Search for the new guy

[jld8996]

Hi All,
I just started learning Perl with Sams Teach Yourself Perl in 24
hours. I have been able to accomplish part of what I wanted to do but
need a little more help. I have 2 files, the first is a list of
element numbers and the 2nd is a large text file (a .f06 file from
NASTRAN for those who are familiar). I started writing a script shown
below to open both files. What I need to do is step through the first
file element by element and look up each element in the 2nd file.
Once found I want to write the line where it is found and the next
line to an output file. Then find the next element of file 1 and so
on. Any help or guidance would be great. Thanks, Jeremy

Here is my script so far:
-----------------------
#!/usr/bin/perl

print "Enter the name of the element control file: \n";
$control=<STDIN>;
open(INPUT,$control) || die "$!";
@input=<INPUT>;
close(INPUT);

print "Enter the name of the .f06 file: \n";
$f06=<STDIN>;
open(STRESS,$f06) || die "$!";
@stresses=<STRESS>;
close(STRESS);

print "Enter the name of the output file: \n";
$output=<STDIN>;
open(OUTPUT,">$output") || die "$!";

-------------------------------------------
so I've only done the easy stuff. I read all of the .f06 file into
the array "stresses". I'm not sure if this was necessary. The actual
file is over 300,000 lines long so this might be slowing me down.

Here's is a sample of file 1, the element file:

-------------------------------------------
119471
119472
119473
119474

--------------------------------------------

And here is a piece of file 2, the .f06 file:

--------------------------------------------

1 MSC.NASTRAN JOB CREATED ON 19-SEP-03 AT 15:32:39
W145
0

S T R E S S E S I N Q U A D R I L A T E R
A
ELEMENT FIBER STRESSES IN ELEMENT COORD SYSTEM
ID. DISTANCE NORMAL-X NORMAL-Y SHEAR-XY
0 119461 -6.250000E-02 -2.809423E+02 -2.226965E+03
7.288095E+02
6.250000E-02 -2.154822E+02 -2.395907E+03
5.431436E+02
0 119462 -6.250000E-02 9.256976E+02 -2.722196E+03
1.075280E+03
6.250000E-02 9.582150E+02 -3.030506E+03
1.060233E+03
0 119463 -6.250000E-02 -5.176458E+01 1.040512E+03
-9.355118E+01
6.250000E-02 -2.901073E+01 1.011089E+03
-2.422135E+01
0 119464 -6.250000E-02 -1.102965E+02 1.167524E+03
5.211899E+01
6.250000E-02 -1.033523E+02 1.100544E+03
1.217817E+02
0 119465 -6.250000E-02 8.531695E+02 3.763257E+02
8.413060E+02
6.250000E-02 7.949519E+02 8.895544E+01
8.487442E+02
0 119466 -6.250000E-02 2.496463E+02 1.557864E+03
7.741991E+02
6.250000E-02 1.736531E+02 1.307298E+03
8.175496E+02
0 119467 -6.250000E-02 -1.153684E+02 1.296065E+03
3.091736E+02
6.250000E-02 -1.198170E+02 1.147716E+03
3.649098E+02
0 119468 -6.250000E-02 1.097371E+03 3.642865E+03
1.958549E+03
6.250000E-02 1.037483E+03 3.431623E+03
1.997865E+03
0 119469 -6.250000E-02 -5.984709E+01 1.383306E+03
1.489364E+02
6.250000E-02 -5.029673E+01 1.210617E+03
1.580537E+02
0 119470 -6.250000E-02 1.908398E+02 1.890122E+03
3.743426E+02
6.250000E-02 1.284111E+02 1.673722E+03
3.498486E+02
0 119471 -1.900000E-01 5.097551E+02 1.409135E+02
-2.752976E+01
1.900000E-01 -1.391517E+03 -3.814824E+02
-3.767153E+02
0 119472 -1.900000E-01 -7.821536E+01 1.430543E+03
-4.919512E+02
1.900000E-01 3.173092E+02 1.755130E+03
-6.183864E+02
0 119473 -1.900000E-01 -4.190353E+02 -9.498234E+02
-3.387274E+02
1.900000E-01 1.031396E+02 2.249585E+02
-2.015406E+02
0 119474 -1.900000E-01 -6.918803E+01 -1.046147E+03
-7.935986E+02
1.900000E-01 7.496266E+02 8.220536E+02
-1.732447E+02
0 119475 -1.900000E-01 -9.534775E+02 -2.848530E+03
-1.332869E+02
1.900000E-01 1.081157E+03 3.724546E+03
4.091841E+02
0 119476 -1.900000E-01 2.104021E+01 -2.328312E+03
-4.683439E+02
1.900000E-01 -3.702834E+01 -3.805241E+02
-1.132028E+02

 

[Gunnar Hjalmarsson]

I just started learning Perl with Sams Teach Yourself Perl in 24
hours. I have been able to accomplish part of what I wanted to do
but need a little more help.

To be honest, you haven't convinced me that you need any help at all.
All you seem to need is some patience.

For instance, you may want to study how to write a while loop
(http://www.perldoc.com/perl5.8.0/pod/perlsyn.html), and the index()
function (http://www.perldoc.com/perl5.8.0/pod/func/index.html) may
come in handy.

What I need to do is step through the first file element by element
and look up each element in the 2nd file. Once found I want to
write the line where it is found and the next line to an output
file. Then find the next element of file 1 and so on.

So, why don't you start working with a program that does that?

Here is my script so far:

Before going on, please learn how to write programs with strictures
and warnings enabled.

use strict;
use warnings;

I read all of the .f06 file into the array "stresses". I'm not
sure if this was necessary.

I'm sure it wasn't.

My suggestion is that you give it a _serious_ try. You are welcome
back here if you would encounter any problem that you are not able to
resolve by help of the Perl documentation.

Good luck!

 

[Tad McClellan]

Subject: Search for the new guy

Please put the subject of your article in the Subject of your article.

I just started learning Perl with Sams Teach Yourself Perl in 24
hours.

That is a really bad Perl book you know. Sell it on E-bay and
get a good one before you pick up bad habits...

Here is my script so far:

It is insulting to be asked to do a machine's work.

Please don't ask us to do that for you, ask the machine before
asking people.

You should put these at the top of every Perl program that you write:

use strict;
use warnings;

You can see the docs for what those do with perldoc, eg:

perldoc strict

$control=<STDIN>;
open(INPUT,$control) || die "$!";

Is the filename supposed to have a newline on the end of it, or
did you just forget to chomp() it after reading?

Did the Sam's book teach you to use useless double quotes like that?

perldoc -q vars

What's wrong with always quoting "$vars"?

so I've only done the easy stuff.

What part, specifically, are you having trouble with?

Getting someone to help you with your code here is pretty easy.

Getting someone to write your program for you will prove a
bit more difficult...

 

[Alan J. Flavell]

That is a really bad Perl book you know. Sell it on E-bay

If it's that bad, surely hanging it in the smallest room would be a
better gift to humanity...?

 

[Bob Walton]

jld8996 wrote:

....

I just started learning Perl with Sams Teach Yourself Perl in 24
hours. I have been able to accomplish part of what I wanted to do but

Yuck. Try the book "Learning Perl", O'Reilly & Associates, Inc.

need a little more help. I have 2 files, the first is a list of
element numbers and the 2nd is a large text file (a .f06 file from
NASTRAN for those who are familiar). I started writing a script shown
below to open both files. What I need to do is step through the first
file element by element and look up each element in the 2nd file.
Once found I want to write the line where it is found and the next
line to an output file. Then find the next element of file 1 and so
on. Any help or guidance would be great. Thanks, Jeremy ....
so I've only done the easy stuff. I read all of the .f06 file into
the array "stresses". I'm not sure if this was necessary. The actual
file is over 300,000 lines long so this might be slowing me down.

You will probably want to do a bit of preliminary parsing of the .f06
file as you read it in. For example, skip to the section headings of
sections that actually have stuff in them you want to retrieve with this
program. It probably *is* a good idea to keep the data you want to
search through in memory, as it looks like you will be doing the
searches a large number of times. You should read through the .f06 file
once, parsing it for element numbers, and, every time an element number
is found, see if it appears in your "element file" (i.e., keep the
elements identified in the element file in memory, rather than the stuff
in the .f06 file). That would reduce the memory requirements and make
everything faster. A Perl hash would be the perfect way to store the
contents of your "element file" -- every time an element ID is located
in the .f06 file, you look it up in the hash to see if it is one of the
elements you want to output.

You will also need to parse the .f06 file enough to be able to
disinguish which number strings are element numbers and which are
something else. In your sample, you could include the space characters
before and after the element number in your match, which would prevent
you from including lines with something like "2.119471E+02" when you're
looking for element 119471. But that probably isn't sufficient to catch
every string that could be confused with an element number. There are
lots of potential output sections with Nastran (which I am familiar
with), so you might have a lot of different parsing rules to do it
right. Good luck.


....

 

[Tad McClellan]

If it's that bad, surely hanging it in the smallest room would be a
better gift to humanity...?

Too tough and inky to touch _my_ "humanity", so I wouldn't
suggest it to others.

 

[jld8996]

Many thanks to Bob for his help. With his help I was able to complete
my script. Many thanks to Gunnar and Tad for well...nothing. Thanks
for taking time out of what must be your very busy day to reply to my
message with your criticism.