Re: display word wrapping

[Kim F. Storm]

David Kastrup <address@hidden> writes:

> address@hidden (Kim F. Storm) writes:
> 
> > David Kastrup <address@hidden> writes:
> > 
> > > If an image type is known, I hope that it is present as just its
> > > name, and _not_ as a cons cell (jpeg . t) or so.
> > > 
> > > If not, we have a major incompatibility to the previous behavior.  
> > 
> > I would not call it a major incompatibility.  Code should not rely
> > on that variable -- they should call image-type-available-p.
> 
> Says who?

I do :-)  

Because that's how the code in image.el checks it.


>      - Variable: image-types
>          This variable contains a list of those image type symbols that are
>          supported in the current configuration.
> 
> 
> image-type-available-p is not even documented anywhere.  And of
> course, stupid as I am, I have followed the Elisp manual's information
> in a package I provide.

We should definitely document image-type-available-p there.

If we change that, it is probably better NOT to maintain image-types
as a (potentially incomplete) list of supported image types.  Then
people will catch problems with the implementation sooner by not
relying on that variable anymore.

In any case, as Juanma also mentions, there is very little code
using image-types (your code being one) so it will not break a
lot of code.

> 
> Since image-types has been the only documented way of accessing that
> information so far, I think there is some justification in not
> changing its meaning before any package provider has had a chance to
> adapt.

True.

> BTW, do you have an idea since which Emacs version
> image-type-available-p has been available?

21.1 

-- 
Kim F. Storm <address@hidden> http://www.cua.dk