An interesting and natural PBT target could be facing off the different container types against each other. That is, for every operation offered by RB, the three container types should have the same behavior. To bypass the usual selection logic maybe test directly on container?
Some performance tricks when using the RoaringBitmap libraryhttps://richardstartin.github.io/posts/roaringbitmap-performance-tricks …
-
-
-
I thought about that before, but they have different preconditions which are maintained elsewhere, which would need to be moved into the test case generator. What the fuzz tests don’t do well is generate data from different distributions to explore more of the state space faster.
- Još 17 drugih odgovora
Novi razgovor -
-
-
That's a really good write-up.
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.