Shell Programming and Scripting

Hello Gurus,

I need to create a file from a .csv file extracting specific columns only.

File structure is

Code:

Column1,Column2,Column3,Column4
abcd,1234,"asdf, tew,123",123456
efgh,234,asdf,654321

My output file should have

Code:

abcd,123456
efgh,654321

Can you pls help me with the code.

Thanks

Moderator's Comments:

Please use code tags, thanks!

What you have tried so far ..

For this case particulary :

Code:

awk -F"," '{print $1","$NF}' inputFile

. Will not work if the last column has more than one field wrapped inside quotes.

I tried cut -d but it takes the commas inside "" as well in count

Code:

while read line
do
   OUTLINE=`echo $line | cut -d, -f1`","`echo $line | cut -d, -f4`
   echo $OUTLINE
done < file.csv

I get following output :

Code:

Column1,Column4
abcd, tew

Thanks

The answer from @Sheel looks good because it finds the first and last comma-separated field (providing that both fields are numbers or strings which contain no comma characters).

Thanks Sheel... But it's possible that I have more that one fields wrapped inside in the last column as well, I need to somehow escape commas inside "". Really appreciate if you can provide an alternative.

I dont think we can do it in a single line. I can suggest you a way though: Filter out the rows having double quotes into a new file and write a command particualry for that file.

---------- Post updated at 07:22 AM ---------- Previous update was at 06:49 AM ----------

Here is another way. Replace the commas withing double quotes with a different character (say pipe) and then read the columns using comma as delimiter.

Command to replace the commas within double quotes

Code:

 awk -F'"' 'BEGIN{OFS="\""}{gsub(",","|",$2);print}' inputFile

You can revert the pipe to comma after getting the desired output.