`[` array indexing fails for non-contiguous array indices

[LTLA]

First, let's mock up a dataset:

library(tiledb)

path <- "mock"
dom <- tiledb_domain(dims = list(
    tiledb_dim("d1", c(1L, 2000L), 2000L, type = "INT32"),
    tiledb_dim("d2", c(1L, 1500L), 1500L, type = "INT32")
))
schema <- tiledb_array_schema(
    dom,
    attrs = list(tiledb_attr("x", type = "FLOAT64")),
    sparse = TRUE
)
tiledb_array_create(path, schema)

smat <- Matrix::rsparsematrix(2000, 1500, density=0.2, repr="T")
arr <- tiledb_array(path, query_type = "WRITE")
arr[] <- data.frame(d1 = smat@i + 1L, d2 = smat@j + 1L, x = smat@x)
tiledb_array_close(arr)

Now, extracting a submatrix with consecutive indices is fine:

arr <- tiledb_array(path, query_type = "READ")
arr[1:10,1:10]
## $d1
##  [1]  4  7 10  4  7  9  1  7  3  5  7 10  9  1 10  5  8  2  3  4  5  6  8  1  6
## [26]  5  7 10
## 
## $d2
##  [1]  1  1  1  2  2  2  3  3  4  4  4  4  5  6  6  7  7  8  8  8  8  8  8  9  9
## [26] 10 10 10
## 
## $x
##  [1] -0.00081 -0.27000 -1.00000  1.30000 -0.31000  1.60000 -1.40000  0.02100
##  [9] -0.99000 -0.82000 -0.81000 -0.81000  1.10000  1.40000  0.49000 -1.40000
## [17] -0.17000 -0.79000  0.80000  0.03000  0.47000  0.36000 -1.40000  1.60000
## [25] -0.30000  1.00000  0.79000  0.76000
## 
## attr(,"query_status")
## [1] "COMPLETE"

But if we have non-contiguous indices, we get an incorrect result. See how there are odd numbers showing up in d1 despite the fact that only even indices in the first dimension were requested. It seems like the request is improperly collapsed to the range of requested indices (i.e., [2, 10]).

arr[c(2,4,6,8,10),1:10]
## $d1
##  [1]  4  7 10  4  7  9  7  3  5  7 10  9 10  5  8  2  3  4  5  6  8  6  5  7 10
## 
## $d2
##  [1]  1  1  1  2  2  2  3  4  4  4  4  5  6  7  7  8  8  8  8  8  8  9 10 10 10
## 
## $x
##  [1] -0.00081 -0.27000 -1.00000  1.30000 -0.31000  1.60000  0.02100 -0.99000
##  [9] -0.82000 -0.81000 -0.81000  1.10000  0.49000 -1.40000 -0.17000 -0.79000
## [17]  0.80000  0.03000  0.47000  0.36000 -1.40000 -0.30000  1.00000  0.79000
## [25]  0.76000
## 
## attr(,"query_status")
## [1] "COMPLETE"

FWIW selected_ranges() does the right thing, but if [ is not intended to support non-consecutive indices, then the method should throw an error rather than silently returning the wrong result.

selected_ranges(arr) <- list(cbind(c(2,4,6,8,10), c(2,4,6,8,10)), cbind(1, 10))
arr[]
## $d1
##  [1]  4 10  4 10 10  8  2  4  6  8  6 10
## 
## $d2
##  [1]  1  1  2  4  6  7  8  8  8  8  9 10
## 
## $x
##  [1] -0.00081 -1.00000  1.30000 -0.81000  0.49000 -0.17000 -0.79000  0.03000
##  [9]  0.36000 -1.40000 -0.30000  0.76000
## 
## attr(,"query_status")
## [1] "COMPLETE"

Session information

R Under development (unstable) (2026-02-19 r89439)
Platform: x86_64-pc-linux-gnu
Running under: Ubuntu 22.04.5 LTS

Matrix products: default
BLAS:   /home/luna/Software/R/trunk/lib/libRblas.so 
LAPACK: /home/luna/Software/R/trunk/lib/libRlapack.so;  LAPACK version 3.12.1

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
 [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8    
 [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
 [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                 
 [9] LC_ADDRESS=C               LC_TELEPHONE=C            
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       

time zone: Australia/Sydney
tzcode source: system (glibc)

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] RcppSpdlog_0.0.28 tiledb_0.33.0    

loaded via a namespace (and not attached):
 [1] zoo_1.8-15          bit_4.6.0           compiler_4.6.0     
 [4] Matrix_1.7-5        tools_4.6.0         RcppCCTZ_0.2.14    
 [7] spdl_0.0.5          Rcpp_1.1.1          nanoarrow_0.8.0    
[10] bit64_4.6.0-1       nanotime_0.3.13     grid_4.6.0         
[13] data.table_1.18.2.1 lattice_0.22-9