复杂的正则表达式场景

Question

I have a scenario where I have multiple items with similar titles and need to simplify the titles so that they each have an identical manifest title.

Here are possible variations:

   Quarter Final Football vs. Loyola #2 -- 46:00
   Quarter Final Football vs. Loyola #1 -- 32:00
   Little Shop Act II -- 39:30 -- 5/22/91
   Little Shop Act I -- 56:15 -- 5/22/91
   B-BALL -- 3/14/86 -- vs. King -- First Half -- 43:15
   B-BALL -- 3/14/86 -- vs. King -- Second Half -- 48:15
   Oak Park 9/26/70 Part 1
   Oak Park 9/26/70 Part 2
   vs. Yale (1958) First Period
   vs. Yale (1958) Second Period

The matching result would ideally be "vs. Yale (1958)" for the bottom two. I tried with the bottom string with this regex pattern, but am feeling a bit lost as to how to handle all these variations programatically.

([.,a-zA-Z0-9\s\(\)]*)\b([#]|Period|period)\b

Thanks for any direction or help.

Edited: Output for each variation-

Quarter Final Football vs. Loyola
Quarter Final Football vs. Loyola
Little Shop
Little Shop
B-BALL -- 3/14/86 -- vs. King
B-BALL -- 3/14/86 -- vs. King
Oak Park 9/26/70
Oak Park 9/26/70
vs. Yale (1958)
vs. Yale (1958)

Answer 1

There does not seem to be a common denominator. If that is the case, you will need a list.

I think you need to consider what might be the start of the various phrases that would be considered surperfluous. The following regex works on your examples; but whether it can be generalized to your real data remains to be seen. I used alternation. If there can be more than two, you might need to add Third, Fourth, etc to the list.

\s*(?:#|act|(?:--)?\s*(?:First|Second)|Part).*

Used perhaps something like:

$result = preg_replace('/\s*(?:#|act|(?:--)?\s*(?:First|Second)|Part).*/i', '', $subject);

Note that I used the case-insensitive mode; and you also want dot to Not match newline.