ExternalArrayReference¶

class asdf.ExternalArrayReference(fileuri, target, dtype, shape)[source]¶

Bases: asdf.types.AsdfType

Store a reference to an array in an external File.

This class is a simple way of referring to an array in another file. It provides no way to resolve these references, that is left to the user. It also performs no checking to see if any of the arguments are correct. e.g. if the file exits.

Parameters

fileuri: `str`: The path to the path to be referenced. Can be relative to the file containing the reference.
target: `object`: Some internal target to the data in the file. Examples may include a HDU index, a HDF path or an asdf fragment.
dtype: `str`: The (numpy) dtype of the contained array.
shape: `tuple`: The shape of the array to be loaded.

Examples

>>> import asdf
>>> ref = asdf.ExternalArrayReference("myfitsfile.fits", 1, "float64", (100, 100))
>>> tree = {'reference': ref}
>>> with asdf.AsdfFile(tree) as ff:
...     ff.write_to("test.asdf")

Attributes Summary

`handle_dynamic_subclasses`
`has_required_modules`
`name`
`organization`
`requires`
`standard`
`supported_versions`
`types`
`validators`
`version`
`yaml_tag`

Methods Summary

`from_tree`(tree, ctx)	Converts basic types representing YAML trees into custom types.
`from_tree_tagged`(tree, ctx)	Converts from tagged tree into custom type.
`incompatible_version`(version)	Indicates if given version is known to be incompatible with this type.
`make_yaml_tag`(name[, versioned])	Given the name of a type, returns a string representing its YAML tag.
`names`()	Returns the name(s) represented by this tag type as a list.
`subclass`(*args[, attribute])	Decorator to enable serialization of a subclass of an existing type.
`subclass_property`(attribute)	Decorator to enable serialization of custom subclass attributes.
`tag_base`()	Returns the base of the YAML tag for types represented by this class.
`to_tree`(data, ctx)	Converts instances of custom types into YAML representations.
`to_tree_tagged`(node, ctx)	Converts instances of custom types into tagged objects.

Attributes Documentation

handle_dynamic_subclasses = False¶

has_required_modules = True¶

name = 'core/externalarray'¶

organization = 'stsci.edu'¶

requires = []¶

standard = 'asdf'¶

supported_versions = []¶

types = []¶

validators = {}¶

version = AsdfVersion('1.0.0')¶

yaml_tag = 'tag:stsci.edu:asdf/core/externalarray-1.0.0'¶

Methods Documentation

classmethod from_tree(tree, ctx)[source]¶

Converts basic types representing YAML trees into custom types.

This method should be overridden by custom extension classes in order to define how custom types are deserialized from the YAML representation back into their original types. Typically the method will return an instance of the original custom type. It is also permitted to return a generator, which yields a partially constructed result, then completes construction once the generator is drained. This is useful when constructing objects that contain reference cycles.

This method is called as part of the process of reading an ASDF file in order to construct an AsdfFile object. Whenever a YAML subtree is encountered that has a tag that corresponds to the yaml_tag property of this class, this method will be used to deserialize that tree back into an instance of the original custom type.

Parameters

treeobject representing YAML tree: An instance of a basic Python type (possibly nested) that corresponds to a YAML subtree.
ctxAsdfFile: An instance of the AsdfFile object that is being constructed.

Returns

An instance of the custom type represented by this extension class,: or a generator that yields that instance.

classmethod from_tree_tagged(tree, ctx)¶

Converts from tagged tree into custom type.

It is more common for extension classes to override from_tree instead of this method. This method should only be overridden if it is necessary to access the _tag property of the Tagged object directly.

Parameters

treeasdf.tagged.Tagged object representing YAML tree
ctxAsdfFile: An instance of the AsdfFile object that is being constructed.

Returns

An instance of the custom type represented by this extension class.

classmethod incompatible_version(version)¶

Indicates if given version is known to be incompatible with this type.

If this tag class explicitly identifies compatible versions then this checks whether a given version is compatible or not (see supported_versions). Otherwise, all versions are assumed to be compatible.

Child classes can override this method to affect how version compatiblity for this type is determined.

Parameters

versionstr or AsdfVersion: The version to test for compatibility.

classmethod make_yaml_tag(name, versioned=True)¶

Given the name of a type, returns a string representing its YAML tag.

Parameters

namestr: The name of the type. In most cases this will correspond to the name attribute of the tag type. However, it is passed as a parameter since some tag types represent multiple custom types.
versionedbool: If True, the tag will be versioned. Otherwise, a YAML tag without a version will be returned.

Returns

str representing the YAML tag

classmethod names()¶

Returns the name(s) represented by this tag type as a list.

While some tag types represent only a single custom type, others represent multiple types. In the latter case, the name attribute of the extension is actually a list, not simply a string. This method normalizes the value of name by returning a list in all cases.

Returns

list of names represented by this tag type

classmethod subclass(*args, attribute='subclass')¶

Decorator to enable serialization of a subclass of an existing type.

Use this method to decorate subclasses of custom types that are already handled by an existing ASDF tag class. This enables subclasses of known types to be properly serialized without having to write an entirely separate tag class for the subclass.

This feature can only be used for tagged types where the underlying YAML representation of the type is an object (i.e. a Python dict). It will not work for nodes that are basic types.

The subclass metadata is stored in a new attribute of the YAML node. By default the attribute name is “subclass”, but it is customizable by using the optional attribute keyword argument of the decorator.

The schema of the base custom type is used for validation. This feature will not work if the base schema disallows additional attributes.

It is incumbent upon the user to avoid name conflicts with attributes that already exist in the representation of the base custom class. For example, a base class may use the attribute “subclass” for some other purpose, in which case it would be necessary to provide a different custom attribute name here.

Parameters

attributestr: Custom attribute name used to store subclass metadata in this node.

classmethod subclass_property(attribute)¶

Decorator to enable serialization of custom subclass attributes.

Use this decorator to serialize attributes that are specific to a subclass of a custom type that is already handled by an existing ASDF tag class. This decorator will only work on subclasses that have been decorated with the subclass decorator.

Methods that are decorated in this way are treated as properties (see property). The name of the property must correspond to a keyword argument of the subclass constructor.

The property will be serialized as a YAML object attribute with the same name. Users are responsible for ensuring that any and all additional subclass properties conform to the schema of the base custom type and do not conflict with existing attributes.

classmethod tag_base()¶

Returns the base of the YAML tag for types represented by this class.

This method returns the portion of the tag that represents the standard and the organization of any type represented by this class.

Returns

str representing the base of the YAML tag

classmethod to_tree(data, ctx)[source]¶

Converts instances of custom types into YAML representations.

This method should be overridden by custom extension classes in order to define how custom types are serialized into YAML. The method must return a single Python object corresponding to one of the basic YAML types (dict, list, str, or number). However, the types can be nested and combined in order to represent more complex custom types.

This method is called as part of the process of writing an AsdfFile object. Whenever a custom type (or a subclass of that type) that is listed in the types attribute of this class is encountered, this method will be used to serialize that type.

The name to_tree refers to the act of converting a custom type into part of a YAML object tree.

Parameters

nodeobject: Instance of a custom type to be serialized. Will be an instance (or an instance of a subclass) of one of the types listed in the types attribute of this class.
ctxAsdfFile: An instance of the AsdfFile object that is being written out.

Returns

A basic YAML type (dict, list, str, int, float, or: complex) representing the properties of the custom type to be serialized. These types can be nested in order to represent more complex custom types.

classmethod to_tree_tagged(node, ctx)¶

Converts instances of custom types into tagged objects.

It is more common for custom tag types to override to_tree instead of this method. This method should be overridden if it is necessary to modify the YAML tag that will be used to tag this object.

Parameters

nodeobject: Instance of a custom type to be serialized. Will be an instance (or an instance of a subclass) of one of the types listed in the types attribute of this class.
ctxAsdfFile: An instance of the AsdfFile object that is being written out.

Returns

An instance of asdf.tagged.Tagged.