search

hypertidy / lazyraster

raster data via GDAL on-demand
https://hypertidy.github.io/lazyraster/
27 stars 1 forks source link

missing values problem #8

Closed mdsumner closed 5 years ago

mdsumner commented 5 years ago

Ok when no missing values: 


s <- vapour::vapour_sds_names(raadfiles::altimetry_daily_files()$fullname[1])$subdataset[4]

plot(lazycrop(lazyraster(s), extent(100, 120, -50, -40)))
Warning messages:
1: In raster_info_cpp(filename = datasourcename, min_max = min_max) :
  GDAL Message 1: The dataset has several variables that could be identified as vector fields, but not all share the same primary dimension. Consequently they will be ignored.
2: In raster_info_cpp(filename = datasourcename, min_max = min_max) :
  GDAL Message 1: The dataset has several variables that could be identified as vector fields, but not all share the same primary dimension. Consequently they will be ignored.

Not ok when we traverse missing pixels (no plot)

plot(lazycrop(lazyraster(s), extent(100, 220, -50, -40)))
Warning messages:
1: In raster_info_cpp(filename = datasourcename, min_max = min_max) :
  GDAL Message 1: The dataset has several variables that could be identified as vector fields, but not all share the same primary dimension. Consequently they will be ignored.
2: In raster_info_cpp(filename = datasourcename, min_max = min_max) :
  GDAL Message 1: The dataset has several variables that could be identified as vector fields, but not all share the same primary dimension. Consequently they will be ignored.
3: In x - r[1] : NAs produced by integer overflow
4: In r[2] - r[1] : NAs produced by integer overflow
mdsumner commented 5 years ago

Fixed in 38c2e58

We'll see if that's a good idea or not

mdsumner commented 5 years ago

Probably better if we get the missing value itself on vapour_raster_info