1429 commits

Author	SHA1	Message	Date
John Miller	f41cfbc410	Merge pull request #487 from dickmao/markdown-no-completion ... Don't attempt completion in markdown cells	2019-03-15 23:14:39 -05:00
John Miller	f6bab5c077	Merge pull request #488 from dickmao/text-plain ... text/plain over text/html	2019-03-15 10:46:03 -05:00
dickmao	22b4e58e56	preserve the no-op for now	2019-03-15 00:50:13 -04:00
dickmao	041a66cf28	need to reorder interactive arguments after function signature change	2019-03-15 00:14:22 -04:00
dickmao	6698b4f563	Don't attempt completion in markdown cells	2019-03-11 22:37:13 -04:00
dickmao	7985387178	:text/plain is also :text ... Notebooks with :text/plain and :text/html outputs opt for :text/html which looks bad in emacs. There is already code to prefer :text over :text/html, so it should prefer :text/plain as well.	2019-03-11 00:02:08 -04:00
John Miller	f2e9193da6	Merge pull request #481 from dickmao/self-diagnose ... Move self-diagnosing information	2019-03-09 20:28:30 -06:00
dickmao	67e82d2f94	Move self-diagnosing information ... Move self-diagnosing information from CONTRIBUTING.md to README.in.rst	2019-03-09 12:52:03 -05:00
dickmao	188b2c0dec	Toggle output (C-c C-e) should also invalidate footer ... Attempt fixing longstanding annoyance whereby `C-c C-e` leaves a peachpuff or pink newline when toggling output with nonempty stderr.	2019-03-08 18:33:15 -05:00
John Miller	bf1bcf12cf	ein-company: Hotfix for #478 ... Do no run ein company backend unless we can find a kernel to do completion against.	2019-03-05 21:50:51 -06:00
John Miller	b790c749d4	Merge pull request #477 from millejoh/connected-buffer-company ... ein-company: Implement company completion in connected buffers.	2019-02-28 14:08:49 -06:00
John Miller	80306130cf	Merge pull request #475 from dickmao/ob-ein-reconnect-kernel ... reconnect kernel if necessary under org-ctrl-c-ctrl-c	2019-02-26 19:51:14 -06:00
dickmao	7ba5a0d5d3	Be less opaque about unfound jupyter executable ... Fixes #473	2019-02-26 18:04:50 -05:00
dickmao	dd01c4578c	reconnect kernel if necessary under org-ctrl-c-ctrl-c ... when `C-c C-c`, reconnect to kernel if it disconnected	2019-02-26 17:58:42 -05:00
John Miller	94655e7738	ein-company: Implement company completion in connected buffers.	2019-02-26 08:55:17 -06:00
John Miller	b9f8854949	Merge pull request #470 from dickmao/org-bug-468 ... Add ob-ein tests	2019-02-22 09:28:24 -06:00
dickmao	4f468ee5f7	Singleton ... We can't be certain when the shared output cell can be safely turned over to the next org block. I have witnessed `display_data` updates being received *after* both the `execute_result` and `execute_reply` messages. As a HACK, we `run-at-time 2` seconds the turning over of the cell to accommodate any eleventh hour straggler display_data's.	2019-02-21 23:14:11 -05:00
dickmao	0f7f8b14cd	Test C-c '	2019-02-20 11:50:46 -05:00
Felipe Lema	9168b2cf8e	sorted items	2019-02-20 12:28:27 -03:00
Felipe Lema	9147159ece	Edit julia code cells. Fixes #471 ... BTW, julia-mode is a dependency for ESS, so wherever you have R, you have Julia (which is the way I implemented this)	2019-02-20 12:28:06 -03:00
dickmao	2f19904fe7	travis buttressing	2019-02-19 15:29:18 -05:00
dickmao	ed6482096e	Add ob-ein tests ... Also, speed up login and other cleanups	2019-02-19 10:06:15 -05:00
John Miller	50b9a8d134	ob-ein: Better output for babel. ... Greg Sexton clearly knows his way around org better than I. For async execution make sure we insert results "the org way."	2019-02-17 21:10:08 -06:00
John Miller	111826958f	Process all output types on json export. ... We were only selectively fixing the mime type for execute_result outputs, now we treat those the same as we do display outputs. Also fixed image/svg+xml MIME type wherever I could find it.	2019-02-09 13:36:37 -06:00
John Miller	b276d6ad0a	Fix mime type for svg images. ... Update, part 1, for #467.	2019-02-09 12:59:55 -06:00
leizhe	46e3fc6247	ignore nonascii completion	2019-02-02 22:32:27 +09:00
dickmao	34b30b5e60	User llcc reports read-only bug in fresh input cell under mingw32	2019-01-23 08:05:09 -05:00
John Miller	297b216a42	Merge pull request #459 from dickmao/turn-off-debug ... turn off --debug in jupyter	2019-01-22 13:13:29 -06:00
dickmao	428f0a1493	turn off --debug in jupyter	2019-01-22 08:15:36 -05:00
dickmao	fbda27ba7f	Surface curl errors ... In light of #425 and #456, provide more guidance on curl errors. Also, clean up eldoc stub. Leveraging elpy's eldoc perfection seems the only sensible choice, however.	2019-01-21 13:42:35 -05:00
John Miller	bcbefbe0b9	ein-notebook: Workaround for issue #454 ... EIN can sometimes lose the reference to the notebook buffer when trying to save "huge" notebooks. This is a temporary (hah!) workaround until the problem is better understood.	2019-01-21 07:54:44 -06:00
Sam Steingold	5ee7ae7d6b	last patch was too aggressive	2019-01-16 18:23:39 -05:00
Sam Steingold	48fa153435	Use standard functionality instead of ad hoc functions ... (ein:filter identity ...) ---> (delete nil ...) ein:filter --> seq-filter hash-vals --> hash-table-values hash-keys --> hash-table-keys	2019-01-15 11:26:09 -05:00
John Miller	37072c3553	Merge pull request #453 from millejoh/dynamic-javascript ... Dynamic javascript	2019-01-15 09:57:19 -06:00
John Miller	2f96fc33ac	ein-skewer: Handle html in the browser. ... If dynamic-javascript is enabled and we see HTML in the display data after executing a cell, then we create a web page for it in the simple-httpd server.	2019-01-14 18:23:22 -06:00
John Miller	4c33a8b057	ein-utils: Make gc-cons-threshold adjustment optional. ... EIN doesn't seem to always return gc-cons-threshold to its original value; this makes for performance problems, particularly on Windows builds, so this behavior is now optional and user-configurable.	2019-01-14 12:56:32 -06:00
John Miller	793d5fa192	ein-skewer: Async javascript eval, better output appending. ... No longer flag if skewer is running, instead check running status of httpd server. EIN also is back to asynchronously executing javascript code and does a much better job of finding where to put the output.	2019-01-14 12:54:16 -06:00
dickmao	9825aa39c0	if C-/ is mapped to anything other than native undo, disable undoing	2019-01-11 08:32:51 -05:00
dickmao	4c60dec927	Disable undo when failure diagnostic triggered	2019-01-07 08:08:11 -05:00
dickmao	126bd6afac	kill-buffer-hook getting cloned by org!	2019-01-02 19:54:46 -05:00
dickmao	1a32c5f832	Less wonky toggle latex fragment ... Using the raw `org-toggle-latex-fragment` function creates some havoc with the overlays, resulting in nonfunctional notebooks. Add some custom logic to improve behaviour.	2019-01-01 21:16:34 -05:00
John Miller	624e9549ef	Merge pull request #444 from dickmao/undo-default-true ... Enable undo by default	2018-12-29 17:15:06 -06:00
John Miller	ed154e91e0	Merge pull request #443 from dickmao/bug-issue-436 ... Respect cursor_start and cursor_end in matches	2018-12-29 17:05:47 -06:00
dickmao	58de1294f8	default undo to true	2018-12-28 22:31:23 -05:00
dickmao	e5fdb91b8d	itertools.accumulate not in python2 ... also fix an install error	2018-12-28 21:45:12 -05:00
dickmao	129afa5fb4	Respect cursor_start and cursor_end in matches ... The :cursor_start and :cursor_end fields tell us whether we need to prepend the prefix ourselves. Fixes #436	2018-12-28 21:14:11 -05:00
John Miller	4c84801be0	Revert "Merge pull request #441 from dickmao/bug-issue-436" ... This reverts commit 1b5d23e7db, reversing changes made to 57cf533466.	2018-12-26 15:26:18 -06:00
John Miller	1b5d23e7db	Merge pull request #441 from dickmao/bug-issue-436 ... [WIP] Respect cursor_start and cursor_end in company	2018-12-26 15:10:12 -06:00
dickmao	fb15e82d1a	WIP Respect cursor_start and cursor_end in matches ... With respect to ein-company, Py2 kernels prepend prefixes, e.g., np.array. Py3 kernels do not, e.g., array. The :cursor_start and :cursor_end fields tell us whether we need to prepend the prefix ourselves. Fixes #436	2018-12-26 03:17:51 -05:00
dickmao	387ceb0046	complete-on-dot is not optional if you agree to auto-completion	2018-12-25 19:12:26 -05:00