Convert a vertical line text file to a csv file with parameters-CodePudding

I tried several ways to transform the text with a column into a summarized csv file based on the contents of the file but I couldn't.

record does not start with comma = first column record that has a comma at the beginning = concatenates with the first

When you find the next record that doesn't have a comma at the beginning, start on another line

Text file to process:

08-ipa_group
,evouth.zip
,zipe.zip
,auth-service.zip
18-ws-api_group
,mocks.zip
,auth-service.zip
,a-service.zip

Output

08-ipa_group,evouth.zip,zipe.zip,auth-service.zip
18-ws-api_group,mocks.zip,auth-service.zip,a-service.zip

CodePudding user response：

Using awk:

awk -v ORS= '
/^[^,]/ {
  eol()
}
END {
  eol()
}
function eol() {
  if (NR > 1)
    print "\n"
}
1' file

CodePudding user response：

Here's a pretty straightforward shell script that:

creates empty string vars line and row
reads the input file line by line, into the var line
if the line starts with a comma (... | grep '^,') it gets appended to the var row
else (it's the first-column line), so do something with row
- if row is not empty (its initial state), echo and append that to output.csv
- else set to the first-col data
finally, append row

rm -rf output.csv

line=''
row=''

while read -r line; do
    [ "$line" = '' ] && continue          # skip blank lines

    if echo "$line" | grep -q '^,'; then  # a "comma" line
        row="${row}${line}"
    else                                  # a "first-col" line
        if [ "$row" != '' ]; then         # row is not in initial state
            echo "$row" >> output.csv
        fi
        row="$line"
    fi
done < input.txt

echo "$row" >> output.csv