* Add container scripts for VSO builds
* Remove relative path from kubectl.exe
* Use a variable and iex for kubectl path
* Use envvar for working dir path
* Try agent directory
* Update kubeconfig path
* Write connection settings json in build script
* Output logging
* remove _work prefix
* End end point instead of service name
* Disable failing tests
* Change service name prefix so it's different than SSMS containers
* Fix a few issues with profiler method handlers
* Filter out profiler polling events
* Add a unit test for profiler events
* Add method comment headers
* Moving logic for adding default values to new rows
* Fixing implementation of script generation to handle default values all around
* Unit tests!
* WIP
* Reworking row create script/command generation to work more cleanly and work on triggered tables
* Addressing some bugs with the create row implementation
* Implementing the trigger table fix for row updates
Some small improvements to the create/update tests.
* Differentiate External and FileTable types in object explorer list
- Add (External) and (File Table) to the labels
- Pre-fetch these properties to avoid perf hit
Note on testing: These are both quite hard to initialize, so adding tests to the integration tests would require always having a server with Polybase enabled / FileTable enabled. Given this I feel it should be OK to use manual testing. Long term, we should have unit-level tests that cover the functionality or figure out how to have a perfect pre-configured server and a set of tests to run the breadth of this logic against it.
* Fix handling of vNext servers to default to latest known version
- Default uplevel / unknown versions to the latest version
- In the IsValidFor check also just see if the server version registers as "All" since this is the default for unknown servers. This is a backup handler, the 1st fix is the primary one
* Refactoring sql script formatting helpers into To and From helpers
* Updates to make error messages for formatting errors more useful
* Fixing dumb breaks in unit tests
* Addressing comments from PR
* Updates to the SR files...
* WIP
* All the new RowCreate tests are working
* Fixing a couple bugs with the row delete and row update tests
* Regenerating localization files
* Fixing multiple iteration in tests
* Support `GO N` syntax to execute multiple times
- Plumbed through the batch execution count from the parser and used in the batch execution code path
- Functionality matches SSMS:
- Outputs loop start/end messages that match SSMS if you're doing multi-batch execution
- Outputs an "ignoring failure" error if an error happens during a batch
- Added tests for this
- Manually verified end to end also
* Fixing test error
- Fixes https://github.com/Microsoft/sqlopsstudio/issues/97.
- This should fix the bulk of the issues reported by users since it's the 1st use of this connection in the query code path.
- Implemented the fix in the connection service. This will always open a connection when calling `GetOrOpenConnection`. I cannot see a reason why the connection returned from this should ever be closed.
- resolve issues in both this code path and the edit data code path since both use this method.
* Moving logic for adding default values to new rows
* Fixing implementation of script generation to handle default values all around
* Unit tests!
* Addressing PR comments
- HandleSimpleExecuteRequest now handles the case where no rows are in a result by cleanly returning a success message but with no rows included. This is handled in the front-end instead and goes through the standard path (with a clean explanation message) instead of showing a `error: no results to return`
- MetadataService was always meant to include the type name in the return result, as otherwise the front end has to guess. In order to fix a bug where this resulted in scripting based on the metadata failing (as front-end used `Procedure` instead of `StoredProcedure`), I'm returning the data here. I'll have a matching front end fix but this is overall a good solution to have.
