PHP HTML DOM Parser : parse, exclude and export

[aviju]

Hello ! I'm new on this board and I need your help !

 

I explain you my problem. I'd like to collect the content of around 1000 URL in a textfile (I use the wget function in a bash). And then I want to parse this textfile in order to pick one type of content up in a csv file.

 

1) So my bash is this one :

 

file=/home/julien/tests/file.txt

for i in $(cat $file)

do

wget $i -O ->> songs_t.txt;

done

It works perfectly and the textfile songs_t.txt is well created. That textfile contains the content of the 1000 URLs.

 

2) Then I make a php script to parse songs_t.txt. I only want to get concert setlists (the setlist is only a part of a the content of each URL). So my approach is to remove tags such as 'a', 'h4', 'Title' and so on and save the rest in a csv file called 'SONGS.csv' An example of a URL can be seen here : http://members.tripod.com/~fun_fun_fun/8-17-63.html

 

My part of the php script dealing with the parsing is this one :

 

$html = file_get_html('songs_t.txt');

foreach ($html->find('title, script, div, center, style, img, noscript, h4, a') as $es)

$es->outertext = 'title, script, div, center, style, img, noscript, h4, a';



$f = fopen('SONGS.csv', "w");

fwrite ($f, $html);

fclose($f);

 

The script works for the 35 first URL (I nearly only get the setlists) but as soon as the script has to deal with more than 35 URL, I have the following error message :

 

Call to a member function find() on a non-object in /home/julien/tests/boys2.php on line 23.

That line 23 corresponds to :

 

foreach ($html->find('title, script, div, center, style, img, noscript, h4, a') as $es)

 

3) In order to test if my html object is valid, I use that code :

 

html = file_get_html('songs_t.txt');

if (!is_object($html)){ 

echo "invalid object"; 

}

And the result is "invalid object". This test is made in a textfile composed of the content of 50 URL. But when I apply that test on textfile composed of 30 URL, I have no error ! So how can I do to parse my HTML even if it's not a entire valid object ?

 

 

Could you help me please ? 

Thanks !