param mixed $force_clear_lines |
|
3154
|
|
|
* @return void |
|
3155
|
|
|
* @author seatle <[email protected]> |
|
3156
|
|
|
* @created time :2016-11-16 11:06 |
|
3157
|
|
|
*/ |
|
3158
|
|
|
public function replace_echo($message, $force_clear_lines = NULL) |
|
3159
|
|
|
{ |
|
3160
|
|
|
static $last_lines = 0; |
|
3161
|
|
|
|
|
3162
|
|
|
if(!is_null($force_clear_lines)) |
|
3163
|
|
|
{ |
|
3164
|
|
|
$last_lines = $force_clear_lines; |
|
3165
|
|
|
} |
|
3166
|
|
|
|
|
3167
|
|
|
// 获取终端宽度 |
|
3168
|
|
|
$toss = $status = null; |
|
3169
|
|
|
$term_width = exec('tput cols', $toss, $status); |
|
3170
|
|
|
if($status || empty($term_width)) |
|
3171
|
|
|
{ |
|
3172
|
|
|
$term_width = 64; // Arbitrary fall-back term width. |
|
3173
|
|
|
} |
|
3174
|
|
|
|
|
3175
|
|
|
$line_count = 0; |
|
3176
|
|
|
foreach(explode("\n", $message) as $line) |
|
3177
|
|
|
{ |
|
3178
|
|
|
$line_count += count(str_split($line, $term_width)); |
|
|
|
|
|
|
3179
|
|
|
} |
|
3180
|
|
|
|
|
3181
|
|
|
// Erasure MAGIC: Clear as many lines as the last output had. |
|
3182
|
|
|
for($i = 0; $i < $last_lines; $i++) |
|
3183
|
|
|
{ |
|
3184
|
|
|
// Return to the beginning of the line |
|
3185
|
|
|
echo "\r"; |
|
3186
|
|
|
// Erase to the end of the line |
|
3187
|
|
|
echo "\033[K"; |
|
3188
|
|
|
// Move cursor Up a line |
|
3189
|
|
|
echo "\033[1A"; |
|
3190
|
|
|
// Return to the beginning of the line |
|
3191
|
|
|
echo "\r"; |
|
3192
|
|
|
// Erase to the end of the line |
|
3193
|
|
|
echo "\033[K"; |
|
3194
|
|
|
// Return to the beginning of the line |
|
3195
|
|
|
echo "\r"; |
|
3196
|
|
|
// Can be consolodated into |
|
3197
|
|
|
// echo "\r\033[K\033[1A\r\033[K\r"; |
|
3198
|
|
|
} |
|
3199
|
|
|
|
|
3200
|
|
|
$last_lines = $line_count; |
|
3201
|
|
|
|
|
3202
|
|
|
echo $message."\n"; |
|
3203
|
|
|
} |
|
3204
|
|
|
|
|
3205
|
|
|
/** |
|
3206
|
|
|
* 展示启动界面, Windows 不会到这里来 |
|
3207
|
|
|
* @return void |
|
3208
|
|
|
*/ |
|
3209
|
|
|
public function display_ui() |
|
3210
|
|
|
{ |
|
3211
|
|
|
$loadavg = sys_getloadavg(); |
|
3212
|
|
|
foreach ($loadavg as $k=>$v) |
|
3213
|
|
|
{ |
|
3214
|
|
|
$loadavg[$k] = round($v, 2); |
|
3215
|
|
|
} |
|
3216
|
|
|
$display_str = "\033[1A\n\033[K-----------------------------\033[47;30m PHPSPIDER \033[0m-----------------------------\n\033[0m"; |
|
3217
|
|
|
//$display_str = "-----------------------------\033[47;30m PHPSPIDER \033[0m-----------------------------\n\033[0m"; |
|
3218
|
|
|
$run_time_str = util::time2second(time() - self::$time_start, false); |
|
3219
|
|
|
$display_str .= 'PHPSpider version:'.self::VERSION.' PHP version:'.PHP_VERSION."\n"; |
|
3220
|
|
|
$display_str .= 'start time:'.date('Y-m-d H:i:s', self::$time_start).' run '.$run_time_str." \n"; |
|
3221
|
|
|
|
|
3222
|
|
|
$display_str .= 'spider name: '.self::$configs['name']."\n"; |
|
3223
|
|
|
if (self::$multiserver) |
|
3224
|
|
|
{ |
|
3225
|
|
|
$display_str .= 'server id: '.self::$serverid."\n"; |
|
3226
|
|
|
} |
|
3227
|
|
|
$display_str .= 'task number: '.self::$tasknum."\n"; |
|
3228
|
|
|
$display_str .= 'load average: '.implode(', ', $loadavg)."\n"; |
|
3229
|
|
|
$display_str .= "document: https://doc.phpspider.org\n"; |
|
3230
|
|
|
|
|
3231
|
|
|
$display_str .= $this->display_task_ui(); |
|
3232
|
|
|
|
|
3233
|
|
|
if (self::$multiserver) |
|
3234
|
|
|
{ |
|
3235
|
|
|
$display_str .= $this->display_server_ui(); |
|
3236
|
|
|
} |
|
3237
|
|
|
|
|
3238
|
|
|
$display_str .= $this->display_collect_ui(); |
|
3239
|
|
|
|
|
3240
|
|
|
// 清屏 |
|
3241
|
|
|
//$this->clear_echo(); |
|
3242
|
|
|
// 返回到第一行,第一列 |
|
3243
|
|
|
//echo "\033[0;0H"; |
|
3244
|
|
|
$display_str .= "---------------------------------------------------------------------\n"; |
|
3245
|
|
|
$display_str .= 'Press Ctrl-C to quit. Start success.'.date('Y-m-d H:i:s').' - '.round(memory_get_usage() / 1024 / 1024, 2).'MB'."\n"; |
|
3246
|
|
|
if (self::$terminate) |
|
3247
|
|
|
{ |
|
3248
|
|
|
$display_str .= "\n\033[33mWait for the process exits...\033[0m"; |
|
3249
|
|
|
} |
|
3250
|
|
|
//echo $display_str; |
|
3251
|
|
|
$this->replace_echo($display_str); |
|
3252
|
|
|
} |
|
3253
|
|
|
|
|
3254
|
|
|
public function display_task_ui() |
|
3255
|
|
|
{ |
|
3256
|
|
|
$display_str = "-------------------------------\033[47;30m TASKS \033[0m-------------------------------\n"; |
|
3257
|
|
|
|
|
3258
|
|
|
$display_str .= "\033[47;30mtaskid\033[0m". str_pad('', self::$taskid_length+2-strlen('taskid')). |
|
3259
|
|
|
"\033[47;30mtaskpid\033[0m". str_pad('', self::$pid_length+2-strlen('taskpid')). |
|
3260
|
|
|
"\033[47;30mmem\033[0m". str_pad('', self::$mem_length+2-strlen('mem')). |
|
3261
|
|
|
"\033[47;30mcollect succ\033[0m". str_pad('', self::$urls_length-strlen('collect succ')). |
|
3262
|
|
|
"\033[47;30mcollect fail\033[0m". str_pad('', self::$urls_length-strlen('collect fail')). |
|
3263
|
|
|
"\033[47;30mspeed\033[0m". str_pad('', self::$speed_length+2-strlen('speed')). |
|
3264
|
|
|
"\n"; |
|
3265
|
|
|
|
|
3266
|
|
|
// "\033[32;40m [OK] \033[0m" |
|
3267
|
|
|
$task_status = $this->get_task_status_list(self::$serverid, self::$tasknum); |
|
|
|
|
|
|
3268
|
|
|
foreach ($task_status as $json) |
|
|
|
|
|
|
3269
|
|
|
{ |
|
3270
|
|
|
$task = json_decode($json, true); |
|
3271
|
|
|
if (empty($task)) |
|
3272
|
|
|
{ |
|
3273
|
|
|
continue; |
|
3274
|
|
|
} |
|
3275
|
|
|
$display_str .= str_pad($task['id'], self::$taskid_length + 2). |
|
3276
|
|
|
str_pad($task['pid'], self::$pid_length + 2). |
|
3277
|
|
|
str_pad($task['mem'].'MB', self::$mem_length + 2). |
|
3278
|
|
|
str_pad($task['collect_succ'], self::$urls_length). |
|
3279
|
|
|
str_pad($task['collect_fail'], self::$urls_length). |
|
3280
|
|
|
str_pad($task['speed'].'/s', self::$speed_length + 2). |
|
3281
|
|
|
"\n"; |
|
3282
|
|
|
} |
|
3283
|
|
|
//echo "\033[9;0H"; |
|
3284
|
|
|
return $display_str; |
|
3285
|
|
|
} |
|
3286
|
|
|
|
|
3287
|
|
|
public function display_server_ui() |
|
3288
|
|
|
{ |
|
3289
|
|
|
$display_str = "-------------------------------\033[47;30m SERVER \033[0m------------------------------\n"; |
|
3290
|
|
|
|
|
3291
|
|
|
$display_str .= "\033[47;30mserver\033[0m". str_pad('', self::$server_length+2-strlen('serverid')). |
|
3292
|
|
|
"\033[47;30mtasknum\033[0m". str_pad('', self::$tasknum_length+2-strlen('tasknum')). |
|
3293
|
|
|
"\033[47;30mmem\033[0m". str_pad('', self::$mem_length+2-strlen('mem')). |
|
3294
|
|
|
"\033[47;30mcollect succ\033[0m". str_pad('', self::$urls_length-strlen('collect succ')). |
|
3295
|
|
|
"\033[47;30mcollect fail\033[0m". str_pad('', self::$urls_length-strlen('collect fail')). |
|
3296
|
|
|
"\033[47;30mspeed\033[0m". str_pad('', self::$speed_length+2-strlen('speed')). |
|
3297
|
|
|
"\n"; |
|
3298
|
|
|
|
|
3299
|
|
|
$server_list_json = queue::get('server_list'); |
|
|
|
|
|
|
3300
|
|
|
$server_list = json_decode($server_list_json, true); |
|
|
|
|
|
|
3301
|
|
|
foreach ($server_list as $server) |
|
3302
|
|
|
{ |
|
3303
|
|
|
$serverid = $server['serverid']; |
|
3304
|
|
|
$tasknum = $server['tasknum']; |
|
3305
|
|
|
$mem = 0; |
|
3306
|
|
|
$speed = 0; |
|
3307
|
|
|
$collect_succ = $collect_fail = 0; |
|
3308
|
|
|
$task_status = $this->get_task_status_list($serverid, $tasknum); |
|
|
|
|
|
|
3309
|
|
|
foreach ($task_status as $json) |
|
|
|
|
|
|
3310
|
|
|
{ |
|
3311
|
|
|
$task = json_decode($json, true); |
|
3312
|
|
|
if (empty($task)) |
|
3313
|
|
|
{ |
|
3314
|
|
|
continue; |
|
3315
|
|
|
} |
|
3316
|
|
|
$mem += $task['mem']; |
|
3317
|
|
|
$speed += $task['speed']; |
|
3318
|
|
|
$collect_fail += $task['collect_fail']; |
|
3319
|
|
|
$collect_succ += $task['collect_succ']; |
|
3320
|
|
|
} |
|
3321
|
|
|
|
|
3322
|
|
|
$display_str .= str_pad($serverid, self::$server_length). |
|
3323
|
|
|
str_pad($tasknum, self::$tasknum_length + 2). |
|
3324
|
|
|
str_pad($mem.'MB', self::$mem_length + 2). |
|
3325
|
|
|
str_pad($collect_succ, self::$urls_length). |
|
3326
|
|
|
str_pad($collect_fail, self::$urls_length). |
|
3327
|
|
|
str_pad($speed.'/s', self::$speed_length + 2). |
|
3328
|
|
|
"\n"; |
|
3329
|
|
|
} |
|
3330
|
|
|
return $display_str; |
|
3331
|
|
|
} |
|
3332
|
|
|
|
|
3333
|
|
|
public function display_collect_ui() |
|
3334
|
|
|
{ |
|
3335
|
|
|
$display_str = "---------------------------\033[47;30m COLLECT STATUS \033[0m--------------------------\n"; |
|
3336
|
|
|
|
|
3337
|
|
|
$display_str .= "\033[47;30mfind pages\033[0m". str_pad('', 16-strlen('find pages')). |
|
3338
|
|
|
"\033[47;30mqueue\033[0m". str_pad('', 14-strlen('queue')). |
|
3339
|
|
|
"\033[47;30mcollected\033[0m". str_pad('', 15-strlen('collected')). |
|
3340
|
|
|
"\033[47;30mfields\033[0m". str_pad('', 15-strlen('fields')). |
|
3341
|
|
|
"\033[47;30mdepth\033[0m". str_pad('', 12-strlen('depth')). |
|
3342
|
|
|
"\n"; |
|
3343
|
|
|
|
|
3344
|
|
|
$collect = $this->get_collect_url_num(); |
|
|
|
|
|
|
3345
|
|
|
$collected = $this->get_collected_url_num(); |
|
|
|
|
|
|
3346
|
|
|
$queue = $this->queue_lsize(); |
|
|
|
|
|
|
3347
|
|
|
$fields = $this->get_fields_num(); |
|
|
|
|
|
|
3348
|
|
|
$depth = $this->get_depth_num(); |
|
|
|
|
|
|
3349
|
|
|
$display_str .= str_pad($collect, 16); |
|
|
|
|
|
|
3350
|
|
|
$display_str .= str_pad($queue, 14); |
|
3351
|
|
|
$display_str .= str_pad($collected, 15); |
|
3352
|
|
|
$display_str .= str_pad($fields, 15); |
|
3353
|
|
|
$display_str .= str_pad($depth, 12); |
|
3354
|
|
|
$display_str .= "\n"; |
|
3355
|
|
|
return $display_str; |
|
3356
|
|
|
} |
|
3357
|
|
|
|
|
3358
|
|
|
/** |
|
3359
|
|
|
* 判断是否附件文件 |
|
3360
|
|
|
* |
|
3361
|
|
|
* @return void |
|
3362
|
|
|
* @author seatle <[email protected]> |
|
3363
|
|
|
* @created time :2016-09-23 17:13 |
|
3364
|
|
|
*/ |
|
3365
|
|
|
//public function is_attachment_file($url) |
|
3366
|
|
|
//{ |
|
3367
|
|
|
//$mime_types = $GLOBALS['config']['mimetype']; |
|
3368
|
|
|
//$mime_types_flip = array_flip($mime_types); |
|
3369
|
|
|
|
|
3370
|
|
|
//$pathinfo = pathinfo($url); |
|
3371
|
|
|
//$fileext = isset($pathinfo['extension']) ? $pathinfo['extension'] : ''; |
|
3372
|
|
|
|
|
3373
|
|
|
//$fileinfo = array(); |
|
3374
|
|
|
//// 存在文件后缀并且是配置里面的后缀 |
|
3375
|
|
|
//if (!empty($fileext) && isset($mime_types_flip[$fileext])) |
|
3376
|
|
|
//{ |
|
3377
|
|
|
//stream_context_set_default( |
|
3378
|
|
|
//array( |
|
3379
|
|
|
//'http' => array( |
|
3380
|
|
|
//'method' => 'HEAD' |
|
3381
|
|
|
//) |
|
3382
|
|
|
//) |
|
3383
|
|
|
//); |
|
3384
|
|
|
//// 代理和Cookie以后实现, 方法和 file_get_contents 一样 使用 stream_context_create 设置 |
|
3385
|
|
|
//$headers = get_headers($url, 1); |
|
3386
|
|
|
//if (strpos($headers[0], '302')) |
|
3387
|
|
|
//{ |
|
3388
|
|
|
//$url = $headers['Location']; |
|
3389
|
|
|
//$headers = get_headers($url, 1); |
|
3390
|
|
|
//} |
|
3391
|
|
|
////print_r($headers); |
|
3392
|
|
|
//$fileinfo = array( |
|
3393
|
|
|
//'basename' => isset($pathinfo['basename']) ? $pathinfo['basename'] : '', |
|
3394
|
|
|
//'filename' => isset($pathinfo['filename']) ? $pathinfo['filename'] : '', |
|
3395
|
|
|
//'fileext' => isset($pathinfo['extension']) ? $pathinfo['extension'] : '', |
|
3396
|
|
|
//'filesize' => isset($headers['Content-Length']) ? $headers['Content-Length'] : 0, |
|
3397
|
|
|
//'atime' => isset($headers['Date']) ? strtotime($headers['Date']) : time(), |
|
3398
|
|
|
//'mtime' => isset($headers['Last-Modified']) ? strtotime($headers['Last-Modified']) : time(), |
|
3399
|
|
|
//); |
|
3400
|
|
|
|
|
3401
|
|
|
//$mime_type = 'html'; |
|
3402
|
|
|
//$content_type = isset($headers['Content-Type']) ? $headers['Content-Type'] : ''; |
|
3403
|
|
|
//if (!empty($content_type)) |
|
3404
|
|
|
//{ |
|
3405
|
|
|
//$mime_type = isset($GLOBALS['config']['mimetype'][$content_type]) ? $GLOBALS['config']['mimetype'][$content_type] : $mime_type; |
|
3406
|
|
|
//} |
|
3407
|
|
|
//$mime_types_flip = array_flip($mime_types); |
|
3408
|
|
|
//// 判断一下是不是文件名被加什么后缀了, 比如 http://www.xxxx.com/test.jpg?token=xxxxx |
|
3409
|
|
|
//if (!isset($mime_types_flip[$fileinfo['fileext']])) |
|
3410
|
|
|
//{ |
|
3411
|
|
|
//$fileinfo['fileext'] = $mime_type; |
|
3412
|
|
|
//$fileinfo['basename'] = $fileinfo['filename'].'.'.$mime_type; |
|
3413
|
|
|
//} |
|
3414
|
|
|
//} |
|
3415
|
|
|
//return $fileinfo; |
|
3416
|
|
|
//} |
|
3417
|
|
|
|
|
3418
|
|
|
//返回当前是否是主进程 |
|
3419
|
|
|
public function is_taskmaster() |
|
3420
|
|
|
{ |
|
3421
|
|
|
return self::$taskmaster; |
|
|
|
|
|
|
3422
|
|
|
} |
|
3423
|
|
|
|
|
3424
|
|
|
//返回当前是否进程ID |
|
3425
|
|
|
public function get_task_id() |
|
3426
|
|
|
{ |
|
3427
|
|
|
return self::$taskid; |
|
3428
|
|
|
} |
|
3429
|
|
|
|
|
3430
|
|
|
//检测子域名数量 |
|
3431
|
|
|
public function sub_domain_count($url) |
|
3432
|
|
|
{ |
|
3433
|
|
|
if (empty($url)) |
|
3434
|
|
|
{ |
|
3435
|
|
|
return 0; |
|
3436
|
|
|
} |
|
3437
|
|
|
$count = 0; |
|
3438
|
|
|
$domain = $this->getRootDomain($url, 'root'); |
|
3439
|
|
|
if (empty($domain)) |
|
3440
|
|
|
{ |
|
3441
|
|
|
return 0; |
|
3442
|
|
|
} |
|
3443
|
|
|
$host = $this->getRootDomain($url, 'host'); |
|
3444
|
|
|
if (empty($host)) |
|
3445
|
|
|
{ |
|
3446
|
|
|
return $count; |
|
3447
|
|
|
} |
|
3448
|
|
|
if (self::$use_redis) |
|
3449
|
|
|
{ |
|
3450
|
|
|
$count = queue::get($domain); |
|
|
|
|
|
|
3451
|
|
|
if ( ! empty(self::$configs['max_sub_num']) and $count > self::$configs['max_sub_num']) |
|
3452
|
|
|
{ |
|
3453
|
|
|
return $count; |
|
3454
|
|
|
} |
|
3455
|
|
|
if (strlen($host) > 32) |
|
3456
|
|
|
{ |
|
3457
|
|
|
$host = md5($host); |
|
3458
|
|
|
} |
|
3459
|
|
|
$hostkey = 'sub_d-'.$host; |
|
3460
|
|
|
$exists = queue::exists($hostkey); |
|
|
|
|
|
|
3461
|
|
|
if ( ! $exists) |
|
|
|
|
|
|
3462
|
|
|
{ |
|
3463
|
|
|
// 子域名数量加一 |
|
3464
|
|
|
$count = queue::incr($domain); |
|
|
|
|
|
|
3465
|
|
|
queue::set($hostkey, 1); |
|
3466
|
|
|
} |
|
3467
|
|
|
} |
|
3468
|
|
|
return $count; |
|
3469
|
|
|
} |
|
3470
|
|
|
|
|
3471
|
|
|
//提取url的根域名 host domain subdomain name tld |
|
3472
|
|
|
public function getRootDomain($url = '', $type = 'root', $domain_check = false) |
|
3473
|
|
|
{ |
|
3474
|
|
|
if (empty($url)) |
|
3475
|
|
|
{ |
|
3476
|
|
|
return $url; |
|
3477
|
|
|
} |
|
3478
|
|
|
$url = trim($url); |
|
3479
|
|
|
if ( ! preg_match('/^http/i', $url)) |
|
3480
|
|
|
{ |
|
3481
|
|
|
$url = 'http://'.$url; |
|
3482
|
|
|
} |
|
3483
|
|
|
//截取限定字符 |
|
3484
|
|
|
$arr = array(); |
|
3485
|
|
|
if (preg_match_all('/(^https?:\/\/[\p{Han}a-zA-Z0-9\-\.\/]+)/iu', $url, $arr)) |
|
3486
|
|
|
{ |
|
3487
|
|
|
$url = $arr['0']['0']; |
|
3488
|
|
|
unset($arr); |
|
3489
|
|
|
} |
|
3490
|
|
|
$url_parse = parse_url(strtolower($url)); |
|
3491
|
|
|
if (empty($url_parse['host'])) |
|
3492
|
|
|
{ |
|
3493
|
|
|
return ''; |
|
3494
|
|
|
} |
|
3495
|
|
|
//host判断快速返回 |
|
3496
|
|
|
if ($domain_check === false and $type == 'host') |
|
3497
|
|
|
{ |
|
3498
|
|
|
return $url_parse['host']; |
|
3499
|
|
|
} |
|
3500
|
|
|
|
|
3501
|
|
|
//结束数组初始化 |
|
3502
|
|
|
$res = array( |
|
3503
|
|
|
'scheme' => '', |
|
3504
|
|
|
'host' => '', |
|
3505
|
|
|
'path' => '', |
|
3506
|
|
|
'name' => '', |
|
3507
|
|
|
'domain' => '', |
|
3508
|
|
|
); |
|
3509
|
|
|
|
|
3510
|
|
|
$urlarr = explode('.', $url_parse['host']); |
|
3511
|
|
|
$count = count($urlarr); |
|
3512
|
|
|
$res['scheme'] = $url_parse['scheme']; |
|
3513
|
|
|
$res['host'] = $url_parse['host']; |
|
3514
|
|
|
if ( ! empty($url_parse['path'])) |
|
3515
|
|
|
{ |
|
3516
|
|
|
$res['path'] = $url_parse['path']; |
|
3517
|
|
|
} |
|
3518
|
|
|
#列举域名中固定元素 |
|
3519
|
|
|
$state_domain = array('com', 'edu', 'gov', 'int', 'mil', 'net', 'org', 'biz', 'info', 'pro', 'name', 'coop', 'aero', 'xxx', 'idv', 'mobi', 'cc', 'me', 'jp', 'uk', 'ws', 'eu', 'pw', 'kr', 'io', 'us', 'cn', 'al', 'dz', 'af', 'ar', 'ae', 'aw', 'om', 'az', 'eg', 'et', 'ie', 'ee', 'ad', 'ao', 'ai', 'ag', 'at', 'au', 'mo', 'bb', 'pg', 'bs', 'pk', 'py', 'ps', 'bh', 'pa', 'br', 'by', 'bm', 'bg', 'mp', 'bj', 'be', 'is', 'pr', 'ba', 'pl', 'bo', 'bz', 'bw', 'bt', 'bf', 'bi', 'bv', 'kp', 'gq', 'dk', 'de', 'tl', 'tp', 'tg', 'dm', 'do', 'ru', 'ec', 'er', 'fr', 'fo', 'pf', 'gf', 'tf', 'va', 'ph', 'fj', 'fi', 'cv', 'fk', 'gm', 'cg', 'cd', 'co', 'cr', 'gg', 'gd', 'gl', 'ge', 'cu', 'gp', 'gu', 'gy', 'kz', 'ht', 'nl', 'an', 'hm', 'hn', 'ki', 'dj', 'kg', 'gn', 'gw', 'ca', 'gh', 'ga', 'kh', 'cz', 'zw', 'cm', 'qa', 'ky', 'km', 'ci', 'kw', 'hr', 'ke', 'ck', 'lv', 'ls', 'la', 'lb', 'lt', 'lr', 'ly', 'li', 're', 'lu', 'rw', 'ro', 'mg', 'im', 'mv', 'mt', 'mw', 'my', 'ml', 'mk', 'mh', 'mq', 'yt', 'mu', 'mr', 'um', 'as', 'vi', 'mn', 'ms', 'bd', 'pe', 'fm', 'mm', 'md', 'ma', 'mc', 'mz', 'mx', 'nr', 'np', 'ni', 'ne', 'ng', 'nu', 'no', 'nf', 'na', 'za', 'aq', 'gs', 'pn', 'pt', 'se', 'ch', 'sv', 'yu', 'sl', 'sn', 'cy', 'sc', 'sa', 'cx', 'st', 'sh', 'kn', 'lc', 'sm', 'pm', 'vc', 'lk', 'sk', 'si', 'sj', 'sz', 'sd', 'sr', 'sb', 'so', 'tj', 'tw', 'th', 'tz', 'to', 'tc', 'tt', 'tn', 'tv', 'tr', 'tm', 'tk', 'wf', 'vu', 'gt', 've', 'bn', 'ug', 'ua', 'uy', 'uz', 'es', 'eh', 'gr', 'hk', 'sg', 'nc', 'nz', 'hu', 'sy', 'jm', 'am', 'ac', 'ye', 'iq', 'ir', 'il', 'it', 'in', 'id', 'vg', 'jo', 'vn', 'zm', 'je', 'td', 'gi', 'cl', 'cf', 'yr', 'arpa', 'museum', 'asia', 'ax', 'bl', 'bq', 'cat', 'cw', 'gb', 'jobs', 'mf', 'rs', 'su', 'sx', 'tel', 'travel', 'shop', 'ltd', 'store', 'vip', '网店', '中国', '公司', '网络', 'co.il', 'co.nz', 'co.uk', 'me.uk', 'org.uk', 'com.sb', '在线', '中文网', '移动', 'wang', 'club', 'ren', 'top', 'website', 'cool', 'company', 'city', 'email', 'market', 'software', 'ninja', '我爱你', 'bike', 'today', 'life', 'space', 'pub', 'site', 'help', 'link', 'photo', 'video', 'click', 'pics', 'sexy', 'audio', 'gift', 'tech', '网址', 'online', 'win', 'download', 'party', 'bid', 'loan', 'date', 'trade', 'red', 'blue', 'pink', 'poker', 'green', 'farm', 'zone', 'guru', 'tips', 'land', 'care', 'camp', 'cab', 'cash', 'limo', 'toys', 'tax', 'town', 'fish', 'fund', 'fail', 'house', 'shoes', 'media', 'guide', 'tools', 'solar', 'watch', 'cheap', 'rocks', 'news', 'live', 'lawyer', 'host', 'wiki', 'ink', 'design', 'lol', 'hiphop', 'hosting', 'diet', 'flowers', 'car', 'cars', 'auto', 'mom', 'cq', 'he', 'nm', 'ln', 'jl', 'hl', 'js', 'zj', 'ah', 'jx', 'ha', 'hb', 'gx', 'hi', 'gz', 'yn', 'xz', 'qh', 'nx', 'xj', 'xyz', 'xin', 'science', 'press', 'band', 'engineer', 'social', 'studio', 'work', 'game', 'kim', 'games', 'group', '集团'); |
|
3520
|
|
|
if ($count <= 2) |
|
3521
|
|
|
{ |
|
3522
|
|
|
#当域名直接根形式不存在host部分直接输出 |
|
3523
|
|
|
$last = array_pop($urlarr); |
|
3524
|
|
|
$last_1 = array_pop($urlarr); |
|
3525
|
|
|
if (in_array($last, $state_domain)) |
|
3526
|
|
|
{ |
|
3527
|
|
|
$res['domain'] = $last_1.'.'.$last; |
|
3528
|
|
|
$res['name'] = $last_1; |
|
3529
|
|
|
$res['tld'] = $last; |
|
3530
|
|
|
} |
|
3531
|
|
|
} |
|
3532
|
|
|
elseif ($count > 2) |
|
3533
|
|
|
{ |
|
3534
|
|
|
$last = array_pop($urlarr); |
|
3535
|
|
|
$last_1 = array_pop($urlarr); |
|
3536
|
|
|
$last_2 = array_pop($urlarr); |
|
3537
|
|
|
$res['domain'] = $last_1.'.'.$last; //默认为n.com形式 |
|
3538
|
|
|
$res['name'] = $last_2; |
|
3539
|
|
|
|
|
3540
|
|
|
//排除非标准 ltd 域名 |
|
3541
|
|
|
if ( ! in_array($last, $state_domain)) |
|
3542
|
|
|
{ |
|
3543
|
|
|
return false; |
|
3544
|
|
|
} |
|
3545
|
|
|
|
|
3546
|
|
|
if (in_array($last, $state_domain)) |
|
3547
|
|
|
{ |
|
3548
|
|
|
$res['domain'] = $last_1.'.'.$last; //n.com形式 |
|
3549
|
|
|
$res['name'] = $last_1; |
|
3550
|
|
|
$res['tld'] = $last; |
|
3551
|
|
|
} |
|
3552
|
|
|
//排除顶级根二级后缀 |
|
3553
|
|
|
if ($last_1 !== $last and in_array($last_1, $state_domain) and ! in_array($last, array('com', 'net', 'org', 'edu', 'gov'))) |
|
3554
|
|
|
{ |
|
3555
|
|
|
$res['domain'] = $last_2.'.'.$last_1.'.'.$last; //n.n.com形式 |
|
3556
|
|
|
$res['name'] = $last_2; |
|
3557
|
|
|
$res['tld'] = $last_1.'.'.$last; |
|
3558
|
|
|
} |
|
3559
|
|
|
//限定cn顶级根二级后缀为'com', 'net', 'org', 'edu', 'gov' |
|
3560
|
|
|
if (in_array($last, array('cn')) and $last_1 !== $last and strlen($last_1) > 2 and ! in_array($last_1, array('com', 'net', 'org', 'edu', 'gov'))) |
|
3561
|
|
|
{ |
|
3562
|
|
|
$res['domain'] = $last_1.'.'.$last; //n.n.cn形式 |
|
3563
|
|
|
$res['name'] = $last_1; |
|
3564
|
|
|
$res['tld'] = $last; |
|
3565
|
|
|
} |
|
3566
|
|
|
} |
|
3567
|
|
|
|
|
3568
|
|
|
//检测和验证返回的是不是域名格式 |
|
3569
|
|
|
if ( ! empty($res['domain']) and preg_match('/^([\p{Han}a-zA-Z0-9])+([\p{Han}a-zA-Z0-9\-])*\.[a-zA-Z\.\p{Han}]+$/iu', $res['domain'])) |
|
3570
|
|
|
{ |
|
3571
|
|
|
if ($type == 'arr') |
|
3572
|
|
|
{ |
|
3573
|
|
|
return $res; |
|
3574
|
|
|
} |
|
3575
|
|
|
elseif ($type == 'host') |
|
3576
|
|
|
{ |
|
3577
|
|
|
return $res['host']; |
|
3578
|
|
|
} |
|
3579
|
|
|
elseif ($type == 'tld') |
|
3580
|
|
|
{ |
|
3581
|
|
|
return $res['tld']; |
|
3582
|
|
|
} |
|
3583
|
|
|
elseif ($type == 'subdomain') |
|
3584
|
|
|
{ |
|
3585
|
|
|
return $res['name']; |
|
3586
|
|
|
} |
|
3587
|
|
|
else |
|
3588
|
|
|
{ |
|
3589
|
|
|
return $res['domain']; |
|
3590
|
|
|
} |
|
3591
|
|
|
} |
|
3592
|
|
|
else |
|
3593
|
|
|
{ |
|
3594
|
|
|
return ''; |
|
3595
|
|
|
} |
|
3596
|
|
|
} |
|
3597
|
|
|
|
|
3598
|
|
|
} |
|
3599
|
|
|
|
ricardosierra /
phpspider
Loading history...
In general, usage of exit should be done with care and only when running in a scripting context like a CLI script.