Command line mode and classification

[daniel]

Hi,

- Nope, there's no other way to colorize a cloud in command line mode right now
- what is the organization of your ASCII file(s)? Because CC is able to automatically validate some evident cases, but as soon as there's an uncertainty it will pop up the confirmation dialog.

[Haowyr]

The source files looks like this

1048.4250 -251.1590 51.1520 101 0 2 1.00 480570.94 83.00 5.00

But it changes during the process as it is processed through both cc command line and canupo command line so some scalar fields are added during this process. I guess I won't have any automatic validation from cc with this type of changing files.

[daniel]

So I guess it's X Y Z R G B and some scalar fields? Indeed the fact that it changes all the time complicates the process.

We could in the best case add an option that tells CC to treat all columns as scalar fields (but the 3 first ones). It would work if you don't use the colors or normals in your process. And if you don't need a header line on output...

[Haowyr]

For now I don't use either of them in my script, more focused on scalar fields created by canupo and some density created via cc and I will clearly never need any header in any of my files.

[sctsmth]

hello,
this question is somewhat related to thread as it's to the command line mode...
trying to run the subsample command on a large file, is .e57 not supported in command line mode for subsample? I get a file recognition type error. if not supported any tips on converting such a large file to a format for CloudCompare command line? or other route to subsampling.
best
Scott

[daniel]

Well e57 files should open fine even in command line mode. Do you manage to open it in graphical mode? What version are you using? On which OS?

[sctsmth]

CloudCompare version 2.6.1 (windows 64 bit) on Windows 7, I haven't been able to open the e57 in graphical mode as it's too large (269 gbs)

[daniel]

269 Gb?! I guess you'll hit the actual limit of 4 billion points (due to 32 bits indexing) anyway...

Still the file recognition type error is a bit weird though (but maybe libE57 sends an error that is misinterpreted by CC).

[sctsmth]

yes it is much too large! but the error pops up straight away, unless cc knows its too large, the file seems ok as lastools (e572las) started a command line processing (splitting off the different scans from the e57) saving as laz files, however then there is the problem of integrating them together. I am trying to split off each building of the whole station ultimately