FluidFramework
FluidFramework copied to clipboard
Some batching improvements part 1 (ScheduleManager) - work in progress
Description
Related to 743 in ADO.
Does this introduce a breaking change?
Yes, ScheduleManager
now requires a callback to fetch the clientId in the constructor.
⯅ @fluid-example/bundle-size-tests: +18.42 KB
Metric Name | Baseline Size | Compare Size | Size Diff |
---|---|---|---|
aqueduct.js | 388.9 KB | 391.94 KB | ⯅ +3.04 KB |
connectionState.js | 680 Bytes | 680 Bytes | ■ No change |
containerRuntime.js | 192.07 KB | 198.93 KB | ⯅ +6.86 KB |
loader.js | 151.12 KB | 151.06 KB | ⯆ -57 Bytes |
map.js | 42.63 KB | 47.38 KB | ⯅ +4.75 KB |
matrix.js | 128.09 KB | 130.87 KB | ⯅ +2.77 KB |
odspDriver.js | 150.23 KB | 150.11 KB | ⯆ -127 Bytes |
odspPrefetchSnapshot.js | 38.39 KB | 38.35 KB | ⯆ -41 Bytes |
sharedString.js | 148.73 KB | 149.95 KB | ⯅ +1.22 KB |
Total Size | 1.24 MB | 1.26 MB | ⯅ +18.42 KB |
Baseline commit: fec2c0edf81f145a96a6aab34fa7aa96692c1964
Generated by :no_entry_sign: dangerJS against a029f6b4227f85f12ee10243c1fde0f47a82ae56
@kian-thompson , @andre4i, can you please chat and see what makes sense here? You both are poking in this area, and I feel like there is some overlap / stepping on each other. Andrei - please also glance at Justus compression PR, it feels like in general we are changing where and how we process things (i.e. it's a pipeline with multiple steps - batch chunking, gathering batch, compression, unpacking), and we likely need to re-evaluate how to isolate and packages code.
@vladsud ack!
This PR can be closed. It was mostly experimental from my pov and I was aiming at refactoring the schedule manager to make it a more explicit state machine and fix some validations as part of ADO:743. None of which make sense in the current codebase though.. Noted about the pipeline!