Seriously out-of-date "UV-CDAT Utilities - CHAPTER 3" documentation

[jypeter]

@doutriaux1 @davis278 A colleague asked some questions about the eofs package and I googled cdat eof after answering, to see what CDAT was providing. I landed on https://cdat.llnl.gov/documentation/utilities/utilities-3.html that gives examples using from eof import Eof and a data file from Paul Dubois (f=cdms.open('/home/dubois/clt.nc'), using cdms and not cdms2, this is old!)

You can load eofs, though

jypeter@obelix3 - ...jypeter - 43 >conda activate cdat-8.1_py2
(cdat-8.1_py2) jypeter@obelix3 - ...jypeter - 44 >python
Python 2.7.15 | packaged by conda-forge | (default, Feb 28 2019, 04:00:11)
[GCC 7.3.0] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import eof
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ImportError: No module named eof
>>> import eofs
>>>

I also failed to import other packages mentioned on this page, but maybe they are now somewhere deeper (dir(cdutil) and dir(genutil) did not show anything, and I did not dig deeper)

>>> import trends
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ImportError: No module named trends
>>> import ort
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ImportError: No module named ort
>>> import asciidata
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ImportError: No module named asciidata
>>> import binaryio
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ImportError: No module named binaryio

[doutriaux1]

@jypeter we stopped supporting these a few years back, and we had no requests for them until today. You can find the source package under the contrib packages directory on the cdat repo:
(see: https://github.com/CDAT/cdat/tree/master/contrib)
Also the conda recipes for most of them can be found there: https://github.com/CDAT/conda-recipes

But again has not been updated in years.

[doutriaux1]

@davis278 when porting cdutil doc to a notebook, it might be worth to mention these links and the fact that these package are no longer part of CDAT as they were external contributions to CDAT and therefore should be maintained via conda by their original developers.

[jypeter]

If these packages are no longer supported, you might as well remove them from the documentation. Trimming the documentation where it does not have anything to do with the current version is the way to go

Oh, I have just had a look at the averager documentation in chapter 1 that we definitely want to keep, and I see lots of references to cdms instead of cdms2 that can probably be easily corrected

[doutriaux1]

@jypeter it is on our todo list to update the cdutil documentation. WE're getting closer. But feel free to help us and take the cdutil doc and put it into a notebook, this will save us a great amount of time!

[jypeter]

I was about to report that problem again and found out that I had already opened this issue!

I wonder if you can just completely remove Chapter 3, if @doutriaux1 can confirm that no User Contributed Package is available any more

You could move the index at the end of Chapter 3 to the end of Chapter 2, but I'm not sure it's worth it, since the Table of Contents is more useful

[doutriaux1]

@jypeter you're right we should remove this.