file(), nulls, extracting data between spaces?

[4rxsid]

Hi all,

I'm reading in a file that is generated by a mainframe job. What I'm trying to do is extract out each number ($piece), from within each "category", so that I can insert each $piece into a mysql db for each corresponding cat-type doc or cat-type line.
So far, the code below works with one exception...the last part of each line echo's out null, but yet it's detected as not null. Any ideas why piece 72 would be echo'd out? b.t.w. there are no tabs between the numbers, no newlines and a variable number of spaces between the numbers.

Input file:

REPORT NO. SOMEONE

PROGRAM ID RPT444

RUN DATE 01/29/04 INPUT CONTROL LIST

--HEADER 1-- --HEADER 2-- --HEADER 3-- --HEADER 4--

CAT TYPE DOCS/LINES DOCS/LINES DOCS/LINES DOCS/LINES TOTALS DOC/LINES

CATEGORY1 4,617 16,859 0 36 21,512

8,031 17,645 0 47 25,723

CATEGORY2 705 2,768 1,058 15 4,546

2,102 14,548 1,058 15 17,723

CATEGORY3 480 1,916 67 66 2,529

2,844 19,262 752 560 23,418

CATEGORY4 11,938 101,336 1,923 407 115,604

34,012 192,574 5,082 728 232,396

TOTAL 68,179 331,617 11,045 751 411,592

119,865 429,442 24,528 1,675 575,510

$lines = file($filePath.$fileIN);
    foreach ($lines as $line) {
        //**** Right now just working on 1 category for testing ****
        if (stristr($line,"CATEGORY1")) {
            $line = preg_replace("/CATEGORY1/","",$line);
            $pieces = explode(" ",$line);
            echo "Pieces Count: ".count($pieces)."\n";
            for ($x=0;$x<count($pieces);$x++) {
            if ($pieces[$x] != null) {
                echo "Piece $x: $pieces[$x]\n";
            }
        }
    }
}

Output is:
Pieces Count: 73
Piece 9: 4,617
Piece 17: 16,859
Piece 30: 0
Piece 45: 36
Piece 62: 21,512
Piece 72:

Why is Piece 72 echo'd out as being non null, when it is null in the file?

Also...any function I'm missing that could clean this up a bit, as far as extracting numbers only in each $line with those variable spaces?

Thanks in advance!

[devinemke]

foreach ($lines as $line)
{ 
	if (stristr ($line, 'CATEGORY1'))
	{
		$line = trim ($line);
		$line = str_replace ('CATEGORY1 ', '', $line);
		$pieces = explode (' ',$line); 
		echo 'Pieces Count: ' . count ($pieces) . '<br><br>';
		foreach ($pieces as $key => $piece) {echo 'Piece ' . $key . ': ' . $piece . '<br>';}
	}
}

[4rxsid]

devinemke...thanks for the reply.

the rtrim() was definately what i was missing for removing the newline.

im wondering, is your code here:

foreach ($pieces as $key => $piece) {echo 'Piece ' . $key . ': ' . $piece . '<br>';} 

I'd still need to check for ! null in yours above.

more efficient than mine?:

for ($x=0;$x<count($pieces);$x++) { 
    if ($pieces[$x] != null) { 
        echo "Piece $x: $pieces[$x]\n"; 
    } 
}

as far as foreach vs for loops?

[devinemke]

yes, you should always use foreach loops to traverse arrays (if you have PHP >= 4.0). foreach loops are faster than for loops.

as far as checking for NULL, maybe i'm still not undertsanding the structure of your text file. why are you checking for NULL? of you simply want to ignore the 0 integers on the text lines then simply do:

foreach ($pieces as $key => $piece)
{
	if ($piece)
	{
		echo 'Piece ' . $key . ': ' . $piece . '<br>';
	}
}

keep in mind that your "'Pieces Count" will be misleading because it will report the total number if integers per line (including the 0's)

[4rxsid]

the pieces count isn't need, it's just for debugging for me and i should've taken it out of the post.

what i need is each individual number that's found on each line wether or not it's =0 or >0, i need all numbers (as individul numbers) from each line.

[devinemke]

what i need is each individual number that's found on each line

then my original code sample will do just that

[4rxsid]

that works great...with one exception, it doesn't include the 0's (zeros) that are in each line. i need to grab the zero's as well, so I'll have to use:

foreach ($pieces as $key => $piece) 
{ 
    if ($piece != "") 
    { 
        echo 'Piece ' . $key . ': ' . $piece . '<br>'; 
    } 
}

thanks for the suggestions/info.