python-igraph API reference

List of all classes, functions and methods in python-igraph

class documentation

class Configuration:

View In Hierarchy

Class representing igraph configuration details.

Note that there is one primary instance of this class, which is used by igraph itself to retrieve configuration parameters when needed. You can access this instance with the instance() method. You may construct other instances by invoking the constructor directly, but these instances will not affect igraph's behaviour. If you are interested in configuring igraph, use igraph.config to get hold of the singleton instance and then modify it.

General ideas

The configuration of igraph is stored in the form of name-value pairs. This object provides an interface to the configuration data using the syntax known from dict:

>>> c = Configuration()
>>> c["general.verbose"] = True
>>> print(c["general.verbose"])
True

Configuration keys are organized into sections, and the name to be used for a given key is always in the form section.keyname, like general.verbose in the example above. In that case, general is the name of the configuration section, and verbose is the name of the key. If the name of the section is omitted, it defaults to general, so general.verbose can be referred to as verbose:

>>> c = Configuration()
>>> c["verbose"] = True
>>> print(c["general.verbose"])
True

User-level configuration is stored in ~/.igraphrc per default on Linux and Mac OS X systems, or in C:\Documents and Settings\username\.igraphrc on Windows systems. However, this configuration is read only when igraph is launched through its shell interface defined in igraph.app.shell. This behaviour might change before version 1.0.

Known configuration keys

The known configuration keys are presented below, sorted by section. When referring to them in program code, don't forget to add the section name, expect in the case of section general.

General settings

These settings are all stored in section general.

  • shells: the list of preferred Python shells to be used with the command-line igraph script. The shells in the list are tried one by one until any of them is found on the system. igraph functions are then imported into the main namespace of the shell and the shell is launched. Known shells and their respective class names to be used can be found in igraph.app.shell. Example: IPythonShell, ClassicPythonShell. This is the default, by the way.
  • verbose: whether igraph should talk more than really necessary. For instance, if set to True, some functions display progress bars.

Application settings

These settings specify the external applications that are possibly used by igraph. They are all stored in section apps.

  • image_viewer: image viewer application. If set to an empty string, it will be determined automatically from the platform igraph runs on. On Mac OS X, it defaults to the Preview application. On Linux, it chooses a viewer from several well-known Linux viewers like gthumb, kuickview and so on (see the source code for the full list). On Windows, it defaults to the system's built-in image viewer.

Plotting settings

These settings specify the default values used by plotting functions. They are all stored in section plotting.

  • layout: default graph layout algorithm to be used.
  • mark_groups: whether to mark the clusters by polygons when plotting a clustering object.
  • palette: default palette to be used for converting integer numbers to colors. See colors.Palette for more information. Valid palette names are stored in colors.palettes.
  • wrap_labels: whether to try to wrap the labels of the vertices automatically if they don't fit within the vertex. Default: False.

Shell settings

These settings specify options for external environments in which igraph is embedded (e.g., IPython and its Qt console). These settings are stored in section shell.

  • ipython.inlining.Plot: whether to show instances of the Plot class inline in IPython's console if the console supports it. Default: True
Class Types Static class for the implementation of custom getter/setter functions for configuration keys
Class Method instance Returns the single instance of the configuration object.
Method __contains__ Checks whether the given configuration item is set.
Method __delitem__ Deletes the given item from the configuration.
Method __getitem__ Returns the given configuration item.
Method __init__ Creates a new configuration instance.
Method __setitem__ Sets the given configuration item.
Method has_key Checks if the configuration has a given key.
Method load Loads the configuration from the given file.
Method save Saves the configuration.
Property filename Returns the filename associated to the object.
Static Method _item_to_section_key Converts an item description to a section-key pair.
Method _get Internal function that returns the value of a given key in a given section.
Class Variable _definitions Undocumented
Class Variable _instance Undocumented
Class Variable _sections Undocumented
Class Variable _types Undocumented
Instance Variable _config Undocumented
Instance Variable _filename Undocumented
@classmethod
def instance(cls):

Returns the single instance of the configuration object.

def __contains__(self, item):

Checks whether the given configuration item is set.

Parameters
itemthe configuration key to check.
Returns
True if the key has an associated value, False otherwise.
def __delitem__(self, item):

Deletes the given item from the configuration.

If the item has a default value, the default value is written back instead of the current value. Without a default value, the item is really deleted.

def __getitem__(self, item):

Returns the given configuration item.

Parameters
itemthe configuration key to retrieve.
Returns
the configuration value
def __init__(self, filename=None):

Creates a new configuration instance.

Parameters
filenamefile or file pointer to be read. Can be omitted.
def __setitem__(self, item, value):

Sets the given configuration item.

Parameters
itemthe configuration key to set
valuethe new value of the configuration key
def has_key(self, item):

Checks if the configuration has a given key.

Parameters
itemthe key being sought
def load(self, stream=None):

Loads the configuration from the given file.

Parameters
streamname of a file or a file object. The configuration will be loaded from here. Can be omitted, in this case, the user-level configuration is loaded.
def save(self, stream=None):

Saves the configuration.

Parameters
streamname of a file or a file object. The configuration will be saved there. Can be omitted, in this case, the user-level configuration file will be overwritten.
@property
filename =

Returns the filename associated to the object.

It is usually the name of the configuration file that was used when creating the object. Configuration.load always overwrites it with the filename given to it. If None, the configuration was either created from scratch or it was updated from a stream without name information.

@staticmethod
def _item_to_section_key(item):

Converts an item description to a section-key pair.

Parameters
itemthe item to be converted
Returns
if item contains a period (.), it is splitted into two parts at the first period, then the two parts are returned, so the part before the period is the section. If item does not contain a period, the section is assumed to be general, and the second part of the returned pair contains item unchanged
def _get(self, section, key):

Internal function that returns the value of a given key in a given section.

_definitions =

Undocumented

_instance =

Undocumented

_sections: tuple[str, ...] =

Undocumented

_types =

Undocumented

_config =

Undocumented

_filename =

Undocumented