error with lasnormalize - output NULL

[mkieser]

I've followed the doc to load a lidar dataset, build a dtm, and then normalize the lidar dataset based on the dtm but the result is "NULL". screen capture

lidar = readLAS("drawno.laz")`
lidar
class        : LAS (LASF v1.2)
memory       : 8.1 Mb
extent       : 278200 , 278300 , 602200 , 602300 (xmin, xmax, ymin, ymax)
area         : 9996.2 m² (convex hull)
points       : 150395 points,  107991 pulses
density      : 15.05 points/m²,  10.8 pulses/m²
field names  : X Y Z gpstime Intensity ReturnNumber NumberOfReturns Classification ScanAngle pulseID
coord. ref.  : NA

dtm = grid_terrain(lidar, method = "kriging", k = 10L)
[using universal kriging]
Warning message:
There were 3 duplicated ground points. Some X Y coordinates were repeated but with different Z coordinates. min Z were retained.
`dtm` 
  X        Y      Z
  1: 278200.5 602200.5 94.656
  2: 278201.5 602200.5 94.768
  3: 278202.5 602200.5 94.833
  4: 278203.5 602200.5 94.754
  5: 278204.5 602200.5 94.668
  9996: 278295.5 602299.5 93.513
  9997: 278296.5 602299.5 93.405
  9998: 278297.5 602299.5 93.389
  9999: 278298.5 602299.5 93.466
  10000: 278299.5 602299.5 93.467

lnorm = lasnormalize(lidar, dtm)
Warning message:
23155 points below 0 found.
lnorm
NULL

[Jean-Romain]

I'm really sorry but I have to repeat myself. The doc is very clear:

The function returns NULL.

So yes lnorm is NULL. It is not a bug it is a feature.

However lidar is normalized. Try plot(lidar) (the topogrphy being relatively flat this is not obvious with this dataset)

[mkieser]

plot(lidar) error: xp_attach_gl_context returned: 2 X Error of failed request: 0 Major opcode of failed request: 149 (GLX) Minor opcode of failed request: 26 (X_GLXMakeContextCurrent) Serial number of failed request: 709 Current serial number in output stream: 709

[Jean-Romain]

That may be a real bug but it is really weird. I have never seen such bug in my whole R carrier. I need a reproducible example of your code as well as the original data (but if it is drawno I already have it).

What Syst.setenv(TZ = 'MST') means ?

[Jean-Romain]

Ok I googled error: xp_attach_gl_context returned: 2 and this sound to be OSX specific.

But your screen shoot shows that plot works isn't it ?

[mkieser]

TZ = 'MST' is just a reoccurring timezone issue with 'sf', trying to resolve separately

I think I got it: dtm = grid_terrain(lidar, method = "kriging", k = 10L) lasnormalize(lidar, dtm)

Yes must be an OS thing R crashes when I use plot(lidar), going to try Docker instead.

[Jean-Romain]

Ok.

Does example(rgl) works. If yes this is a bug in lidR if no this is a bug in rgl with OSX and openGL