FYI question on Stackoverflow on i/o on sagecell

[kcrisman]

For anyone who might be interested in I/O problems on Sagecell, apparently someone has found a race condition of some kind:
http://stackoverflow.com/questions/37692165/sagemathcell-valueerror-when-printing-string-sometimes

[William Stein]

When I click on his link to the code, I don't see any code! Chrome + OS X.

> --
> You received this message because you are subscribed to the Google Groups
> "sage-notebook" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to sage-noteboo...@googlegroups.com.
> Visit this group at https://groups.google.com/group/sage-notebook.
> For more options, visit https://groups.google.com/d/optout.



--
William (http://wstein.org)

[Andrey Novoseltsev]

On Thursday, 9 June 2016 08:02:02 UTC-6, William Stein wrote:

When I click on his link to the code, I don't see any code!  Chrome + OS X.

I saw this happening - something is off with CodeMirror initialization but I haven't figured out what exactly yet. It is high priority and hopefully will be fixed soon (= in a couple of weekends).

On Thu, Jun 9, 2016 at 9:59 AM, kcrisman <kcri...@gmail.com> wrote:
> For anyone who might be interested in I/O problems on Sagecell, apparently
> someone has found a race condition of some kind:
> http://stackoverflow.com/questions/37692165/sagemathcell-valueerror-when-printing-string-sometimes

There were relatively frequent problems with this that were fixed upstream (I am using ipykernel master rather than Sage for the moment). I can't say right away if it is another similar problem upstream or something should be fixed in SageMathCell itself, but will look into it.

[Andrey Novoseltsev]

On Thursday, 9 June 2016 23:36:41 UTC-6, Andrey Novoseltsev wrote:

On Thursday, 9 June 2016 08:02:02 UTC-6, William Stein wrote:

When I click on his link to the code, I don't see any code!  Chrome + OS X.

I saw this happening - something is off with CodeMirror initialization but I haven't figured out what exactly yet. It is high priority and hopefully will be fixed soon (= in a couple of weekends).

I think I have finally fixed it (the problem was using CodeMirror before CSS loads).

On Thu, Jun 9, 2016 at 9:59 AM, kcrisman <kcri...@gmail.com> wrote:
> For anyone who might be interested in I/O problems on Sagecell, apparently
> someone has found a race condition of some kind:
> http://stackoverflow.com/questions/37692165/sagemathcell-valueerror-when-printing-string-sometimes

There were relatively frequent problems with this that were fixed upstream (I am using ipykernel master rather than Sage for the moment). I can't say right away if it is another similar problem upstream or something should be fixed in SageMathCell itself, but will look into it.

The original problem seems to be worse now - I more often get the error than not. Maybe due to the "proper" use of RequireJS which made something faster/more efficient and some race condition is easier to trigger now.

 

[Andrey Novoseltsev]

On Friday, 8 July 2016 23:48:58 UTC-6, Andrey Novoseltsev wrote:

On Thu, Jun 9, 2016 at 9:59 AM, kcrisman <kcri...@gmail.com> wrote:
> For anyone who might be interested in I/O problems on Sagecell, apparently
> someone has found a race condition of some kind:
> http://stackoverflow.com/questions/37692165/sagemathcell-valueerror-when-printing-string-sometimes

There were relatively frequent problems with this that were fixed upstream (I am using ipykernel master rather than Sage for the moment). I can't say right away if it is another similar problem upstream or something should be fixed in SageMathCell itself, but will look into it.

The original problem seems to be worse now - I more often get the error than not. Maybe due to the "proper" use of RequireJS which made something faster/more efficient and some race condition is easier to trigger now.

 

As it turned out my "fix" by using ipykernel master was not applied anymore as the master version was behind the latest release (in terms on version number, but not commits). Fixed by https://github.com/sagemath/sagecell/commit/9db1d42094c0f3a4509b613ac21a14964b7204e7