The classes and functions in this module simplify basic access to CPOs.
Typically, to get a CPO from the UAL, one writes:
import ual
itm = ual.itm(17151, 4)
itm.open_env('klingshi', 'aug', '4.09a')
itm.edge.getSlice(2.5, ual.ualdef.CLOSEST_SAMPLE)
te = itm.edge.fluid.te.value.array[0].scalar[0]
Instead, you can write:
import itm.grid.cpo_tools as ct
c = ct.Cpo(17151, 4, 'edge', 2.5, 'klingshi', 'aug', '4.09a')
te=c.cpo.fluid.ne.value.array[0].scalar[0]
If the data in the UAL changes, you can reload the latest state by calling
c.reload()
The Cpo class makes handling complex grids and data stored on them easy.
c = ct.Cpo(17151, 4, 'edge', 2.5, 'klingshi', 'aug', '4.09a')
c.grid
returns a itm.grid.base.Grid object wrapping the grid for the CPO,
c.list_data()
returns a list of ScalarData objects for all fields in the CPO that are filled with data. Likewise,
c.list_data_lists()
returns a list of :class:’~itm.grid.data.ScalarDataList’ objects for all arrays of data fields that contain data.
Helper class wrapping a CPO data structure to add high-level functionality.
Creates an object wrapping an CPO with the given parameters. Shot number, run number and cpo name (e.g. ‘equilibrium’) have to be given. For time-dependent CPOs, time has to be given.
If user, tokamak, version are omitted, the values set in the environment are used (variables USER, TOKAMAKNAME, DATAVERSION).
The doOpen parameter controls whether UAL access is done immediately when creating the Cpo object. If it is set to False, UAL access is deferred to the first access to CPO data.
The useHDF5 property controls whether UAL access is done through HDF5. If set to True, the parameters user, tokamak and version have no effect.
Returns an identifier string for the CPO
Format: shot/run/time/name for time-dependent CPOs, shot/run/name for non-time-dependent CPOs.
The CPO data object associated with this CPO object as provided by the UAL Python interface.
Return a list of all complexgrid_scalar objects in this CPO that contain data as itm.grid.data.ScalarData objects.
If the optional parameter addPrefix=True is given, the CPO identifier as obtained by str() is prepended to the object paths.
Return a list of all arrays of complexgrid_scalar objects in this CPO that contain data as itm.grid.data.ScalarDataList objects.
If the optional parameter addPrefix=True is given, the CPO identifier as obtained by str() is prepended to the object paths.
The CpoDescriptor class describes a set of CPOs, and can create the Cpo objects for them.
Say you have a specific shot, for which different versions of the edge cpo were stored with different run values. The following will give you a list of three Cpo objects for the runs 2, 3 and 4:
In [1]: import itm.grid.cpo_tools as ct
In [2]: cd = ct.CpoDescriptor(17151, (2,3,4), 'edge', 2.5, 'klingshi', 'aug')
CpoDescriptor then behaves like a sequence of Cpo objects, and individual Cpo objects can be accessed by their (zero-based, as usual in Python) index:
In [3]: len(cd)
Out[3]: 3
In [10]: print cd[1]
17151/3/2.5/edge
In [4]: for c in cd:
...: print str(c)
...:
17151/2/2.5/edge
17151/3/2.5/edge
17151/4/2.5/edge
For every parameter to the CpoDescriptor you can give a list of values. For example, to get the same CPO for different users:
cd = ct.CpoDescriptor(17151, 'edge', 2.5, ('klingshi', 'coster'), 'aug')
Helper class describing one or more CPOs and presenting them as a sequence.
It holds combinations of shot/run number(s), CPO name(s), time stamp(s), user/tokamak/version.
Create a CPO Descriptor.
Every parameter to the constructor can be either a single value or a tuple. The CpoDescriptor then describes all possible combinations of these parameters. The order of the CPOs is obtained by iterating the leftmost parameter first.
If one of user, tokamak, version are omitted, the resulting CPO will take the values set in the environment (variables USER, TOKAMAKNAME, DATAVERSION).
If doOpen = False is given, UAL access for the resulting CPOs is deferred to the first method call that accesses CPO data.
If useHDF5 = True is specified, UAL access is done via HDF5 instead of MDSPlus. In this case, user/tokamak/version have no effect.
Return a string representation in the form shot/run/time/cpo name/user/tokamak/version, where individual values are either scalars or tuples.