Question

I am using Beautiful Soup to identify a specific tag and its contents. The contents are html-links and I want to extract the text of these tags.

The problem is that the text is made up of different numbers according to a specific pattern. I am only interested in number such as “61993J0417” and “61991CJ0316” and I need the regexp to match both when the number has a “J” and “CJ” in the middle.

I have used this code to achieve this:

soup.find_all(text=re.compile('[6][1-2][0-9]{3}[J]|[CJ][0-9]{4}'))

The soup variable is the contents of the specific tag. This code works in 9 out of 10 cases. However, when I run this script on one of my source files, it also matches numbers such as “51987PC0716”.

I cannot understand why so I turn to you for assistance.

Answer 1

You haven’t specified what the | applies to; by default it’s the entire regex, meaning you have asked for either

[6][1-2][0-9]{3}[J]

(which is the same thing as 6[12][0-9]{3}J) or

CJ[0-9]{4}

(not [CJ], which means “either C or J”). Use parentheses to specify what the alternatives are:

^6[12][0-9]{3}(J|CJ)[0-9]{4}$

which is better written

^6[12][0-9]{3}C?J[0-9]{4}$