* adding network file validator contract
* updating to dotnet 5
* reverting change in pipeline
* updating target framework to 5.0
* updating srgen and using target framework var
* reverting srgen back to 3.1
* Adding dotnet 3.1 to build
* Fixing task def
* removing global json
* updating other 3.1 refs to 5.0
* removing netcore folder from build.cake
* Moving and renaming dotnet 3.1 task
* Remove xunit dependency from testdriver
* swap expected/actual as needed
* Convert Test.Common to nunit
* port hosting unit tests to nunit
* port batchparser integration tests to nunit
* port testdriver.tests to nunit
* fix target to copy dependency
* port servicelayer unittests to nunit
* more unit test fixes
* port integration tests to nunit
* fix test method type
* try using latest windows build for PRs
* reduce test memory use
* Update form .NET Core 2.2 to .NET Core 3.1
- Global variable for projects
- Change TFMs from netcoreapp2.2 to netcoreapp3.1
- Update global.json
- Update build.json
- Remove direct framework cake TestCore task
- Update travis dotnet version
- Update azure pipline file
- Update vscode launch.json
* Add Central Package Management
* Fix xUnit Breaking Change for MemberData type
* Fix xUnit breaking change for duplicate test method name
* Fix Rang/Index type conflict with System.Rang/Index
* Update vscode tasks.json
* Change serviceHostExecutable path in ServiceTestDriver.cs
* Downgrade SDK version (https://github.com/appveyor/ci/issues/3440)
- Appveyor hasn't installed latest SDK therefore I downgrade it until they install it.
* Dump Microsoft.SqlServer.DACFx
* Revert "Update SMO (#875)"
This reverts commit 96593d66e3.
* Revert "Switch Tools Service to new SMO and Microsoft.Data.SqlClient driver (#865)"
This reverts commit 9d140b53f3.
* Bump SMO to 160.1910315.0-preview
* switch to ambient props and targets files
* build against Microsoft.Data.SqlClient
* build tests
* fix test bug
* temporarily add SMO nuget to the repo
* update to released Microsoft.Data package
* Switch to the unified SMO NuGet so all binaries are strong named and signed.
* use one AssemblyLoader instance for all loads
* Revert "use one AssemblyLoader instance for all loads"
This reverts commit 48c59ffd5c57152de281c87acdbcad7ddf7ab760.
* Stop creating multiple AssemblyLoadContext objects during composition, per https://github.com/dotnet/coreclr/issues/19632
* restore high entropyva property
* Improve the comment
* Update SMO to 150 and fix scripting for objects with a single quote in their name.
* restore newtonsoft reference
* Revert "fixed perf tests and added more scenarios (#683)"
This reverts commit aa2b30f486.
* Revert "Switch to the unified SMO NuGet so all binaries are strong named and signed. (#682)"
This reverts commit 69961992bb.
* Upgrade SMO packages to fix issue with scripting tables when Script Dependencies is true on the script options.
* use linux style import with case matching file name
* scripting working with race conditions
* new service works with no race conditions
* use new scripting service and commented out tests
* refactored peek definition to use mssql-scripter
* fixed peek definition tests
* removed auto gen comment
* fixed peek definition highlighting bug
* made scripting async and fixed event handlers
* fixed tests (without cancel and plan notifs)
* removed dead code
* added nuget package
* CR comments + select script service implementation
* minor fixes and added test
* CR comments and script select
* added unit tests
* code review comments and cleanup
* Bump SMO to 140.2.5 to pick-up private XEvent binaries
* Pick up SMO binaries from the build lab
* Add ProfilerService class placeholder
* Update SMO nuget package to include DB Scoped XEvents
* Stage changes
* Stage changes
* Update SMO to use RTM dependencies and remove separate SqlScript package
* Stage changes
* Iterate on profiler service
* Fix post-merge break in localization
* More refactoring
* Continue iterating on profiler
* Add test profiler listener
* Address a couple of the code review feedback
* Fix AppVeyor build break
* Use self-cleaning test file
* Manual port of latest 'feature/mssq-scripter' branch
* Bumpnuget package SqlScriptPublishModel.140.2.0 to Microsoft.SqlServer.Management.SqlScriptPublishModel.140.2.3
* In TestDriver, fix the path to Microsoft.SqlTools.ServiceLayer.exe after move to .NET Core 2.0
* Initial .net core 2.0 conversion
* Convert a few more projects to .net core 2.0
* Convert a few more projects to .net core 2.0
* Fix build.cmd errors
* Add mising nuget package
* Remove dead code
* Add checked in references to workaround nuget package issues
* Update SLN file to refer to correct csproj files
* Rename applications to workaround .net core tooling bug
* Update nuget package with SQL Parser changes
* Add PreserveCompliationContext to avoid MEF bug
* Update smo version to pickup .net core 2 changes
* Pickup latest SMO changes to fix merge break
* Actually pickup correct SMO binaries
* Add support for SLES 12.2
* Fix break running archiving on Linux
* Revert "Add support for SLES 12.2"
This reverts commit 95cdb6d0e35a425be5c0081345d214079cbdc3db.
* Update to latest SMO build
* Install .Net Core 2 during install phase
* Move .Net Core install
* Try to reference dotnet.exe directly
* Fix code coverage script for CSPROJ instead of project.json
* Turn off test that is unreliable in AppVeyor builds.
* Fix appveyor.yml line feed.
* Turn off another flaky test failing in AppVeyor
Update the ScriptingService to expose new scripting JSON-RPC APIs that use the SqlScriptPublishModel for script generation.
The SqlScriptPublishModel is the model behind the SSMS scripting wizard. To enable scripting for CLI tools, we've ported SqlScriptPublishModel to .NET Core. The SqlScriptPublishModel wraps the SMO scripting APIs for .sql script generation.
1) Added three new requests to the ScriptingService: ScriptingRequest, ScriptingListObjectsRequest, ScriptingCancelRequest.
2) Generating scripts are long running operations, so the ScriptingRequest and ScriptingListObjectsRequest kick off a long running scripting task and return immediately.
3) Long running scripting task reports progress and completion, and can be cancelled by a ScriptingCancelRequest request.
4) Bumped the SMO nuget package to 140.17049.0. This new version contains a signed SSMS_Rel build of SMO with the SqlScriptPublishModel.
5) For testing, adding the Northwind database schema
TODO (in later pull requests)
1) Integrate the new ScriptingService APIs with the ConnectionService
2) Integrate with the metadata support recently added
* Initial credential service files
* Clean-up hostloader
* Update build scripts to build credentials archive
* Move hosting files to new assembly
* Add credentials files to MS.SqlTools.Credentials
* Remove duplicate files
* Update namespace in program.cs
* Fix test build breaks
* Update extensions visibility.
* Remove unused resource strings
* Add xproj files to SLN for appveyor builds
* Fix appveyor build break in test project
* Fix extensibility tests
* Fix various typos in latest iteration
* Add settings for Integration build
* Fix codecoverage.bat to use full pdb for new projects
* Fix bug when packing in folder with native images
* Fix typos in xproj
* Reset XLF to fix build.cmd
- TSqlFormatterService with support for formatting document and text range inside document
- Settings support for all formatting options.
- Extensibility support so that the service can be initialized using MEF extensibility, and can find all necessary TSqlFormatters using the same process
Fix Initialize request error on startup
- Messages were being read from the input channel before all request handlers were registered
- In particular, the Initialize request which is key for any server to talk to the client was getting lost because the message reader thread begins consuming, and we take an extra few hundred milliseconds due to MEF startup before we register the handler
- The solution is to initialize the message handler so request handlers can register, but not actually start processing incoming messages until all handers are ready. This is a safer way to go and should improve reliability overall
Improvements from internal prototype:
- Normalizing baselines to handle the line ending differences on Mac & Linux vs. Windows
- Significantly shortened most lines by implementing base class methods to wrap common objects from Visitor.Context and removing unnecessary "this." syntax
- Refactored the SqlCommonTableExpressionFormatter and related classes to reduce code count significantly. This provides a pattern to follow when refactoring other classes for similar clarity. It's likely a lot of common logic could be found and reused across these.
- Reduced overall code size by adding utility methods
This is a small API addition that allows us to execute queries directly as strings. This will make it easier to execute queries outside the confines of a workspace like VS Code.
* Refactor execution requests and events are now named less redundantly and moved into a separate namespace for organization. This is the bulk of the changes.
* QueryExecuteBatchNotification -> ExecuteRequests/BatchEvents
* QueryExecuteMessageNotification -> ExecuteRequests/MessageEvent
* QueryExecuteCompleteNotification -> ExecuteRequests/QueryCompleteEvent
* QueryExecuteResultSetCompleteNotification -> ExecuteRequests/ResultSetEvents
* QueryExecuteSubsetRequest -> SubsetRequest.cs
* Creating an inheritance pattern where
* `ExecuteRequestParamsBase` has execution options and ID for a query execute request
* `ExecuteDocumentSelectionParams` inherits from `ExecuteRequestParamsBase` and provides a document selection
* `ExecuteStringParams` inherits from `ExecuteRequestParamsBase` and provides the query text
* Adding a helper method to get SQL text based on request type
* Through the AWESOME POWER OF POLYMORPHISM, we are able to create a request for executing straight SQL basically for free.
* **Breaking change:** query/execute => query/executeDocumentSelection to make it more obvious what is expected.
* Adding unit tests for the code that gets SQL text
* Refactoring of execute contracts into their own namespace
* Refactoring application
* Adding new request for executing queries as strings
* Adding forgotten string request
* Changing the logic for checking the request param types
* Removing redundant declarations
This change is a reworking of the way that messages are sent to clients from the service layer. It is also a reworking of the protocol to ensure that all formulations of query send back events to the client in a deterministic ordering. To support the first change:
* Added a new event that will be sent when a message is generated
* Messages now indicate which Batch (if any) generated them
* Messages now indicate if they were error level
* Removed message storage in Batch objects and BatchSummary objects
* Batch objects no longer have error state
* Implemented function signature help, and added tests
* Incremental commit of changes from code review feedback
* Rename test
* Added check to make sure intellisense is enabled
* Use HoverTimeout instead of BindingTimeout
This is another large code review. I want to make a few more changes, but since these changes will stand on their own, I'll hold back on making this change set any larger than it already is.
Changes in this request:
To address Microsoft/vscode-mssql#326, instead of doing taskkill on the service layer when WaitForExit is executed, we now make an educated guess at which service layer was spawned when the test starts and do a Process.Kill on it when we shut down the test.
All the perf tests have been moved into a new project. This was done to keep them easily separated from code coverage test runs. At the same time the perf tests were separated into separate classes for logical categorization. This process will likely be repeated on the stress tests. The tests can still easily be ran from Visual Studio Test Explorer
To address Microsoft/vscode-mssql#349, a new SelfCleaningFile class was created to allow for easy cleanup of temporary files generated for integration tests via using blocks.
Due to some of the refactoring done while moving the perf tests to a new project, the TestBase class had to be switched to more of a helper class style. As such, all tests that use inherit from TestBase now create a TestBase object on start via a using block. This also simplifies the cleanup at the end of the test.
* Solution for hanging code coverage runs
Code coverage runs would hang in certain scenarios if a test failed before
the service process could be spawned. The taskkill command would fail to
find the service process. The test would then wait for opencover to exit,
but it would not since the service process it had spawned would still be
running, causing the test run to hang indefinitely.
Solution was to capture the service process after it launched and
explicitly kill it when shutting down the test driver.
* Setting the test name in the propery in the class and removign the parameter from each method
* New project for perf tests
* Reworking integration tests to cleanup temp files
* Changes as per @llali review comments
* Adding copyright notices
* Renaming TestBase => TestHelper
* Renaming SelfCleaningFile => SelfCleaningTempFile
* Removing code that sets TestName property
* Fixing compilation error due to removed code
Using CallerMemberName attribute to get the test name for generating the test result instead of having a name passed as a parameter for each test method
