The Attach To was showing only localhost for SQL, since we overrode the standard kernels from SQL with the ones from Jupyter.
Fix is to save all standard kernels.
Also, added dispose handling for some events I found during debugging and removed unused imports
Recent changes in the VS Code layout for that viewlet caused the min-height property to affect the display (the property was set on that class before but the layout prevented it from actually being displayed as it was).
I removed the style for the messages class completely since I couldn't find a place that we actually used that ourselves - and anyways having styling on such a common name like messages isn't really a good practice since it applies to everything in ADS.
* engine check when install extension
* gallery install/update and vsix install
* FIX COMMENTS
* Fix the detail not loading issue when version is invalid
* add more comments and adress PR comments
* add install telemetry for install from vsix scenario
* correct the name of the version property for telemetry
* Moved dacfx wizard into separate extension
* updating to use azdata
* one more azdata change
* bump import extension version
* renaming extension to dacpac
* check to see whether options is available
* removing options state and using _modelOptions
* removed additional space
* making options getter private as not intended to expose
* adding context
* apply extension changes
* shimming disconnect
* add data explorer context menu and add disconnect to it
* clean up shim code; better handle errors
* remove tpromise
* simplify code
* add node context on data explorer
* formatting
* fix various errors with how the context menus work
* Added verify OE child nodes tests for sandalone and BDC instance
* msg changes
* Added standalone OE node test
* Resolved PR comments
* Change env name
* Added scripts to set env for integration test
* Fixed#4181 and #4167
The problems are:
- When change Kernel, setProviderIdForKernel switches Kernel first. So when we switch kernel across provider for exmple from PySpark3 to SQL. The Kernel is set SQL already in ClientSession.changeKernel. Then we lost oldKernel info.
The fix is cache the old session in model before switch and use it to get the correct context
- SQL Kenerl could make mulitple connects from "Add new connection". While we didn't track them, those connections didn't close properly when close the notebook
The fix is saving the connections made from "Add new connection" in model. and close them and activeConnection when close notebook
Problem is not solved yet in this PR:
- Didn't shutdown Jupytper when swich kernel from spark kernel to SQL.
* adding context
* apply extension changes
* shimming disconnect
* add data explorer context menu and add disconnect to it
* clean up shim code; better handle errors
* remove tpromise
* simplify code
* Add azdata.d.ts for new extensibility APIs
* Update azdata typing files for connection API proposal
* Add implementation for azdata module
* Fix build break in agent
