On Thursday, August 15, 2024 at 7:01:28 PM UTC-7 Kwankyu Lee wrote:
On Friday, August 16, 2024 at 10:50:34 AM UTC+9 Matthias Koeppe wrote:
On Thursday, August 15, 2024 at 4:54:29 PM UTC-7 Kwankyu Lee wrote:
Why don't we upload only sage-10.x.tar.gz as assets? It contains all other
files upload
On Friday, August 16, 2024 at 11:01:28 AM UTC+9 Kwankyu Lee wrote:
On Friday, August 16, 2024 at 10:50:34 AM UTC+9 Matthias Koeppe wrote:
On Thursday, August 15, 2024 at 4:54:29 PM UTC-7 Kwankyu Lee wrote:
Why don't we upload only sage-10.x.tar.gz as assets? It contains all other
files uploa
On Friday, August 16, 2024 at 10:50:38 AM UTC+9 Matthias Koeppe wrote:
On Thursday, August 15, 2024 at 4:10:27 PM UTC-7 Kwankyu Lee wrote:
> those tarballs should be reachable by a click from the installation guide.
That's exactly where the existing link in the installation guide leads,
after th
On Friday, August 16, 2024 at 10:50:34 AM UTC+9 Matthias Koeppe wrote:
On Thursday, August 15, 2024 at 4:54:29 PM UTC-7 Kwankyu Lee wrote:
Why don't we upload only sage-10.x.tar.gz as assets? It contains all other
files uploaded there.
Our download procedure for upstream tarballs likes to dow
On Thursday, August 15, 2024 at 4:10:27 PM UTC-7 Kwankyu Lee wrote:
> those tarballs should be reachable by a click from the installation guide.
That's exactly where the existing link in the installation guide leads,
after the mirror selection.
Note that the decision to de-emphasize use of the se
On Thursday, August 15, 2024 at 4:54:29 PM UTC-7 Kwankyu Lee wrote:
Why don't we upload only sage-10.x.tar.gz as assets? It contains all other
files uploaded there.
Our download procedure for upstream tarballs likes to download individual
files.
--
You received this message because you are
For comparison: The current total size is 1.4 gigabytes.
For the sizes of the release tarballs since 9.0, see
https://mirrors.mit.edu/sage/src/index.html
According to
https://docs.github.com/en/repositories/releasing-projects-on-github/about-releases,
there is 2GB limit on the size of asse
Do you mean this
https://github.com/sagemath/sage/archive/refs/tags/10.4.tar.gz ? This one
also does not have "upstream/" directory.
No, that one is also automatically generated by GitHub and contains the
same files as the .zip file.
This one:
https://github.com/sagemath/sage/releases/dow
For comparison: The current total size is 1.4 gigabytes.
For the sizes of the release tarballs since 9.0, see
https://mirrors.mit.edu/sage/src/index.html
To watch continuously and for non-developer users, those tarballs should be
reachable by a click from the installation guide.
--
You re
On Wednesday, August 14, 2024 at 7:33:49 PM UTC-7 Kwankyu Lee wrote:
... we discussed the size of these wheel files previously in the thread
"Current status of possibility of integrating libraries written in Rust
into Sage" (
https://groups.google.com/g/sage-devel/c/JiSryImYGe0/m/LeNyLi6dFAAJ).
The installation guide does not tell me where I can download
sage-10.4.tar.gz that contains upstream/ directory. Where is it?
See
https://doc.sagemath.org/html/en/installation/source.html#installation-steps
Yes, there it is though quite buried among many files. Thanks.
You may recall tha
On Wednesday, August 14, 2024 at 3:08:39 PM UTC-7 Kwankyu Lee wrote:
On Thursday, August 15, 2024 at 4:11:50 AM UTC+9 Matthias Koeppe wrote:
On the other hand, the "sage-10.4.tar.gz" tarball that is built by the
Release Manager and put on the Sage mirrors (as well as the
"sage-10.4.tar.gz" tarb
On Thursday, August 15, 2024 at 4:11:50 AM UTC+9 Matthias Koeppe wrote:
On the other hand, the "sage-10.4.tar.gz" tarball that is built by the
Release Manager and put on the Sage mirrors (as well as the
"sage-10.4.tar.gz" tarball that is built using the same procedure and put
on GitHub Release
On the other hand, the "sage-10.4.tar.gz" tarball that is built by the
Release Manager and put on the Sage mirrors (as well as the
"sage-10.4.tar.gz" tarball that is built using the same procedure and put
on GitHub Releases - https://github.com/sagemath/sage/releases/tag/10.4)
does contain an u
On Wednesday, August 14, 2024 at 7:50:52 AM UTC-7 Kwankyu Lee wrote:
There are multiple "tarballs" (actually wheel files) listed in
checksums.ini and uploaded/downloaded to/from GitHub Releases and the Sage
mirrors
Together they cover the supported platforms.
Thanks. I confess that I misunde
Yes. There are multiple "tarballs" (actually wheel files) listed in
checksums.ini and uploaded/downloaded to/from GitHub Releases and the Sage
mirrors
Together they cover the supported platforms.
Thanks. I confess that I misunderstood things about installing sage from
source. I thought, fo
On Tuesday, August 13, 2024 at 4:36:18 PM UTC-7 Kwankyu Lee wrote:
> Does this [platform-dependent standard package] mean that there will be
different sage tarballs for supported platforms?
Yes. There are multiple "tarballs" (actually wheel files) listed in
checksums.ini and uploaded/downloade
> Does this [platform-dependent standard package] mean that there will be
different sage tarballs for supported platforms?
I want to here an answer for this question.
--
You received this message because you are subscribed to the Google Groups
"sage-devel" group.
To unsubscribe from this group
Thanks, Eric, for the link. I've
opened https://github.com/sagemath/sage/pull/38503 for mentioning it in our
review checklist.
On Monday, August 12, 2024 at 1:01:35 AM UTC-7 Eric Gourgoulhon wrote:
> Le lundi 12 août 2024 à 02:45:03 UTC+2, Matthias Koeppe a écrit :
>
>
> I think it would be be
Le lundi 12 août 2024 à 02:45:03 UTC+2, Matthias Koeppe a écrit :
I think it would be be valuable to prepare a checklist for making such
updates and add it to our developer documentation. I think Eric Gourgoulhon
has some Jupyter notebooks that are a good test of graphics and
typesetting, some
On Sunday, August 11, 2024 at 8:05:42 AM UTC-7 Marc Culler wrote:
The real driver of the complexity, then, is the number of different
versions of python which Sage is obliged to support
Sure. My current implementation of the "sage --package create --pypi" for
these types of packages does not e
Oh. I wasn't paying very close attention when I looked at that file. I am
familiar with manylinux wheels, of course. I guess I was too busy being
impressed by the size of the file.
The real driver of the complexity, then, is the number of different
versions of python which Sage is obliged to su
On Friday, August 9, 2024 at 7:37:34 AM UTC-7 Marc Culler wrote:
I was naively looking at this from the point of view of someone maintaining
a port of Sage to a single platform, which also happens to allow a single
binary wheel to work for multiple versions of the OS. Looking at that
linked fil
As far as I can tell, that is quite different from what I was suggesting.
It involves a separate package directory for each binary wheel, not one
directory for many binary wheels which would all get installed together at
the end of the build process in one big pip run.
However, I was naively l
24 matches
Mail list logo
