Hi Nathan,

That approach sounds good. What do you think about this implementation?
import h5py
import yt 
 
ds = yt.load('RD0009/RD0009')
hc_file = h5py.File('halo_catalogs/catalog/catalog.0.h5') 
 
# Include unit_registry as an attribute
hc_file.attrs['unit_registry'] = ds.unit_registry.to_json() 
 
# Or as a string dataset
unit_registry_json = ds.unit_registry.to_json()
str_type = h5py.special_dtype(vlen=str)
unit_registry_h5 = hcfile.create_dataset('unit_registry_json', shape=(1,), dtype=str_type)
unit_registry_h5[:] = unit_registry_json 
 
hc_file.close() 
 
# Regenerate unit_registry when loading the HaloCatalog?
halos_ds = yt.load('halo_catalogs/catalog/catalog.0.h5')
hc_file = h5py.File('halo_catalogs/catalog/catalog.0.h5')
unit_registry_json = hc_file.attrs['unit_registry']
halos_ds.unit_registry = yt.units.unit_registry.UnitRegistry.from_json(unit_registry_json)


I'm not sure if including the unit_registry as an attribute or string dataset is a better idea -- it seems to be an attribute in other parts of the codebase. 

Then, how do we get it to take the place of the halos_ds unit_registry? Should this be done on creation in HaloCatalogDataset? (i.e., check if the h5 file has a unit_registry attr, if so, load that, if not use the current defaults?)


Thanks,

Rasmi

On Wed, Jun 1, 2016 at 1:01 PM, Nathan Goldbaum <nathan12343@gmail.com> wrote:


On Wed, Jun 1, 2016 at 11:40 AM, Rasmi Elasmar <re2300@columbia.edu> wrote:
Hi all,

Greg and I found a bug involving halo catalog unit handling:

>>> halo.quantities['particle_position_x']
0.495074982741 cm
>>> halo.quantities['particle_position_x'].in_units('code_length')
0.495074982741 code_length
>>> halo.quantities['particle_position_x'].in_units('cm')
0.495074982741 cm
>>> ds.unit_registry['code_length']
(9.195880139956267e+25, (length))
>>> halos_ds.unit_registry['code_length']
(1.0, (length))

The halos_ds mixes up cm and code_length units when the HaloCatalog object is created from a saved halo catalog. The halo catalog values are saved in code_length, but the HaloCatalog object assumes they are in cm. 

Here is where the code_length units are written out after halo-finding is done (this is confirmed with an h5ls).
Here is where the halos_ds for the HaloCatalog is created. The length unit is set in cm -- the catalog is assumed to be in cgs.
The HaloCatalog fields also assume cgs.

In theory, the HaloCatalog could just parse the code_length units of the halos_ds, but this isn't necessarily known at the time of creation, so the ideal fix may be to save the halo catalog length units in cm instead of in code_length. Then the assumptions that are made about length being in cm when creating a HaloCatalog object from a halo catalog would be correct. Any thoughts on this approach or other approaches?


It would probably be better just to save the unit registry into the hdf5 file. You might find the UnitRegistry.to_json() and UnitRegistry.from_json() to be useful here - the json data could be saved in the HDF5 output file as a string dataset.



Thanks,

Rasmi


_______________________________________________
yt-dev mailing list
yt-dev@lists.spacepope.org
http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org



_______________________________________________
yt-dev mailing list
yt-dev@lists.spacepope.org
http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org