AiiDA internals

Node

The Node class is the basic class that represents all the possible objects at the AiiDA world. More precisely it is inherited by many classes including (among others) the Calculation class, representing computations that convert data into a different form, the Code class representing executables and file collections that are used by calculations and the Data class which represents data that can be input or output of calculations.

Methods & properties

In the sequel the most important methods and properties of the Node class will be described.

Node subclasses organization

The Node class has two important variables:

• _plugin_type_string characterizes the class of the object.
• _query_type_string characterizes the class and all its subclasses (by pointing to the package or Python file that contain the class).

The convention for all the Node subclasses is that if a class B is inherited by a class A then there should be a package A under aiida/orm that has a file __init__.py and a B.py in that directory (or a B package with the corresponding __init__.py)

An example of this is the ArrayData and the KpointsData. ArrayData is placed in aiida/orm/data/array/__init__.py and KpointsData which inherits from ArrayData is placed in aiida/orm/data/array/kpoints.py

This is an implicit & quick way to check the inheritance of the Node subclasses.

General purpose methods

• __init__(): The initialization of the Node class can be done by not providing any attributes or by providing a DbNode as initialization. E.g.:

  dbn = a_dbnode_object
  n = Node(dbnode=dbn.dbnode)
  

• ctime() and mtime() provide the creation and the modification time of the node.

• is_stored() informs whether a node is already stored to the database.

• query() queries the database by filtering for the results for similar nodes (if the used object is a subclass of Node) or with no filtering if it is a Node class. Note that for this check _plugin_type_string should be properly set.

• computer() returns the computer associated to this node.

• _validate() does a validation check for the node. This is important for Node subclasses where various attributes should be checked for consistency before storing.

• get_user() returns the user that created the node.

• _increment_version_number_db(): increment the version number of the node on the DB. This happens when adding an attribute or an extra to the node. This method should not be called by the users.

• copy() returns a not stored copy of the node with new UUID that can be edited directly.

• uuid() returns the universally unique identifier (UUID) of the node.

• pk() returns the principal key (ID) of the node.

• dbnode() returns the corresponding Django object.

• get_computer() & set_computer() get and set the computer to be used & is associated to the node.

Annotation methods

The Node can be annotated with labels, description and comments. The following methods can be used for the management of these properties.

Label management:

• label() returns the label of the node. The setter method can be used for the update of the label.
• _update_db_label_field() updates the label in the database. This is used by the setter method of the label.

Description management:

• description(): the description of the node (more detailed than the label). There is also a setter method.
• _update_db_description_field(): update the node description in the database.

Comment management:

• add_comment() adds a comment.
• get_comments() returns a sorted list of the comments.
• _get_dbcomments() is similar to get_comments(), just the sorting changes.
• _update_comment() updates the node comment. It can be done by verdi comment update.
• _remove_comment() removes the node comment. It can be done by verdi comment remove.

Link management methods

Node objects and objects of its subclasses can have ancestors and descendants. These are connected with links. The following methods exist for the processing & management of these links.

• _has_cached_links() shows if there are cached links to other nodes.
• add_link_from() adds a link to the current node from the ‘src’ node with the given label. Depending on whether the nodes are stored or node, the linked are written to the database or to the cache.
• _add_cachelink_from() adds a link to the cache.
• _replace_link_from() replaces or creates an input link.
• _remove_link_from() removes an input link that is stored in the database.
• _replace_dblink_from() is similar to _replace_link_from() but works directly on the database.
• _remove_dblink_from() is similar to _remove_link_from() but works directly on the database.
• _add_dblink_from() adds a link to the current node from the given ‘src’ node. It acts directly on the database.

Listing links example

Assume that the user wants to see the available links of a node in order to understand the structure of the graph and maybe traverse it. In the following example, we load a specific node and we list its input and output links. The returned dictionaries have as keys the link name and as value the linked node. Here is the code:

In [1]: # Let's load a node with a specific pk

In [2]: c = load_node(139168)

In [3]: c.get_inputs_dict()
Out[3]:
{u'code': <Code: Remote code 'cp-5.1' on daint, pk: 75709, uuid: 3c9cdb7f-0cda-402e-b898-4dd0d06aa5a4>,
 u'parameters': <ParameterData: uuid: 94efe64f-7f7e-46ea-922a-fe64a7fba8a5 (pk: 139166)>,
 u'parent_calc_folder': <RemoteData: uuid: becb4894-c50c-4779-b84f-713772eaceff (pk: 139118)>,
 u'pseudo_Ba': <UpfData: uuid: 5e53b22d-5757-4d50-bbe0-51f3b9ac8b7c (pk: 1905)>,
 u'pseudo_O': <UpfData: uuid: 5cccd0d9-7944-4c67-b3c7-a39a1f467906 (pk: 1658)>,
 u'pseudo_Ti': <UpfData: uuid: e5744077-8615-4927-9f97-c5f7b36ba421 (pk: 1660)>,
 u'settings': <ParameterData: uuid: a5a828b8-fdd8-4d75-b674-2e2d62792de0 (pk: 139167)>,
 u'structure': <StructureData: uuid: 3096f83c-6385-48c4-8cb2-24a427ce11b1 (pk: 139001)>}

In [4]: c.get_outputs_dict()
Out[4]:
{u'output_parameters': <ParameterData: uuid: f7a3ca96-4594-497f-a128-9843a1f12f7f (pk: 139257)>,
 u'output_parameters_139257': <ParameterData: uuid: f7a3ca96-4594-497f-a128-9843a1f12f7f (pk: 139257)>,
 u'output_trajectory': <TrajectoryData: uuid: 7c5b65bc-22bb-4b87-ac92-e8a78cf145c3 (pk: 139256)>,
 u'output_trajectory_139256': <TrajectoryData: uuid: 7c5b65bc-22bb-4b87-ac92-e8a78cf145c3 (pk: 139256)>,
 u'remote_folder': <RemoteData: uuid: 17642a1c-8cac-4e7f-8bd0-1dcebe974aa4 (pk: 139169)>,
 u'remote_folder_139169': <RemoteData: uuid: 17642a1c-8cac-4e7f-8bd0-1dcebe974aa4 (pk: 139169)>,
 u'retrieved': <FolderData: uuid: a9037dc0-3d84-494d-9616-42b8df77083f (pk: 139255)>,
 u'retrieved_139255': <FolderData: uuid: a9037dc0-3d84-494d-9616-42b8df77083f (pk: 139255)>}

Understanding link names

The nodes may have input and output links. Every input link of a node should have a unique name and this unique name is mapped to a specific node. On the other hand, given a node c, many output nodes may share the same output link name. To differentiate between the output nodes of c that have the same link name, the pk of the output node is added next to the link name (please see the input & output nodes in the above example).

Input/output related methods

The input/output links of the node can be accessed by the following methods.

Methods to get the input data

• get_inputs_dict() returns a dictionary where the key is the label of the input link.
• get_inputs() returns the list of input nodes
• inp() returns a NodeInputManager() object that can be used to access the node’s parents.
• has_parents() returns true or false whether the node has parents

Methods to get the output data

• get_outputs_dict() returns a dictionary where the key is the label of the output link, and the value is the output node.
• get_outputs() returns a list of output nodes.
• out() returns a NodeOutputManager() object that can be used to access the node’s children.
• has_children() returns true or false whether the node has children.

Navigating in the ``node`` graph

The user can easily use the NodeInputManager() and the NodeOutputManager() objects of a node (provided by the inp() and out() respectively) to traverse the node graph and access other connected nodes. inp() will give us access to the input nodes and out() to the output nodes. For example:

In [1]: # Let's load a node with a specific pk

In [2]: c = load_node(139168)

In [3]: c
Out[3]: <CpCalculation: uuid: 49084dcf-c708-4422-8bcf-808e4c3382c2 (pk: 139168)>

In [4]: # Let's traverse the inputs of this node.

In [5]: # By typing c.inp. we get all the input links

In [6]: c.inp.
c.inp.code                c.inp.parent_calc_folder  c.inp.pseudo_O            c.inp.settings
c.inp.parameters          c.inp.pseudo_Ba           c.inp.pseudo_Ti           c.inp.structure

In [7]: # We may follow any of these links to access other nodes. For example, let's follow the parent_calc_folder

In [8]: c.inp.parent_calc_folder
Out[8]: <RemoteData: uuid: becb4894-c50c-4779-b84f-713772eaceff (pk: 139118)>

In [9]: # Let's assign to r the node reached by the parent_calc_folder link

In [10]: r = c.inp.parent_calc_folder

In [11]: r.inp.__dir__()
Out[11]:
['__class__',
 '__delattr__',
 '__dict__',
 '__dir__',
 '__doc__',
 '__format__',
 '__getattr__',
 '__getattribute__',
 '__getitem__',
 '__hash__',
 '__init__',
 '__iter__',
 '__module__',
 '__new__',
 '__reduce__',
 '__reduce_ex__',
 '__repr__',
 '__setattr__',
 '__sizeof__',
 '__str__',
 '__subclasshook__',
 '__weakref__',
 u'remote_folder']

In [12]: r.out.
r.out.parent_calc_folder         r.out.parent_calc_folder_139168

In [13]: # By following the same link from node r, you will get node c

In [14]: r.out.parent_calc_folder
Out[14]: <CpCalculation: uuid: 49084dcf-c708-4422-8bcf-808e4c3382c2 (pk: 139168)>

Attributes related methods

Each Node() object can have attributes which are properties that characterize the node. Such properties can be the energy, the atom symbols or the lattice vectors. The following methods can be used for the management of the attributes.

• _set_attr() adds a new attribute to the node. The key of the attribute is the property name (e.g. energy, lattice_vectors etc) and the value of the attribute is the value of that property.
• _del_attr() & _del_all_attrs() delete a specific or all attributes.
• get_attr() returns a specific attribute.
• iterattrs() returns an iterator over the attributes. The iterators returns tuples of key/value pairs.
• attrs() returns the keys of the attributes.

Extras related methods

Extras are additional information that are added to the calculations. In contrast to files and attributes, extras are information added by the user (user specific).

• set_extra() adds an extra to the database. To add a more extras at once, set_extras() can be used.
• get_extra() and get_extras() return a specific extra or all the available extras respectively. extras() returns the keys of the extras. iterextras() returns an iterator (returning key/value tuples) of the extras.
• del_extra() deletes an extra.

Folder management

Folder objects represent directories on the disk (virtual or not) where extra information for the node are stored. These folders can be temporary or permanent.

• folder() returns the folder associated to the node.
• get_folder_list() returns the list of files that are in the path sub-folder of the repository folder.
• _repository_folder() returns the permanent repository folder.
• _get_folder_pathsubfolder() returns the path sub-folder in the repository.
• _get_temp_folder() returns the node folder in the temporary repository.
• remove_path() removes a file/directory from the repository.
• add_path() adds a file or directory to the repository folder.
• get_abs_path() returns the absolute path of the repository folder.

Store & deletion

• store_all() stores all the input nodes, then it stores the current node and in the end, it stores the cached input links.
• _store_input_nodes() stores the input nodes.
• _check_are_parents_stored() checks that the parents are stored.
• _store_cached_input_links() stores the input links that are in memory.
• store() method checks that the node data is valid, then check if node‘s parents are stored, then moves the contents of the temporary folder to the repository folder and in the end, it stores in the database the information that are in the cache. The latter happens with a database transaction. In case this transaction fails, then the data transfered to the repository folder are moved back to the temporary folder.
• __del__() deletes temporary folder and it should be called when an in-memory object is deleted.

DbNode

The DbNode is the Django class that corresponds to the Node class allowing to store and retrieve the needed information from and to the database. Other classes extending the Node class, like Data, Calculation and Code use the DbNode code too to interact with the database. The main methods are:

• get_aiida_class() which returns the corresponding AiiDA class instance.
• get_simple_name() which returns a string with the type of the class (by stripping the path before the class name).
• attributes() which returns the all the attributes of the specific node as a dictionary.
• extras() which returns all the extras of the specific node as a dictionary.

Folders

AiiDA uses Folder and its subclasses to add an abstraction layer between the functions and methods working directly on the file-system and AiiDA. This is particularly useful when we want to easily change between different folder options (temporary, permanent etc) and storage options (plain local directories, compressed files, remote files & directories etc).

Folder

This is the main class of the available Folder classes. Apart from the abstraction provided to the OS operations needed by AiiDA, one of its main features is that it can restrict all the available operations within a given folder limit. The available methods are:

• mode_dir() and mode_file() return the mode with which folders and files should be writable.
• get_subfolder() returns the subfolder matching the given name
• get_content_list() returns the contents matching a pattern.
• insert_path() adds a file/folder to a specific location and remove_path() removes a file/folder
• get_abs_path() returns the absolute path of a file/folder under a given folder and abspath() returns the absolute path of the folder.
• create_symlink() creates a symlink pointing the given location inside the folder.
• create_file_from_filelike() creates a file from the given contents.
• open() opens a file in the folder.
• folder_limit() returns the limit under which the creation of files/folders is restrained.
• exists() returns true or false depending whether a folder exists or not.
• isfile() and py:meth:~aiida.common.folders.Folder.isdir return true or false depending on the existence of the given file/folder.
• create() creates the folder, erase() deletes the folder and replace_with_folder() copies/moves a given folder.

RepositoryFolder

Objects of this class correspond to the repository folders. The RepositoryFolder specific methods are:

• __init__() initializes the object with the necessary folder names and limits.
• get_topdir() returns the top directory.
• section() returns the section to which the folder belongs. This can be for the moment a workflow or node.
• subfolder() returns the subfolder within the section/uuid folder.
• uuid() the UUID of the corresponding node or workflow.

SandboxFolder

SandboxFolder objects correspond to temporary (“sandbox”) folders. The main methods are:

• __init__() creates a new temporary folder
• __exit__() destroys the folder on exit.