<?php
/**
* 采集药监局国内药品库
*
*/
require dirname(__DIR__) . '/conf.php';
for($i=1;$i<=11248;$i++){
//采集标题和内页url
$commend = 'curl "http://app1.sfda.gov.cn/datasearch/face3/search.jsp" -H "Host: app1.sfda.gov.cn" -H "User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:34.0) Gecko/20100101 Firefox/34.0" -H "Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8" -H "Accept-Language: zh-cn,zh;q=0.8,en-us;q=0.5,en;q=0.3" --compressed -H "Cache-Control: no-cache" -H "Content-Type: application/x-www-form-urlencoded; charset=UTF-8" -H "Referer: http://app1.sfda.gov.cn/datasearch/face3/base.jsp?tableId=25&tableName=TABLE25&title="%"B9"%"FA"%"B2"%"FA"%"D2"%"A9"%"C6"%"B7&bcId=124356560303886909015737447882" -H "Cookie: JSESSIONID=275C0B7B0067B4817BC0A321587B4A2F.7" -H "Connection: keep-alive" -H "Pragma: no-cache" --data "tableId=25&State=1&bcId=124356560303886909015737447882&State=1&tableName=TABLE25&State=1&viewtitleName=COLUMN167&State=1&viewsubTitleName=COLUMN166,COLUMN170,COLUMN821&State=1&curstart='.$i.'&State=1&tableView="%"25E5"%"259B"%"25BD"%"25E4"%"25BA"%"25A7"%"25E8"%"258D"%"25AF"%"25E5"%"2593"%"2581&State=1"';
$result = exec($commend, $output);
$result1 = array();
foreach($output as $value){
if(strpos($value,'国药')){
$result1[] = trim($value);
}
}
//采集 批准文号 和 剂量
foreach($result1 as $value1){
$detail = array();
$html = phpQuery::newDocumentHTML($value1);
preg_match("/<a href=(.*?)>(.*)/i", $value1, $match);
$arr1 = explode("'", $match[1]);
$commend2 = 'curl "http://app1.sfda.gov.cn/datasearch/face3/'.$arr1[1].'" -H "Host: app1.sfda.gov.cn" -H "User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:34.0) Gecko/20100101 Firefox/34.0" -H "Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8" -H "Accept-Language: zh-cn,zh;q=0.8,en-us;q=0.5,en;q=0.3" --compressed -H "Content-Type: text/html;encoding=gbk" -H "Referer: http://app1.sfda.gov.cn/datasearch/face3/base.jsp?tableId=25&tableName=TABLE25&title="%"B9"%"FA"%"B2"%"FA"%"D2"%"A9"%"C6"%"B7&bcId=124356560303886909015737447882" -H "Cookie: JSESSIONID=275C0B7B0067B4817BC0A321587B4A2F.7" -H "Connection: keep-alive"';
exec($commend2, $output2);
foreach($output2 as $key2 => $value2){
if(trim(strip_tags($value2)) == '批准文号'){
$detail['approve_code'] = trim(strip_tags($output2[$key2+1]));
}
if(trim(strip_tags($value2)) == '剂型'){
$detail['dosage'] = trim(strip_tags($output2[$key2+1]));
}
}
$id = $db->insert('39_sfda_db', $detail);
if($id){
echo $id."抓取成功\r\n";
}else{
echo $id."抓取失败\r\n";
}
}
}
浙公网安备 33010602011771号