-
-
Notifications
You must be signed in to change notification settings - Fork 97
Conversation
Uploads of html and rst worksheets were also affected by the bug. The new (optimal?) solution also fix this cases. |
A simple sample rst worksheet for testing upload. =============
Test Notebook
=============
Let's do some computations::
sage: a = '<';a
'<'
::
sage: ratpoly.<t> = PolynomialRing(QQ)
|
I plan to look at this soon (today?) but have a concern. This link and this link seem to suggest that we might run into trouble with utf-8 inputs. Like |
E.g. http://lucumr.pocoo.org/2013/7/2/the-updated-guide-to-unicode/ seems to explain Flask's take on this, anyway, though mostly from the Py3 perspective. Hopefully I'm just making much about nothing. Though you do say
So how can we ensure this? Will |
The class In fact, if you remove all the unicode thing, the example that you propose will fail. In my Notebook refactor/rewrite, (currently in progress), I plan to convert all the relevant parts of the notebook to unicode_literals as previous step to the python 3 migration. |
I don't even get anything correct when I do this: =============
Test Notebook
=============
Let's do some computations::
sage: a = 5 |
So maybe I am doing something wrong. But I do have the fix for the other stuff, so I have the right branch, I guess. |
So I don't know why mine isn't working. Everything else seems to work, I guess? |
It is a very disturbing problem. Your output points to a problem with the rst->html conversion. |
I'll note that HTML seems to upload correctly. |
Right, but you are not having the problem. I have 6.10.beta6 but the only difference relevant to this should be in sagenb, right? |
Your output shows crude rst. It seems that the |
Yes, that is exactly what is happening, it's just this when I click "Edit" after uploading.
|
And yes, docutils is there.
|
I've directly tested |
I can confirm I have this problem whether or not I use your branch. |
Probably, the problem is not related with the notebook, but with the sage packages. |
Yes, most likely. This output seems okay, I guess? |
I obtain the same ouput as you, but try
|
This is long...
|
By the way, I did |
It seems to be fine. |
Huh. I have to go soon but may try to track this down with a separate Sage install, maybe on SMC or something. Not sure why this isn't working. |
Doing this with Sage 6.9 yields the same result for me. Not sure what the problem is. |
No idea. I can't reproduce the problem. |
I'm just going to rebuild 6.9 from scratch and then do this all again. So annoying. I'm sure it's fine, I've been having other weird troubles as well. |
Okay, I see what happened - indeed something local. I'm embarrassed to even say what it was - wrong file extension! I specifically made my Mac change it to |
So I will merge this. By the way, are you having any trouble evaluating cells in the live documentation e.g. http://localhost:8080/doc/live/tutorial/tour_rings.html ? I tried removing my cache and who knows what else but nothing seems to work in any browser. @jdemeyer says here that it works so I'm hoping that I'm just imagining things. |
Live docs problems with autoescaping
It works in my system (with the current master branch after merging). |
See #357 for followup. |
As @kcrisman pointed out in #324, the new
autoescape
default for jinja2 templates breaks some live docs. Here is a possible solution to be tested. This solution does not break the fix for the hidden bugs found in #324 during the discussion of the problem.