Parse multiple sections of data into separate files

Question

I have a concatenated log file with multiple logs inside that I'm trying to parse out into individual log files. I will later rename them to the date/time of each. Each log is separated by "--- LOG REPORT ---".

So far I have:

sed -n '/--- LOG REPORT ---/,/--- LOG REPORT ---/p' logname.log > test.out

However, as you can imagine, that only outputs the first instance of the pattern. I looked over the man page for sed and I'm not convinced it can output multiple files. Perhaps I could keep extracting from a file until it's empty but that seems like too much work. How I can achieve this? Maybe I should be using awk instead?

Example of input file filename.log

--- LOG REPORT ---
Mary
Had
A
Little
Lamb
--- LOG REPORT ---
Her
Fleece
Was
White
As
Snow

Desired output:

In filename_1.log

--- LOG REPORT ---
Mary
Had
A
Little
Lamb

In filename_2.log

--- LOG REPORT ---
Her
Fleece
Was
White
As
Snow

Answer 1

How about this command?

csplit logname.log /---\ LOG\ REPORT\ ---/ {*}

Testing

cat logname.log
--- LOG REPORT ---
Mary
Had
A
Little
Lamb
--- LOG REPORT ---
Her
Fleece
Was
White
As
Snow

After running the above command, the output I get is,

cat xx01
--- LOG REPORT ---
Mary
Had
A
Little
Lamb
cat xx02
--- LOG REPORT ---
Her
Fleece
Was
White
As
Snow

Answer 2

How about something like

awk '/--- LOG REPORT ---/ {n++;next} {print > "test"n".out"}' logname.log

Answer 3

Wrong tool, use either perl or awk e.g.

cat > splitFileByLogReport.pl <<EOF
#!/usr/bin/perl

undef $/;
$_ = <>;
$n = 0;

for $match (split(/(?=--- LOG REPORT ---)/)) {
      open(O, ">$ARGV[$argnum]" . ++$n);
      print O $match;
      close(O);
}

EOF

Then to run:

perl splitFileByLogReport.pl yourFile.txt