Comparison of two files..

[clearguy02]

Hi folks,

I have two files:
a.txt has 100 unique log_id's (one id per line);
all.txt has 5000 entries (each line has six entries seperated by a
tab and the first entry on each line is the login ID and then full
name, country etc).

Now I want to match both files and get the output with all 100 full
entries and ignore the rest.

Here is the code I am working on.. for some reason, I see more 160
entries instead of the exact 100 entries.

++++++++++++++++++++
my %myconfig = (
input1 => 'a.txt',
input2 => 'all.txt',
matching => 'required.txt',
non_matching => 'ignore.txt',
);

my %fields2;
{
open my $input, '<', $myconfig{input1} or die "Cannot open
'$myconfig{input1}': $!";
while ( <$input> )
{
if ( /^(\w+)/ )
{
$fields2{ $1 } = 1;
}
}
close $input or die "Cannot close '$myconfig{input1}': $!";
}
open my $input, '<', $myconfig{input2} or die "Cannot open
'$myconfig{input2}': $!";
open my $matching, '>', $myconfig{matching} or die "Cannot open
'$myconfig{matching}': $!";
open my $non_matching, '>', $myconfig{non_matching} or die "Cannot
open '$myconfig{non_matching}': $!";

while ( <$input> )
{
if ( /^(\w+)/ )
{
if ( exists $fields2{ $1 } )
{
print $matching "$_\n";
}
else
{
print $non_matching "$_\n";
}
}
}

++++++++++++++++++++++++++++++++++++

What I am doing wrong here? Or is there any alternative way of doing
it?

Thanks,
J

 

[Jim Gibson]

Hi folks,

I have two files:
a.txt has 100 unique log_id's (one id per line);
all.txt has 5000 entries (each line has six entries seperated by a
tab and the first entry on each line is the login ID and then full
name, country etc).

Now I want to match both files and get the output with all 100 full
entries and ignore the rest.

Here is the code I am working on.. for some reason, I see more 160
entries instead of the exact 100 entries.

What does "I see more 160 entries ..." mean? Do you mean you see more
than 160 lines output to required.txt when you only expected 100? What
constitutes the excess lines? Are there duplicates in required.txt? Are
there lines in required.txt that do not have corresponding entries in
a.txt?

++++++++++++++++++++
my %myconfig = (
input1 => 'a.txt',
input2 => 'all.txt',
matching => 'required.txt',
non_matching => 'ignore.txt',
);

my %fields2;
{
open my $input, '<', $myconfig{input1} or die "Cannot open
'$myconfig{input1}': $!";
while ( <$input> )
{
if ( /^(\w+)/ )
{
$fields2{ $1 } = 1;
}
}
close $input or die "Cannot close '$myconfig{input1}': $!";
}
open my $input, '<', $myconfig{input2} or die "Cannot open
'$myconfig{input2}': $!";
open my $matching, '>', $myconfig{matching} or die "Cannot open
'$myconfig{matching}': $!";
open my $non_matching, '>', $myconfig{non_matching} or die "Cannot
open '$myconfig{non_matching}': $!";

while ( <$input> )
{
if ( /^(\w+)/ )
{
if ( exists $fields2{ $1 } )
{
print $matching "$_\n";
}
else
{
print $non_matching "$_\n";
}
}
}

++++++++++++++++++++++++++++++++++++

What I am doing wrong here? Or is there any alternative way of doing
it?

There doesn't appear to be anything wrong with your code (nothing
obvious anyway). While there are certainly alternate ways of doing
this, you seem to have stumbled upon a good solution that uses a hash.
Without seeing your exact input and output data, it is difficult to do
any further analysis of your problem.

If you can answer the questions above, it might help. If you can
isolate the problem to a few anomalous test cases, you can post those.

 

[Jürgen Exner]

Now I want to match both files and get the output with all 100 full
entries and ignore the rest.

Here is the code I am working on.. for some reason, I see more 160
entries instead of the exact 100 entries. [...]
What I am doing wrong here? Or is there any alternative way of doing
it?

Your code logic looks alright to me and I can't spot any glaring issues
with it.
Did you consider, that some IDs might appear more than once in the
second file? If you got duplicates that would explain the mismatch.

jue

 

[Eric Pozharski]

*SKIP*

There doesn't appear to be anything wrong with your code (nothing
obvious anyway). While there are certainly alternate ways of doing

Looking at that --

perl -wle '
q|x| =~ m/(x)/; print $1;
q|y| =~ m/(x)/; print $1;'
x
x

I suppose, that OP doesn't show his code.

*CUT*