Search Results

This may work, at least on your simplified samples:awk '/<AAAA>/,/<\/AAAA>/ {X=X"\n"$0} /<\/AAAA>/ && X ~ /123456789/ {print X; X=""}' file

<AAAA>
abc
<id>123456789</id>
ddd
</AAAA>

X should be cleared at the end of every block
awk '/<AAAA>/,/<\/AAAA>/{X=X"\n"$0}
/<\/AAAA>/ {if (X~/123456789/) print X; X=""}' file

Try...nawk 'c-->0;$0~s{if(b)for(c=b+1;c>1;c--)print r[(NR-c+1)%b];print;c=a}b{r[NR%b]=$0}' b=2 a=4 s="string" file1...where "b" and "a" are the number of lines to print before and after string "s".
...

This is not correct. These are three steps; you have to run them one at a time:
./configurethen
makethen
make installAs for the garbled command... hmm... interesting. It is working ok, with my...

If there are no empty lines, you need something more complicated like this, or you can preprocess first like mirni suggested...
/usr/xpg4/bin/awk '$0~"<"resp{sect=1}
sect{p=p (p?RS:x) $0}
...

In your posted input, there is no 'response2', there is only 'reponse2'. Is this a typo, or is there really "reponse" in the input?