Formatting text outputs in unix - unix

Question

Hi I have a list here:
list_1.txt
Alpha
Bravo
Charlie
and files in a directory with the following filenames and contents:
Alpha_123.log
This is a sample line in the file
error_log "This is error1 in file"
This is another sample line in the file
This is another sample line in the file
This is another sample line in the file
error_log "This is error2 in file"
This is another sample line in the file
This is another sample line in the file
error_log "This is error3 in file"
This is another sample line in the file
This is another sample line in the file
Alpha_123.sh
This is a sample line in the file
This is another sample line in the file
This is another sample line in the file
error_log "This is errorA in file"
This is another sample line in the file
This is another sample line in the file
This is another sample line in the file
This is another sample line in the file
error_log "This is errorB in file"
This is another sample line in the file
This is another sample line in the file
error_log "This is errorC in file"
This is another sample line in the file
Bravo.log
Charlie.log
contents of Bravo.log and Charlie.log is similar to Alpha.log
I would like to have an output like this:
Alpha|"This is error1 in file"
Alpha|"This is error2 in file"
Alpha|"This is error3 in file"
Alpha|"This is errorA in file"
Alpha|"This is errorB in file"
Alpha|"This is errorC in file"
Any inputs is greatly appreciated. Thank you!
So basically, I would like to find first the files with names containing the string patterns in list_1.txt, then find the error messages and output with |

Answer 1

awk to the rescue!
awk '{gsub(/^error_log /,FILENAME"|")}1' $(awk '{print $0".log"}' list_1.txt)
UPDATE
Based on the updated info, I think that's what you're looking for.
awk '/^error_log/ {split(FILENAME,f,"_");
gsub(/^error_log /,f[1]"|")}' $(awk '{print $0"_*"}' list_1.txt)

Answer 2

If I understood correctly, this should do:
awk -vOFS=\| 'FNR==1{file=FILENAME;sub("[.]log$","",file)}{sub("^error_log ","");print file,$0}' *.log
explanation:
-vOFS=\| sets the output field separator to |. (The \ is needed to escape the | from the shell (which would treat it as pipe). You could use -vOFS='|' instead.)
FNR==1{...} makes sure this code is run only once per input file: FNR is number of records (i.e lines) read by awk so far from the current file. So this is only equal to 1 when processing the very first line of each file.
file=FILENAME just stores the filename of the currently processed input file in a variable for later editing.
sub("[.]log$","",file) removes .log (the [...] escape the dot (.) from being interpreted as any character in the regular expression. You could use \\. instead.) from the end (that's what the $ stands for) of the filename.
{...} runs the code for every record/line of each input file.
sub("^error_log ","") removes "error_log " (note the trailing space!) from the beginning (that's what the ^ stands for) of each line ("record") of the input.
print file,$0 prints the remainder of each record (i.e. line) prefixed by the corresponding filenames. Note that the comma (,) will be replaced by the output field separator specified earlier. You could use print file "|" $0 instead without specifying the OFS.
*.log will make every file ending in .log in the current directory an input file for the awk command. You could specify Alpha.log Bravo.log Charly.log explicitely instead.
Here is an alternative using your list.txt to construct the filenames:
awk -vOFS=\| '{file=$0;logfile=file ".log";while(getline < logfile){sub("^error_log ","");print file,$0}}' list.txt
explanation:
file=$0 saves the current line (record) from list.txt in a variable.
logfile=file ".log" appends .log to it to get the corresponding log filename.
while(getline < logfile){...} will run the code for each line/record in the current log file.
The rest should be clear from the above example.