How to create extractor pattern that uses parts of the text it wishes to extract?
I would like to extract a date and time from within a bunch of HTML. A lot of the HTML looks very similar, and it might change occasionally change. This is the format of what I would like to extract: 1:15PM 1/21/10 I am searching for this field because it changes periodically. Is it possible to set up the etractor pattern to look for the colon, AM or PM, and the 2 slashes? The slashes will shift depending on if the month or day has 1 or 2 digits. And it will either show "AM" or "PM" depending on the time of day.
Can you create an extractor pattern that uses parts of the text it wishes to extract? How would you do that?
Gary, Try the
Gary,
Try the following.
Example to scrape:
1:15PM 1/30/10Extractor pattern token:
~@time@~ ~@date@~Regular expression for "time" token:
\d{1,2}:\d{2}AM|PM{1}Regular expression for "date" token (same as pre-loaded "U.S. Date" regex):
\d{1,2}[-/. ]+\d{1,2}[-/. ]+\d{2,4}If you have stuff other than HTML tags surrounding the date/time then you may need to modify the regex somewhat.
-Scott