Searching a file for multiple patterns recursively with IF condition

kk2018 · 11-02-2018, 03:41 AM

Lets say we have three strings Str1, Str2 and Str3 among many other strings in a huge file.

Str1, Str2 and Str3 can occur multiple times in the file.

But Str2 and Str3 occurs only after Str1. The pattern would be something like

Str1
Str2
Str3

Str1
Str2
Str3

So, I need to search this file for Str1, whenever this occurs, I need to search for Str2 and Str3 until I find Str1 then again I need to search for Str2 and Str3 and this repeats till the end of the file.

I have used SED, AWK and GREP for simple search. I have already spent some time and read about RE and understood a bit of it but this seems to be complex

berndbausch · 11-02-2018, 03:55 AM

The description is not quite clear. Does Str2 come right after Str1 or anywhere after Str1? Are you searching for entire lines or are the strings embedded in text lines? Is there anything to do once you have found a str1-str2-str3 pattern?

Also, can you share what you have done so far? Perhaps the most promising attempt at a solution.

chrism01 · 11-02-2018, 04:01 AM

Well, something like this might be what you want

Code:

# test file t.t content
trx
str1
str2
str3
strc
strv
str1
str2
str3
strf

# now the code & o/p
sed -nE '/str1/,/str3/p' t.t

str1
str2
str3
str1
str2
str3

but as above, there are some ambiguities in your spec..

pan64 · 11-02-2018, 04:02 AM

in awk you can do something like this:

Code:

awk '
/Str1/ { s1=1 }   # set variable s1 to 1 if Str1 found.
/Str2/ { if ( s1 == 1 ) s2=1 }   # similar to Str2
...

But this is just an idea, you need to specify what you need better.