One more data point: calling latex (or show or view) on such an expression containing a held integral *in a hold context* gives a RecursionError.
HTH, Le dimanche 9 août 2020 17:38:10 UTC+2, Emmanuel Charpentier a écrit : > > On Cocalc > <https://cocalc.com/projects/d9f9cc53-525c-4d92-a0a7-bc5941b8c734/files/?session=default>, > > I have checked that the problem can be reproduced on all available Sage > kernels back to 8.2. > > Looks more like an error from me, rather than a bug… > > Suggestions most welcome ! > > > Le dimanche 9 août 2020 16:31:47 UTC+2, Emmanuel Charpentier a écrit : >> >> The problem can be reproduced on [sagecell]( >> https://sagecell.sagemath.org/?z=eJxVi0EKhDAMAO99SSKu4t5zEg-5-wFxQy2ILU1W8nx72cPCwMDAcKQ9KzgGNuI4pMsk1s0EvH-V1DeOfH5orV_BMCtNI7y7kjAsTrN2bKHUNsHZHgc2xL-weAu6RRluqZry9fMDXHAoTw==&lang=sage&interacts=eJyLjgUAARUAuQ==), >> >> which is currently at 9.1. >> >> HTH, >> >> Le dimanche 9 août 2020 16:00:11 UTC+2, Emmanuel Charpentier a écrit : >>> >>> *Note : for once, this is a cross-post from ask.sagemath.org >>> <https://ask.sagemath.org/question/52922/latexing-showing-or-viewing-a-held-integral-triggers-its-evaluation/>, >>> >>> for ticket advice…* >>> >>> Exploring this ask.sagemath question >>> <https://ask.sagemath.org/question/52913/error-during-integration-not-of-type-unsigned-byte-15/> >>> >>> led me to discover what I think is a serious bug : it seems that latexing >>> an expression *containing* a held (= unevaluated) integral triggers the >>> latter’s evaluation ; ditto for viewing or showing… >>> >>> Minimal case: >>> >>> sage: Ig=sin(x) >>> sage: It=Ig.integrate(x,-pi,pi,hold=True) >>> sage: Cs=1/(2*pi) >>> sage: latex(It) >>> \int_{-\pi}^{\pi} \sin\left(x\right)\,{d x} >>> >>> So far, so good. But : >>> >>> sage: Ex=Cs*It >>> sage: Ex >>> 1/2*integrate(sin(x), x, -pi, pi)/pi >>> sage: latex(Ex) >>> \frac{0}{2 \, \pi} >>> >>> Aaaarghh ! And, of course : >>> >>> sage: show(Ex) >>> \newcommand{\Bold}[1]{\mathbf{#1}}\frac{0}{2 \, \pi} >>> >>> And view(Ex) proudly displays this crap… >>> >>> IMHO, this is a serious bug, entailing a risk of inadvertently >>> triggering long, impossible or never ending computations. >>> >>> Unless someone points me to the direction and magnitude of my stupidity, >>> I intend to file this as a *critical* (maybe *blocker*) ticket; >>> however, the target is not clear to me : I do not thoink that the rendering >>> system is the only culprit (nor even the main one…) ; there may be a more >>> fundamental problem in the evaluation system. >>> >>> Advice ? >>> >>> >> -- You received this message because you are subscribed to the Google Groups "sage-support" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-support+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-support/d59390ca-7240-48d0-9992-15aedbfb3691o%40googlegroups.com.
