求用WWW::Mechanize模块提交表单，字段为中文提交后乱码的解决办法？

iamlimeng

Perl有部分模块对中文支持不到位，比如GD，楼主的问题也可能是因为WWW::Mechanize对中文支持不好。用LWP实现了一下，代码有点啰嗦，将就一下吧：

1. #!/usr/bin/perl
   
2. 
   
3. use strict;
   
4. use warnings;
   
5. use LWP;
   
6. use HTTP::Cookies;
   
7. use HTTP::Response;
   
8. use LWP::ConnCache;
   
9. use Encode;
   
10. 
    
11. my $username = "abcdef";
    
12. my $password = "123456";
    
13. my $recipient = "美女";
    
14. my $topic = "测试 by 立虻 20100905";
    
15. my $content = "明月几时有，把酒问清天，不知天上宫阙，今夕是何年!";
    
16. 
    
17. my $lwp = LWP::UserAgent->new;
    
18. my $conncache = new LWP::ConnCache;
    
19. $lwp->conn_cache($conncache);
    
20. 
    
21. ######################## LOGIN #########################
    
22. my $login_url = "http://passport.libaclub.com/login.php?act=submit";
    
23. my %login_form = (
    
24.         "username" => $username,
    
25.         "password" => $password,
    
26. );
    
27. my $login_response = $lwp->post($login_url,\%login_form);
    
28. my $login_cookie;
    
29. my $sessionhash;
    
30. if ($login_response->is_success) {
    
31.         my $cookie_jar = HTTP::Cookies->new;
    
32.         $cookie_jar->extract_cookies($login_response);
    
33.          my $cookie_temp = $cookie_jar->as_string();
    
34.          ($sessionhash) = ($cookie_temp =~ /.*?sessionhash=(.*?)\;.*/i);
    
35.          ($login_cookie) = ($cookie_temp =~ /Set-Cookie3: (.*)/i);
    
36.          chomp($login_cookie);
    
37. }
    
38. else {
    
39.         print "Login Error: ".$login_response->status_line;
    
40. }
    
41. #print "sessionhash: $sessionhash\n";
    
42. 
    
43. ######################## GET FORM URL #########################
    
44. my $form_url = "http://bbs.sh.libaclub.com/message.php?action=write";
    
45. $lwp->default_header('Cookie'=>$login_cookie);
    
46. my $form_response = $lwp->get($form_url);
    
47. my $passid;
    
48. if ($form_response->is_success) {
    
49.         my $cookie_jar = HTTP::Cookies->new;
    
50.         $cookie_jar->extract_cookies($form_response);
    
51.          my $cookie_temp = $cookie_jar->as_string();
    
52.         ($passid) = ($cookie_temp =~ /.*?PHPSESSID=(.*?)\;.*/i);
    
53. }
    
54. else {
    
55.         print "Form url request error: ".$form_response->status_line;
    
56. }
    
57. #print "passid $passid\n";
    
58. 
    
59. ######################## SEND MESSAGE #########################
    
60. my $send_url = "http://bbs.sh.libaclub.com/message.php?action=doSend";
    
61. my %send_form = (
    
62.          "name" => "form1",
    
63.          "receiverName" => $recipient,
    
64.          "topic" => $topic,
    
65.          "content" => $content,
    
66. );
    
67. my $send_cookie = qq~ sessionhash=$sessionhash; PHPSESSID=$passid~;
    
68. #print "Send_Cookie $send_cookie\n";
    
69. $lwp->default_header('Cookie'=>$send_cookie);
    
70. my $send_response = $lwp->post($send_url,\%send_form);
    
71. if ($send_response->is_success) {
    
72.         my $html = $send_response->decoded_content;
    
73.         if (encode('gbk',$html) =~ /发送成功/) {
    
74.                  print "Data send sucessful!";
    
75.          }
    
76.          else { print "Data send failed!"; }
    
77. }
    
78. else {
    
79.         print "Data send error: ".$send_response->status_line;
    
80. }
    
81. print "\7";
    
82. 
    
83. <STDIN>;

复制代码

lanparky

这是由于提交时HTML::Form自动编码的问题,如果没有指定编码格式,默认是utf-8,如果指定了gbk,接收的地址必须有对应的解码程序才可以正常显示中文.所以修改下模块HTML::Form (先把文件的只读取消),在HTML::Form里找到

1. eval { require Encode };
   
2. $Encode_available = !$@;

复制代码

在后面加上

1. $Encode_available=0;

复制代码

把HTML:Form自动编码的功能关闭.提交程序代码改为

1. 
   
2. use WWW::Mechanize;
   
3. use Encode;
   
4. use LWP::ConnCache;
   
5. 
   
6. 
   
7. $url="http://bbs.sh.libaclub.com/message.php?action=write";
   
8. 
   
9. my $mech=WWW::Mechanize->new();
   
10. $mech->conn_cache(LWP::ConnCache->new);
    
11. $mech->agent_alias('Windows IE 6');
    
12. $mech->add_header(Referer=>' http://bbs.sh.libaclub.com/message.php');
    
13. $mech->add_header("Accept-Charset" =>'GBK;q=0.7,*;q=0.7');
    
14. $mech->add_header("Accept-Language"=>'zh-cn;q=0.5');
    
15. 
    
16. $mech->add_header(Cookie=>'sessionhash=fa55180844a4152e93b64fc17482b53f; __utma=87665927.1972509085.1283527516.1283616492.1283653233.7; __utmz=87665927.1283527516.1.1.utmcsr=(direct)|utmccn=(direct)|utmcmd=(none); __utmc=87665927; mytopictabs=0; navtoptabs=0; PHPSESSID=1c4aafe2bad88da7cb26b89cd514dd42; __utmb=87665927.6.10.1283653233');
    
17. $mech->get($url);
    
18. my $res = $mech->res;
    
19. ##关闭decoded_content会自动解码,把返回的内容按页面原样输出
    
20. $html= $res->decoded_content(charset => 'none');
    
21. $mech->update_html( $html );
    
22. $form=$mech->form_name("form1");  
    
23. $mech->field("receiverName","sx98083714");
    
24. my $subject="测试题目";
    
25. $mech->field("topic",$subject);
    
26. my $content="明月几时有，把酒问清天，不知天上宫阙，今夕是何年！";
    
27. $mech->field("content",$content);
    
28. $mech->submit();
    
29. $res = $mech->res;
    
30. $html= $res->decoded_content(charset => 'none');
    
31. $mech->update_html( $html );
    
32. print $mech->dump_all();  
    

复制代码

sx98083714

特感谢 iamlimeng  、lanparky 两位兄弟的热心相助！！！

感谢之情不知如何言表，刚好快中秋了，祝二位兄弟中秋节：全家团圆，工作顺利！！！