I'm using ensembldb to construct some ranges, and then was using select() to select some metadata columns.

I noticed that the transcripts I was working with have a metadata column width, not the same as the width of the range. This then interferes with select():

> gr <- GRanges("chr1",IRanges(1:3,width=10))
> mcols(gr)$width <- 7
> mcols(gr)$test <- 0
> mcols(gr)$test2 <- "a"
> gr
GRanges object with 3 ranges and 3 metadata columns:
      seqnames    ranges strand |     width      test       test2
         <Rle> <IRanges>  <Rle> | <numeric> <numeric> <character>
  [1]     chr1      1-10      * |         7         0           a
  [2]     chr1      2-11      * |         7         0           a
  [3]     chr1      3-12      * |         7         0           a
  -------
  seqinfo: 1 sequence from an unspecified genome; no seqlengths
> gr %>% select(test2)
GRanges object with 3 ranges and 1 metadata column:
      seqnames    ranges strand |     test2
         <Rle> <IRanges>  <Rle> | <numeric>
  [1]     chr1      1-10      * |         0
  [2]     chr1      2-11      * |         0
  [3]     chr1      3-12      * |         0
  -------
  seqinfo: 1 sequence from an unspecified genome; no seqlengths

This is using current release/devel branch.

I wonder if select() should complain about metadata columns with names shared with core variables at the outset to avoid this issue?