Python RegEx: match words in a string only if they are not preceded by a specific character

For example:

something foo somethingelse // matches: [something, foo, somethingelse]
something1 something.foo-somethingelse // matches: [something1, something, foo, somethingelse]

but not if preceded by ‘>’

something1 something>foo-.somethingelse // matches: [something1, something, somethingelse]

I did develop this regex:

(?(?<!>)(\w+)|\w+)
(? - if
 (?<! - is not preceded by (negative lookbehind)
  > - literal '>'
 )
 (\w+) - capture a group of 1 or more word characters
 | - else
 \w+ - match 1 or more word characters without capturing them
)

which worked in regex101 (words starting with > weren’t included in group 1), but that’s when I learned that conditional regex is either not a thing in Python, or it doesn’t work like in Perl.

Also correct me please if that is not true.

>Solution :

You don’t have to use a conditional, you can use your negative lookbehind assertion and start with a word boundary:

\b(?<!>)\w+

See a regex demo

If you want to use the capture group:

\b(?<!>)(\w+)|\w+

See another regex demo