Picking a Substring in a .ksh script

[tbohon]

I have a file of records that look like this:

03/02/2005|oru_in|BAD MEDREC ID|medrec='12345',enc='111'

My script needs to separate these out into two separate files and the only difference is in the length of the medrec value (12345 above) ... one group has values of length 10, the other has values of length 7.

I can parse the line out based on the field separator '|' but can't figure out how to check the length of the medrec value (again, 12345 in the example above).

This will be operating on the input file and creating the two output files all at the same time although I could create intermediate working files if necessary.

Ideas?

Tnx.

Tom

"My mind is like a steel whatchamacallit ...

 

[xmb]

Similiar to bash:

str="03/02/2005|oru_in|BAD MEDREC ID|medrec='12345',enc='111'"
medrec=${str##*rec=\'} medrec=${medrec%%\'*}

echo medrec: $medrec, length: ${#medrec}

awk -F'[ \t"\x27]*[=|,][ \t"\x27]*' '/medrec=/ {
  printf "Medrec: %s (length %d)\n", $5, length($5)
}'

. Mac for productivity
.. Linux for developement
... Windows for solitaire

 

[zaxxon]

I hope I got it right - maybe not the nicest code but works:

#!/bin/ksh
# where $1 is the input file, of course
sed 's/.*medrec=.\(.*\).,.*/\1/' $1| awk '{ if ((length($0)) == 7) print ($0)}' >> 7-file
sed 's/.*medrec=.\(.*\).,.*/\1/' $1| awk '{ if ((length($0)) == 10) print ($0)}' >> 10-file

Input file looked like this:

03/02/2005|oru_in|BAD MEDREC ID|medrec='1234567',enc='111'
03/02/2005|oru_in|BAD MEDREC ID|medrec='1234567890',enc='111'
03/02/2005|oru_in|BAD MEDREC ID|medrec='1234567',enc='111'
03/02/2005|oru_in|BAD MEDREC ID|medrec='1234567890',enc='111'
03/02/2005|oru_in|BAD MEDREC ID|medrec='1234567',enc='111'
03/02/2005|oru_in|BAD MEDREC ID|medrec='1234567890',enc='111'
03/02/2005|oru_in|BAD MEDREC ID|medrec='1234567890',enc='111'

laters
zaxxon

 

[Ygor]

Try...

awk -F\' '{print > "outfile" length($2)}' infile

Tested...

==> outfile10 <==
03/02/2005|oru_in|BAD MEDREC ID|medrec='1234567890',enc='111'
03/02/2005|oru_in|BAD MEDREC ID|medrec='1234567890',enc='111'
03/02/2005|oru_in|BAD MEDREC ID|medrec='1234567890',enc='111'
03/02/2005|oru_in|BAD MEDREC ID|medrec='1234567890',enc='111'

==> outfile7 <==
03/02/2005|oru_in|BAD MEDREC ID|medrec='1234567',enc='111'
03/02/2005|oru_in|BAD MEDREC ID|medrec='1234567',enc='111'
03/02/2005|oru_in|BAD MEDREC ID|medrec='1234567',enc='111'

 

[Ygor]

Or...

awk 'match($0, "medrec=[^,]*") {print > "outfile" RLENGTH-9}' infile

 

[tbohon]

Thanks, all ... greatly appreciate it!

Tom

"My mind is like a steel whatchamacallit ...