Shell Programming and Scripting

Hi,

My input file looks like below:

Code:

XXXXXXXXX,XXXXXXXXX,MARR,"EDWIN, III",BCCT,I,XGC0125M20091,XXXXXX,20110201,,M,MARR,EDWIN,19780312,039528630,1,
XXXXXXXXX,XXXXXXXXX,MARR,"EDWIN, III",EXPWL,I,0125M20091,WL7A,20110201,,R,MARR,EDWIN,19780312,039528630,1,1

I want to replace every occurence of "," to ":" to make it ":" delimited file, except when "," is a part of the name like "EDWIN, III". Here if I replace "," with ":", they will be treated as two different column, which I do not want to happen. I will be passing this file to awk script later.

I tried:
sed '/"[A-Z ]*,[A-Z ]*"/!s/,/:/g' input > output

But it simply ingnored these two lines and replace "," with ":" in all other lines.

Any idea how this can be done? I assume here that such name should be enclosed in " and " only.

regards,
juzz4fun

Try:

Code:

awk 'NR%2{gsub(/,/,":")}1' RS=\" ORS=\" file

or

Code:

awk '{for (i=1; i<=NF; i+=2) gsub(/,/,":",$i)}1' FS=\" OFS=\" file

Scrutinizer, sorry but I do not understand your solution.
FYI .. the input file contains few thousand records, not all records have this situation where "," appears in the last name of the person. In short, most of the rows have 17 columns while few (as shown) records have more than 17 columns. Sorry for not making it clear in my original post

In that case try the 2nd suggestion...

Sorry no luck

Getting below errors:

Code:

awk: syntax error near line 1
awk: illegal statement near line 1
awk: syntax error near line 1
awk: bailing out near line 1

If possible, could you please explain me the logic behind the script?

On Solaris use /usr/xpg4/bin/awk rather than awk

--
The script works by using the double quote as both input and output field separator.1 means print the record..

Amazing. This worked.
You are something !! Thanks.