Array_merge() Question

[Tyrone Slothrop]

I am writing a script to upload data to a delimited text file. The
data is in the format:
recordID \t the rest of the data ...

It is significant that the recordID's are unique and that the newly
uploaded data replaces the old data in the event of duplicate
recordID's.

The solution I came up with (trying to avoid having to loop through
all of the records of both files looking for dupes) was to:
* treat both files as arrays using file()
* assign the keys of each line on each array using the alpha+recordID
(prevents renumbering of key when ...)
* running array_merge ($old_data, $new_data)
* running ksort() on merged data to sort by recordID
* imploding the data on \n and writing to the data file

This seems to be working as planned. However, I am really concerned
that the old data will not be overwritten by new when running
array_merge.

Comments, suggestions, snark, flames? TIA!

The (simplified) code:

$data_file = "D:\\pathto\\data.txt";
$dfile = file($data_file);
$data_in = file(addslashes($_FILES['upload']['tmp_name']));
// I hate Windows servers!

foreach ($dfile as $line) {
$line = trim($line);
if (!empty($line)) { list ($id, $unused) = explode ("\t", $line); }
$odata['a'.$id] = $line;
}

foreach ($data_in as $line) {
$line = trim($line);
if (!empty($line)) { list ($id, $unused) = explode ("\t",$line); }
$ndata['a'.$id] = $line;
}

$mdata = array_merge ($odata, $ndata);

ksort ($mdata);

$fp = fopen ($data_file, 'w');
fwrite ($fp, implode("\n", $mdata));
fclose ($fp);

[Andrew Hutchings]

Tyrone Slothrop wrote:

I am writing a script to upload data to a delimited text file. The
data is in the format:
recordID \t the rest of the data ...

It is significant that the recordID's are unique and that the newly
uploaded data replaces the old data in the event of duplicate
recordID's.

Why aren't you using something like SQLite?

--
Andrew Hutchings - LinuxJedi - http://www.linuxjedi.co.uk/
Windows is the path to the darkside...Windows leads to Blue Screen. Blue
Screen leads to downtime. Downtime leads to suffering...I sense much
Windows in you...

[gosha bine]

Tyrone Slothrop wrote:

I am writing a script to upload data to a delimited text file. The
data is in the format:
recordID \t the rest of the data ...

It is significant that the recordID's are unique and that the newly
uploaded data replaces the old data in the event of duplicate
recordID's.

The solution I came up with (trying to avoid having to loop through
all of the records of both files looking for dupes) was to:
* treat both files as arrays using file()
* assign the keys of each line on each array using the alpha+recordID
(prevents renumbering of key when ...)
* running array_merge ($old_data, $new_data)
* running ksort() on merged data to sort by recordID
* imploding the data on \n and writing to the data file

This seems to be working as planned. However, I am really concerned
that the old data will not be overwritten by new when running
array_merge.

Comments, suggestions, snark, flames? TIA!

The (simplified) code:
[snip]

I've looked into your problem and I think I can suggest simpler code

//

function merge($old, $new) {
$new = str_replace("\t", "~\t", $new);
$all = array_merge($old, $new);
sort($all);
$all = implode("\n", $all);
$all = preg_replace('/(\d+) .* \n \1 ~ (.*)/x', "$1$2", $all);
$all = str_replace('~', '', $all);
return $all;
}

#test

$a = array(
'1 bbb',
'2 ccc',
'7 vvv',
'5 zzz',
);

$b = array(
'3 XXX',
'2 CCCC',
'5 ZZZ',
'9 TTTT',
);
echo merge($a, $b);

//
Although I agree with Andrew you should use databases for this

--
gosha bine

extended php parser ~ http://code.google.com/p/pihipi
blok ~ http://www.tagarga.com/blok