Variants: can there be fewer keys for a given pattern than there are selectors in the preamble?

[mihnita]

[mihnita]

Comments migrated from the slides

---

Slides comment, Mihai Nita (@mihnita), 2:06 PM Apr 21

How is this valid?
Does it mean
foo1 _
or
_ foo1
?
Why do this arbitrary decision?
Ambiguity to save 2 characters (space & '_')

[markusicu]

The syntax needs to be unambiguous and prevent confusion. The number of selectors and the number of keys in each variant need to match.

[eemeli]

My preference would be to allow for fewer keys, extending with * at the end if necessary. The primary benefit of this would be that this removes an error condition from the parsing and message handling.

However, this is not a particularly strong preference. Hence the resolve-candidate label, to see if we might reach agreement on this.

[macchiati]

I think it is better to alert the author that there is a problem: they could have inadvertently forgotten why of the keys.

…

On Tue, May 31, 2022, 01:15 Eemeli Aro ***@***.***> wrote: My preference would be to allow for fewer keys, extending with * at the end if necessary. The primary benefit of this would be that this removes an error condition from the parsing and message handling. However, this is not a particularly strong preference. Hence the resolve-candidate label, to see if we might reach agreement on this. — Reply to this email directly, view it on GitHub <#254 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACJLEMFMOYS2NWNHBR6WZVTVMXDD5ANCNFSM5VWCW5UA> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[stasm]

The develop syntax allowed for fewer keys primarily because of how some selectors weren't actually used for selecting, but only for local variable definitions. #252 has more details about this, and why we shouldn't do it.

Assuming we fix #252, there's still a decision to be made about whether to allow for a syntax sugar in form of filling in for missing keys with *. What's nice about this is that because it's strictly syntax sugar, we can safely punt it to MF2.x, if there's demand for it.

For 2.0, I agree with resolve-candidate and with requiring that the count of selectors match the count of variant keys on each variant.

[macchiati]

sounds good.

…

On Wed, Jun 1, 2022 at 8:07 AM Stanisław Małolepszy < ***@***.***> wrote: The develop syntax allowed for fewer keys primarily because of how some selectors weren't actually used for selecting, but only for local variable definitions. #252 <#252> has more details about this, and why we shouldn't do it. Assuming we fix #252 <#252>, there's still a decision to be made about whether to allow for a syntax sugar in form of filling in for missing keys with *. What's nice about this is that because it's strictly syntax sugar, we can safely punt it to MF2.x, if there's demand for it. For 2.0, I agree with resolve-candidate and with requiring that the count of selectors match the count of variant keys on each variant. — Reply to this email directly, view it on GitHub <#254 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACJLEMEQ2AFI4ZITAKJTNMLVM54EXANCNFSM5VWCW5UA> . You are receiving this because you commented.Message ID: ***@***.***>

[romulocintra]

Consensus : Have same number of keys and selectors