As suggested in a comment to patch2233.

1 patch for repository http://darcs.net/screened:

patch 6795d973c28d5194908b8e4e621b588ba407df6d
Author: Ben Franksen <ben.franksen@online.de>
Date:   Sun Jun 25 12:35:08 CEST 2023
  * group --external-merge with the other conflict handling options

  This also slightly refactors the internal representation of these options.

Attachments

• group-__external_merge-with-the-other-conflict-handling-options.dpatch
• patch-preview.txt

Let me drop a hint here. Whenever you see a patch that remains in 
status "needs-screening" for more than a day, then this often means: I 
am not sure if this is really a good idea, please take a look and 
offer some feedback.

After glancing at the code:
 - I didn't see any red flags for the idea
 - It looks like it simplifies some logic, particularly the case 
statement at `hunk ./src/Darcs/Repository/Merge.hs 281`

But I'd have to review it properly to express any stronger opinions.

Slightly rebased and now screened.

1 patch for repository http://darcs.net/screened:

patch 1248a6c005afd13e69da36941dfc7bbb82bf7ab7
Author: Ben Franksen <ben.franksen@online.de>
Date:   Sun Jun 25 12:35:08 CEST 2023
  * group --external-merge with the other conflict handling options

  This also slightly refactors the internal representation of these options.

Attachments

• group-__external_merge-with-the-other-conflict-handling-options.dpatch
• patch-preview.txt

Looks good.

>  NoAllowConflicts -> error "impossible" -- was handled in 
announceConflicts

There's some scope to remove this case by having announceConflicts
return Maybe ResolveConflicts instead of Bool, but I'm not sure
it's worth it.

(The old code just treated this impossible case the same as one
of the possible cases, so this code is clearer as it's more explicit).