Question

I have a CSV file that has rows, where data for some columns only appear in one of those rows, with other columns repeating their value:

Heading1, Heading2, Heading3, Heading4
1       , 2       ,         , 4
1       ,         , 3       , 4

How can I end up with:

Heading1, Heading2, Heading3, Heading4
1       , 2       , 3       , 4

I want to group on Heading1 and Heading4 as they are unique to the repeated rows and get the first non-blank value for all other rows, ending up with a single string[].

I’ve got as far as grouping on new { Header1, Header4 } to get a group of rows, but I’m having a hard time turning that into something where I can select the first non null value for each column, then turning it back into a single row (string[]).

Ideally I’d like a function that works with any number of columns as in the actual file there are a large number.

Answer 1

• Make an array of length 4 with values initialized to some to empty strings (or zeros as appropriate).
• For each non-header row, loop over the fields, storing the value in the corresponding array position whenever the field value is not blank.
• Write-out the values in the array to the new CSV file.