|
Ran
|
Jobs
6
|
Files
23
|
Run time
1min
|
Badge
README BADGES
|
push
github
Add more filters (#154) * Document the Filter interface * Move filters to a folder Same rationale as the other changes :D - just for cleanliness and clarity. * Factor out variable-length filters to a new file * Add docstrings to filter API functions * Add a Fletcher32 filter and test * re-add the dictionary entries for the vlen filters * Semi-working fixed scale offset filter * Add FixedScaleOffset tests * Add shuffle filter (buggy in the last few bytes, indexing issues) * WIP quantize filter * ShuffleFilter working and tested * Semi working quantize filter * Format tests better * Complete interface and test quantize * Uncomment the FixedScaleOffset tests * fix getfilter syntax * Add delta filter * Adapt for Kerchunk playing fast and loose with the spec - Kerchunk often encodes the compressor as the last filter, so we check that the compressor isn't hiding in the filters array if the compressor is null. - Similarly, the dtype is often unknown in this case, or the transform is not encoded correctly, so we ensure that the datatypes of `data` and `a2` remain the same by reinterpreting. * Fix the delta and quantize JSON.lower * Change the tests to be more sensible/Julian and avoid truncation errors * Fix the FixedScaleOffset filter materializer * Fix decoding for fill values to use `reinterpret` on unsigned -> integer * If `getfilter` fails, show the filter name and then throw an error * Apply reinterpret before multiplication in fixed-scale-offset filter * Only reinterpret negative integers when decoding fill values to unsigned * Revert "Only reinterpret negative integers when decoding fill values to unsigned" This reverts commit 24a68e6e3. * let Fletcher32 operate on n-dimensional arrays not just vectors, as it was previously constrained to * fix FixedScaleOffset in many ways - Never use reinterpret - use array comprehensions... (continued)
126 of 160 new or added lines in 8 files covered. (78.75%)
908 of 1054 relevant lines covered (86.15%)
14286.37 hits per line
| Lines | Coverage | ∆ | File |
|---|---|---|---|
| 1 |
97.78 |
src/Filters/vlenfilters.jl | |
| 2 |
94.59 |
src/Filters/shuffle.jl | |
| 3 |
77.65 |
-2.6% | src/metadata.jl |
| 4 |
71.43 |
src/Filters/Filters.jl | |
| 4 |
76.47 |
src/Filters/delta.jl | |
| 5 |
88.1 |
src/Filters/fletcher32.jl | |
| 5 |
75.0 |
src/Filters/quantize.jl | |
| 10 |
52.38 |
src/Filters/fixedscaleoffset.jl |
| ID | Job ID | Ran | Files | Coverage | |
|---|---|---|---|---|---|
| 1 | 12200200130.1 | 23 |
85.11 |
GitHub Action Run | |
| 2 | 12200200130.2 | 23 |
85.34 |
GitHub Action Run | |
| 3 | 12200200130.3 | 23 |
85.34 |
GitHub Action Run | |
| 4 | 12200200130.4 | 23 |
85.11 |
GitHub Action Run | |
| 5 | 12200200130.5 | 23 |
85.11 |
GitHub Action Run | |
| 6 | 12200200130.6 | 23 |
85.34 |
GitHub Action Run |
| Coverage | ∆ | File | Lines | Relevant | Covered | Missed | Hits/Line |
|---|