继续 新闻抓取

抓取一例，以

#!/usr/bin/perl
use DBI;
use Encode;
# Connect to target DB
my $dbh = DBI->connect("DBI:mysql:database=news_db;host=127.0.0.1","root","", {'RaiseError' => 0});

`rm -f fenghuan_tmp`;
`wget -O fenghuan_tmp "http://news.ifeng.com/" -o .log`;
my %hash;
my $domain = "ifeng.com";
my $category = decode("utf-8","滚动新闻");
#滚动新闻
open(FH,"fenghuan_tmp");
while($line=<FH>){
    if($line=~/<h3 class="sysNW fz20">/){
        while($line=<FH>){
            if($line=~/^href=/"([^/"]+)/" [^>]+>([^<]+)/){
                my $url = $1;
                my $con = $2;
                if($url=~//.shtml$/){
                    $hash{$con} = $url;
                    #print $url."/t".$con."/n";
                }
            }elsif($line=~/<!--左侧 end-->/){
                last;
            }
        }
        last;
    }
}
close(FH);
foreach my $k (keys %hash){
    my $url = $hash{$k};
    &get_page($url,$k,$category,$domain);
}
undef %hash;