This module contains functions and classes related to file input and output.

It primarily provides two high-level functions for reading and writing external files:

In addition, the module contains the FileSource class, which is a data source for OVITO’s data pipeline system reading its input data from an external file.

ovito.io.import_file(location, **params)

This high-level function imports external data from a file.

This Python function corresponds to the Load File command in OVITO’s user interface. The format of the imported file is automatically detected. However, depending on the file’s format, additional keyword parameters may need to be supplied to to specify how the data should be interpreted. These keyword parameters are documented below.

Parameters:location (str) – The file to import. This can be a local file path or a remote sftp:// URL.
Returns:The ObjectNode that has been created for the imported data.

The function creates and returns a new ObjectNode, which provides access the imported data or allows you to apply modifiers to it.


Note that the newly created ObjectNode is not automatically inserted into the three-dimensional scene. That means it won’t appear in the interactive viewports of OVITO or in rendered images. However, you can subsequently insert the node into the scene by calling the add_to_scene() method on it.

Sometimes it may be desirable to reuse an existing ObjectNode. For example if you have already set up a modification pipeline and just want to replace the input data with a different file. In this case you can call node.source.load(...) instead on the existing ObjectNode to select another input file while keeping the applied modifiers.

File columns

When importing XYZ files or binary LAMMPS dump files, the mapping of file columns to OVITO’s particle properties must be specified using the columns keyword parameter:

import_file("file.xyz", columns = 
  ["Particle Identifier", "Particle Type", "Position.X", "Position.Y", "Position.Z"])

The length of the list must match the number of columns in the input file. See the list of particle properties for standard property names. You can also specify a custom property name, in which case a user-defined particle property with that name is created from the corresponding file column. For vector properties, the component must be appended to the property base name as demonstrated for the Position property in the example above. To skip a file column during import, specify None instead of a property name at the corresponding position in the columns list.

For text-based LAMMPS dump files it is also possible to explicitly specify a file column mapping using the columns keyword. Overriding the default mapping can be useful, for example, if the file columns containing the particle positions do not have the standard names x, y, and z (e.g. when reading time-averaged atomic positions computed by LAMMPS).

File sequences

You can import a sequence of files by passing a filename containing a * wildcard character to import_file(). There may be only one * in the filename (and not in a directory name). The wildcard matches only to numbers in a filename.

OVITO scans the directory and imports all matching files that belong to the sequence. Note that OVITO only loads the first file into memory though.

The length of the imported time series is reported by the num_frames field of the FileSource class and is also reflected by the global AnimationSettings object. You can step through the frames of the animation sequence as follows:

from ovito import dataset
from ovito.io import import_file

# Import a sequence of files.
node = import_file('simulation.*.dump')

# Loop over all frames of the sequence.
for frame in range(node.source.num_frames):    
    # Let the node load the corresponding file into its cache.
    # Access the loaded data of the current frame,
    print("Number of atoms at frame %i is %i." % 
          (node.source.loaded_frame, node.source.number_of_particles))

Multi-timestep files

Some file formats can store multiple frames in a single file. OVITO cannot know in some cases (e.g. XYZ and LAMMPS dump formats) that the file contains multiple frames (because, by default, reading the entire file is avoided for performance reasons). Then it is necessary to explicitly tell OVITO to scan the entire file and load a sequence of frames by supplying the multiple_frames option:

node = import_file("file.dump", multiple_frames = True)

You can then step through the contained frames in the same way as for sequences of files.

LAMMPS atom style

When trying to load a LAMMPS data file which is using an atom style other than “atomic”, the atom style must be explicitly specified using the atom_style keyword parameter. The following LAMMPS atom styles are currently supported by OVITO: angle, atomic, body, bond, charge, dipole, full, molecular.

ovito.io.export_file(node, file, format, **params)

High-level function that exports the results of the data pipeline to a file. See the Data export section for an overview.

  • node (ObjectNode) – The object node that provides the data to be exported.
  • file (str) – The output file path.
  • format (str) –

    The type of file to write:

    • "txt" – A text file with global quantities computed by OVITO (see below)
    • "lammps_dump" – LAMMPS text-based dump format
    • "lammps_data" – LAMMPS data format
    • "imd" – IMD format
    • "vasp" – POSCAR format
    • "xyz" – XYZ format
    • "fhi-aims" – FHI-aims format
    • "ca"Text-based format for storing dislocation lines
    • "povray" – POV-Ray scene format

The function first evaluates the modification pipeline of the given ObjectNode to obtain the data to be exported. This means it is not necessary to call ObjectNode.compute() before calling export_file(), but it doesn’t hurt either.

The format parameter determines the type of file written by the export function; the filename suffix is ignored. However, for files that end in .gz, automatic gzip compression is activated if the file format is text-based.

Depending on the selected format, additional keyword arguments must be provided:

File columns

When writing files in one of the formats lammps_dump, xyz, or imd, you must specify the set of particle properties to be exported using the columns keyword parameter:

export_file(node, "output.xyz", "xyz", columns= 
  ["Particle Identifier", "Particle Type", "Position.X", "Position.Y", "Position.Z"]

See the list of standard particle properties for valid names. Additionaly, user-defined properties can be listed if they exist in the output data. For vector properties, the component name must be appended to the base name as demonstrated for the Position property in the example above.

Exporting several simulation frames

By default, only the current animation frame (given by the current_frame global variable) is exported. To export a different frame, pass the frame keyword parameter to the export_file() function. Alternatively, you can export all frames of the current animation sequence at once by passing multiple_frames=True. Refined control of the exported frames is possible using the keyword arguments start_frame, end_frame, and every_nth_frame.

The lammps_dump and xyz file formats can store multiple frames in a single output file. For all other formats, or if you intentionally want to generate one file per frame, you must pass a wildcard filename to export_file(). This filename must contain exactly one * character as in the following example, which will be replaced with the animation frame number:

export_file(node, "output.*.dump", "lammps_dump", multiple_frames=True)

The above call is equivalent to the following Python loop:

for i in range(node.source.num_frames):
    export_file(node, "output.%i.dump" % i, "lammps_dump", frame=i)

LAMMPS atom style

When writing files in the lammps_data format, the LAMMPS atom style “atomic” is used by default. If you want to create a data file with a different atom style, the style can be selected using the atom_style keyword parameter:

export_file(node, "output.data", "lammps_data", atom_tyle="bond")

The following LAMMPS atom styles are currently supported by OVITO: angle, atomic, bond, charge, dipole, full, molecular, sphere.

Global attributes

The txt file format allows you to export global quantities computed by OVITO’s data pipeline to a text file. For example, to write out the number of FCC atoms identified by a CommonNeighborAnalysisModifier as a function of simulation time, one would do the following:

export_file(node, "data.txt", "txt", 
    columns=["Timestep", "CommonNeighborAnalysis.counts.FCC"], 

See the documentation of each analysis modifier to find out which global quantities it computes. At runtime, you can determine which attributes are available for export as follows:

class ovito.io.FileSource
Base class:ovito.data.DataCollection

This object serves as a data source for modification pipelines and is responsible for reading the input data from one or more external files.

You normally do not create an instance of this class yourself. The ovito.io.import_file() function does it for you and assigns the file source to the source attribute of the returned ObjectNode. This file source loads data from the external file given by the source_path attribute. The ObjectNode then takes this data and feeds it into its modification pipeline.

You typically don’t set the source_path attribute directly. Instead, use the FileSource.load() method to load a different input file and hook it into an existing modification pipeline:

from ovito.io import import_file
from ovito.modifiers import ColorCodingModifier

# This creates a new node with an empty modification pipeline:
node = import_file('first_file.dump')

# Populate the pipeline with a modifier:
node.modifiers.append(ColorCodingModifier(particle_property='Potential Energy'))

# Call FileSource.load() to replace the input data with a different file
# but keep the node's current modification pipeline:

File sources are also used by certain modifiers to load a reference configuration, e.g. by the CalculateDisplacementsModifier, whose reference attribute also contains a FileSource.

Data access

The FileSource class is derived from the DataCollection base class. This means the file source also stores the data loaded from the external file, and you can access this data through the DataCollection base class interface. Note that the cached data represents the outcome of the most recent successful loading operation and may change every time a new simulation frame is loaded (see loaded_frame).

from ovito.io import import_file

# This creates a node with a FileSource, which also is a DataCollection.
node = import_file('simulation.dump')

# Access data cached in the DataCollection.

A flag that controls whether the animation length in OVITO is automatically adjusted to match the number of frames in the loaded file or file sequence.

The current length of the animation in OVITO is managed by the global AnimationSettings object. The number of frames in the external file or file sequence is indicated by the num_frames attribute of this FileSource. If adjust_animation_interval is True, then the animation length will be automatically adjusted to match the number of frames provided by the FileSource.

In some situations it makes sense to turn this option off, for example, if you import several data files into OVITO simultaneously, but their frame counts do not match.

load(location, **params)

Loads a new external file into this data source object.

The function auto-detects the format of the file.

The function accepts additional keyword arguments that are forwarded to the format-specific file importer. See the documentation of the import_file() function for more information on this.

Parameters:location (str) – The local file or remote sftp:// URL to load.

The path or URL of the data file that is currently loaded and kept in memory by the FileSource (read-only).


The zero-based frame index that is currently loaded and kept in memory by the FileSource (read-only).

The data of this frame is accessible through the inherited DataCollection interface.


The total number of frames the imported file or file sequence contains (read-only).


The path or URL of the imported file or file sequence. Unlike the path return by loaded_file, the source path may contain a ‘*’ wildcard character when a file sequence has been imported.