Така и така го написах зада пробвам нещо, реших да Ви го дам с перспективата да помогне на някои (с малко преправяне).
traffic:/tmp# cat wikisearch.php
<?php
$DIR="/var/www/dcableusers/amri/wikisearch";
$SDB="$DIR/search_words.txt";
$CDB="$DIR/correct_words.html";
unlink($CDB);
function utf8_to_cp1251($s)
{
for ($c=0;$c<strlen($s);$c++)
{
$i=ord($s[$c]);
if ($i<=127) $out.=$s[$c];
if ($byte2){
$new_c2=($c1&3)*64+($i&63);
$new_c1=($c1>>2)&5;
$new_i=$new_c1*256+$new_c2;
if ($new_i==1025){
$out_i=168;
} else {
if ($new_i==1105){
$out_i=184;
} else {
$out_i=$new_i-848;
}
}
$out.=chr($out_i);
$byte2=false;
}
if (($i>>5)==6) {
$c1=$i;
$byte2=true;
}
}
return $out;
}
function chkw($sword)
{
$ch = curl_init("http://bg.wikipedia.org/wiki/$sword");
curl_setopt ($ch,CURLOPT_POST, 1);
curl_setopt($ch,CURLOPT_USERAGENT,"Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.1.10) Gecko/20071115 Firefox/2.0.0.10");
curl_setopt($ch,CURLOPT_REFERER,"http://www.google.bg/search?q=wikipedia&ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-US
fficial&client=firefox-a");
curl_setopt($ch,CURLOPT_RETURNTRANSFER, 1);
$data=curl_exec($ch);
$data=utf8_to_cp1251($data);
$left=explode('var wgPageName = "',$data);
$right=explode('"',$left[1]);
$editlink=$right[0];
curl_close ($ch);
if(eregi("Уикипедия все още не разполага със статия с точно това име","$data"))
$error=1;
else
$error=0;
return "$error $editlink";
}
$lines = file("$SDB");
$fp = fopen("$CDB", 'a');
fwrite ($fp, "<html><head><title>Съществуващи записи във Wikipedia</title><meta http-equiv=\"REFRESH\" content=\"300;URL=correct_words.html\"><meta http-equiv=\"Content-Type\" Content=\"text/html; charset=windows-1251\"></head><body><center><h2><font color='green'>Съществуващи записи във Wikipedia</font></h2><br><table border='1'><tr><td><b>N</b></td><td><b>Wikipedia Link</b></td><td><b>Copytext wikipedia</b></td></tr>");
$i=0;
foreach ($lines as $line_num => $dword)
{
$dword=trim(str_replace(" ","_",$dword));
$infoword=explode(" ",chkw($dword));
if ( $infoword[0] != 1 )
{
$i++;
$all="<tr><td><b>$i</b></td><td><a href='http://bg.wikipedia.org/w/index.php?title=$infoword[1]&action=edit' target='_blank'>$dword</a></td><td>* [http://www.kakvo.org/translate_word/$dword Английско-Български превод на $dword] (бълг.)</td></tr>";
fwrite ($fp, $all);
};
$random=rand(0,5);
sleep($random);
echo "[>$i<] $dword [->wait: $random secs\n";
}
fwrite ($fp, "</table></body></html>");
fclose($fp);
?>
traffic:/tmp#
traffic:/tmp# cat wikisearch.php
<?php
$DIR="/var/www/dcableusers/amri/wikisearch";
$SDB="$DIR/search_words.txt";
$CDB="$DIR/correct_words.html";
unlink($CDB);
function utf8_to_cp1251($s)
{
for ($c=0;$c<strlen($s);$c++)
{
$i=ord($s[$c]);
if ($i<=127) $out.=$s[$c];
if ($byte2){
$new_c2=($c1&3)*64+($i&63);
$new_c1=($c1>>2)&5;
$new_i=$new_c1*256+$new_c2;
if ($new_i==1025){
$out_i=168;
} else {
if ($new_i==1105){
$out_i=184;
} else {
$out_i=$new_i-848;
}
}
$out.=chr($out_i);
$byte2=false;
}
if (($i>>5)==6) {
$c1=$i;
$byte2=true;
}
}
return $out;
}
function chkw($sword)
{
$ch = curl_init("http://bg.wikipedia.org/wiki/$sword");
curl_setopt ($ch,CURLOPT_POST, 1);
curl_setopt($ch,CURLOPT_USERAGENT,"Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.1.10) Gecko/20071115 Firefox/2.0.0.10");
curl_setopt($ch,CURLOPT_REFERER,"http://www.google.bg/search?q=wikipedia&ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-US
fficial&client=firefox-a");curl_setopt($ch,CURLOPT_RETURNTRANSFER, 1);
$data=curl_exec($ch);
$data=utf8_to_cp1251($data);
$left=explode('var wgPageName = "',$data);
$right=explode('"',$left[1]);
$editlink=$right[0];
curl_close ($ch);
if(eregi("Уикипедия все още не разполага със статия с точно това име","$data"))
$error=1;
else
$error=0;
return "$error $editlink";
}
$lines = file("$SDB");
$fp = fopen("$CDB", 'a');
fwrite ($fp, "<html><head><title>Съществуващи записи във Wikipedia</title><meta http-equiv=\"REFRESH\" content=\"300;URL=correct_words.html\"><meta http-equiv=\"Content-Type\" Content=\"text/html; charset=windows-1251\"></head><body><center><h2><font color='green'>Съществуващи записи във Wikipedia</font></h2><br><table border='1'><tr><td><b>N</b></td><td><b>Wikipedia Link</b></td><td><b>Copytext wikipedia</b></td></tr>");
$i=0;
foreach ($lines as $line_num => $dword)
{
$dword=trim(str_replace(" ","_",$dword));
$infoword=explode(" ",chkw($dword));
if ( $infoword[0] != 1 )
{
$i++;
$all="<tr><td><b>$i</b></td><td><a href='http://bg.wikipedia.org/w/index.php?title=$infoword[1]&action=edit' target='_blank'>$dword</a></td><td>* [http://www.kakvo.org/translate_word/$dword Английско-Български превод на $dword] (бълг.)</td></tr>";
fwrite ($fp, $all);
};
$random=rand(0,5);
sleep($random);
echo "[>$i<] $dword [->wait: $random secs\n";
}
fwrite ($fp, "</table></body></html>");
fclose($fp);
?>
traffic:/tmp#
