[Axiom-developer] Re: [fricas-devel] mathml patch

[Waldek Hebisch]

> Hi Tim, Waldek,<br><br>This patch covers a few typos in the documentation
> and the correction for<br>"continuedFraction(3,repeating [1], repeating
> [3,6])".
> Waldek your patch<br>for this one missed the last "ZAG" in the case of
> a finite continued fraction<br>and replaced it with an ellipsis.
> I've added a test to see if the last argument<br>is a ZAG or an ellipsis.
> If neither then I give an error but still deliver valid<br>MathML, i.e.
> the error is contained in an mtext element

I am not sure why do you want to deliver valid MathML in this case.
Namely, getting to formatZag with something else is a bug -- either
in code genereting unexpected thing or in formatZag.  So, the output
is incorrect anyway.  But if you only deliver text inside MathML
it may easily be overlooked.

In general, I feel that instead of guessing what given construct
means you should just throw errors: once an unhandled construct
appears it is easy to catch error in the debugger and find out
what it means.  If you try to mask errors debugging becomes much
harder...


> Would you like this as a series of patches<br>or one big one at the end?
> Or since the package may end up being largely<br>rewritten maybe I should
> just send the rewritten package?

I prefer small patches -- it makes discussing changes much easier.
But if you fell that changes are mutually dependent then combining
them is OK.

-- 
                              Waldek Hebisch
address@hidden