Running just step 4 and 5 crashes.

[GoogleCodeExporter]

drj@millstone:~/ccc-gistemp$ time nice python tool/run.py -s 4,5
====> STEPS 4, 5  ====
No more recent sea-surface data files.

No land data; using ocean data only
Step4: closing output file
WARNING: Bad mix of land and ocean data.
  Land range from 9998-12 to 0000-01; Ocean range from 1880-01 to 2010-12.
Traceback (most recent call last):
  File "tool/run.py", line 280, in <module>
    sys.exit(main())
  File "tool/run.py", line 261, in main
    data = step_fn[step](data)
  File "tool/run.py", line 104, in run_step5
    result = step5.step5(data)
  File "/home/drj/ccc-gistemp/code/step5.py", line 541, in step5
    land,ocean,mixed = as_boxes(subboxes)
  File "/home/drj/ccc-gistemp/code/step5.py", line 36, in as_boxes
    return land_ocean_boxes(meta, data)
  File "/home/drj/ccc-gistemp/code/step5.py", line 121, in land_ocean_boxes
    return iterland(), iterocean(), itermixed()
  File "/home/drj/ccc-gistemp/code/step5.py", line 118, in itermixed
    (land_meta.gridding_radius, str(year_min), str(year_max)))
AttributeError: 'SubboxMetaData' object has no attribute 'gridding_radius'

real    0m16.480s
user    0m15.289s
sys 0m0.540s

Original issue reported on code.google.com by drj...@googlemail.com on 8 Feb 2011 at 9:00

[GoogleCodeExporter]

Happens because running Step 4 first synthesizes a blank land stream (and this 
was how we used to do "ocean only" analyses).  The synthesized stream does not 
have the gridding_radius set on its metadata.

We don't need a special hack for generating ocean only analyses now.  They are 
produced routinely by Step 5.  Change Step 4 so that it copies land data from 
the SBBX file.

Original comment by drj...@googlemail.com on 8 Feb 2011 at 9:02

[GoogleCodeExporter]

This issue was closed by revision r695.

Original comment by d...@ravenbrook.com on 8 Feb 2011 at 11:02

• Changed state: Fixed