Shell Programming and Scripting

Hi everyone,

I have a very big git log, and I want to extract the part
between 2 DIFFERENT markers(the marker is something like
a serial number d1938d11e7eb4d845cfe1f3c6bc5899b5ac538a7
)

beginning of the log
<marker1>
part of the log I'm interessed in
<marker2>
rest of the log

apparently I cannot do it using grep, and I have not managed to do it
using awk. Could someone help me do it please?

Code:

perl -ne 'print if /^<marker1>$/../^<marker2>$/;' yourfile

Note that this will print to the console, so redirect it to a file if needed. Also, both marker lines will be included.

If you know the sha1's (markers), then you could try

Code:

$ sed -n '/marker1/,/marker2/p' logfile

Try:

Code:

awk '/^<marker2>/{p=0}p;/^<marker1>/{p=1}' infile

output:

Code:

part of the log I'm interessed in

Could you give some link describing that perl syntax? Is ".." part of regex or if statement?

I think I'm gonna cry :_), I tried to do it for one hour, one hour and a half

Thank you all. I used Scrutinizer technique as I was trying to do it
with awk. Now I can extract easily my push between release.


Should I edit the thread to add [solved] ?

@solal: Yes please.

@bartus11: neither. Usually, two dots mean a range, as in 1..5. However, since TMTOWTDI, it can also act as a flip-flop (yes, just like in electronics). In this case, the first regex sets the flip-flop to always return true (thus printing a line), untill the second regex sets it back to false (thus disabling printing). See Range Operators in perlop.