parsing a page line by line

sarmenhb · 12-03-2008, 05:55 PM

hello,

i have a page which contains about 5000 lines.

each line is in this format

vocab n. definition

how can i have php read the first word and place it into a tables column then the second word/char which is n and place that into another column and the last word into another column?

liveordie · 12-03-2008, 09:49 PM

PHP Code:


			
<?php

//Read the file line by line into an array
$lines = file( '/path/to/file.txt' );

//Loop over the array by line
foreach( $lines as $line )
{
    $parts = array();

    //Explode each line into exactly 3 parts by a space.
    //Allows the definition to have multiple words separated by spaces.
    $parts = explode( ' ', $line, 3 );

    //Print out the three resulting parts
    echo 'Word: ' . $parts[0] . '<br />';
    echo 'Type: ' . $parts[1] . '<br />';
    echo 'Definition: ' . $parts[2] . '<br />';
}

If the second part is an abbreviation like n. v. adj. or something of that sort, you can strip the period of the second part that is returned from the explode function

Salathe · 12-03-2008, 11:19 PM

The only trouble with using file is that you're loading the entire file into memory. Much better is to read the file line by line and do with it what you want.

Here are two approaches using this line of thought.

Plain old fopen/fgets

PHP Code:


			
$file = fopen('dictionary.txt', 'r');

while ( ! feof($file))

{

    $line = fgets($file, 1024);

    list($word, $type, $definition) = explode(' ', rtrim($line), 3);

}

SPL File Object

PHP Code:


			
$file = new SplFileObject('dictionary.txt');

foreach ($file as $line)

{

    list($word, $type, $definition) = explode(' ', rtrim($line), 3);

}

liveordie · 12-04-2008, 01:58 AM

I hadn't even considered the memory usage when I quickly typed that out.

I've never seen SplFileObject used before that is a nice class.
Any idea why it hasn't made it into the official docs yet?