Bill Page wrote:
>
> I think there is a potential in the 'latex' routines to produce a more
> "semantic" style of LaTeX output. See for example:
>
> https://www.researchgate.net/publication/216797904_Using_LaTeX_as_a_Semantic_Markup_Format
There is potential to put more information into latex output.
The main issue is what kind of information we want to encode
(somewhat related is that we currently in some cases allow
lossy output). The main thing we can add relatively easily
is type information and that is in fact present if one considers
full output (with "Type:" part). Concerning 'latex' versus
coercion to OutputForm, the natural way to build such functions
is recursive. In case of 'latex' natural recursion has
_strings_ for parts and builds form correspondig to the
whole. In such framework it is very inconvenient to
transfer information form parts to whole (essentially we
would have to parse freshly build expression). That is
why 'latex' as standalone function is not a good idea.
OutputForm works at tree level and here we can stash
extra info in additional tree nodes. And beside semantic
latex we may wish to have more semantics in MathML.
With separate formatters we would have to redo the work.
Extending OutputForm we can handle both.
Extra remark: one thing worth checking is if we can
reconstruct types for OutputForm. That is given
current OutputForm and type of the whole expression
annotate subexpressions with correct types.
--
Waldek Hebisch
--
You received this message because you are subscribed to the Google Groups
"FriCAS - computer algebra system" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/fricas-devel.
For more options, visit https://groups.google.com/d/optout.
