UNIX for Advanced & Expert Users

Hi All,

Could some one help me on one of my requirement below:

I have a sequential file with 4fields in it and it is a comma (,) seperated file.

Delimeter is 'comma'.

But in of the file column for ex: 3rd column it is 'Description' (column name) I am getting the values with commas.

Since this file is comma delimeted file ,the comma mentioned in the 'Description' filed is also treating as a delimeter.

Now I need to replace this comma with 'Pipe' symbol in the data for the field 3 (i.e. 'Description')...

How can I do this in unix?

ex: I have a file called 'emp.csv'

Code:

empno,empname,description,salary
'1','sam','clerk,accounts','3000'
'2','Peter','Manager,accounts','5000'

In the above example I have total 4 fields in my file and it is comma seperated file.But in the 3rd filed(descrption) filed the value it self has the comma in it.

My requirement is to find the comma in 3rd field and replace them into pipe (|)

Please advise?

Thanks in advance

Code:

perl -MText::ParseWords -nle'
  @cols = parse_line(",",0, $_);
  $cols[2] =~ s/,/|/;
  print join ",", @cols;
  ' infile

Another way:

Code:

awk -F"'" '{sub(",","|",$6)}1' OFS=\' file

Thanks for responding my query.

But I am not getting $6 is ?

My column position is 3...means I need to replace the commas with pipe in the 3rd field's data.....

The discription column is the 6th field if the fieldseparator is a single quote:

Code:

cat file
'1','sam','clerk,accounts','3000'
'2','Peter','Manager,accounts','5000'
$ awk -F\' '{sub(",","|",$6)}1' OFS=\' file
'1','sam','clerk|accounts','3000'
'2','Peter','Manager|accounts','5000'

Thanks Franklin52 for the reply..

I have one more qn on the same as what if the data has double quotes(") as a field seperator?
What could be the description column position ,if the field seperator is double quotes(") ?

Thanks for the patience...

Just replace the field separators with a double quote:

Code:

awk -F\" '{sub(",","|",$6)}1' OFS=\" file