380575 - Clarify the usefulness of trailing letters in the l20n spec, or whack 'em

Status: RESOLVED FIXED

(L20n :: General, defect)

Description

[Wil Clouser [:clouserw]]

[migrating from email conversation]

In the l20n grammar ( http://people.mozilla.com/~axel/l20n/grammar.html ), the expander is defined as:

 	'${' expression '}' ('s' | 'i') ;

I'm curious what the 's' and 'i' represent.  I suspect 'string' and 'int' but I don't understand how that would be beneficial to us.  If they aren't useful, I'd suggest removing them.

Comment 1

[Robert Kaiser]

I actually don't think we really need them - and we should try to keep the grammar as simple as possible while still fulfilling its requirements.

Do we have any requirements that make them necessary? Do we have use cases where they are strictly needed?

I haven't seen any yet, but Axel might know more :)

Comment 2

[Axel Hecht [:Pike]]

I have put them in because all other printf-like formats have them.

I'm not sure that 's' and 'i' are enough, though, they're really placeholders for some wider discussion on how much of printf we should support. http://docs.python.org/lib/typesseq-strings.html has more on what python does. We should think about precision and width specifiers, too.

Comment 3

[Zibi Braniecki [:zbraniecki][:gandalf]]

In the series of "back from dead"

Axel: do we want to make a final decision on this now?

Comment 4

[Axel Hecht [:Pike]]

I'm not set about placeable formats, I could go for {{ var|filter }} like in django, too. Where the 's' and 'i' are basically filters.

Comment 5

[Zibi Braniecki [:zbraniecki][:gandalf]]

fixed long ago http://l20n.github.io/spec/grammar.html