Andreas Abel

Great! So the failure list is unchanged over the PLY target. I'll have another look.

N.B. The CI failures are caused by dependencies: - 8.2.2: https://github.com/seereason/Diff/issues/16 - 8.6.5 / 8.8.4: https://github.com/sol/doctest/issues/460

> **Testsuite results:** The example grammars work (no layout supported and internal has no effect). The regression tests yield: I went through all of the failed tests: ### Parameterized tests:Python:Python:100_coercion_lists...

@AiStudent wrote: > It seems that the length of the regex decides the precedence, not the lengths of the matched tokens. But this seems then to be a bug in...

Thanks for the report! I tried both examples with Agda 2.5.1.1 - 2.6.4.3 (all 2.5 and 2.6 versions I have installed), and they pass. So at least the release notes...

Agda dev meeting 2024-09-11: market this as a bug fix. Add the example(s) to the testsuite.

Also, `_→_` isn't a _thing_ in Agda as `→` can only appear "fully applied" so far, so we would have to 1. reserve an abstract name for it 2. create...

> Here the comment (referring to #2964) N.B.: This issues was closed for 2.6.0 but the regression is in 2.6.2.

Yeah, you are probably right about this, bisection only returns the generic commit: > 2f212bf5568a8d408f3b3a8d3cb87a6fa6e6b3a1 is the first bad commit > Author: Ulf Norell > Date: Thu May 28 09:54:29...

ATTN maintainer (should be one of @fizruk @snoyberg @phadej): https://hackage.haskell.org/package/http-api-data/maintainers/