The following changes are made:
`jupyter-initialize-connection` -> `jupyter-comm-initialize`
`jupyter-connect-client` -> `jupyter-comm-add-handler`
`jupyter-disconnect-client` -> `jupyter-comm-remove-handler`
`jupyter-comm-client-loop` -> `jupyter-comm-handler-loop`
* README.org: Do it.
* jupyter-channel-ioloop-comm: Do it.
* jupyter-client.el: Do it.
* jupyter-comm-layer.el: Do it.
(jupyter-comm-layer): Rename `clients` slot to `handlers`. Update all uses.
* jupyter-kernel-process-manager.el: Do it.
* jupyter-repl.el: Do it.
* jupyter-server.el: Do it.
* jupyter-zmq-channel-comm.el: Do it.
* test/jupyter-server-test.el: Do it.
* test/jupyter-test.el: Do it.
That also happens to use a ZMQ function.
* jupyter-rest-api.el (jupyter-api-upload-large-file): Do it.
* jupyter-tramp.el: Remove mention of `jupyter-api-upload-large-file` in a
comment.
* README.org: Add notes on behavior when a kernel's language name has spaces
and uppercase characters.
* jupyter-base.el (jupyter-canonicalize-language-string): New function.
* jupyter-client.el (jupyter-kernel-info): Use it.
* jupyter-kernelspec.el (jupyter-available-kernelspecs): Use it.
* ob-jupyter.el (org-babel-jupyter-aliases-from-kernelspecs): Use it.
* test/jupyter-test.el (jupyter-canonicalize-language-string): New test.
`jupyter-wait-until-idle` can return immediately if an idle message has already
been received, but the function `jupyter-wait-until` will eventually raise an
error in that case. Avoid the error by checking
`jupyter-request-idle-received-p` before calling `jupyter-wait-until`.
* jupyter-client.el (jupyter-wait-until-idle): Do it.
* ob-jupyter.el (org-babel-execute:jupyter): Account for changes in
`jupyter-wait-until-idle`.
* jupyter-widget-client.el (jupyter-widgets-start-websocket-server)
(jupyter-widgets--initialize-client): New functions.
(jupyter-handle-comm-open): Use them.
* jupyter-repl.el (jupyter-repl--deactivate-interaction-buffers): New function.
(jupyter-repl-kill-buffer-query-function): Extract code that deactivates
`jupyter-repl-interaction-mode` in connected buffers to the new function.
(jupyter-repl-mode): Add the new function to `kill-buffer-hook`.
* jupyter-org-client.el (jupyter-drop-request): Reset
`org-babel-jupyter-current-src-block-params` if its value matches the dropped
request's `block-params`.
The described behavior for when `jupyter-repl-echo-eval-p` is `t` and the REPL buffer is not visible was not working, i.e., no pop-up buffer was being show.
Move the whitespace that was originally concatenated to the prefix variable `jupyter-eval-overlay-prefix` to the variable itself, so that it can be easily customized. For instance, in the case where one does not want any prefix, with the changes, it can be easily removed by setting the prefix variable to the empty string. Originally one would have to also modify the function `jupyter-eval-ov--propertize` to achieve the same result.
* jupyter-server.el (jupyter-server--kernel-list-entries):
Add missing jupyter-time text property to activity string.
(jupyter-server-list-kernels): Set `tabulated-list-sort-key`.
* jupyter-base.el: Don't add `jupyter-server-mode--unset-client-soon` to
`server-switch-hook` at toplevel
(jupyter-server-mode-set-client): Add `jupyter-server-mode--unset-client-soon`
to `server-switch-hook` if necessary.
* jupyter-client.el: (jupyter--run-callbacks): Use `when-let`.
(jupyter--set-callback): Remove.
(jupyter--add-callback): Use `add-function`.
* jupyter-ioloop.el (jupyter-ioloop-add-arg-type): Use `setf` along with `alist-get`.
(jupyter-ioloop--replace-args): Use `pcase`.
(jupyter-ioloop--event-dispatcher): Extract out event cases into variable `user-events`.
(jupyter-ioloop--body): New function.
(jupyter-ioloop--function): Use it.
* jupyter-kernel-process-manager.el
(jupyter--kernel-died-process-sentinel)
(jupyter-stop-channels): Use `when-let`.
* jupyter-org-client.el (jupyter-org-interaction-mode): Use `cl-callf2?`.
* jupyter-server.el: Add TODO.
* jupyter-tramp.el: Remove unused function declaration.
Since we are using keyword to represent channels on the client level, use a
plist of channels on the comm layer level instead of channel slots. This allows
us to remove `jupyter-comm-channel` and simplify the relevant code somewhat.
Only do so when asking the user for a URL and no server exists at the URL
provided. See #183.
* jupyter-server.el (jupyter-server-non-existent): New error.
(jupyter-current-server): Signal `jupyter-server-non-existent` when asking the
user for a server URL and a server doesn't exist at URL.
TRAMP sets its own value of `signal-hook-function` which interferes with
`signal` data when handling errors in `condition-case` so unbind
`signal-hook-function` until we have a chance to look at `signal` data.
* jupyter-server.el (jupyter-kernel-alive-p): Use `nth`
* jupyter-tramp.el (jupyter-tramp--get-directory-or-file-model)
(jupyter-tramp-delete-directory): Unbind `signal-hook-function`.
Verify that the `major-mode` of the `buffer-base-buffer` is derived from
`org-mode` if the `current-buffer` happens to be an indirect one. This should
allow `emacs-jupyter` to work better with `polymode`. See #171.
When the `:display` header argument is used and a particular message type
doesn't return a mime bundle with one of the requested mime types, include the
message type in the warning that is displayed.
This should clarify the situation where an `:execute-result` and
`:display-data` message both return data to display, but one of them does not
have one of the mimetypes requested in the `:display` header argument.
Closes#144.
This is needed because `org-mode` merges many sources of source block params so
its not enough to just re-compute the parameters in `jupyter-generate-request`
using `org-babel-get-src-block-info` or `org-babel-log-get-info`.
