Closed eric-czech closed 4 years ago
Merging #48 into master will increase coverage by
0.10%
. The diff coverage is100.00%
.
@@ Coverage Diff @@
## master #48 +/- ##
==========================================
+ Coverage 94.89% 95.00% +0.10%
==========================================
Files 10 10
Lines 392 400 +8
Branches 75 78 +3
==========================================
+ Hits 372 380 +8
Misses 10 10
Partials 10 10
Impacted Files | Coverage Δ | |
---|---|---|
rechunker/algorithm.py | 82.45% <ø> (ø) |
|
rechunker/api.py | 92.53% <100.00%> (+0.47%) |
:arrow_up: |
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 e454e85...ac03623. Read the comment docs.
Thanks @tomwhite. Let me know if there's anything else I can do to get this merged @TomAugspurger / @rabernat.
Thanks.
Thanks for all your contributions @eric-czech. It's great to have you involved in rechunker!
Sure thing @rabernat!
Fixes https://github.com/pangeo-data/rechunker/issues/46
This adds options to
rechunk
that are passed to zarr create methods. The only use cases I have in mind for this are allowing for overwrites and compression, but it may be useful in other ways. Otherwise it may eventually be worth lifting those parameters into therechunk
signature given that allowing any option to be passed is a bit of a footgun.Note: I had to modify one of the existing test fixtures since it was doing almost what I wanted for a compression test except that the chunks were too small for compression to take effect. I didn't know that occurred with zarr but I can't find any other explanation for why specifying a compressor makes no difference when an array is too small. I have no idea where the cutoff is but perhaps it is related to compression block sizes? Let me know if anybody understands that better.