[SOLVED] Sed: capture the text between the first ( and )

Issue

I have the following badly formatted text:

<h1 id="page-title">ABCD TEXT TEXT ( QQQ-10-123-01)</h1>
<h1 id="page-title">ABCD TEXT TEXT (QQQ-10-123-02)</h1>
<h1 id="page-title">ABCD TEXT TEXT (QQQ-10-123-03 (QWERTY))</h1>

and need to get from it:

QQQ-10-123-01
QQQ-10-123-02
QQQ-10-123-03 (QWERTY)

I.e. get only text between the first "(" and ")", at the moment doing the following:

sed -n "s/.*<h1 id=\"page-title\">.*(\(.*\))<\/h1>.*/\1/p" ./file.txt

and get:

 QQQ-10-123-01
QQQ-10-123-02
QWERTY)

As you can see only the second line is being processed properly, since this line is most accurate. There are problems with ignoring possible whitespace and dealing with double entry "(" and ")". Can somebody give the right direction for solving the problems?

P.S. I need to parse over 2k lines; would there be a big difference in performance between sed and awk? As far as I have been reading and understood, sed should have a little benefit in speed. Is that really so?

Solution

Using sed

$ sed 's/[^(]*([[:space:]]\?\([^)]*)\?\)).*/\1/' input_file
QQQ-10-123-01
QQQ-10-123-02
QQQ-10-123-03 (QWERTY)

$ sed -E 's/[^(]*\([[:space:]]?([^)]*\)?)\).*/\1/' input_file
QQQ-10-123-01
QQQ-10-123-02
QQQ-10-123-03 (QWERTY)

Answered By - HatLess

Answer Checked By - Robin (WPSolving Admin)