Parsing word docs without COM object?

[envytomdead]

Has anyone done this or have any ideas? Would another LAMP language be better to attempt this?

 

-Thanks

[corbin]

Under some legal thing, MS had to release the specifications of the format of all of their formats (Word 03 is what I mean...  Word 07 is a known format), so you could make something if you're desperate and have a ton of time.  Other than that, I guess just try googling.

 

 

 

By the way, a "LAMP" language doesn't make sense since LAMP is generally Linux Apache MySQL PHP.

[envytomdead]

Sometimes the P in LAMP can stand Perl or Python. That is what I meant by another LAMP language. So I will rephrase that to ask would I be better off trying to do this with a perl or python script.

[corbin]

If I had to guess, I would say you're best off with PHP, then Python then Perl.  I can't imagine a Word class in Perl or Python, although I think finding one in Python would be more likely than Perl.

[Maq]

Sometimes the P in LAMP can stand Perl or Python. That is what I meant by another LAMP language. So I will rephrase that to ask would I be better off trying to do this with a perl or python script.

 

You're right, but since you posted in the PHP Help section it's assumed you're using PHP.

[Mark Baker]

I've seen at least one library on phpClasses for reading word documents in pure PHP. Can't comment on whether it's any good, and you'd have to search for it, but it might provide a starting point for you

[envytomdead]

Ok I found this bit of code that works for the most part.  However there are still some issues with the first line getting lost.

 

The code checks to see if the line contains a NULL character or if the string length is 0 if it doesn't meet either of those conditions it cleans up the line and adds it to the output.

The problem is there is occasionally good data in a line that contains 0x00. I found a way rid of the garbage characters. See the commented out $mytemp lines. However there are a lot of random strings that would look like good data (z @YYYY0@UnknownGTimes New Roman5Symbol3Arial qh2fSF Nkg4XNx) any ideas?

 

Thanks

 

function parseWord($userDoc)

{

    $fileHandle =fopen($userDoc,"r");

    $line = @fread($fileHandle,filesize($userDoc)); 

    $lines = explode(chr(0x0D),$line);

    //var_dump($lines);

    $outtext = "";

    $count=0;

    foreach($lines as $thisline)

    {

      $pos =strpos($thisline,chr(0x00));

        if(($pos !== FALSE) || (strlen($thisline) == 0))

          {

          // $mytemp=strtr($thisline,'','');

          //$mytemp=preg_replace('/[^\b\s]/','',$mytemp);

          // $mytemp=preg_replace("/[^a-zA-Z0-9\s\,\.\-\n\r\t@\/\_\(\)]/","",$thisline);

         

        //  if($count==0 ){

          // echo "The position in line ".$count." is ".$pos."<br />";

          //echo $mytemp."<br />";

          // }

           

          }else{

            $outtext .= $thisline . " ";

          }

      $count++;

    }

    $outtext =preg_replace("/[^a-zA-Z0-9\s\,\.\-\n\r\t@\/\_\(\)]/","",$outtext);

    return $outtext;

}