UNIX for Dummies Questions & Answers

Hi all,
I am new to unix and I am trying hard to get this requirement, but no luck.
I am trying to compare two cloumns in two files and if it matches, the last column in file1 must be divided by file2 and the output must be written in a new file. To elaborate the 2nd column in file1 (EUR) must be compared with 1st column in file2 (EUR) and if it matches the last coulmn of EURO rows in file1 must be divided with last column in file2 of EURO row and the output must be similar to file1 with calculated values. And if there is no match between file1 and file2 the file1 unmatched records must be as it is. This must be similar for all matching strings.
Please help, thanks in advance.
Eg:

Code:

(7,050.00)/0.73351=9611.32
(57,412.00)/0.98250=58434.60

File1:

Code:

0000001100 EUR 2008-02-04 Vendor Tooling (7,050.00)
0000001100 CHF 2008-03-20 Vendor Tooling (57,412.00)
0000001100 U.S. Dollar 2010-04-22 Vendor Tooling (63,203.00)

File2:

Code:

EUR   0.73351
CHF   0.98250
Output:
0000001100 EUR 2008-02-04 Vendor Tooling (9611.32)
0000001100 CHF 2008-03-20 Vendor Tooling (58434.60)
0000001100 U.S. Dollar 2010-04-22 Vendor Tooling (63,203.00)

A couple of questions...

1) Is this a homework question?
2) Are you doing line number matches? That is, is line 1 of file1 always compared to line 1 of file2? Or are you instead doing, "if EUR appears in column 2 of any line of file 1 and in the first column of any line of 2, perform the given calculation on those lines, whatever their numbers may be"? Both can be done, but the first is considerably easier.

thanks for the reply treesloth.

As I am new, I was searching for awk, grep and sed commands functionality, but not able to get how the calculation can be done.

please help.

I am trying to calculate the values in last column, if EUR appears in column 2 of any line of file 1 and in the first column of any line of 2, perform the given calculation on those lines, likewise for other currencies also and if there are no matches, those records must be as is in output file.

Thanks again treesloth
I am trying to calculate the numbers

In that case, something like this may work:

Code:

#! /usr/bin/awk -f

BEGIN   {
        while ("cat file2" | getline)
                {
                currency[$1] = $NF
                }
        }

        {
        gsub(",","",$NF)
        }

currency[$2] != ""       {
        gsub("[()]","",$0)
        $NF = "(" $NF / currency[$2] ")"
        }

        {
        print
        }

To use this, you'll need to make sure that the first line points correctly to the location of your awk binary, and that set it to be executable. Also note that "file2" is used within the program. If that's not the actual name of your file, change the program accordingly. Then, run it:

Code:

$ ./compute.awk file1
0000001100 EUR 2008-02-04 Vendor Tooling (9611.32)
0000001100 CHF 2008-03-20 Vendor Tooling (58434.6)
0000001100 U.S. Dollar 2010-04-22 Vendor Tooling (63203.00)

Hi treesloth,
The code is not working in my server, however it is working on different server. Might be issue with awk binary.
Anyhow you are awesome. Thanks a lot for your timely help. I will go through the forum whenever i will get time to update my knowlede.

Srini

I'm not exactly sure what would cause syntax errors in this script, but let's start with these:

1) All of the lines in the error use double quotes. Sometimes copying and pasting from a web page messes those up. You can go through and manually replace all double quotes.
2) Does the script have the user execute bit set? This seems unlikely to be the cause, really, since the expected error would be something more like "Permission denied", but we might as well check it.
3) What version of awk are you using? Do awk --version.
4) What version of Unix/Linux are you running?

---------- Post updated at 09:21 AM ---------- Previous update was at 09:13 AM ----------

Ah, just saw your last reply. If it's a question of the awk binary version, you might want to move this to the shell scripting section of the forums. There are some real wizards that hang out there that can probably tell you very quickly the little things that cause some versions of awk to choke. I'm afraid I only know what I come up with on my systems, generic though they may be.

It seems the issue with awk binary version.

Once again thanks for your timely help.