mjhjmtu

This one is tricky... input has multiple occurrences per line over several lines. many "id=" lines exist, but I only want ones starting with ai-c " id="

Example input (two long lines):

d-fl ai-c" id="3af3a73f26a63a-list-" lima zulu bravo hotel d-fl ai-c" id="7ab8273f273e7273f82c2-list-" alpha id="8372364fa2361ab1-list-" hotel foxtrox mike d-fl ai-c" id="973a72b72c72df12a2-list-" uniform hotel

start of second line golf november d-fl ai-c" id="83hd826ab712cc71-list-" bravo mike mike d-fl ai-c" id="7612364bab72a-list-" golf mike

Output desired is:

id="3af3a73f26a63a"

id="7ab8273f273e7273f82c2"

id="973a72b72c72df12a2"

id="83hd826ab712cc71"

id="7612364bab72a"

All of the script examples I've seen only seem to work for the first occurrence in a line.

One way to do this is:

grep -oP 'hai-c"hKid="[^-]*' input.file | sed -e 's/$/"/'

Where we use GNU grep with the Perl feature turned on and the -o option to output just the matche(s), rather than the default whole line when a match is found.

Output:

id="3af3a73f26a63a"
id="7ab8273f273e7273f82c2"
id="973a72b72c72df12a2"
id="83hd826ab712cc71"
id="7612364bab72a"

Another way:

sed ' s/d-fl ai-c" /n/g; s/-/"n/g; :L; /^id=/ P; D; bL; ' file
id="3af3a73f26a63a"
id="7ab8273f273e7273f82c2"
id="973a72b72c72df12a2"
id="83hd826ab712cc71"
id="7612364bab72a"