podpac.data.DataSource

class podpac.data.DataSource(**kwargs)[source]

Bases: podpac.core.node.Node

Base node for any data obtained directly from a single source.

Parameters
  • source (Any) – The location of the source. Depending on the child node this can be a filepath, numpy array, or dictionary as a few examples.

  • native_coordinates (podpac.Coordinates) – The coordinates of the data source.

  • interpolation (str, dict, optional) –

    {interpolation}

    If input is a string, it must match one of the interpolation shortcuts defined in podpac.data.INTERPOLATION_SHORTCUTS. The interpolation method associated with this string will be applied to all dimensions at the same time.

    If input is a dict, the dict must contain one of two definitions:

    1. A dictionary which contains the key 'method' defining the interpolation method name. If the interpolation method is not one of podpac.data.INTERPOLATION_SHORTCUTS, a second key 'interpolators' must be defined with a list of podpac.interpolators.Interpolator classes to use in order of uages. The dictionary may contain an option 'params' key which contains a dict of parameters to pass along to the podpac.interpolators.Interpolator classes associated with the interpolation method. This interpolation definition will be applied to all dimensions.

    2. A dictionary containing an ordered set of keys defining dimensions and values defining the interpolation method to use with the dimensions. The key must be a string or tuple of dimension names (i.e. 'time' or ('lat', 'lon') ). The value can either be a string matching one of the interpolation shortcuts defined in podpac.data.INTERPOLATION_SHORTCUTS or a dictionary meeting the previous requirements (1). If the dictionary does not contain a key for all unstacked dimensions of the source coordinates, the podpac.data.INTERPOLATION_DEFAULT value will be used. All dimension keys must be unstacked even if the underlying coordinate dimensions are stacked. Any extra dimensions included but not found in the source coordinates will be ignored.

    If input is a podpac.data.Interpolation class, this interpolation class will be used without modification.

  • nan_vals (List, optional) – List of values from source data that should be interpreted as ‘no data’ or ‘nans’

  • coordinate_index_type (str, optional) – Type of index to use for data source. Possible values are ['list', 'numpy', 'xarray', 'pandas'] Default is ‘numpy’

Notes

Custom DataSource Nodes must implement the get_data() and get_native_coordinates() methods.

Methods

__init__(**kwargs)

Do not overwrite me

create_output_array(coords[, data])

Initialize an output data array

eval(coordinates[, output])

Evaluates this node using the supplied coordinates.

eval_group(group)

Evaluate the node for each of the coordinates in the group.

find_coordinates()

Get the available native coordinates for the Node.

from_url(url)

Create podpac Node from a WMS/WCS request.

get_cache(key[, coordinates])

Get cached data for this node.

get_data(coordinates, coordinates_index)

This method must be defined by the data source implementing the DataSource class.

get_native_coordinates()

Returns a Coordinates object that describes the native coordinates of the data source.

has_cache(key[, coordinates])

Check for cached data for this node.

init()

Overwrite this method if a node needs to do any additional initialization after the standard initialization.

put_cache(data, key[, coordinates, overwrite])

Cache data for this node.

rem_cache(key[, coordinates, mode])

Clear cached data for this node.

Attributes

base_definition

Base node defintion for DataSource nodes.

base_ref

Default pipeline node reference/name in pipeline node definitions

cache_ctrl

A trait whose value must be an instance of a specified class.

cache_output

A boolean (True, False) trait.

cache_update

A boolean (True, False) trait.

coordinate_index_type

An enum whose value must be in a given sequence.

definition

Full pipeline definition for this node.

dtype

A trait which allows any value.

hash

interpolation

A trait type representing a Union type.

interpolation_class

Get the interpolation class currently set for this data source.

interpolators

Return the interpolators selected for the previous node evaluation interpolation.

json

definition for this node in json format

json_pretty

nan_vals

An instance of a Python list.

native_coordinates

A trait whose value must be an instance of a specified class.

pipeline

Create a pipeline node from this node

source

A trait which allows any value.

style

A trait whose value must be an instance of a specified class.

units

A trait for unicode strings.

Members

__init__(**kwargs)

Do not overwrite me

property base_definition

Base node defintion for DataSource nodes.

Returns

Dictionary containing the location of the Node, the name of the plugin (if required), as well as any parameters and attributes that were tagged by children.

Return type

OrderedDict

coordinate_index_type

An enum whose value must be in a given sequence.

eval(coordinates, output=None)[source]

Evaluates this node using the supplied coordinates.

The native coordinates are mapped to the requested coordinates, interpolated if necessary, and set to _requested_source_coordinates with associated index _requested_source_coordinates_index. The requested source coordinates and index are passed to get_data() returning the source data at the native coordinatesset to _requested_source_data. Finally _requested_source_data is interpolated using the interpolate method and set to the output attribute of the node.

Parameters
  • coordinates (podpac.Coordinates) –

    The set of coordinates requested by a user. The Node will be evaluated using these coordinates.

    An exception is raised if the requested coordinates are missing dimensions in the DataSource. Extra dimensions in the requested coordinates are dropped.

  • output (podpac.UnitsDataArray, optional) – Default is None. Optional input array used to store the output data. When supplied, the node will not allocate its own memory for the output array. This array needs to have the correct dimensions, coordinates, and coordinate reference system.

Returns

Unit-aware xarray DataArray containing the results of the node evaluation.

Return type

podpac.UnitsDataArray

Raises

ValueError – Cannot evaluate these coordinates

find_coordinates()[source]

Get the available native coordinates for the Node. For a DataSource, this is just the native_coordinates.

Returns

coords_list – singleton list containing the native_coordinates (Coordinates object)

Return type

list

get_data(coordinates, coordinates_index)[source]

This method must be defined by the data source implementing the DataSource class. When data source nodes are evaluated, this method is called with request coordinates and coordinate indexes. The implementing method can choose which input provides the most efficient method of getting data (i.e via coordinates or via the index of the coordinates).

Coordinates and coordinate indexes may be strided or subsets of the source data, but all coordinates and coordinate indexes will match 1:1 with the subset data.

This method may return a numpy array, an xarray DaraArray, or a podpac UnitsDataArray. If a numpy array or xarray DataArray is returned, podpac.data.DataSource.evaluate() will cast the data into a UnitsDataArray using the requested source coordinates. If a podpac UnitsDataArray is passed back, the podpac.data.DataSource.evaluate() method will not do any further processing. The inherited Node method create_output_array can be used to generate the template UnitsDataArray in your DataSource. See podpac.Node.create_output_array() for more details.

Parameters
  • coordinates (podpac.Coordinates) – The coordinates that need to be retrieved from the data source using the coordinate system of the data source

  • coordinates_index (List) – A list of slices or a boolean array that give the indices of the data that needs to be retrieved from the data source. The values in the coordinate_index will vary depending on the coordinate_index_type defined for the data source.

Returns

A subset of the returned data. If a numpy array or xarray DataArray is returned, the data will be cast into UnitsDataArray using the returned data to fill values at the requested source coordinates.

Return type

np.ndarray, xr.DataArray, podpac.UnitsDataArray

Raises

NotImplementedError – This needs to be implemented by derived classes

get_native_coordinates()[source]

Returns a Coordinates object that describes the native coordinates of the data source.

In most cases, this method is defined by the data source implementing the DataSource class. If method is not implemented by the data source, it will try to return self.native_coordinates if self.native_coordinates is not None.

Otherwise, this method will raise a NotImplementedError.

Returns

The coordinates describing the data source array.

Return type

podpac.Coordinates

Notes

Need to pay attention to: - the order of the dimensions - the stacking of the dimension - the type of coordinates

Coordinates should be non-nan and non-repeating for best compatibility

Raises

NotImplementedError – Raised if get_native_coordinates is not implemented by data source subclass.

interpolation

A trait type representing a Union type.

property interpolation_class

Get the interpolation class currently set for this data source.

The DataSource interpolation property is used to define the podpac.data.Interpolation class that will handle interpolation for requested coordinates.

Returns

Interpolation class defined by DataSource interpolation definition

Return type

podpac.data.Interpolation

property interpolators

Return the interpolators selected for the previous node evaluation interpolation. If the node has not been evaluated, or if interpolation was not necessary, this will return an empty OrderedDict

Returns

Key are tuple of unstacked dimensions, the value is the interpolator used to interpolate these dimensions

Return type

OrderedDict

nan_vals

An instance of a Python list.

native_coordinates

A trait whose value must be an instance of a specified class.

The value can also be an instance of a subclass of the specified class.

Subclasses can declare default classes by overriding the klass attribute

source

A trait which allows any value.