On 30/06/2014 19:28, Joseph Wright wrote:
>> In order to address this, we have now added a new set of experimental
>> functions to l3candidates:
>> - \tl_upper_case:n(n)
>> - \tl_lower_case:n(n)
>> - \tl_mixed_case:n(n)
> A question raised elsewhere
> (http://chat.stackexchange.com/transcript/message/16351207#16351207) is
> of course whether "tl" is the right place for such functions at all.
> It's arguable that they can be regarded a "text" functions, so perhaps a
> "text manipulation" module would be a better location. That does not of
> course preclude discussing the detail of how they should also work, but
> may be worth consideration. Feedback here also welcome!
One argument here is that *at present* it's not clear what might be a
'better' location for case changing, while the need for the
functionality is apparent and an implementation is doable 'now'. Thus we
might argue that adding to tl with the possibility of a (well-defined)
move to another module could occur at some stage in the future. This
approach avoids adding new modules which turn out to be poorly defined.
There is a tension there of course with 'stability': we are aiming not
to make changes without good reason, but at the same time are trying to
have have mechanisms which do allow for some change where this makes sense.