-
Notifications
You must be signed in to change notification settings - Fork 18
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
fix Vector{Bool} branches and factorize interped_data
#95
Conversation
Alternatively, I could just undo the factorization of function interped_data(rawdata, rawoffsets, ::Type{T}, ::Type{Nojagg}) where {T<:Bool}
return map(ntoh,reinterpret(T, rawdata))
end but then it's less nice to read the long if-else chain. |
Codecov Report
@@ Coverage Diff @@
## master #95 +/- ##
==========================================
+ Coverage 84.61% 84.63% +0.02%
==========================================
Files 11 11
Lines 1287 1328 +41
==========================================
+ Hits 1089 1124 +35
- Misses 198 204 +6
Continue to review full report at Codecov.
|
interped_data
Tamas probably uses more |
I am back on track. Thanks for this PR, I will check it and report back, but I think it should be fine since the test coverage is pretty good for custom stuff. That's why I wanted to leave it there in first place ;) |
I tried it but I get ambiguities. I probably need to change the way I deal with
|
Ah wait, I am dumb. This is colliding with an existing implementation in the core library... |
OK, I refactored |
I screwed up the merge, so I replaced this PR with #106 . ;) |
The change in #93 results in
BitVector
instead ofVector{Bool}
which breaks LazyBranch. We didn't have a test for it, but I added one here. I split up the largeinterped_data
into smaller chunks for non-, singly-, doubly-jagged branches so that it's easier to specialize on Bool. Evidently this breaks some LV64 test cases which I'm still figuring out.