- Do not reuse parameter variables `start` and `end`. Create new variables with meaningful names and rename them to `startUtc` and `endUtc`.
- Use `EachTradeableDayInTimeZone` to calculate `tradableDates`.
- Adds another test/assertion in the regression algorithms to ensure tests in the scheduled event were performed.
The tradable days of the history request should respect the data time zone since the data source files also do.
Upgrade `BasicTemplateFuturesHistoryAlgorithm` to a regression algorithm and add a schedule event to test history requests every hour.
* The original implementation was made with the idea in mind that bars
have a time range that they encompass. When the `Time` property is
rounded down, the period added to `EndTime` ensures that the data
will not have look-ahead bias.
However, multiple data sources do not have a time range that the data
applies to, and is actually point-in-time. For these kinds of
data, rounding of the data is not preferred since they are supposed
to represent a point in time.
- Add a comment explaning why we need to round down in data time zone
and not exchange timezone.
- Keep subscription EndTime RoundDown by DataResolution since it won't
change and it's expensive.
- Replace Enum to string operations on Options enums, since it's
expensive
- Replace Enum IsDefined check since it uses reflection
- Improve OptionFilterUniverse linq operations
- Improve performance of SecurityIdentifier properties to avoid having
to extract them from properties always
- Avoid calling ToList for already list data collections (it creates a
copy)
- Avoid iterating over securities for which we have no holdings when
calculating TPV or MarginUsed
- Refactor ZipDataCacheProvider to avoid value types thread issues, it
will now use a timer. Add missing dispose calls.
- Synchronizer won't share the `SubscriptionFrontierTimeProvider`
instance since it's not thread safe and shouldn't be called
- Fix BacktestingResultHandler `_daysProcessed` thread safety
- Fixing `RealTimeScheduleEventService` thread safety
- Adding new ReferenceWrapper for structs, value types, to avoid thread
race conditions while reading and writting. In C# reference type
assignments are atomic, so it allows us avoid using locks.
- Add more fill forward unit test including a weekend using different
combinations of resolutions
- Add required AIG second data file
- Avoid list in most common FF case
- Add StreamReader Reader for Tick data type.
- Adding stream reader GetString extension, adding tests.
- BacktestingBrokerage will not create unnecessary order events
- AlgorithmPythonWrapper will directly call base OnFrameworkData()
implementation skipping going through python and it's overhead
- Small performance improvement for adding Tick data points into a Ticks
collection
- For python always wrap slice with PythonSlice, so that slice.Get()
works even when no custom data is present, adding test.
- Reduce ResultHandler amount of packets sent and their size.
- Only send order and orderEvents packet if there is actually data in them.
- Don't send Holdings fields if value is 0
- Don't send AlphaRuntimeStatistics fields if default values
- Adding new ConcatEnumerator that will join enumerators together
secuentially. Adding unit tests
- SubscriptionDataReaderHistoryProvider will consume an optional
intraday enumerator that will be concatenated with the existing
SubscriptionDataReader
- Algorithms logs of the same type will be batched together when sent
- Increasing backtesting update to 3 seconds same as LiveTrading
- Increasing insight sending interval from 1 second to 3
- Reducing code duplication for retrieving and sending algorithm logs
- Adding new SerializedOrderEvent and SerializedOrder with new
JsonConverters
- Specifying OrderEvent json converter when storing, streaming data
- Adding unit tests
Martin Molinero