Shell Programming and Scripting

hi guys,

I want to do pattern matching with awk or sed but I don't know how. here's what I want:

I have a line number for a pattern that I have already found using grep, and I know a pattern like "---" that happens a few lines above that certain line number. I want to print out the chunk between "---" and that line number. But here's the catch. the pattern "---" can be between 1 and 20 lines above that line number and I don't know how many lines above ... Also, another catch is that, "---" can happen multiple times in the file but I want the "CLOSEST" to the line number. Here's an example:

Here's the file:

I found the pattern say qqq to be on line 20. I want to find the first --- before pattern qqq (line 20) and I want to print the entire chunk from --- till the qqq pattern printed out.

Can someone help me out here please?
Thanks

Should be something like:

Code:

awk '
/---/{c=0}
{a[++c]=$0}
/qqq/{for(i=1;i<=c;i++){print a[i]}exit}
' file

Regards