I've noticed few bugs in the current version of Xcas (stable branch), which are listed below.
1. When I try to solve limit in Example 61 in http://math.cmu.edu/~bkell/lhopital.pdf using the command
Code : Tout sélectionner
assume(a>0); limit((sqrt(2*a^3*x-x^4)-a*root(3,a^2*x))/(a-root(4,a*x^3)),x=a)
2. The function "plotarea" does not work for straight lines. When I enter, for example, plotarea(x,x=1..2), i get the following error message.
Code : Tout sélectionner
"Project-Id-Version: giac 0.9.0
Report-Msgid-Bugs-To:
POT-Creation-Date: 2017-07-17 11:09+0200
PO-Revision-Date: 2010-06-30 10:24+0200
Last-Translator: parisse <bernard.parisse@ujf-grenoble.fr>
Language-Team: English
Language:
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Plural-Forms: nplurals=2; plural=(n != 1);
Error: Bad Argument Value"
Code : Tout sélectionner
%{1,2,3%} minus %{1,2,3%}
Code : Tout sélectionner
%{cos(k*pi/6-pi/3)$(k=1..100)%} minus %{cos(k*pi/6+pi/3)$(k=1..100)%}
4. The function "plotinequation" sometimes does not fill the resulting area although it is expected to do so. For example, a circle will be filled only if its radius is not an integer. The command
Code : Tout sélectionner
plotinequation(x^2+y^2<=3)
Code : Tout sélectionner
plotinequation(x^2+y^2<=4)
5. The unit prefix "micro" is not working.
6. (This is maybe not a bug). When I define F:=Int(x^2,x), I get integral of x^2, as expected ("Int" is an inert function). But when F is part of some subsequent command, the integral gets evaluated, for example when I try to do variable substitution with "subs". It would be nice to get unevaluated integral to show the result of the substitution. Is it possible somehow?