Need help comparing lines in two files

[Stuart Clemons]

This very green newbie would like to compare two files, let's say File1
and File2. I
want to put the difference from File2 only, into a new file, File3.

For example:

File1.txt
oranges
apples
bananas

File2.txt
apples
kiwi
bananas

The result I want for File3 is the new entry in File2, which is kiwi. (I
don't care that oranges was in File1 and not File2.)

I tried using a nested foreach loop structure, but I can't get that to
work and I have a feeling using nested foreach's is not the way to go.

I'm guessing somehow I should use hashs, but I've never used a hash for
anything and I don't really know how to use a hash. Can someone help ?

Here's my feeble attempt:

my $file1;
my $file2;

my @file1 = qw(oranges apples bananas);
my @file2 = qw(apples kiwi bananas);

foreach $file2 (@file2){
foreach $file2 (@file2){
#print "$mastervob $tempvob \n";
if ($file2 eq $file1) {
last; # I would like to go up to the
toplevel "foreach" here, but I don't know how to do it
} # and I'm not sure this would even
work.
else{
print "$file2 \n";
}
}
}

[Dan Anderson]

Lets say file 1 is:

foo
bar
.... continues on for 100 lines

And file 2 is:

foo
baz
bar
.... continues on exactly the same 100 lines as file 1

Would file 2 be different from file 1 from line 2 and down? Or would it
be different for line 2 and 3?

Also, the keywords:

next; Brings you to the next iteration in a loop
last; leaves the loop

Should help you iterate through a while loop (or empty loop)

i.e.

{
# this is a loop, just two sets of brackets
# put a last statement and it will leave.
# put one of these in your for loops, or outside of your for loops.
}

Also you can get tricky by naming loops, i.e.:

FOO:
{
print "foo";
BAR:
{
last FOO;
}
# anything below here never executes
print "bar";
}

[Dan Anderson]

One more thing, those loops I was telling you about, just using a pair
of brackets, also keep their scope. It's a good way to clean up with
yourself, i.e.

my $foo = 40;
{
my $foo = 50;
print $foo; # prints 50
# garbage collector called on all declarations before here
}
print $foo; # prints 40

Also:

use strict;
use warnings;

Should ALWAYS be at that op of your scripts until you know enough Perl
to know when to bend or break this rule.

-Dan

[Wolf Blaum]

> This very green newbie would like to compare two files, let's say File1

> and File2. I
> want to put the difference from File2 only, into a new file, File3.

I had a very simliar problem about a week ago, which James answerd here:

[url]http://groups.google.com/groups?q=Perl+looping+(a+lot+of[/url])
+files&hl=en&lr=&ie=UTF-8&selm=28A16704-4AD3-11D8-9A03-000A95BA45F8%
40grayproductions.net&rnum=1

or try google groups "perl looping through (a lot of) files"

The only really difference is that I didnt want to compare one FILE2 to FILE1
but 500.

However, be carefull on your filesize:
I settled reading one file into mem (as an array) and looping through the
other ones using a while (<FILE2>) reading the 500 files line by line.

> For example:
>
> File1.txt
> oranges
> apples
> bananas
>
> File2.txt
> apples
> kiwi
> bananas
>
> The result I want for File3 is the new entry in File2, which is kiwi. (I
> don't care that oranges was in File1 and not File2.)
>
> I tried using a nested foreach loop structure, but I can't get that to
> work and I have a feeling using nested foreach's is not the way to go.

why not?

> I'm guessing somehow I should use hashs, but I've never used a hash for
> anything and I don't really know how to use a hash. Can someone help ?

do you need to associate the contens of the line with a filename ore
something? if not, use an array.

> Here's my feeble attempt:
>
> my $file1;
> my $file2;
>
> my @file1 = qw(oranges apples bananas);
> my @file2 = qw(apples kiwi bananas);
>

As Dan showed:

FILE2:

> foreach $file2 (@file2){
> foreach $file2 (@file2){

you may want
foreach my $file1(@file1){
here

> #print "$mastervob $tempvob \n";
> if ($file2 eq $file1) {
> last; # I would like to go up to the
> toplevel "foreach" here, but I don't know how to do it
> } # and I'm not sure this would even
> work.

as Dan said:

next FILE2;

will do the job.

> else{
> print "$file2 \n";
> }
> }
> }

This doesent do what I assume you want: when you place the print in the inner
loop.
Just look at the link above.

Hope thats a start, Wolf

[Drieux]

On Jan 22, 2004, at 4:52 PM, [email]stuart_clemons@us.ibm.com[/email] wrote:

> This very green newbie would like to compare two files, let's say File1
> and File2. I
> want to put the difference from File2 only, into a new file, File3.
>
> For example:
>
> File1.txt
> oranges
> apples
> bananas
>
> File2.txt
> apples
> kiwi
> bananas
>
> The result I want for File3 is the new entry in File2, which is kiwi.
> (I
> don't care that oranges was in File1 and not File2.)

in theory then file2.txt could have been

oranges
apples
kiwi
banana

what about
apples
kiwi
wombat
bananas

you would want to have kiwi and wombat

One strategy would be say:
my @file1 = qw(oranges apples bananas);
my @file2 = qw(apples kiwi bananas frodo bagins);

my @list = get_diff_list(\@file1,\@file2);

print "we see $_\n" foreach(@list);

#------------------------
#
sub get_diff_list
{
my ($list1, $list2 ) = @_;

my %hash = map { $_ => 1 } @$list1;

my @ret_list;
foreach (@$list2)
{
if ( $hash{$_} )
{
delete( $hash{$_} );
} else {
push(@ret_list,$_);
}
}
# if you wanted to have the remaining bits that were
# in list1 and not in list2
#push(@ret_list,$_) foreach(keys(%hash));
@ret_list;

} # end of get_diff_list

or how about

sub get_diff_list
{
my ($list1, $list2 ) = @_;

my %hash = map { $_ => 1 } @$list1;
grep { $_ if ( ! exists($hash{$_})) } @$list2;
}


ciao
drieux

---

[Stuart Clemons]

Thank you Dan and Wolf ! With the suggested changes, my foreach loop
script now works as I hoped it would. (My first script did have a typo,
as you pointed out, though my logic was still wrong.) I'm glad to be able
to set aside my study of hashes for another day. I needed to get this
problem solved so that I can get some other work done.

To correct the script, I added LABELS, used the next statement with a
LABEL, and moved the $print statement out of the inner loop, and waalaa,
it worked properly. I quickly went over the logic of the working program
and it makes sense. It's funny how things seem so clear once they're
solved !

My files are actually probably only 30 to 35 lines each, so size isn't a
problem. The real data I'm comparing has email addresses in them. File2
will either match File1, or have new email addresses in them. I then do
stuff with the new email addresses.

Thanks again. I really appreciate the help.

Here's the working program:

use strict;
use warnings;

my $file1;
my $file2;

my @file1 = qw(oranges apples bananas);
my @file2 = qw(apples kiwi bananas);

FILE2: foreach $file2 (@file2){
FILE1: foreach $file1 (@file1){
if ("$file2" eq "$file1") {
next FILE2;
}
}
print "$file2 \n";
}

The output is "kiwi", which is exactly right.
kiwi




wolf blaum <wolf.blaum@charite.de>
01/22/2004 08:38 PM

To
[email]stuart_clemons@us.ibm.com[/email], [email]beginners@perl.org[/email]
cc

Subject
Re: Need help comparing lines in two files

> This very green newbie would like to compare two files, let's say File1
> and File2. I
> want to put the difference from File2 only, into a new file, File3.

I had a very simliar problem about a week ago, which James answerd here:

[url]http://groups.google.com/groups?q=Perl+looping+(a+lot+of[/url])
+files&hl=en&lr=&ie=UTF-8&selm=28A16704-4AD3-11D8-9A03-000A95BA45F8%
40grayproductions.net&rnum=1

or try google groups "perl looping through (a lot of) files"

The only really difference is that I didnt want to compare one FILE2 to
FILE1
but 500.

However, be carefull on your filesize:
I settled reading one file into mem (as an array) and looping through the
other ones using a while (<FILE2>) reading the 500 files line by line.

> For example:
>
> File1.txt
> oranges
> apples
> bananas
>
> File2.txt
> apples
> kiwi
> bananas
>
> The result I want for File3 is the new entry in File2, which is kiwi. (I
> don't care that oranges was in File1 and not File2.)
>
> I tried using a nested foreach loop structure, but I can't get that to
> work and I have a feeling using nested foreach's is not the way to go.

why not?

> I'm guessing somehow I should use hashs, but I've never used a hash for
> anything and I don't really know how to use a hash. Can someone help ?

do you need to associate the contens of the line with a filename ore
something? if not, use an array.

> Here's my feeble attempt:
>
> my $file1;
> my $file2;
>
> my @file1 = qw(oranges apples bananas);
> my @file2 = qw(apples kiwi bananas);
>

As Dan showed:

FILE2:

> foreach $file2 (@file2){
> foreach $file2 (@file2){

you may want
foreach my $file1(@file1){
here

> #print "$mastervob $tempvob \n";
> if ($file2 eq $file1) {
> last; # I would like to go up to the
> toplevel "foreach" here, but I don't know how to do it
> } # and I'm not sure this would

even

> work.

as Dan said:

next FILE2;

will do the job.

> else{
> print "$file2 \n";
> }
> }
> }

This doesent do what I assume you want: when you place the print in the
inner
loop.
Just look at the link above.

Hope thats a start, Wolf

[Eurospace Szarindar]

Hi Stuart,

Have a look on CPAN ([url]www.cpan.org[/url]) there are two wonderfull packages to do
exactely what you are dreaming of :

Algorithm::Diff
Text::ParagraphDiff


Have a nice day
Michel



-----Message d'origine-----
De: Dan Anderson [mailto:dan@mathjunkies.com]
Date: vendredi 23 janvier 2004 02:17
À: [email]stuart_clemons@us.ibm.com[/email]
Cc: Perl Beginners
Objet: Re: Need help comparing lines in two files


Lets say file 1 is:

foo
bar
... continues on for 100 lines

And file 2 is:

foo
baz
bar
... continues on exactly the same 100 lines as file 1

Would file 2 be different from file 1 from line 2 and down? Or wouldit
be different for line 2 and 3?

Also, the keywords:

next; Brings you to the next iteration in a loop
last; leaves the loop

Should help you iterate through a while loop (or empty loop)

i.e.

{
# this is a loop, just two sets of brackets
# put a last statement and it will leave.
# put one of these in your for loops, or outside of your for loops.
}

Also you can get tricky by naming loops, i.e.:

FOO:
{
print "foo";
BAR:
{
last FOO;
}
# anything below here never executes
print "bar";
}


--
To unsubscribe, e-mail: [email]beginners-unsubscribe@perl.org[/email]
For additional commands, e-mail: [email]beginners-help@perl.org[/email]
<http://learn.perl.org/> <http://learn.perl.org/first-response>

[Stuart Clemons]

Thanks Michael. I'll take a look at those modules and see if my Perl
skills are sufficient to understand how to use them. Thanks again.




EUROSPACE SZARINDAR <EUROSPACE.SZARINDAR@space.eads.net>
01/23/2004 02:47 AM

To
[email]stuart_clemons@us.ibm.com[/email]
cc
Perl Beginners <beginners@perl.org>
Subject
RE: Need help comparing lines in two files







Hi Stuart,

Have a look on CPAN ([url]www.cpan.org[/url]) there are two wonderfull packages to do
exactely what you are dreaming of :

Algorithm::Diff
Text::ParagraphDiff


Have a nice day
Michel



-----Message d'origine-----
De: Dan Anderson [mailto:dan@mathjunkies.com]
Date: vendredi 23 janvier 2004 02:17
À: [email]stuart_clemons@us.ibm.com[/email]
Cc: Perl Beginners
Objet: Re: Need help comparing lines in two files


Lets say file 1 is:

foo
bar
.... continues on for 100 lines

And file 2 is:

foo
baz
bar
.... continues on exactly the same 100 lines as file 1

Would file 2 be different from file 1 from line 2 and down? Or would it
be different for line 2 and 3?

Also, the keywords:

next; Brings you to the next iteration in a loop
last; leaves the loop

Should help you iterate through a while loop (or empty loop)

i.e.

{
# this is a loop, just two sets of brackets
# put a last statement and it will leave.
# put one of these in your for loops, or outside of your for loops.
}

Also you can get tricky by naming loops, i.e.:

FOO:
{
print "foo";
BAR:
{
last FOO;
}
# anything below here never executes
print "bar";
}


--
To unsubscribe, e-mail: [email]beginners-unsubscribe@perl.org[/email]
For additional commands, e-mail: [email]beginners-help@perl.org[/email]
<http://learn.perl.org/> <http://learn.perl.org/first-response>