Regex capturing not working as one would expect (Failed to compose renaming scheme)

[gr4nt3d]

I wanted to rename files according to the following scheme:

"original data"	"more data"	>	rename
first middle surname of authors	title	>	surname_f(m)_title
Author McAuthentic	How to waste time with regex	>	mcauthentic_a_how_to_waste... (depends on cutoff)
Adam Adams; Bob Bobsen	Our-Very-Important Book	>	adams_a_bobsen_b_our_very_important_book
...	...	...	the middle names are optional and I already gave up on

I tried to get information from the website, but the description on how users can define wildcards did not cover all necessary information. Googling and a code search was unsuccessful. ChatGPT helped a little but did not get the problem. My regex should do the job according to regex101.com -- compare below -- but I am too inexperienced to rule out a simple error on my behalf. Anyways, Zotero / ZotFile does not accept it and just gives me the full authorLastG that I used, or doesn't capture all necessary info.

Zotero-Version: 6.0.33 (homebrew)
ZotFile-Version: 5.1.2
OS: macOS

Ideal fix: Please provide other means to make such formats easily achievable.

---

My test-case for the regex capturing:

I am aware that the test-case does not suffice but should still represent a relevant MWE. If it works a sequence of regex's, append's and finally replace and toLower would be enough (hence it must work in sequence). In the end the information from authorLastG (eg. _surname,_firstname_surname2,_firstname2_m.) should be extracted and the format would look something like this: {%1_}{%t}

My regex:

([a-zA-Z]+)\,_([a-zA-Z])?(?:[a-zA-Z_\.]+?(?=[a-zA-Z]+\,|$))?(?:([a-zA-Z]+)\,_([a-zA-Z])?(?:[a-zA-Z_\.]+?(?=[a-zA-Z]+\,|$))?)?

ChatGPTs try:

([A-Za-z]+)\s*,\s*([A-Za-z])?(?:[A-Za-z_\.]+?(?=[A-Za-z]+\s*,|$))?(?:\s*,\s*([A-Za-z]+)\s*([A-Za-z])?(?:[A-Za-z_\.]+?(?=[A-Za-z]+\s*,|$))?)?

Test:

{
  "2": {
      "field": "authorLastG",
      "operations": [
          {
              "function": "exec",
              "regex": "([a-zA-Z]+)\\,_([a-zA-Z])?(?:[a-zA-Z_\\.]+?(?=[a-zA-Z]+\\,|$))?(?:([a-zA-Z]+)\\,_([a-zA-Z])?(?:[a-zA-Z_\\.]+?(?=[a-zA-Z]+\\,|$))?)?",
              "group": 2
          },
          {
              "function": "toLowerCase"
          }
      ]
  }
}

Some authorLastG test-cases for regex101:

brunton,_steven_l._kutz,_j._nathan_
bishop,_christopher_m._
sutton,_richard_s._barto,_andrew_g._
cai,_qingpeng_et_al_
konda,_vijay_tsitsiklis,_john_
konda,_vijay_tsitsiklis,_john
li,_yuxi_
lillicrap,_timothy_p._et_al_
mcauthentic,__
tsitsiklis,_john_n_van_roy,_benjamin_
sutton,_richard_s._et_al
sutton,_richard_s._barto,_andrew_g.
mitchell,_thomas
szepesvari,_c.