Arbritrary Information is wrongly taken from Filename for Metadata

[SomeRandomDude870]

Describe the issue
I have a D:\test folder with:
folder: [359637][uten ameka][[Uten Ameka] Koukan Renshuu Zenpen _ Exchange Practice - First Part (COMIC Anthurium 2020-10) [English] [RedLantern] [Digital]]
[359685][ziran][[ZIRAN]辣妹母……（K记翻译）].cbz
[359688][asahi][[NTR Chronicle(Huskar)] Cover + illustration [Chinese]].cbz
[359705][shirokuma a][(C97) [Kuma no Ana (Shirokuma A)] Shitara-chan to no Kyuujitsu (Alice Gear Aegis) [Chinese] [爱弹幕汉 化组]].cbz
[Zucchini] (C97) [Z.A.P. (Zucchini)] Artoria wa Makerarenai. (Fate／Grand Order) [Chinese] [黎欧x新桥月白日语社汉化] (356060)

The problem is that the Metadata is completely randomly filled:

Shirokuma -> correct
NTR chronicle -> correct
[359685][ziran][[ZIRAN]辣妹母……（K记翻译）] -> Here the number is instead taken as the artist ID and the link goes to nhentai /358775, then the site is filled with the Metadata from this Link. Not even the name is correctly taken.
Koukan -> the number is again taken as artist id and nothing else is filled. And even after deleting the artist Id it is not possible to query for Metadata.
Zucchini -> Nearly everything is taken correctly, however out of some reason it has taken the English version instead of Chinese. It´s also linking to the english one.

To Reproduce
See above, I have used the Scan Tab with D:\test

Expected behavior
Well, that the data is correctly taken and that in a logical way.
I don´t see why once it correctly takes the id as the nhentai galleryId and then another time as the artist Id.
What is the naming scheme? I cannot find this anywhere.

Logs
If applicable, include your error.log or debug.log found in [HPX]\logs\ to help identifying your problem.

Screenshots
If applicable, add screenshots to help explain your problem.

Environment:
win10
firefox latest
13.3

Additional context
Add any other context about the problem here.