`align_move()` fails for projected data with identical track length

[robe2037]

When input data are projected, align_move() can fail if the number of interpolated points along each track in the same.

Reprex:

library(moveVis)
#> Find out about new features added to moveVis at
#> https://movevis.org/news/index.html

data("move_data")

# Create sample dataset with identical track lengths:
d <- move_data |> 
  dplyr::group_split(track) |> 
  purrr::map(~ .x[1:10, ]) 

ts <- seq.POSIXt(
  as.POSIXct("2018-05-15 08:00:00", tz = "UTC"), 
  as.POSIXct("2018-05-24 08:00:00", tz = "UTC"), 
  length.out = 10
)

d[[1]]$timestamp <- ts
d[[2]]$timestamp <- ts
d[[3]]$timestamp <- ts

d <- move2::mt_as_move2(
  purrr::list_rbind(d), 
  time_column = "timestamp", 
  track_id_column = "track"
)

# WGS84 succeeds
x <- align_move(d)
#> Temporal resolution of 1 [d] is used to align trajectories.

# Projected CRS fails because CRS is dropped
x <- align_move(sf::st_transform(d, "epsg:32637"))
#> Temporal resolution of 1 [d] is used to align trajectories.
#> Error: arguments have different crs

^{Created on 2025-10-15 with reprex v2.1.1}

If the projected data don't end up with the same number of interpolated points, align_move() works as expected, so this is an edge-case issue. Still, it seems like it would be easy to resolve in these lines:

moveVis/R/align_move.R

Lines 147 to 151 in cf88cdc

	if(all(st_is_longlat(m), sf_use_s2())){
	m_aligned <- mapply(.m = m_sf_lines, .nd = nd, function(.m, .nd) st_as_sf(s2_interpolate_normalized(st_geometry(.m), .nd)), SIMPLIFY = T)
	} else{
	m_aligned <- mapply(.m = m_sf_lines, .nd = nd, function(.m, .nd) st_line_interpolate(st_geometry(.m), .nd), SIMPLIFY = T)
	}

For projected CRS, st_line_interpolate() returns objects of "sfc_POINT". When each of the tracks has the same number of these points, mapply() is willing to coerce them to a rectangular matrix given that SIMPLIFY = T, but this drops CRS information. For geographic coordinates, st_as_sf() coerces the output elements to sf objects, which mapply() doesn't simplify to a matrix.

Forcing the interpolation code to return an sf object for projected coordinates would likely fix the issue.