* Store secrets in the notebook
* Automatically grants access to secrets
* Put notebook secrets on session
* Shows secrets as a list
* Grant access message box
* Grant access confirmation on select
* Applying suggestions - grant access confirmation
* Handles unavailable secrets
* Toggle secrets
* Session only secrets
* Sync secrets
* Fix delete runtime secret
* Clean up
* Component helpers
* Does not store secrets in notebooks
* Store Livebook secrets
* Fix sync secrets
* Tests for secrets
* Doesn't refetch livebook_secrets
* Removes unused function
* More test for secrets
* Livebook secrets as maps
* Fix secret tests
* Applying suggestions
* All secrets as a map
* Shows grant access for missing secret errors
* Unifies grant access
* Fix set_runtime_secrets
* Applying suggestions
* Updates sync secrets tests
* Fix active secret
* Unifies prefill secret name
* Define .input_wrapper and .hex_color_input components
* Define Tailwind variants from Phoenix v1.6 to help styling
* Make sure single-action modals redirect on save
* Add pill to session secrets in the "Secrets" session sidebar
* Fix alignments of "Users" in the session sidebar
* Add secrets to the sidebar
* Get and put secrets (naive)
* Secrets list - shortcut
* Validates secret name
* Livebook secrets as environment variables
* Restores secrets when necessary
* Minor adjustments
* Implements add_system_envs for NoopRuntime
* Moves secrets to session data
* Granular operations for secrets
* Minor adjustments
* Applying suggestions
* Test for add secrets
* Type for secret
* Applying suggestions
* Prevents duplicates
* Refactor to live_render
* Update lib/livebook/session/data.ex
Co-authored-by: Jonatan Kłosko <jonatanklosko@gmail.com>
* Sort secrets
* Applying suggestions
Co-authored-by: Jonatan Kłosko <jonatanklosko@gmail.com>
macOS removes files from the temporary directory but
keeps the directory. This would cause LiveView to
misbehave as it would assume all assets are checked out.
* Synchronize smart cell source before evaluation
* Sync JS view only if present
* Synchronize with smart cell only when started
* Update references to kino packages
* Update the Smart cells notebook
* Rename add dependnecy to add package
* Update changelog
* Update references to kino
* Update changelog
* Make the Chart cell section branching
* Apply suggestions from code review
Co-authored-by: José Valim <jose.valim@dashbit.co>
* Capitalization
* Add link to the Iris dataset
* Pass keys to Process.info/2
Co-authored-by: José Valim <jose.valim@dashbit.co>
* Add windows version of ci
Config autocrlf to input on Windows
start epmd in background
* Update .github/workflows/test.yaml
* Update .github/workflows/test.yaml
* Update .github/workflows/test.yaml
* Fix tests
* Fix ownership of cached files
* Fix tests
* Increase timeouts
* Run tests on Windows only on main
Co-authored-by: Jonatan Kłosko <jonatanklosko@gmail.com>
* Fix horizontal scrollbar on smaller screens
* Apply navigation shortcuts without an additional roundtrip
* Shorten the data element selector
* Fix URL in changelog
* Return reference from handle_intellisense
* Automate searching and adding new dependencies
* Restructure runtime management
* Leave search restrictions up to the runtime
* Support predefined dependencies in Embedded runtime dependency search
* Use the embedded runtime for dependency search test
* Fix typo
* Use module for loading embedded runtime dependencies
* Improve naming
* List predefined smart cells and automatically add their dependencies
* Generalize dependency insertion to multiple entries
* Add runtime setup modal
* Don't evaluate setup when restarting the runtime fails
* Automatically add vega_lite for chart builder
* Improve confirmation modal actions when the action is not destructive
* Keep binding in reversed evaluation order
* Treat rebound names as new
* Add test for the default order
* Optimise binding reorder
* Update lib/livebook/runtime/evaluator.ex
Co-authored-by: José Valim <jose.valim@dashbit.co>
* Apply review comments
Co-authored-by: José Valim <jose.valim@dashbit.co>
* Add support for scanning evaluation context from smart cells
* Rename prev_locator to base_locator to clarify the meaning
* Adjustments
* Add data tests
* Adjustments
* Test smart cell intearaction with RuntimeServer
* Serialize binding scanning and leave sending to the user
* Monitor the scanning process
* Write to file out-of-band
It won't block the immediate write but it may block
a subsequent write. A sync function was added for
testing purposes.
* Update lib/livebook/storage/ets.ex
Co-authored-by: Jonatan Kłosko <jonatanklosko@gmail.com>
Co-authored-by: Jonatan Kłosko <jonatanklosko@gmail.com>
* Group all runtime messages under a common prefix
* Move output specification ot the Runtime protocol
* Update doc comments
* Livebook.Evaluator -> Livebook.Runtime.Evaluator
* Livebook.ContentLoader -> Livebook.Notebook.ContentLoader
* Livebook.UniqueTask -> Livebook.Utils.UniqueTask
* Add import/export delegates to Livebook.LiveMarkdown
* check if section has valid parent + tests
* formatting
* apply review feedback
* fix the forgotten test
* apply feedback
* apply feedback
* produce a warning if the section points to itself + tests
* Apply suggestions from code review
Co-authored-by: Jonatan Kłosko <jonatanklosko@gmail.com>
* Support downloading source on a per session basis
* Update session_list_component.ex
* Add test
* Update lib/livebook_web/live/home_live/session_list_component.ex
* Update test/livebook_web/live/home_live_test.exs
Co-authored-by: José Valim <jose.valim@gmail.com>
Co-authored-by: Jonatan Kłosko <jonatanklosko@gmail.com>
* Export file with title or file name
* Export with title or file name
* Add Session.file_name_for_download/1
* Compute the name without calling the server
Co-authored-by: Jonatan Kłosko <jonatanklosko@gmail.com>
* Skip XML.decode!(body) if body is empty
Google Cloud Storage has almost S3 compatible XML API,
but some responses have `content-type: text/html; charset=UTF-8`
and empty body. This change prevents XML decode error and
enables work fine with Google Cloud Storage.
* Handle Google Cloud Storage XML API response
* Rename S3.encode to S3.decode
* Improve a test case description
Co-authored-by: Jonatan Kłosko <jonatanklosko@gmail.com>
* Make S3.decode/1 let any content type pass
* mix format
Co-authored-by: Jonatan Kłosko <jonatanklosko@gmail.com>
1. Allow multiple sessions to close or disconnect at once,
and then request the system resources to be updated
2. Make sure that closing a session happens synchronously,
otherwise a race condition could still list the closed
session in a dashboard
3. Ensure close and disconnect can happen more than once
(for example if two users click the same button at the
same time)
* Minimize race condition in the frame update test
* Use defmacrop for building intellisense context
* Remove unnecessary cell view computation
* Fix nested assets resolution
* Fix typing errors
* Add missing async attribute to test suites
* Improve rendering synchronization
* Up
* Initial implementation to close multiple sessions
* Sessions: bulk actions with components
* Rename Disconnect sessions to Disconnect runtime
* Select all and disabled when nothing is selected
* Styled checkbox
* Renames toggle events
* Warning about not persisted notebooks
* Adds disconnect runtime option for a single session
* Edit sessions on right
* Fix: typos and plural
* Minor adjustments
* Removes the loop for rendering the menu
* Menus with fixed width
* Minor adjustments
* Pluralize as global helper
* Bulk actions form on client side
* Track bulk actions buttons state
* Fix: home live tests
* Doctests for pluralize
* Fix: bulk actions buttons losing state on session update
* Fix: format
* Minor adjustment on toggle_edit
* Review-based adjustments
* Reset the Edit state after single-session actions
* Minor adjustments
* Fixes bulk action events
* Submit the bulk action form directly
* Tests for bulk actions
* Indentation
* Update lib/livebook_web/live/home_live/close_session_component.ex
Co-authored-by: José Valim <jose.valim@gmail.com>
Co-authored-by: José Valim <jose.valim@gmail.com>
* Address race condition on cancel timer
* Include memory measurement as part of evaluation metadata
* Move periodic resource computation to a single process
* Have a explicit call out for total memory
* Utils to fetch and format system and node memory usage
* Order by memory
* Track memory on session
* Show memory usage on runtime sidebar
* Shows memory usage percentage on home
* Layout adjustments
* Sidebar design adjustments to match Figma
* Home design adjustments to show the memory information
* Move memory calculations to utils
* Shows disconnected notebooks as consuming 0mb on home
* Simplifies the data structure of memory usage
* Node memory tracker on runtime
* Clean up
* Renames node memory to runtime memory
* Standardizes the data structure of memory usage
* Sends evaluation_finished to the runtime to update the memory usage after an evaluation
* Fix: The evalutor does not notify when there is no notify_to option
* Adds a test with the notify_to option to the evaluator
* Documents the notify_to option
* Minor fixes on runtime and runtime_server
* Minor fixes on sessions
* Minor adjustments
* Updates docs and specs on Utils
* Minor adjustments on session_live
* Fix total memory used by sessions on home
* Put duplicated functions on helpers
* Better filter by memory
* Fix the tooltip text for memory information on sidebar
* Minor alignment adjustment on home
* Restructure frame output
* Reboot JSComponent when ref changes
* Prune outputs from LV, handle frame and stdout updates via component messaging
* Keep ANSI modifiers as keyword list
* Don't re-render pruned outputs
* Improve empty frame styling
* Use more specific id
* Add clarification on trailing CR
* Handle output group borders with CSS
* Keep only relevant outputs in memory
* Adjust borders
* Update lib/livebook/live_markdown/import.ex
Co-authored-by: José Valim <jose.valim@dashbit.co>
* Apply review comments
Co-authored-by: José Valim <jose.valim@dashbit.co>
* Show deprecated and since doc metadata on mouse over
* Make deprecated indication stronger
Co-authored-by: Jonatan Kłosko <jonatanklosko@gmail.com>
Co-authored-by: Jonatan Kłosko <jonatanklosko@gmail.com>
* Introduce broadcast communication from runtime
* Return broadcast target from group leader
* Increase timeout
* Make the channel test async
* Decouple base topic and ref
* override tab name when notebook is renamed
* set page_title in after operation function
* add custom page title to all live pages
* Revert "override tab name when notebook is renamed"
This reverts commit 90303e08a8.
* add test for update
* Add completion for struct keys
Largely ported from `IEx.Autocomplete`.
* Add test for __exception__ field in struct
* Fix exception test
* Fix exceptions assetion
* Create `:in_struct_field` identifier
Along with a refactor
* Fix typespecs for `:map_field`
* Address feedback
* Update lib/livebook/intellisense/identifier_matcher.ex
Co-authored-by: José Valim <jose.valim@gmail.com>
* Use markdown snippet for both docs
* Fix tests
Co-authored-by: José Valim <jose.valim@gmail.com>
The currently suggested vim modeline formatting string to allow .livemd
files to render as markdown on GitHub works, but confuses vim itself.
```
<!-- vim: syntax=markdown -->
```
When opening a .livemd file with the suggested modeline in vim it errors
when it tries to parse `-->` as a modeline option.
e.g.
```
"2021/day2.livemd" 81L, 1616B
Error detected while processing modelines:
line 1:
E518: Unknown option: -->
```
The fix is to replace the suggested vim modeline
```
<!-- vim: syntax=markdown -->
```
With a modeline declaration format that indicates when the modeline
options are complete, allowing vim to ignore the rest of the line.
```
<!-- vim: set syntax=markdown: -->
```
Reference
- vim issue: https://github.com/vim/vim/issues/1648
- especially this comment: https://github.com/vim/vim/issues/1648#issuecomment-296386153Fixes#773
* Make cell status italic when content changed
* Add Ctrl+Shift+Enter for evaluating all cells
* Improve the behaviour of evaluating all cells
* Fix typo
* Add keywords to completion
* Fix signature request caching for call without parentheses
* Don't insert parentheses for def* macros
* Don't trigger missing runtime message when auto completion is enabled
* Don't insert parentheses for keyword macros
* Improve completion of env macros
* Apply review comments
* Update locals without parentheses
* Apply suggestions from code review
Co-authored-by: José Valim <jose.valim@dashbit.co>
* Format
Co-authored-by: José Valim <jose.valim@dashbit.co>
* Remove Code.Fragment backport
* Fix tests compatibility with Elixir 1.13
* Implement signature intellisense
* Don't show signatures on module attributes
* Add tests for calls with do-end block
* Unify spec formatting
* Insert parentheses when completing a function call
* Send all text until cursor in signature request
* Add configuration for completion/signature popups (#693)
* Add editor settings form
* Add configuration for intellisense defaults
* Read fresh settings when editor mounts
* Scope attribute names
* Fix disabled button styling
* Simplify signature box and enable by default
* Split settings into system and user sections
* Update lib/livebook_web/live/settings_live.ex
Co-authored-by: José Valim <jose.valim@dashbit.co>
* Update lib/livebook_web/live/settings_live.ex
Co-authored-by: José Valim <jose.valim@dashbit.co>
Co-authored-by: José Valim <jose.valim@dashbit.co>
* Fix spacing in documentation tests
Co-authored-by: José Valim <jose.valim@dashbit.co>
