factorize interpretation

[aminnj]

Split up the large interped_data into smaller chunks for non-, singly-, doubly-jagged branches in case we want to cleanly specialize further (like for Bool).

Replaces #95 .

[codecov]

Codecov Report

Merging #106 (e608734) into master (b32d2cb) will decrease coverage by 0.24%.
The diff coverage is 96.15%.

@@            Coverage Diff             @@
##           master     #106      +/-   ##
==========================================
- Coverage   89.84%   89.59%   -0.25%     
==========================================
  Files          11       11              
  Lines        1319     1326       +7     
==========================================
+ Hits         1185     1188       +3     
- Misses        134      138       +4     

Impacted Files	Coverage Δ
src/custom.jl	93.82% <75.00%> (-4.88%)	⬇️
src/root.jl	91.11% <100.00%> (+0.12%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update b32d2cb...e608734. Read the comment docs.

[tamasgal]

OK thanks! I just run some random script, just to make sure everything is fine ;)

[tamasgal]

Alright, it works nicely!

I am just wondering how we should deal with code repetitions in such cases:

function interped_data(rawdata, rawoffsets, ::Type{Vector{_KM3NETDAQHit}}, ::Type{Nojagg})
    UnROOT.splitup(rawdata, rawoffsets, _KM3NETDAQHit, skipbytes=10)
end
function interped_data(rawdata, rawoffsets, ::Type{Vector{_KM3NETDAQHit}}, ::Type{Offsetjagg})
    UnROOT.splitup(rawdata, rawoffsets, _KM3NETDAQHit, skipbytes=10)
end

[aminnj]

I had originally tried to make a Union like this

-function interped_data(rawdata, rawoffsets, ::Type{Vector{_KM3NETDAQHit}}, ::Type{Nojagg})
-    UnROOT.splitup(rawdata, rawoffsets, _KM3NETDAQHit, skipbytes=10)
-end
-function interped_data(rawdata, rawoffsets, ::Type{Vector{_KM3NETDAQHit}}, ::Type{Offsetjagg})
-    UnROOT.splitup(rawdata, rawoffsets, _KM3NETDAQHit, skipbytes=10)
-end
+function interped_data(rawdata, rawoffsets, ::Type{Vector{_KM3NETDAQHit}}, ::Type{Union{Nojagg,Offsetjagg}})

but I saw tons of unit test failures like this

  ArgumentError: cannot reinterpret `UInt8` as `Vector{UnROOT._KM3NETDAQHit}`, type `Vector{UnROOT._KM3NETDAQHit}` is not a bits type
  Stacktrace:
   [1] (::Base.var"#throwbits#241")(S::Type, T::Type, U::Type)
     @ Base ./reinterpretarray.jl:16
   [2] reinterpret(#unused#::Type{Vector{UnROOT._KM3NETDAQHit}}, a::Vector{UInt8})
     @ Base ./reinterpretarray.jl:36
   [3] interped_data(rawdata::Vector{UInt8}, rawoffsets::Vector{Int32}, #unused#::Type{Vector{UnROOT._KM3NETDAQHit}}, #unused#::Type{UnROOT.Nojagg})
     @ UnROOT ~/.julia/dev/UnROOT/src/root.jl:201

so the interped_data in root.jl was called instead of the one in custom.jl, presumably because the JaggType parameter is more specific in root.jl. Any suggestions?

[tamasgal]

Hmm... maybe some more fine-graded hierarchy for the JaggType could do it, but I kind of struggle to group them in a logical way in the sense that I have no idea what would be the supertype of Nojagg and OffsetJagg. It's a bit annoying that the Union loses the dispatch-battle.

[aminnj]

We don't want this right? :)

for J in (:Nojagg,:Offsetjagg)
    T = _KM3NETDAQHit # can loop over this to get the other struct too
    @eval function UnROOT.interped_data(rawdata, rawoffsets, ::Type{Vector{$T}}, ::Type{$J})
        UnROOT.splitup(rawdata, rawoffsets, $T, skipbytes=10)
    end
 end

[tamasgal]

I don't know, probably not 😆

[Moelf]

maybe it's time to figure out a more sane customization interface

[aminnj]

I'm also not opposed to keeping the main interped_data as in master (as long as constant propagation compiles the right part of the if-else chain, which is hopefully the case...)

[Moelf]

ehh, as always I fucked up the rebase

[codecov]

Codecov Report

❌ Patch coverage is 96.15385% with 2 lines in your changes missing coverage. Please review.
✅ Project coverage is 89.95%. Comparing base (527f04b) to head (d9fd45a).
⚠️ Report is 256 commits behind head on main.

Files with missing lines	Patch %	Lines
src/custom.jl	75.00%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #106      +/-   ##
==========================================
- Coverage   90.15%   89.95%   -0.21%     
==========================================
  Files          11       11              
  Lines        1605     1612       +7     
==========================================
+ Hits         1447     1450       +3     
- Misses        158      162       +4     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.