Get all links by reading the source files for a site

Read the source file of a site, and then use regular analysis of its source code, to get all the links /**********qiushuiwuhen (2002-5-20) ***********/ if (empty ($url)) $url = "http://www.csdn.net/expert/";//Set URL $site =substr ($url, 0,strpos ($url, "/", 8));//Site $base =substr ($url, 0,strrpos ($url, "/") 1);//directory where the file is located $fp = fopen ($url, "R");//Open URL while (!feof ($FP)) $contents. =fread ($FP, 1024);// $pattern = "|href=["]? ([^ '\"] ) ['\" ]| U "; Preg_match_all ($pattern, $contents, $REGARR, preg_set_order);//Match All href= for ($i =0; $i if (!eregi ("://", $REGARR [$i][1])//is the relative path, that is, whether there is:// if (substr ($regArr [$i][1],0,1) = = "/")//is the root directory of the site echo "link". ($i 1). ":". $site. $regArr [$i][1]. " ";//root directory Else echo "link". ($i 1). ":". $base. $regArr [$i][1]. " ";//Current directory Else echo "link". ($i 1). ":" $regArr [$i][1]. " ";//relative path } Fclose ($FP); ?> Copy Code

