Redshift regexp_substr

问题

I want to replicate this regex pattern to regexp_substr. I want to capture the second group.

'(\?)(.*?)(&|$)'

I have tried this

regexp(my_url, '\\?.*?&|$')

And some similar variations of the above, but I have been getting the errror: ERROR: XX000: Invalid preceding regular expression prior to repetition operator. The error occured while parsing the regular expression: '\?.*?>>>HERE>>>&|$'.

回答1:

Since Amazon Redshift supports only POSIX regex, you need to use greedy quantifiers rather than lazy ones, but with a negated character class:

regexp(my_url, '\\?([^&]*)')

The pattern matches:

• \? - a question mark
• ([^&]*) - Group 1: zero or more chars other than &

来源：https://stackoverflow.com/questions/46495580/redshift-regexp-substr