Define the variables that are in-scope when a property is expanded

[xatapult]

In the specification it is not defined what variables are in-scope when a property is expanded. So what variables is a <value-of> in a property allowed to reference?

I would expect only the variables that are lexically in scope, meaning, for properties, only global variables defined before the property is defined? Or also the variables that are defined in the pattern/rule that fired and caused the property to be expanded?

[rjelliffe]

Good question. In general the right answer should be the straightforward thing that gives the user the most value, I suggest. In this case, it would clearly be that the property is evaluated in the context of the assert etc. and so the variables should be visible. The same would be true of diagnostics. The reason that properties and diagnostics are declared outside the assert statements is to keep asserts centred on the assertion text and not, by allowing nested features, complicate their mark up and so dilute their "essense", if that makes sense. This is the same design rationale (which is really "separation of concerns") why phases are siblings to patterns rather than containing them, and why rules cannot contain rules or patterns cannot contain patterns. A three-level structure (pattern/rule/assert) is already complex, but a five-level structure (phase/pattern/rule/assert/property) has too many extraneous things to explain to someone just interested in the assert. Regards Rick Jelliffe

…

On Wed, 30 Jun 2021, 5:30 pm Erik Siegel, ***@***.***> wrote: In the specification it is not defined what variables are in-scope when a property is expanded. So what variables is a <value-of> in a property allowed to reference? I would expect only the variables that are lexically in scope, meaning, for properties, only global variables defined before the property is defined? Or also the variables that are defined in the pattern/rule that fired and caused the property to be expanded? — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#13>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AF65KKIICVJ6XFYHBRXO33LTVLB2HANCNFSM47RXQPRQ> .

[dmj]

In this case, it would clearly be that the property is evaluated in the
context of the assert etc. and so the variables should be visible.

The same would be true of diagnostics.

I agree. Properties and diagnostics belong to the same "logical category", so to speak: They encode information relating to a failed assertion or successful report.

[AndrewSales]

Clarified this in latest draft for both properties and diagnostics.