Index into a CartesianIndices with a Block

[jishnub]

After this, indexing into a CartesianIndices with a Block produces a CartesianIndices if the ndims match.

julia> B = BlockArray(reshape([1:9;],3,3), [2,1], [2,1])
2×2-blocked 3×3 BlockMatrix{Int64}:
 1  4  │  7
 2  5  │  8
 ──────┼───
 3  6  │  9

julia> C = CartesianIndices(B)
CartesianIndices((1:1:3, 1:1:3))

julia> C[Block(1,1)]
CartesianIndices((1:2, 1:2))

This addresses the comment in #346 (comment)

The indexing operation C[Block(1), Block(1)] should also work identically, but this is WIP. works identically:

julia> C[Block(1), Block(1)]
CartesianIndices((1:2, 1:2))

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 94.47%. Comparing base (ea33af1) to head (53a2f7d).
Report is 8 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #354      +/-   ##
==========================================
+ Coverage   92.66%   94.47%   +1.81%     
==========================================
  Files          16       16              
  Lines        1499     1503       +4     
==========================================
+ Hits         1389     1420      +31     
+ Misses        110       83      -27     

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