Source code for MDAnalysis.core._get_readers
# -*- Mode: python; tab-width: 4; indent-tabs-mode:nil; coding:utf-8 -*-
# vim: tabstop=4 expandtab shiftwidth=4 softtabstop=4
#
# MDAnalysis --- http://www.MDAnalysis.org
# Copyright (c) 2006-2015 Naveen Michaud-Agrawal, Elizabeth J. Denning, Oliver Beckstein
# and contributors (see AUTHORS for the full list)
#
# Released under the GNU Public Licence, v2 or any higher version
#
# Please cite your use of MDAnalysis in published work:
# N. Michaud-Agrawal, E. J. Denning, T. B. Woolf, and O. Beckstein.
# MDAnalysis: A Toolkit for the Analysis of Molecular Dynamics Simulations.
# J. Comput. Chem. 32 (2011), 2319--2327, doi:10.1002/jcc.21787
#
"""Functions for fetching Readers
These functions officially live in in topology/core (parsers) and
coordinates/core (all others). They are declared here to avoid
circular imports.
"""
import copy
import inspect
from .. import (_READERS, _READER_HINTS,
_PARSERS, _PARSER_HINTS,
_MULTIFRAME_WRITERS, _SINGLEFRAME_WRITERS, _CONVERTERS)
from ..lib import util
[docs]
def get_reader_for(filename, format=None):
"""Return the appropriate trajectory reader class for `filename`.
Parameters
----------
filename
filename of the input trajectory or coordinate file. Also can
handle a few special cases, see notes below.
format : str or :class:`Reader` (optional)
Define the desired format. Can be a string to request a given
Reader.
If a class is passed, it will be assumed that this is
a Reader and will be returned.
Returns
-------
:class:`Reader`
A Reader object
Raises
------
ValueError
If no appropriate Reader is found
Notes
-----
There are a number of special cases that can be handled:
- If `filename` is a numpy array,
:class:`~MDAnalysis.coordinates.memory.MemoryReader` is returned.
- If `filename` is an MMTF object,
:class:`~MDAnalysis.coordinates.MMTF.MMTFReader` is returned.
- If `filename` is a ParmEd Structure,
:class:`~MDAnalysis.coordinates.ParmEd.ParmEdReader` is returned.
- If `filename` is an iterable of filenames,
:class:`~MDAnalysis.coordinates.chain.ChainReader` is returned.
Automatic detection is disabled when an explicit `format` is provided,
unless a list of filenames is given, in which case
:class:`~MDAnalysis.coordinates.chain.ChainReader` is returned and `format`
passed to the :class:`~MDAnalysis.coordinates.chain.ChainReader`.
.. versionchanged:: 1.0.0
Added format_hint functionalityx
"""
# check if format is actually a Reader
if inspect.isclass(format):
return format
# ChainReader gets returned even if format is specified
if _READER_HINTS['CHAIN'](filename):
format = 'CHAIN'
# Only guess if format is not specified
if format is None:
for fmt_name, test in _READER_HINTS.items():
if test(filename):
format = fmt_name
break
else: # hits else if for loop completes
# else let the guessing begin!
format = util.guess_format(filename)
format = format.upper()
try:
return _READERS[format]
except KeyError:
errmsg = (
"Unknown coordinate trajectory format '{0}' for '{1}'. The FORMATs \n"
" {2}\n"
" are implemented in MDAnalysis.\n"
" See https://docs.mdanalysis.org/documentation_pages/coordinates/init.html#id1\n"
" Use the format keyword to explicitly set the format: 'Universe(...,format=FORMAT)'\n"
" For missing formats, raise an issue at "
"https://github.com/MDAnalysis/mdanalysis/issues".format(
format, filename, _READERS.keys()))
raise ValueError(errmsg) from None
[docs]
def get_writer_for(filename, format=None, multiframe=None):
"""Return an appropriate trajectory or frame writer class for `filename`.
The format is determined by the `format` argument or the extension of
`filename`. If `format` is provided, it takes precedence over the
extension of `filename`.
Parameters
----------
filename : str or ``None``
If no *format* is supplied, then the filename for the trajectory is
examined for its extension and the Writer is chosen accordingly.
If ``None`` is provided, then
:class:`~MDAnalysis.coordinates.null.NullWriter` is selected (and
all output is discarded silently).
format : str (optional)
Explicitly set a format.
multiframe : bool (optional)
``True``: write multiple frames to the trajectory; ``False``: only
write a single coordinate frame; ``None``: first try trajectory (multi
frame writers), then the single frame ones. Default is ``None``.
Returns
-------
:class:`Writer`
A Writer object
Raises
------
ValueError:
The format could not be deduced from `filename` or an unexpected value
was provided for the `multiframe` argument.
TypeError:
No writer was found for the required format or the required `filename`
argument was omitted.
.. versionchanged:: 0.7.6
Added `multiframe` keyword; the default ``None`` reflects the previous
behaviour.
.. versionchanged:: 0.14.0
Removed the default value for the `format` argument. Now, the value
provided with the `format` parameter takes precedence over the extension
of `filename`. A :exc:`ValueError` is raised if the format cannot be
deduced from `filename`.
.. versionchanged:: 0.16.0
The `filename` argument has been made mandatory.
"""
if filename is None:
format = 'NULL'
elif format is None:
try:
root, ext = util.get_ext(filename)
except (TypeError, AttributeError):
# An AttributeError is raised if filename cannot
# be manipulated as a string.
# A TypeError is raised in py3.6
# "TypeError: expected str, bytes or os.PathLike object"
errmsg = f'File format could not be guessed from "{filename}"'
raise ValueError(errmsg) from None
else:
format = util.check_compressed_format(root, ext)
if format == '':
raise ValueError((
'File format could not be guessed from {}, '
'resulting in empty string - '
'only None or valid formats are supported.'
).format(filename))
format = format.upper()
if multiframe is None:
# Multiframe takes priority, else use singleframe
options = copy.copy(_SINGLEFRAME_WRITERS) # do copy to avoid changing in place
options.update(_MULTIFRAME_WRITERS) # update overwrites existing entries
errmsg = "No trajectory or frame writer for format '{0}'"
elif multiframe is True:
options = _MULTIFRAME_WRITERS
errmsg = "No trajectory writer for format '{0}'"
elif multiframe is False:
options = _SINGLEFRAME_WRITERS
errmsg = "No single frame writer for format '{0}'"
else:
raise ValueError("Unknown value '{0}' for multiframe,"
" only True, False, None allowed"
"".format(multiframe))
try:
return options[format]
except KeyError:
raise TypeError(errmsg.format(format)) from None
def get_parser_for(filename, format=None):
"""Return the appropriate topology parser for `filename`.
Automatic detection is disabled when an explicit `format` is
provided.
Parameters
----------
filename : str or mmtf.MMTFDecoder
name of the topology file; if this is an instance of
:class:`mmtf.MMTFDecoder` then directly use the MMTF format.
format : str
description of the file format
Raises
------
ValueError
If no appropriate parser could be found.
.. versionchanged:: 1.0.0
Added format_hint functionality
"""
if inspect.isclass(format):
return format
# Only guess if format is not provided
if format is None:
for fmt_name, test in _PARSER_HINTS.items():
if test(filename):
format = fmt_name
break
else:
format = util.guess_format(filename)
format = format.upper()
try:
return _PARSERS[format]
except KeyError:
try:
rdr = get_reader_for(filename)
except ValueError:
errmsg = (
"'{0}' isn't a valid topology format, nor a coordinate format\n"
" from which a topology can be minimally inferred.\n"
" You can use 'Universe(topology, ..., topology_format=FORMAT)'\n"
" to explicitly specify the format and\n"
" override automatic detection. Known FORMATs are:\n"
" {1}\n"
" See https://docs.mdanalysis.org/documentation_pages/topology/init.html#supported-topology-formats\n"
" For missing formats, raise an issue at \n"
" https://github.com/MDAnalysis/mdanalysis/issues".format(
format, _PARSERS.keys()))
raise ValueError(errmsg) from None
else:
return _PARSERS['MINIMAL']
def get_converter_for(format):
"""Return the appropriate topology converter for ``format``.
Parameters
----------
format : str
description of the file format
Raises
------
TypeError
If no appropriate parser could be found.
.. versionadded:: 1.0.0
"""
try:
writer = _CONVERTERS[format]
except KeyError:
errmsg = f'No converter found for {format} format'
raise TypeError(errmsg) from None
return writer