Using RegEx I am trying to extract all sentences from an article containing the word "Figure" and I have this:

((?<=^|\s)[A-Za-z0-9][^!?.]*(Figure)[^.]*(?=\.|\!|))

which will, in the case of the sentence, "This effect (Smith et al., 2008) was seen in 0.0001% of samples (Figure 1b).", will give me, "of samples (Figure 1b)"

How could I modify my code so it will allow me to capture the decimals and references as well, thereby including the entire sentence from start to end?

CodePudding user response：

Assuming that your sentences are well formed (which given they look like they're from a scientific journal they should be) and always start with a capital letter preceded by a space or beginning of string, you can use this regex:

(?:^|(?<=[.!?]\s))(?=[A-Z])(?:[^.!?]|[.!?](?!$|\s[A-Z]))*Figure.*?[.!?](?=$|\s[A-Z])

This matches:

• (?:^|(?<=[.!?]\s)) : either start of string or a lookbehind that asserts a .,?, or ! followed by a space
• (?=[A-Z]) : a lookahead asserting a capital letter (we use a lookahead here so we can match Figure if it's the first word in the sentence)
• (?:[^.!?]|[.!?](?!$|\s[A-Z]))* some number of either a non-sentence ending character or a sentence ending character that is not followed by end-of-string or a space and a capital letter
• Figure : the word Figure
• .*?[.!?] : a minimal number of characters followed by a sentence ending character
• (?=$|\s[A-Z]) : a lookahead that asserts either end of string or a space and a capital letter (i.e. the start of a new sentence)

Regex demo on regex101