find and replace

StevenSm

I am fairly new to PHP, and am trying to accomplish the following:

I have two files.

1) File A contains master data comprising of records in the form of
<domainname.com>::<IPaddress>

2) File B contains a list of names that are a subset of a.

I need to replace items in B that are also in A with items in A, but the rest of the data in B should be left alone.

ie/
if A contains

abc.com::66.225.222.11
cde.com:: 66.225.222.11

and B contains

cde.com::192.225.111.22,

I need the script to replace the cde.com entry in file A with that of file B (192.225.11.22)

My question is, what is the best function to use for 'find and replace"? CompareFiles() ?

Thanks in advance

Balthazar

Depending on how big the files are you might just want to use file() to read each line in, then foreach through that and $lista = explode("::", $filea);
$listb = explode("::", $fileb);

and compare the [0] value of the array and next($fileb);

More of a concept than exact code, hopefully you understand what I'm talking about 🙂

if $list1[0] == $listb[0]
{
$lista[1] = $listb[1];
}

would take car of your updating A with B.

Anyone with a more elegant way please feel free 🙂

StevenSm

Thanks for your reply.

The file is about 4,000 lines long..The only problem with this code is that the script would need to compare Line 1 of file B with all 4,000 lines in file A, and Line 2 of file B with 4,000 lines, etc...

It seems like this would take a long time to process, but maybe that's the only way it can be done..

Balthazar

No is its that long I doubt I would use that method.

I really havent tried to do that with a large file so off the top of my head I cant honestly say whats going to be fastest.

Weedpacket

Originally posted by StevenSm
Thanks for your reply.

The file is about 4,000 lines long..The only problem with this code is that the script would need to compare Line 1 of file B with all 4,000 lines in file A, and Line 2 of file B with 4,000 lines, etc...

It seems like this would take a long time to process, but maybe that's the only way it can be done..

Well, since you want to compare each line of file B with every line in file A, then you would have to do something like that.

This is where a proper database would be a lot more useful than a plain text file, because then the files you store the data in would be much better organised and also indexed to speed tasks such as this up.

In this case, you could read both files in as arrays of individual entries, with the entries in a comparable form (so you can tell if a line in B is also in A with a simple == test), and then use some combination of array algebra functions ([man]array_diff[/man], [man]array_intersect[/man], [man]array_merge[/man], for example) to get (elements of B that are in A), (elements of B that are not in A...), (elements of A that are in 😎 etc., and merge the appropriate bits back together. The result might need to be resorted afterward depending on whether the order in which entries appear in the file is important.