UnifiedIORegistry#

class gwpy.io.registry.UnifiedIORegistry[source]#

Bases: UnifiedIORegistry

UnifiedIORegistry that can handle reading files in parallel.

Methods Summary

identify_format(origin, data_class_required, ...)

Loop through identifiers to see which formats match.

register_identifier(data_format, data_class, ...)

Associate an identifier function with a specific data type.

Methods Documentation

identify_format(
origin: Literal['read', 'write'],
data_class_required: type,
path: FileSystemPath | None,
fileobj: FileLike | None,
args: tuple,
kwargs: dict,
) → list[str][source]#

Loop through identifiers to see which formats match.

Parameters:
originstr

A string "read or "write" identifying whether the file is to be opened for reading or writing.

data_class_requiredobject

The specified class for the result of read or the class that is to be written.

pathstr or path-like or None

The path to the file or None.

fileobjfile-like or None.

An open file object to read the file’s contents, or None if the file could not be opened.

argssequence

Positional arguments for the read or write function. Note that these must be provided as sequence.

kwargsdict-like

Keyword arguments for the read or write function. Note that this parameter must be dict-like.

Returns:
valid_formatslist

List of matching formats.

register_identifier(
data_format: str,
data_class: type,
identifier: IdentifyProtocol,
force: bool = False,
) → None[source]#

Associate an identifier function with a specific data type.

Parameters:
data_formatstr

The data format identifier. This is the string that is used to specify the data type when reading/writing.

data_classclass

The class of the object that can be written.

identifierfunction

A function that checks the argument specified to read or write to determine whether the input can be interpreted as a table of type data_format. This function should take the following arguments:

  • origin: A string "read" or "write" identifying whether the file is to be opened for reading or writing.

  • path: The path to the file.

  • fileobj: An open file object to read the file’s contents, or None if the file could not be opened.

  • *args: Positional arguments for the read or write function.

  • **kwargs: Keyword arguments for the read or write function.

One or both of path or fileobj may be None. If they are both None, the identifier will need to work from args[0].

The function should return True if the input can be identified as being of format data_format, and False otherwise.

forcebool, optional

Whether to override any existing function if already present. Default is False.

Examples

To set the identifier based on extensions, for formats that take a filename as a first argument, you can do for example

from astropy.io.registry import register_identifier
from astropy.table import Table
def my_identifier(*args, **kwargs):
    return isinstance(args[0], str) and args[0].endswith('.tbl')
register_identifier('ipac', Table, my_identifier)
unregister_identifier('ipac', Table)