Widget to customize number of seeds to generate

[rhinopotamus]

This commit makes a new IntText widget on the dashboard so you can customize the number of seeds to generate.
The value of this widget passes into output.generate_exercises(), which now has a new signature including an amount to generate.
generate_exercises() feeds this value to the sage command in wrapper/__init__.py, and the indexing of sys.argv in wrapper.sage is modified accordingly.

(Please feel free to ignore the version number change in checkit/__init__.py -- I put that in there so I could make sure I was using the local dev copy of dashboard instead of the global one.)

[StevenClontz]

Looks good at a glance, modulo my notes. I'd be happy to incorporate this in a new release.

[StevenClontz]

This seems about ready. (I don't have a test suite like I should, but if it's working for you...)

If you want to be super cool, since you have a working dev environment and I actually don't right now, I can set you up to push a release candidate to PyPI, and if that works successfully I wouldn't mind letting you make the full release.

[StevenClontz]

amount=10 will be ignored by the sage script (see next comment)

[rhinopotamus]

Ok, yep, this works. (I was worried that the signature wouldn't match but it's ok because we handle a default amount in wrapper/init.py.)

[StevenClontz]

here we ignore the amount if we're set to "preview". So I think it's better to simply omit the amount=10 earlier as it's misleading (it will use the default 1000 instead, but that's happily ignored here).

OR previews could support variable #s of seeds (maybe you really want to preview just 1 seed, or 100). But that's more work for minimal benefit?

[rhinopotamus]

FYI I've been feeling kinda poorly since Saturday but I'll do more stuff when I can.

…

On Fri, Jul 21, 2023, 11:16 AM Steven Clontz ***@***.***> wrote: ***@***.**** requested changes on this pull request. This seems about ready. (I don't have a test suite like I should, but if it's working for you...) If you want to be super cool, since you have a working dev environment and I actually don't right now, I can set you up to push a release candidate to PyPI, and if that works successfully I wouldn't mind letting you make the full release. ------------------------------ In dashboard/checkit/outcome.py <#79 (comment)>: > @@ -50,7 +50,8 @@ def to_dict(self,regenerate=False): def preview_exercises(self): preview_json = os.path.join(self.build_path(),"preview.json") - sage(self,preview_json,preview=True,images=True) + # hardcode: previews generate 10 seeds + sage(self,preview_json,amount=10,preview=True,images=True) amount=10 will be ignored by the sage script (see next comment) ------------------------------ In dashboard/checkit/wrapper/wrapper.sage <#79 (comment)>: > @@ -206,15 +206,15 @@ if len(sys.argv) >= 4: if sys.argv[3].lower() == "preview": here we ignore the amount if we're set to "preview". So I think it's better to simply omit the amount=10 earlier as it's misleading (it will use the default 1000 instead, but that's happily ignored here). OR previews could support variable #s of seeds (maybe you really want to preview just 1 seed, or 100). But that's more work for minimal benefit? — Reply to this email directly, view it on GitHub <#79 (review)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACDYL6WGJBDBZDX6UJ5U3YTXRK2PFANCNFSM6AAAAAA2TBK64I> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[StevenClontz]

No rush - I'm happy to help you make a 0.2.7 release with this feature if that's more convenient for you, or do it myself once someone else makes the same request.

[rhinopotamus]

I am more or less back. I wonder if before we push out a 0.2.7 official release, we may also want to attempt to get latex markdown preview working again. Do you know where / when it broke?

…

On Wed, Jul 26, 2023 at 9:29 AM Steven Clontz ***@***.***> wrote: No rush - I'm happy to help you make a 0.2.7 release with this feature if that's more convenient for you, or do it myself once someone else makes the same request. — Reply to this email directly, view it on GitHub <#79 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACDYL6WBOJQSWPDQVGY2HNDXSEZUHANCNFSM6AAAAAA2TBK64I> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[StevenClontz]

Ah. That's caused by using non-standard Jupyter that CoCalc supported, but probably not JupyterLab et al. Let me see something...

[StevenClontz]

My hunch is that changing the \(, \), \[, \] at https://github.com/StevenClontz/checkit/blob/main/dashboard/checkit/static/html.xsl#L96 to $ and $$ as appropriate will fix the Jupyter notebook stuff. What's unclear offhand is whether that might break the CheckIt viewer app. So you could give that a try and report back.

[rhinopotamus]

Unfortunately that change in html.xsl is no immediate dice.
I wonder if jupyter "knows" that there is html (and thus mathjax) inside that output cell, or if we have to tell it so in some fancy way.