Helper function to turn diverse inputs into neuroglancer segment ids

ngl_segments<- replaces neuroglancer segments in a neuroglancer scene parsed by ngl_decode_scene.

Usage

ngl_segments(
  x,
  as_character = TRUE,
  include_hidden = FALSE,
  must_work = TRUE,
  unique = FALSE,
  ...
)

ngl_segments(x) <- value

Arguments

x: Neuroglancer ids either as a vector of ids (character vector or integer64 recommended) OR a scene specification as raw JSON. format (character vector), the path to a file on disk, a neuroglancer scene URL (which embeds a JSON scene specification in a single URL), or an R list generated by parsing one of the above.
as_character: Whether to return segments as character rather than numeric vector (the default is character for safety).
include_hidden: Whether to include hiddenSegments (typically for flywire scenes).
must_work: if TRUE, the default, then an error will be generated if the scene has no segments or if invalid ids are present. Explicit NAs will trigger this error when must_work=T, but will be returned as id 0 without error when must_work=F.
unique: When TRUE drops any duplicated ids with a warning
...: Additional arguments passed to ngl_decode_scene
value: Segment ids in any form understandable by ngl_segments. This can include character/numeric/int64 ids, a URL, parsed neuroglancer scene etc.

Value

Numeric (or character) vector of segment ids, taken from the first segmentation layer (with a warning) if the scene contains more than one.

Details

For simple vector inputs, ngl_segments will treat the values "NA", "NAN", "NULL" (all case insensitive) and "" as NA.

ngl_segments<- chooses the FlyWire style segmentation_with_graph layer if it exists otherwise the first visible segmentation layer. Note that hiddenSegment will be removed in this process.

Examples

# -> character
ngl_segments(c(10950626347, 10952282491, 13307888342))
#> [1] "10950626347" "10952282491" "13307888342"
# turns these into numeric
ngl_segments(c("10950626347", "10952282491", "13307888342"), as_character=FALSE)
#> [1] 10950626347 10952282491 13307888342

# \donttest{
u="https://ngl.flywire.ai/?json_url=https://globalv1.flywire-daf.com/nglstate/5409525645443072"
ngl_segments(u, as_character = TRUE)
#> [1] "720575940621039145" "720575940626877799"
sc=ngl_decode_scene(u)
# set segments
ngl_segments(sc) <- c("720575940621039145")
# or a shortcut to add ids
sc=sc+c("720575940621039145", "720575940626877799")
sc
#> neuroglancer scene with 2 layers and 2 segments (of which 2 shown)
#>                                 name                    type visible nsegs
#> 1                   Production-image                   image    TRUE     0
#> 2 Production-segmentation_with_graph segmentation_with_graph    TRUE     2
#>   nhidden
#> 1       0
#> 2       0
#>                                                                        source
#> 1 precomputed://gs://microns-seunglab/drosophila_v0/alignment/image_rechunked
#> 2          graphene://https://prodv1.flywire-daf.com/segmentation/1.0/fly_v31
if (FALSE) { # \dontrun{
# paste resultant URL to clipboard to use in neuroglancer
clipr::write_clip(as.character(sc))
} # }

# you can also modify the URL directly
ngl_segments(u)=c("720575940621039145", "720575940626877799")
# }

if (FALSE) { # \dontrun{
browseURL(u)

## Summary of different classes of input
# from clipboard
ngl_segments(clipr::read_clip())
# URL
ngl_segments("<ngl-scene-url>")
# path to file on disk
ngl_segments("/path/to/scene.json")
# R list
ngl_segments(scenelist)
} # }