This page is part of the web mail archives of SRFI 29 from before July 7th, 2015. The new archives for SRFI 29 contain all messages, not just those from before July 7th, 2015.
> > I know that Scheme is supposed to be written > interactively, but I > wouldn't like to exclude users of the likes of > Stalin or Bigloo > either. They'll be using offline compilation, and > splitting the > code like above is the best way I can see to > support this. I'd > like this behaviour to be explicitly allowed in an > implementation. This is only an issue as to how bundles get declared and then subsequently stored. Bundles would necessarily be stored at compile time, and just read at runtime. > One other point which I thought was badly specified > was > current-locale-details. I think implementers are > given so much > latitude as to make it useless. I'd suggest making it > manipulate an > alist instead of a list of symbols, and have the same > keys as fields in > the C struct lconv (see the localeconv or setlocale > man page). current-locale-details should provide a list of attributes of the locale, such as encodings, that are less important than the country and language of the locale. They don't specify information about how to format data for the locale (as struct lconv does). Thats the job of the translator defining message templates. Scott
Description: PGP signature