Open BBurkhardt opened 3 years ago
hm. I am not able to reproduce this on commit bded0e5 or on master.
This is similar to my note #386 in the scimax section. I did some experiments and tried to checkout older versions of ox-ipynb. github is still a black box to me, but the following should have worked I hope
git checkout 8d62d420f8cd0ba6e6696a99de63a5ad2f29024b
this particular commit is from Oct 2018, and I know for certain that tables worked until at least August this year. However, I get the same result.
If I used the checkout command correctly, this implies that the problem with the tables is elsewhere. Is this a possibility?
It might not be enough to simply do the checkout. That will update the elisp files, but not reload them in emacs. What I would do is
then try the export.
The order of those may be important. ox-ipynb is a submodule in scimax, and if you open scimax after you do a git checkout, it may revert it back to the committed version before loading it, which will make it look like nothing changed.
oh ok. that explains it. Starting scimax does indeed update the ox-ipynb directory. If I do it your way, the export succeeds as expected.
On Wed, Nov 4, 2020 at 10:17 AM John Kitchin notifications@github.com wrote:
It might not be enough to simply do the checkout. That will update the elisp files, but not reload them in emacs. What I would do is
- open emacs
- In a shell do these: a. cd /path/to/ox-ipynb b. git checkout 8d62d42 https://github.com/jkitchin/ox-ipynb/commit/8d62d420f8cd0ba6e6696a99de63a5ad2f29024b
- Back in emacs, open ox-ipynb.el and M-x eval-buffer
then try the export.
The order of those may be important. ox-ipynb is a submodule in scimax, and if you open scimax after you do a git checkout, it may revert it back to the committed version before loading it, which will make it look like nothing changed.
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/jkitchin/ox-ipynb/issues/36#issuecomment-721793019, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABWSVAR7NR33SWBFCYDATXTSOFWB3ANCNFSM4TCOLOYA .
-- Ulrich G. Wortmann https://www.es.utoronto.ca/people/faculty/wortmann-ulrich/ Dept. of Earth Sciences Fax : 416 978 3938 University of Toronto Phone: 416 978 7084 22 Ursula Franklin Street, Toronto, ON, Canada M5S 3B1
This still does not work on master and appears to be due to commit https://github.com/jkitchin/ox-ipynb/commit/c9834ae695c782f5bc0f663aad5c15052174eb84
I'm able to "fix" it by replacing (if (not (= 0 nrules))
with (if (> nrules 1)
because for some reason this part is run twice and then the second time nrules
is not 0 anymore and the separator isn't added, but I guess there is a more correct way to do it?
Ok, I think I see what was happening, and I think the fix above fixes it. It might not solve all problems, and is limited to the simplest markdown for tables. anything fancier probably requires an html table. I will save that for another day. Let me know if this works for you.
Yes, this works again for simple tables with one horizontal rule. Thank you!
Dear John,
I ran into the following unexpected behavior while exporting some Org-Tables.
Exporting the following table:
| column-a | column-b | |----------+----------| | test1 | test2 | | test3 | test4 |
produces the following MarkDown Code:
| column-a|column-b| | test1|test2| | test3|test4|
Since the table head separator is missing, the table is not rendered properly in Jupyter Notebook.