2

I have multiple comma-separated files that I need first to merge the rows into a single row then I can append that row to append to one output.csv file.

file-1.txt:

CV-1999-0001,
CV-1999-0002,
CV-1999-0003,
CV-1999-0004,
CV-1999-0005,
  ...

file-2.txt:

CV-2000-0006,
CV-2000-0007,
CV-2000-0008,
....

output.csv:

IDs
CV-1999-0001,CV-1999-0002,CV-1999-0003,CV-1999-0004,CV-1999-0005
CV-2000-0006,CV-2000-0007,CV-2000-0008,CV-2000-0009,CV-2000-0010

I have tried these but none seem to achieve what I need. I still end up with multiple rows

echo $(cat /Users/...../"$ROWS".txt) | sed 's/ //g' > "$ROWS_new".txt


echo $(cat paste -s -d "" files-1.txt) | sed 's/ //g' >> output.csv
Improve this question
3
  • Where did CV-2000-0009,CV-2000-0010 come from? Please edit your question to make sure that the expected output you provide is exactly the output you expect given the same input you provide so it accurately demonstrates your requirements and we have something we can copy/paste to test a potential solution with. And get rid of those ... lines that are cluttering up your example. Also, if you don't REALLY want IDs printed at the top of your output file then don't show it at the top of your output file. Commented Jun 13, 2022 at 11:54
  • @EdMorton the files are generated by an external app. I simply access the folders from a shared location and used the data. Commented Jun 13, 2022 at 23:59
  • I'm sorry - I'm not asking where your input file comes from, I'm asking where did the strings CV-2000-0009 and CV-2000-0010 in your posted expected output come from since they aren't in your posted sample input? Are they from some 3rd input file or do you need this tool to generate them or is it a mistake in your sample input/output? If the former please edit your question to explain it, if the latter please edit your question to fix that. Commented Jun 14, 2022 at 0:18

4 Answers 4

Reset to default
0

With GNU awk for multi-char RS and reading each file into memory:

$ awk -v RS='^$' -F',\n' -v OFS=',' '{$1=$1}1' file-1.txt file-2.txt
CV-1999-0001,CV-1999-0002,CV-1999-0003,CV-1999-0004,CV-1999-0005,
CV-2000-0006,CV-2000-0007,CV-2000-0008,

or with GNU awk for multi-char RS and ENDFILE reading 1 line at a time:

$ awk -v RS=',\n' -v ORS=',' '1; ENDFILE{printf "\n"}' file-1.txt file-2.txt
CV-1999-0001,CV-1999-0002,CV-1999-0003,CV-1999-0004,CV-1999-0005,
CV-2000-0006,CV-2000-0007,CV-2000-0008,

or with any awk reading 1 line at a time:

$ awk '
    FNR==1 { printf "%s%s", prev, ors; ors=ORS }
    FNR>1  { printf "%s", prev }
    { prev=$0 }
    END { print prev }
' file-1.txt file-2.txt
CV-1999-0001,CV-1999-0002,CV-1999-0003,CV-1999-0004,CV-1999-0005,
CV-2000-0006,CV-2000-0007,CV-2000-0008,

If the last line of your input files end in , it'd be easy to remove it but since the sample input in your question ends with ... we don't know if your input files really end with ,s or not so I left that as an exercise.

Improve this answer
1
  • Thank you @EdMorton the file ends with , Commented Jun 14, 2022 at 0:06
0

Do you try:

for ROW in ${ROWS}; do
    tr -d "\n" < "/Users/...../${ROW}.txt" \
    | sed -e 's/,$//'
done >> "output.txt"
Improve this answer
2
  • Thank you @Levallois-Perret for the response. I have tried this but it builds multiple rows for each value in each file and not a row for each files values. Commented Jun 13, 2022 at 9:54
  • Paset bin output this is the output sample. Commented Jun 13, 2022 at 10:22
0

With a sed that offers the -z (--null-data; separate lines by NUL characters) option, and thus reads the entire input stream, try

sed -z 's/\n//g; s/\r/\n/' file1 <(echo $'\r') file2
CV-1999-0001,CV-1999-0002,CV-1999-0003,CV-1999-0004,CV-1999-0005,
CV-2000-0006,CV-2000-0007,CV-2000-0008,

Without that option, try

sed -f chabo.sed  file1 <(echo $'\r') file2
CV-1999-0001,CV-1999-0002,CV-1999-0003,CV-1999-0004,CV-1999-0005,
CV-2000-0006,CV-2000-0007,CV-2000-0008,

with chabo.sed being

:L
N
$!bL
s/\n//g
s/^M/\  
/

and ^M the <cr> (0x0D) character.

Improve this answer
0

What's wrong with this solution?

╰─ → $ paste -sd " " file1.txt file2.txt
CV-1999-0001, CV-1999-0002, CV-1999-0003, CV-1999-0004, CV-1999-0005,
CV-2000-0006, CV-2000-0007, CV-2000-0008,
Improve this answer

You must log in to answer this question.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.