I was getting quite far via regex101 but now I am stuck.

I want to extract a string between "markers" using Regex from Python 3.9. In the following example lines I will get the foobar back for each line. The "marker" is =. But that marker does have some edge cases.

1. lore =foobar= ipsum (there is space before and after =)
2. lore =foobar=.
3. =foobar= ipsum
4. lore =foobar=

This is what shouldn't not match because the =x is not allowed.

5. lore =foobar=x

That is the regex I am using (Python 3.9)

=(.*?)=[ .] (see a space in the beginning!)

I can handle the characters following after the second marker; allowed is a space or a period.

Number 1 and 2 are working. But 3 and 4 are missing.

The no character or line ending is missing.

Also in the beginning I don't now how to check for no character before = OR .

CodePudding user response：

You could write the pattern as:

(?:^| )=(.*?)=(?:[ .]|$)

• (?:^| ) Non capture group with an alternation | matching either a space or assert the start of the string
• = Match literally
• (.*?) Capture group 1, match any character as least as possible
• = Match literallt
• (?:[ .]|$) Match either a space or dot, or assert the end of the string

Regex demo

If there can not be any equals sign in between, you might also write the pattern as:

(?<!\S)=([^=\n]*)=(?:[ .]|$)

• (?<!\S) Assert a whitspace boundary to the left
• = Match literally
• ([^=\n]*) Capture group 1, match any character except = or a newline
• = Match literally
• (?:[ .]|$) Match either a space or dot, or assert the end of the string

Regex demo