As a static function that could be called from anywhere in the code
(directly or indirectly), 'Configuration.Initialize' was vulnerable to
race bugs. The "Configuration.get<T>()" method, whose calls are
ubiquitous in our code, would initialize config if it hadn't been done
so already. Without a reference to the configServer param, this param
would effectively be ignored/overridden with null.
It's important that login only be performed once, for obvious reasons
(our exes that connect to sparkd are generally designed to process data
for a single user). Furthermore, not every routine that requires a
successful login has a reference to the credentials needed to do it. To
address this, we now pass the ISparkdLoginService singleton to these
routines so they can wait until the singular login task has completed
before they begin.
The former is the base class for the latter, so this will widen the umbrella to catch a few corner cases
that would otherwise been handled by the generic "Exception" catch block. One corner case I know
of is when an op is canceled via "CancellationToken.ThrowIfCancellationRequested". In that case,
an OperationCanceledException is thrown (not a TaskCanceledException).
https://stackoverflow.com/questions/34359129/operationcanceledexception-vs-taskcanceledexception-when-task-is-canceled
Pnl window settings are stored with the workspace on the server (like all shell
window settings), and may originate from an older version of Spark with a
different set of columns. We shouldn't crash if this happens.
It is initialized from data stored in shared memory (not a normal struct). The
cs build emits a warning because it detects a field that "is never written to
and will always have a default value of 0."
This code hasn't changed for years. It suddenly becoming a warning-as-error is
likely related to a recent Visual Studio update.
Bugs (fixed):
* SettingsPersistenceModel drops items after initial update in filtered
subscriptions.
Cause: The SettingsPersistenceModel subscribe fn only publishes the "added"
items in each update, and omits pre-existing items. From the observer's
perspective, pre-existing items appear deleted.
Binary search is used to add, remove, and change rows in response to
incremental updates. Needless to say, if the rows become unsorted,
everything breaks.
Bugs (fixed):
* The fills window crashes sometimes, possibly related to client
reconnecting to server.
Cause: This was a 2 part bug.
1) Changes to existing rows are applied BEFORE the new sort index
is determined via binary search. If the change affects a
property in the current sorting criteria, the rows could
become unsorted and thus invalidate the binary search.
2) When an existing row is changed so that its new sort index is
greater than its current one, the new index is not decremented
by one (it must be since the row is removed before it is
re-inserted).
There was a typo when determining "deletes" and "undeletes" on synth
changes (i.e., when deleted changes false -> true, and true-> false
respectively).
This new structure provides several benefits:
* Allows us to install multiple CMake projects in the same destination
dir (without include path conflicts).
- This makes shipping the trading algo sdk a trivial excercise. Simply
copy the lib and include dirs from the install dest. That's it!
* Puts us in a position to break up the monolithic libspark into smaller
modules, which are easier to manage.
Windows specific notes:
* Get rid of unnecessary script vs_filters.cmake
* Define WebsocketConnectionGateway directly in cs (instead of proto)
- The proto would no longer be in the include path passed to protoc.
Moving it to c# because there wasn't a compelling reason for it to
be a proto in the first place.
* Set spark output prefix to "lib" on windows
- So we get "libspark.dll" instead of "spark.dll", which conflicts
with the csharp lib, "Spark.dll".
* Copy libspark.dll to net8.0 & net8.0-windows in a single cmake build
target
fix build: Copy libspark to both dirs net8.0, net8.0-windows
Previously, animations were expressed through polymorphism (derived from the
base class CellAnimation), but the new impl uses composition. This 'OfType' call
was leftover from the polymorphism implementation, an attempt to linq cast
elements of base class (CellAnimation) to BackgroundAnimation.