* add new exception test
* Make memory tests and file path tests be explicit
* set the default write parallelism to 1
* set to single reader for caching channel
* format
* Try to have consistent DB locked error test
* always a single reader of the channel
* Remove extra snapshot
* Revert "Try to have consistent DB locked error test"
This reverts commit 93669c57a3.
* remove extra test that doesn't do anything
* First pass of ObjectSaver and better in-memory usage
* fix some tests
* add commit to match deserialize process
* correct more tests
* format
* make a deserialize factory
* fix tests? and format
* use distinct
* Fix mismerge
* Fix serialization issues with tests
* fix merges
* follow copilot suggestions
* remove disables
* Serialize now waits for the scheduler to be completed before returning
* formatting
* wait in deserialize too
* Pass cancellation token to download, refactor how things are passed/created
* responses are now streamed and cancel them earlier
* format
* add manual empty message for sending that definitely knows when the channels are empty
* Fix configure awaits
* more configure await false
* Don't log cancelling
* redo exception handling for receive
* remove null test case
* clean up with Id/Json and more cancels
* Change the exception stacks
* fix serialization test
* make a custom scrubber for internalized exceptions
* clean up
* fix namespaces again :(
* adjust the scrubber
* try to make tests more predictable
* rework exceptions again
* strip out compile files used
* formatting
* custom exception validation
* fix init
* Move serialization to own class
* save serialize test
* add deep clean
* add cancellation test on save to cache
* cancellation tests
* format
* do DI correctly
* receive cancel works
* Add exception handling for SerializeProcess with CancellationTokenSource
* formatting
* add exception test to make sure we handle a server exception
* add extra exception and handling to stop
* add comment and another test
* one last chance for user to cancel
* formatting
* add ServerObjectManagerFactory
* add usage of a command pool
* add more disposal
* save saving increase
* fix tests
* fixes
* push out concurrency and disposablity
* Add a custom task scheduler
* Better usage, don't wait to enqueue to save to channels
* Completely pre-cal batch size to avoid spinning issues
* Try to fix cache counting
* properly dispose things
* format
* clean up
* adjust count and save on current thread
* move batch it's own file
* update a few packages
* fix build and add batch tests
* revert and format
* Revert "save saving increase"
This reverts commit 3b50c857fb.
* revert change
* adjust and add tests
* Dispose sqlite manager properly
* Make Batch a IMemoryOwner to allow for pooling
* Fix tests
* Upgrade some deps
* try to make tests more explicit
* remove return value
* Use named tuple for all objects
* refactor options to use on factories, add dummies
* format
* reduce lists to yield return
* get primitives first to avoid to string
* fmt
* add cache base option
* Can debug dependencies
* Different exceptions
* Uses root id only after we found it to signal the end
* DataChunks are created later and need to be accounted for
* format
* use app ids in tests and references
* check sqlite cache after serialize
* use dummy to go through channels to end
* fmt
* Extend channel lib to batch by size
* fmt
* build fix
* adjust limits
* FIx sending
* Optimize reference generation
* more
* remove tolist
* rework closures to be constant and serializer only deals with current....references bases are cached
* fix chunk creation
* another bug fix
* clean up with factories
* add deserializer factory
* Needed to reference interface
* add Id and Json types
* Fix tests
* remove optional
* add another missing id write
* fix merge
* Fix new code
* Can debug dependencies
* Different exceptions
* Uses root id only after we found it to signal the end
* DataChunks are created later and need to be accounted for
* format
* use app ids in tests and references
* check sqlite cache after serialize
* use dummy to go through channels to end
* fmt
* Extend channel lib to batch by size
* fmt
* build fix
* adjust limits
* FIx sending
* Optimize reference generation
* more
* remove tolist
* rework closures to be constant and serializer only deals with current....references bases are cached
* fix chunk creation
* another bug fix
* clean up with factories
* add deserializer factory
* Needed to reference interface
* move around streamId
* some clean up
* Use StringBuilder pool on serialization to reduce memory pressure
* remove extra
* remove extra clears
* Fix a flaw in batchsize
* use default complete
* format
* loader should use 1 writer that is batched
* remove redundant ref gen
* Fix graphql commands by adding project id
Adam Hathcock