some issues with installation

[enriqu...@gmail.com]

I wanted to share some minor issues that I have with the classical build of sage. If they can be solved somehow, it may be a good idea to update the installation guide. In a forthcoming message I plan to post more relevant issues concerning the meson install. BTW, is it better to put them here or as an issue in github?

The first minor issue concern variables. These are some lines of the log:

[sagelib-10.9.beta1] [spkg-install]   ../../src/sage/meson.build:168: WARNING: The variable(s) 'PACKAGE_VERSION' in the input file 'src/sage/ext_data/notebook-ipython/kernel.json.in' are not prese
nt in the given configuration data.
[sagelib-10.9.beta1] [spkg-install]   Configuring config.py using configuration
[sagelib-10.9.beta1] [spkg-install]   ../../src/sage/meson.build:178: WARNING: The variable(s) 'FOURTITWO_CIRCUITS', 'FOURTITWO_GRAVER', 'FOURTITWO_GROEBNER', 'FOURTITWO_HILBERT', 'FOURTITWO_MARKO
V', 'FOURTITWO_PPI', 'FOURTITWO_QSOLVE', 'FOURTITWO_RAYS', 'FOURTITWO_ZSOLVE', 'SAGE_MATHJAX_DIR', 'SAGE_ROOT', 'SAGE_VENV', 'configure_input' in the input file 'src/sage/config.py.in' are not present in the given configuration data.

The first minor issue concern variables. For the last ones, if the variable are not needed maybe they should disappear. For PACKAGE_VERSION I wonder if it is behind the fact that some tuning must be done to use sage in system jupyter. Up to know, it installs two useless GB with an incomplete kernel.json file.

The second issue concerns documentation. Using meson it is possible to disable it with an option. Using ./configure and make, what I do is to comment out the last line of src/meson.build; what it is strange is with this line uncommented even if one does not build the documentation, one has two wait for several minutes. Moreover, if one decides to build the documentation with make doc, then one has to wait again, apparently the previous setup is lost.

As I said, there are minor issues, but it is probably not a big deal to patch them.

Anyway, happy new year and thanks for the work making this software useful. Best, Enrique.

[Tobia...@gmx.de]

Thanks for the report!

The ipython kernel spec is indeed mostly broken. There is an open issue on github for it already https://github.com/sagemath/sage/issues/40605 (linking mostly to a very helpful discussion with one of the maintainers of ipython). Based on that input, it's clear what needs to be done but nobody found the time yet to create a PR.

I cannot say anything about the FOURTITWO variables. The Mathjax variable should be fixed by https://github.com/sagemath/sage/pull/40404, and for the other variables I've just opened https://github.com/sagemath/sage/pull/41358.

Fixing the annoying issue with the documentation is high on my todo list. See https://github.com/sagemath/sage/pull/41156 for work-in-progress.