KhiCAS, le moteur formel Giac/Xcas pour toutes TI-Nspire

[parisse]

log_10(10^x) returns x*ln(10)/ln(10) on my calc, that's unsimplified but correct.

[stevenmbiz]

log_10(10^x) returns x*ln(10)/ln(10) on my calc, that's unsimplified but correct.

I mean log(10^x) not log_10(10^x), because the base 10 can be omitted. Sorry about that.

[parisse]

That's a different convention, and I can't change that. log is a synonym for natural logarithm in the standard C library, and many scientific softwares adopt this convention. It's unfortunate that the template or log key adds a 10 when displayed, while the 10 is not added to the arguments to the log function.
Is there any way to catch the log key in the lua code and return log10 instead?

[Adriweb]

Is there any way to catch the log key in the lua code and return log10 instead?

I tried some things, but so far, no luck... (it's because of the MathBox inside the D2Editor, I don't think we can control things that far...)

[parisse]

Perhaps, I can redirect log to log10 for the nspire only, and make log10 act like logb.

[parisse]

Yes, it works. Archive updated on my webpage.

[Adriweb]

Good

[stevenmbiz]

byparts.test.jpg

Yes, the log function works now. A similar issue arose in integration by parts as shown above, if I omit the * between e^x and x the answer is undef. But in the cas software this syntax error can be corrected automatically and return the right answer. I'm wondering if this situation can be dealt with in the same way.

[Adriweb]

Well, it's not a syntax error, it's "just" that implicit multiplication is understood for non-ambiguous cases (there are some where it is, and then there is an error message).
On Xcas, it seems that some simple implicit multiplications are handled, but not all, like this one. And parisse will probably not change the parser as he stated here.

But who knows, let's wait for his answer

[stevenmbiz]

Understood, I'm interested about the answer too.