Login and notebooklist requests have a tendency to fail with the curl request
backend when there is no cookie jar file. Request will create the cookiejar, but
apparently with the asynchronous nature of ein this happens too slowly causing
content REST queries to fail. My tactic here is to repeat the call once, and
only once when a failure is detected. It is a difficult problem to reliably
create so I'm not sure if this is a good fix or not. Time will tell.
When the customizable variable `ein:enable-keepalive` is non-nil EIN will
automatically call `ein:notebooklist-enable-keepalive` when calling
`ein:notebooklist-open`. Keepalive has also been tweaked to make sure it does
not create multiple timers.
When calling `ein:jupyter-server-stop` ask to user if they are sure and also
give them the option to save any unsaved notebooks before killing the server.
Before closing the server also close any open notebook buffers.
For the moment we only support current version of jupyter (i.e. 4.3.1 or
greater).
The special commands are `ein:jupyter-server-start` to start a notebook server
and `ein:jupyter-server-stop` to, clearly, stop a notebook server.
On starting a server EIN will try to determine the url and token for accessing
the notebook server, login and open the notebook list automatically.
Changed description of manual installation in "usage" section.
Made it more explicit that you only need to manually put files in
the load path when not using MELPA.
Initial steps integrating magit's sections into ein. Also an attempt to make
notebooklist buffers more "stable" by adding a slight pause during session
queries.
Display value of object's __repr__, link to source when available, and
documentation.
Adding command and keybinding (C-ci) to notebook buffer keymap.
Get rid of unnecessary api check when rendering notebooklist buffer.
Mistyped some accessors for worksheet slot accessors. Correct
`*running-ipython-version*` hash to accept strings (i.e. URL's) as well as just
port numbers.
Automatically calling ein:notebooklist-login when a content API calls is a BAD
IDEA. Much better to warn and suggest, and if debugging is enabled throw the
error so hopefully a backtrace can be generated.
Taking some initial steps away from eieio and towards cl-generic.
The function `ein:notebooklist-enable-keepalive` and
`ein:notebooklist-disable-keepalive`, respectively enable a timer that
ocassionally (on the order of hours) sends a query to the notebook server. The
goal is to keep cookies and sessions from expiring in long running notebooks.
We'll see if it works.
Some believe that alerting the user in the *Messages* buffer when there is an
error trying to run the before-save-hooks is overly verbose and a nuisance. EIN
should still warn what is happening, I think, but instead of `warn`ing the
message no goes to the *ein:log-all* buffer.
Saving in a cell editing buffer now actually causes the cell in the worksheet to
be dynamically updated (oh how we have waited for this moment).
Try to get cell undo actions to be less likely to trigger an error.
Post Ipython 2.0 there are no longer separate websocket connections for iopub
and shell, they now share the same websocket connect. This is a simple
convenience function for finding the websocket log buffer for the new-style
connection.
ob-ein will alert the user if they are generating an image in the ob-ein source
block and no :image header has been supplied. Added some documentation for good
measure, though people will probably still get tripped up by this.
Return the results of execution as a single string instead of a list of strings.
This allows some more flexibility when dealing with :results parameters and
seems to fix a bug with inserting matplotlib images. Hint: try using `:results
value raw`.
Using alabaster for the documentation theme. Building documentation now relies
on having cask present, which does a better job of tracking dependencies than
the old way.
Content api tries to handle redirects, pointing new content towards the
redirected url instead of the one supplied by the user. Login with redirects not
working, though.
Other fixes:
- Fixed bug in ein-ac that was generating errors outside of ein buffers.
- New function `ein:pyools-set-figure-size` as shortcut for modifying
matplotlib figsize rcParam.
- ob-ein will output tracebacks when ein source block generates an error.
Finally do the right thing when clear_output is received. Closes issue #24. Also
fix a nasty bug that might occurr when trying to open a notebook with no
associated kernelspec.
Prettier kernelspec names in notebooklist buffers. Only load pytools for python
kernels, which hopefully will help with connecting to non-python kernels.
tkf/emacs-request#60 was recently fixed, so no longer need to defadvice
`request--netscape-cookie-parse. Changed names of some obsolete variables. Also
trying to be a bit more clever in error handling by judicious use of
condition-case.
