fixdat does not respect --single

[ozboss]

Paste the command

npx -y igir@latest fixdat
--input fixdat_input
--dat "No-Intro/Nintendo - Game Boy (20240914-013134).dat"
--filter-language DE,EN
--single
--prefer-language DE \

Describe the bug

Running the above command filters the languages correctly, but still has lots of clones in the created .dat file, i.e. entries with "cloneof=..." information.

Expected behavior

Creates a .dat file without any clones.

Debug logs

igir.log

DAT(s) used

No response

igir version

v3.0.0

Node.js version

v20.12.2

Operating system

Fedora 40

Additional context

No response

[IDmedia]

"No-Intro/Nintendo - Game Boy (20240914-013134).dat"

Have you tried using the Parent-Clone dat files? They yield better results with --single.

[ozboss]

Have you tried using the Parent-Clone dat files? They yield better results with --single.

Same behavior with P/C XMLs.

[emmercm]

@ozboss is right, I see why this would be unexpected, and it's the same root issue as https://github.com/emmercm/igir/issues/1341.

Igir is currently performing 1G1R on found input files, not on ROMs in DATs. Meaning if a game has these three clones:

Game (USA)
Game (Europe)
Game (JPN)

and you provide --prefer-region USA, but only have a copy of Game (Europe), it will consider that a match. The thinking is having any copy of the game is better than having no copy of the game.

But that thinking has other implications, such as this. Because Igir is willing to consider any of the clones as a match it will/must keep all of them in fixdats and reports.

[ozboss]

I see now, --filter-language DE, EN correctly removed Game (JPN) and kept the other two valid options. I commented my opinion on this already in #1341 so this can be closed as duplicate.

[github-actions[bot]]

:lock: Inactive issue lock

This issue has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.

Comment generated by the GitHub Lock Issues workflow.