Re: [sage-devel] Different behavior in normal usage and when building the doc (decimal separator / tachyon syntax error)

On Mon, Jan 06, 2020 at 06:37:25PM +0100, Jean-Florent Raymond wrote: > > I have been investigating a bug that happened when I tried to build the > documentation of sage 9.0 on a new machine. > It seems to be the same as described in > > https://groups.google.com/d/topic/sage-devel/5jajeJiJNiY/di

Re: [sage-devel] Re: drop python2 compatibility in 9.1 ?

On Mon, Jan 6, 2020 at 7:23 PM Travis Scrimshaw wrote: > > I agree that the Python3 only version of Sage should be called 10.0 given the > large backwards incompatible changes that result from the Python3 change. > Furthermore, I also concur that we should release a version 9.1 (on an > acceler

Re: [sage-devel] Re: drop python2 compatibility in 9.1 ?

On Mon, Jan 6, 2020 at 7:30 PM Eric Gourgoulhon wrote: > > Le lundi 6 janvier 2020 14:21:56 UTC+1, E. Madison Bray a écrit : >> >> >> I agree with Nils. There should be at least a one release deprecation >> period. Also, while I don't think we use any kind of real semantic >> versioning, I think

Re: [sage-devel] Re: drop python2 compatibility in 9.1 ?

On Tue, Jan 7, 2020 at 12:06 AM rjf wrote: > > just curious when this ends. Python 4 awaits. It's already ended. Python 4 is not going to be the major backwards-compatibility breaker that Python 3 was. It's just going to be the next release after 3.9, at which point we can also hopefully final

Re: [sage-devel] Re: drop python2 compatibility in 9.1 ?

William, thanks for this relevant (if not directly, as you say) point of view. For CoCalc (run by a company), we have many different versions of Sage > installed for projects to use. For Jupyter notebooks, users just > explicitly select whatever version they want (via kernel selection), and i

Re: [sage-devel] Re: drop python2 compatibility in 9.1 ?

Le mardi 7 janvier 2020 13:25:04 UTC+1, E. Madison Bray a écrit : > > On Mon, Jan 6, 2020 at 7:30 PM Eric Gourgoulhon > wrote: > > > > On the other hand, for the end user the major backwards-incompatibility > change already happened: a Python 2-only piece of code will break > immediately in an

Re: [sage-devel] Re: drop python2 compatibility in 9.1 ?

Why 8.9? One can build 9.0 with python2 just fine. On Tue, Jan 7, 2020 at 1:16 PM Eric Gourgoulhon wrote: > > Le mardi 7 janvier 2020 13:25:04 UTC+1, E. Madison Bray a écrit : >> >> On Mon, Jan 6, 2020 at 7:30 PM Eric Gourgoulhon wrote: >> > >> > On the other hand, for the end user the major bac

Re: [sage-devel] Re: drop python2 compatibility in 9.1 ?

Le mardi 7 janvier 2020 14:35:45 UTC+1, Dima Pasechnik a écrit : > > Why 8.9? One can build 9.0 with python2 just fine. > > Yes I know, but I was speaking about the time and energy to actually build those and distribute them. IMHO, there are more pressing issues, like the handling of ipython7

Re: [sage-devel] Re: drop python2 compatibility in 9.1 ?

On Tue, Jan 7, 2020 at 2:16 PM Eric Gourgoulhon wrote: > > Le mardi 7 janvier 2020 13:25:04 UTC+1, E. Madison Bray a écrit : >> >> On Mon, Jan 6, 2020 at 7:30 PM Eric Gourgoulhon wrote: >> > >> > On the other hand, for the end user the major backwards-incompatibility >> > change already happened

Re: [sage-devel] Re: drop python2 compatibility in 9.1 ?

On Tue, 7 Jan 2020, 06:04 William, wrote: > This is just some clarification and remarks related to what Karl wrote, > and not really super relevant to this thread... > > On Monday, January 6, 2020 at 7:17:16 AM UTC-8, kcrisman wrote: >> >> Because a lot of Sage users are also developers (presumab

Re: [sage-devel] Re: drop python2 compatibility in 9.1 ?

Just a silly question. How do we know that the code remains python2 compatible in the first place? As far as I can see all the patchbots run python3 so it is very easy to break something. 9.0 is supposed to be fully python2 compatible, but are the doctests being tested for that? If we don't te

Re: [sage-devel] Re: drop python2 compatibility in 9.1 ?

On Tuesday, January 7, 2020 at 2:17:54 PM UTC-5, Jonathan Kliem wrote: > > Just a silly question. How do we know that the code remains python2 > compatible in the first place? > > As far as I can see all the patchbots run python3 so it is very easy to > break something. 9.0 is supposed to be fu

[sage-devel] Removing reverse graph from CGraph

Dear all, currently the sparse graph backend keeps a reversed copy of the graph. However, the SparseGraph itself does not have access to it and thus the reversed structure should be moved there for obvious optimizations. See #28904 . As the sparse graph

Re: [sage-devel] Re: drop python2 compatibility in 9.1 ?

The question I have is the folllowing. Suppose a user have 10k lines of SageMath code (notebooks, files, etc.) working well in 8.9 which obviously will get broken in the default 9.0 mostly because of: - print - comparisons - iterkeys, iteritems - and few other particular changes 1. What is t