You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This is a bit weird because in the above example, "Firefox" is the real value of the message and nominative and genitive are the value's facets. However the value field of the message remains empty due to the fact that the facets are defined as traits. These traits also need to remain private which in turn necessitates a way to mark public traits as public; we currently use namespaces for this purpose.
The proposed change is to allow defining facets of a value as variants of a select-expression which doesn't have a selector:
This would also allow to remove namespaces from member keys (keywords). The member-expression would change the meaning: it would now be used to access a named variant of the first placeable in the value of the message. If that placeable isn't a select-expression or already has a selector, normal evaluation would follow.
about1 = About { brand-name }
about2 = About { brand-name[nominative] }
I'm looking at this from the POV of a developer writing an ftl file in an editor, and from that view, it just looks as if we added two characters that I can get wrong.
I'd love to look at this proposal once we know what ot do with #6 and #7, and look at the impact then.
@Pike, I'd argue that from the POV of a developer this change is about putting the value into the value field instead of abusing traits. And with #6 it should be clear that attributes are just that: attributes.
Goal
Define variants of a message which can be accessed from other strings.
Description
We currently define variants of the message which are local and specific to the natural language using traits. For instance, we'd write:
This is a bit weird because in the above example, "Firefox" is the real value of the message and nominative and genitive are the value's facets. However the
value
field of the message remains empty due to the fact that the facets are defined as traits. These traits also need to remain private which in turn necessitates a way to mark public traits as public; we currently use namespaces for this purpose.The proposed change is to allow defining facets of a value as variants of a
select-expression
which doesn't have a selector:This would also allow to remove namespaces from member keys (keywords). The
member-expression
would change the meaning: it would now be used to access a named variant of the first placeable in the value of the message. If that placeable isn't aselect-expression
or already has a selector, normal evaluation would follow.Dependencies
#1.
Discussion
https://groups.google.com/forum/#!topic/mozilla.tools.l10n/dhWfBXHzuZI
The text was updated successfully, but these errors were encountered: