This is a partial fix that lays groundwork for full "Prompt to connect" if a kernel needs a connection.
I am waiting on Yurong's refactoring of connection handling before doing any of the prompt work.
- Adds kernel metadata about whether a connection is required.
- For Jupyter, only Spark kernels are listed as requiring a connection
- If this is true and there's no active connection, will show notification and not call execute
In the future, this path will still be used if user is prompted to connect and cancels out.
The future change will be to inject a "connect" handler from notebook.component to the cell callback and use to set connection context
Add back check for textDocuments with same name, should've been there anyhow
On rehydration files show as text docs before clicking as only get
changed by customInputConverter code path.
We should look at this long term - ideally we'd update notebookDocuments
with correct values on initial start. #4958 opened to track this.
* Fixed#4800 need to use ConnectionProfile in order to get the correct connection
* Go back to create connect in run cell to avoid to fail to run cell or close the connection used by other.
* more strict null checks in base browser code
* revert changes to radiobutton
* fix some more minor things, enable strict null check in pipelines
* formatting
* fix compile errors
* make null undefined
* more null to undefined
* runAllCells API
* add comment
* more run cells fixes
* Add integration test
* Add multiple cell SQL notebook test
* Comment out python tests as they fail in the lab
* remove unused imports
* PR comments
* Remove localize
* Return true instead of promise.resolve(true)
* add cluster name to page
* implement profile page -1
* fix compilation error due to new method
* profile page 0328
* summary page
* make divcontainer accessible
* handle disposable
* add support for "coming soon" cards
* Change attach to show user name. Use connectionProfile.title to display the same info as OE server node
* Fixed some potential profile leak
* Removed unused import
* Resolve PR comments
* change to fs to write file
* change random range to 1 to 100
* Added more env variables check as well
* Add azure and standalone env variables
* Add Azure instance to unblock mac pipeline testing
* Fixed some merge issue and improve delete file
* Added standalone test only run on windows
* change to fs to write file
* Added more env variables check as well
* Resolve PR comments
* Merge master
* Add file name exist check and delete it after test is done
* Fix workspaceRoot macro for insights
The workspaceRoot macro wasn't working correctly for finding the queryFile. There were a couple of issues :
1. The path separators were hardcoded as / which wasn't xplat-compatible
2. They required the first section of the path was one of the folders in the workspace - e.g. if the workspace contained a folder named foo you'd have to specify ${workspaceRoot}\foo\myfile.sql. This is inconsistent with the folder logich was just appends the path after ${workspaceRoot} to the folder that's currently open
I changed the logic to just append the relative part of the path to every folder currently open in the workspace and choose the first one that it found that contained the file we were told to look for - which follows the convention the folder logic uses. If the file doesn't exist it'll just fall back to using the path without the macro (which is likely to not resolve and thus will display an error, but there's nothing we can do at that point anyways)
* Switch to using VS Code resolver (support for more than just workspaceRoot) and move resolution code into helper method so it can be used by the multiple places it's called. Added tests for the methods.
* Add test for invalid param
* Change resolveQueryFilePath to be a standalone exported function. Change it to throw if the file can't be resolved/found so callers can display error correctly. Added more tests to covery new scenarios. Switch to using pfs instead of fs for file existance checks.
* Add extra param to InsightsDialogController construction in test
* Fix formatting and test errors.
* Change to suiteSetup and suiteTeardown so the setup/teardown is only ran once instead of once per test - we don't need unique files and this stops a race condition error with deleting the test folder.
* spaces -> tabs
