Le dimanche 26 mars 2023 à 14:24 +0200, David Kastrup a écrit :
> Patterns like
>
> ("flageolet" .
>
> (acons "flageolet"
>
> (assoc "flageolet"Why handle alists with articulation-type keys but not ``` (list "flageolet" ...) (cons "flageolet" ...) (hash-ref xxx "flageolet") ``` etc. The possibilities are just endless. I think it has a low time-saved-for-the-user/time-spent-writing-and-testing-the-rule ratio to hunt for them, especially if you want to restrict the heuristics to contexts where “articulation-type” is referenced (how to determine them?). > How bad is the chance for false positives? Considering all articulation > types, > probably non-trivial. Yes. Especially if you also consider that many of the articulation types can equally be used with code like `(string-append "scripts." type)` to construct Emmentaler glyph names.
signature.asc
Description: This is a digitally signed message part
