Shell Programming and Scripting

num desc ind code

11 hi,feather y food
121 edible,oil y food
100 meal-pack y food
010 health,prod
120 lunch,pack


The above one is a flat file with the first line as column name.This is a comma delimited file with quote character as double quotes.

ie desc column has fields like "Hi,Feather" or may be "Hi,Fea"ther"

in short the record might look like

"10","Hi,Feather","y","Food"

this should go into the file but as in the example file above.Since delimiter is comma here ,I wud like to know if anyone can give me a script which can parse this the way i showed above and put into the file.ie Hi,Feather shud be in one column.Note: Its a vARIABLE LENGTH RECORD.

Because you have commas embedded inside the double-quoted text, this is not trivial.

Have a look a CVSparse and see if you can use it:

http://www.visi.com/~hawkeyd/csvutils.3.html

is this what you want:

"10","Hi,Feather","y","Food"

into

10 Hi,Feather y Food

if that is the case you can simply do a sed...

sed -e 's/\",\"/ /g' -e 's/\'//g' filename

The first expression is clear to me it identifies < "," > expressions and substitutes with space. Let me know what are you trying to achieve with second expression -e 's/\'//g'

typo error

should be:

sed -e 's/\",\"/ /g' -e 's/\"//g' filename

only human

Thanks All, I was really busy figuring out similar pbm at my work place, will let you know my feedbak asap