Re: [sage-devel] Re: Demote SageTeX to an optional package?

2021-12-06 Thread Nathan Dunfield
On Monday, December 6, 2021 at 8:39:23 AM UTC-6 Michael Orlitzky wrote: > > Does "sage -i" not work either? (Is there a ticket for fixing it?) > It doesn't with the recommend macOS binaries, nor does it if you install SageMath via conda ("make: *** No rule to make target 'all-toolchain'. Stop"

Re: [sage-devel] Re: Demote SageTeX to an optional package?

2021-12-06 Thread Dima Pasechnik
On Mon, Dec 6, 2021 at 6:06 PM Matthias Koeppe wrote: > > -1 on demoting it from standard. It's tiny, and installing it as part of the > distribution is unproblematic. it's possible to make it into a PyPI package - e.g. switch from using make to using ninja to build the package (I even have a p

Re: [sage-devel] Re: Naming conflict between Givaro and Factory

2021-12-06 Thread Dima Pasechnik
On Mon, 6 Dec 2021, 18:55 Matthias Koeppe, wrote: > Since https://trac.sagemath.org/ticket/29024 Sage uses suitable system > installations of Singular, so it cannot be assumed that you can patch away > this problem > one can install a patched header into SAGE_LOCAL/include/ and use it instead of

[sage-devel] Re: Demote SageTeX to an optional package?

2021-12-06 Thread Matthias Koeppe
-1 on demoting it from standard. It's tiny, and installing it as part of the distribution is unproblematic. -1 on adding a Sage-specific installation procedure for TeX. Actually, we already have one but there is no good reason to use it other than possibly as part of the binary distribution; h

Re: [sage-devel] Re: Naming conflict between Givaro and Factory

2021-12-06 Thread Clement Pernet
For the record, I reported the problem https://www.singular.uni-kl.de/forum/viewtopic.php?f=10&t=2965&start=0 and a fix has quickly been commited. I will include this fix as a patch to singular in the branch of https://trac.sagemath.org/ticket/32959 where the conflict occured. Clément

Re: [sage-devel] Re: Naming conflict between Givaro and Factory

2021-12-06 Thread Dima Pasechnik
On Mon, 6 Dec 2021, 14:42 Clement Pernet, wrote: > Thanks, I also recently realized it was coming from Singular. > > > Le 03/12/2021 à 16:10, Maarten Derickx a écrit : > > > Not really sure why they #define IntegerDomain 1 on line 25 there. But I > guess that doesn't matter.> It is just an occasi

Re: [sage-devel] Re: Demote SageTeX to an optional package?

2021-12-06 Thread Michael Orlitzky
On Mon, 2021-12-06 at 01:55 -0800, Emmanuel Charpentier wrote: > > Having a *standard* way to integrate Sage’s results in a document is > **crucial*. > > So -1. Modulo the binary package issue, I'm only suggesting that we change the word "standard" in build/pkgs/sagetex/type to "optional", not

Re: [sage-devel] Re: Naming conflict between Givaro and Factory

2021-12-06 Thread Clement Pernet
Thanks, I also recently realized it was coming from Singular. Le 03/12/2021 à 16:10, Maarten Derickx a écrit : > Not really sure why they #define IntegerDomain 1 on line 25 there. But I > guess that doesn't matter.> It is just an occasion of having to different > libraries accidentally using t

Re: [sage-devel] Re: Demote SageTeX to an optional package?

2021-12-06 Thread Michael Orlitzky
On Sun, 2021-12-05 at 20:03 -0800, Nathan Dunfield wrote: > I'd be extremely hesitant to remove something that's been a standard > package for over a decade unless it was causing major headaches. I could > be wrong, but isn't SageTeX just a couple files totaling under 25K? > > The vast majority

[sage-devel] Re: Demote SageTeX to an optional package?

2021-12-06 Thread kcrisman
Having a *standard* way to integrate Sage’s results in a document is > **crucial*. > > Since Sage is not (yet) in the position of R to have many user packages in a standardized place like CRAN, and since things like knitr are (hah!) crucial in that ecosystem, we almost certainly should continu

Re: [sage-devel] Re: Demote SageTeX to an optional package?

2021-12-06 Thread Dima Pasechnik
It can become an optional spkg as soon as we figure out a robust way to install it with the binary distributions. The main problem is that it's not a very standard Python package, as its installation involves generation of Python source from .dtx files by TeX. So, if we make it into a PyPI package

[sage-devel] Re: Demote SageTeX to an optional package?

2021-12-06 Thread Emmanuel Charpentier
Having a *standard* way to integrate Sage’s results in a document is **crucial*. So -1. Counter-proposal : add a pseudo-package à la tinytex (which tests for a local TeX installation, and install a minimal TeX if necessary). Other