UNIX for Dummies Questions & Answers

I need help on following script:



I need to print the lines which are in bold letters in separate file as

record

string("|") emp_name;

string("|") emp_id;

decimal("|") emp_salary;

string("|") emp_status;

string("\n") emp_proj;

end


Above lines are related to file ${_AB_PROXY_DIR}"'/rf_emp_status_emp_proj-3.dml which is present in 19 th line in below script.

The issue is i need to print only specific lines which are in bold letters in separate file. In below script you will find one field record which will be repeating more times . I need to select all that that lines fromrecord field to end of file ${_AB_PROXY_DIR}"'/rf_emp_status_emp_proj-3.dml.



Please can you help on sending that bold letter fields into separate file.
I am highliting all that lines in below script wich i want in seperate file, in red color


print -r -- 'record

string("|") emp_name;

string("|") emp_id;

decimal("\n") emp_salary;

end' > "${_AB_PROXY_DIR}"'/in_emp_data-1.dml'

print -r -- '/*Reformat operation*/

out::reformat(in) =

begin

out.*:: in.*;

out.emp_status :: if(next_in_sequence()<3) "A" else "D";

out.emp_proj :: "ABC";

end;' > "${_AB_PROXY_DIR}"'/rf_emp_status_emp_proj-2.xfr'

print -r -- 'record

string("|") emp_name;

string("|") emp_id;

decimal("|") emp_salary;

string("|") emp_status;

string("\n") emp_proj;

end;' > "${_AB_PROXY_DIR}"'/rf_emp_status_emp_proj-3.dml'

print -r -- '/*Reformat operation*/

out::reformat(in) =

begin

out.*:: in.*;

out.emp_loc :: "USA";

end;' > "${_AB_PROXY_DIR}"'/rf_emp_location-4.xfr'

print -r -- 'record

string("|") emp_name;

string("|") emp_id;

string("|") emp_salary;

string("|") emp_status;

string("|") emp_proj;

string("\n") emp_loc;

end;' > "${_AB_PROXY_DIR}"'/rf_emp_location-5.dml'

Thanks,
gangadhar.

What's the criterion for deciding which part of the file to extract? The name of the output file?

Code:

perl -0777 -pe "s%;' > \"\\$\\{_AB_PROXY_DIR\\}\"'/rf_emp_status_emp_proj-3.dml'.*%%s;
s/^.*print -r -- '//s" file

This reads in the whole file, and replaces everything from the matching file name through the end of file with nothing, then everything up through the last "print -r -- '", and prints whatever remains. The quoting got a bit tricky, with lots of ugly backslashes; there are ways to make that more elegant, but hopefully this should at least get you started.