What is the easiest and fastest way to parse this file?

Question

bash

Good afternoon,
tell me the easiest way to parse this file on bash

xxx:[<b>yyyyy1</b>] yyy:[xxxx]
xxx:[<b>yyyyy2</b>] yyy:[xxxx]
....

As a result, you need to get:
yyyyy1
yyyyy2

Answer 1

grep/sed

$ printf "xxx:[<b>yyyyy1</b>] yyy:[xxxx]\nxxx:[<b>yyyyy2</b>] yyy:[xxxx]" | grep -oE '<b>+</b>' | sed -E 's/<\/?b>//g'

yyyyy1
yyyyy2

perl one-liner:

$ printf "xxx:[<b>yyyyy1</b>] yyy:[xxxx]\nxxx:[<b>yyyyy2</b>] yyy:[xxxx]" | perl -e 'while(<>){ if ($_ =~ m/<b>(.*)<\/b>/){ print $1."\n"; } }'
yyyyy1
yyyyy2

Answer 2

cat /tmp/33.sh | egrep -o "<b>\w+" | sed 's/<b>//g'