Parsing to Individual Variables

[tbohon]

I have a Unix text file containing records in the following format:

r1f1|r1f2|r1f3!r2f1|r2f2| ... etc.

I can easily break up the individual 'lines' in this record by using split on the '!' character (actually a chr(13)). This puts each line into an array entry in @segs.

What I now need to do is pull out an individual field from an individual line and put it into a variable.

What I've coded so far (am leaving a lot of extra 'stuff' out here):

my @segs = ();
my @parts = ();

$ptname = "";

if (open(MYFILE, "xxx"))
{
   $line = <MYFILE>;
   while ($line ne "")
   {
     @segs = &breakup($line);

     foreach $seg (@segs)
     {
       @parts = &breakupsegment($seg);
       ...
      }
     ...
     $line = <MYFILE>;
}

sub breakup
{
  $CR = chr(13);

  @segs = split(/$CR/, $line);
}

sub breakupsegment
{
  @parts = split(/\|/, $seg);
}

What I need to do is assign the value of, say, parts(3) to the variable ptname - but everything I've tried fails.

Thoughts/comments appreciated.

Tnx.

Tom

"My mind is like a steel whatchamacallit ...

 

[PaulTEG]

open FH, "myfile.txt" or die "$!";
while (<FH>) {
  @parts=split /\m/, $_; #IIRC \m is chr(13)
[COLOR=red]  myvar=$parts[3];[/color]

   ...

}
close FH;

The code above should do what you want, the bit in red is the assignment part you asked for

Paul
------------------------------------
Spend an hour a week on CPAN, helps cure all known programming ailments ;-)

 

[tbohon]

I'll give that a shot when I get back to work Wednesday. I've been using a '$' variable, i.e., $myvar=$parts[3]; and it doesn't like that in the least.

Will let you know how it works out - and thanks!!!

"My mind is like a steel whatchamacallit ...

 

[KevinADC]

PaulTEG's code should be:

$myvar=$parts[3];

 

[tbohon]

Which is exactly what I coded and tested and which doesn't work. Ergo, it's not this statement but something else that's causing the problem.

<sigh> I just love computers!

Thanks to both of you.

Best,

Tom

"My mind is like a steel whatchamacallit ...

 

[ishnid]

This line is strange:

@segs = &breakup($line);

You're already assigning the result of the split to the @segs array in the breakup() subroutine, so it doesn't add anything too assign it again there. Also, you're passing in $line as an argument to that subroutine, even though you never access the passed arguments in the subroutine itself, preferring to work directly on the package variable $line.

I'd write the subroutine like this:

sub breakup
{
  $CR = chr(13);

  return split(/$CR/, shift);
}

It's the same situation for your breakupsegment subroutine too.

When you say that it fails, what does happen? Do you get any errors/warnings? Do you have warnings turned on?

 

[tbohon]

isnhid:

I don't have error checking turned on - probably should.

Understand the duplication - copied this from another routine that works (written by someone else a long time ago) and didn't understand why the syntax was as shown. I'll make the changes.

As to failing, I get nothing in the @parts array as far as I can tell - at least nothing shows up for me to check and assign to the local variable.

Let me work with it a bit more today, incorporate your suggested changes, and see what happens.

Thanks - greatly appreciate the information!

Best,

Tom

"My mind is like a steel whatchamacallit ...

 

[NullTerminator]

In PaulTEG's code,

@parts should be split on the '|' character not the \m

All the record would wind up in $parts[0] otherwise.

jeb

 

[tbohon]

Let me step back a second and regroup.

Here's the source code (there's a bunch of initialization before this but not germane here):

...

if (open(MYFILE, "tmp1ready"))
{
  $line = <MYFILE>;

  while ($line ne "")
  {
    @segs = &breakup($line);

    foreach $seg (@segs)
    {

      print "segment is\n";
      print "$seg\n";

      @parts = split(/\|/, $seg);

      $SEGTYPE = substr($seg, 0, 3);

      $ptname = $parts[6];
      [red]print "pt name is:  $ptname\n";
      print "directly from array:  $parts[6]\n"; [/red]
    }  


    $line = <MYFILE>;
  }
}

sub breakup
{

  $CR = chr(13);

  $line =~ s/^(.*)MSH/MSH/;
  @segs = split(/$CR/, $line);
}

The input file contains HL7 segments (if you don't know about HL7, it's a hospital inter-system transfer protocol) with a chr(13) (^M) between segments. Each segment has a series of fields separated by a '|' character. In other words it looks something like this:

r1f1|r1f2|r1f3^Mr2f1|r2f2|r2f3^Mr3f1 ...

The subroutine breakup successfully breaks on the ^M characters so that I have the individual segments in the @segs array. What I'm trying to do now is to look at each segment in @segs and break it up into the individual fields in @parts so I can extract specific fields. Once I've processed the entire record, I'll then put it back together using just the extracted values.

What I have above works without error - but it doesn't print anything from the two lines in [red]red[/red] that I put in to see what was being put into the @parts.

Does this clarify? Am I doing something horribly wrong here? It's pretty simple Perl code and, as I said earlier, the core is taken from another analyst's routine which works just fine.

Thanks again.

Tom

"My mind is like a steel whatchamacallit ...

 

[tbohon]

I fixed the issue - the file I was using as input was corrupted somehow - regenerating it and running the routine above as written solved the problem.

Sorry to have wasted everyone's time on this non-issue ... I should have thought to check the input file one last time before putting out the mayday call.

Now if I just knew how it had become corrupted ... :-(

Appreciate all the input - take care and thanks!

Best,

Tom

"My mind is like a steel whatchamacallit ...