DefTree API

Element

class deftree.Element(name)

Element class. This class defines the Element interface

add_attribute(name, value)

Creates an Attribute instance with name and value as a child to self.

add_element(name)

Creates an Element instance with name as a child to self.

append(item)

Inserts the item at the end of this element’s internal list of children. Raises TypeError if item is not a Element or Attribute

attributes([name, value])

Creates a tree iterator with the current element as the root. The iterator iterates over this element. Only Attributes. Name and value are optional and used for filters

clear()

Resets an element. This function removes all children, clears all attributes

copy()

Returns a deep copy of the current Element.

elements([name])

Creates a tree iterator with the current element as the root. The iterator iterates over this element and all elements below it, in document (depth first) order. Only Element whose name equals name are returned from the iterator

get_attribute(name[, value])

Returns the first Attribute instance whose name matches name and if value is not None whose value equal value. If none is found it returns None.

get_element(name)

Returns the first Element whose name matches name, if none is found returns None.

get_parent()

Returns the parent of the current Element

index(item)

Returns the index of the item in this element, raises ValueError if not found.

insert(index, item)

Inserts the item at the given position in this element. Raises TypeError if item is not a Element or Attribute

iter()

Creates a tree iterator with the current element as the root. The iterator iterates over this element and all elements below it, in document (depth first) order. Both Element and Attribute are returned from the iterator.

iter_elements(name[, value])

Creates a tree iterator with the current element as the root. The iterator iterates over this element and all elements below it, in document (depth first) order. Only Element are returned from the iterator.

remove(child)

Removes child from the element. Compares on instance identity not name. Raises TypeError if child is not a Element or Attribute

set_attribute(name, value)

Sets the first Attribute with name to value.

Attribute

class deftree.Attribute(parent, name, value)

Attribute class. This class defines the Attribute interface.

get_parent()

Returns the parent element of the attribute.

DefTree

class deftree.DefTree

DefTree class. This class represents an entire element hierarchy.

dump()

Write the the DefTree structure to sys.stdout. This function should be used for debugging only.

from_string(text[, parser])

Parses a Defold document section from a string constant which it returns. parser is an optional parser instance. If not given the standard parser is used. Returns the root of DefTree.

get_document_path()

Returns the path to the parsed document.

get_root()

Returns the root Element

parse(source[, parser])

Parses a Defold document into a DefTree which it returns. source is a file_path. parser is an optional parser instance. If not given the standard parser is used.

write(file_path)

Writes the element tree to a file, as plain text. file_path needs to be a path.

Helpers

deftree.parse(source)

Parses a Defold document into a DefTree which it returns. source is a file_path. parser is an optional parser instance. If not given the standard parser is used.

deftree.from_string(text[, parser])

Parses a Defold document section from a string constant which it returns. parser is an optional parser instance. If not given the standard parser is used. Returns the root of DefTree.

deftree.to_string(element[, parser])

Generates a string representation of the Element, including all children. element is a Element instance.

deftree.dump(element[, parser])

Write element tree or element structure to sys.stdout. This function should be used for debugging only. element is either an DefTree, or Element.

deftree.validate(string, path_or_string[, verbose])

Verifies that a document in string format equals a document in path_or_string. If Verbose is True it echoes the result. This function should be used for debugging only. Returns a bool representing the result

Using DefTree

If you are not familiar with Defold files this is how the syntax looks, it is in a Google Protobuf format.

elementname {
  attributename: attributevalue
  position {
    x: 0.0
    y: 0.0
    z: 0.0
  }
  type: TYPE_BOX
  blend_mode: BLEND_MODE_ALPHA
  texture: "atlas/logo"
  id: "logo"
}

Parsing Defold Documents

Parsing from a file is done by calling the parse method

import deftree
tree = deftree.parse(path)  # parse the document into a DefTree
root = tree.get_root()      # returns the root from the tree

Or alternatively we can first create a DefTree instance

tree = deftree.Deftree()
root = tree.parse(path)

We can also parse a document from a string

tree = deftree.from_string(document_as_string)  # parse the document into a DefTree
root = tree.get_root()      # returns the root from the tree

Finding interesting elements

Element has some useful methods that help iterate recursively over all the sub-tree below it (its children, their children, and so on). For example, Element.iter():

for child in root.iter():
    print(child.name)

We can also iterate only elements by calling Element.iter_elements()

for child in root.iter_elements():
    print(child.name)

for child in root.iter_elements("nodes"):  # iter_elements also supports filtering on name
    print(child.name)

Element.get_attribute() finds the first attributes with the given name in that element.

attribute = element.get_attribute("id")

Modifying existing scenes

DefTree provides a simple way to edit Defold documents and write them to files. The DefTree.write() method serves this purpose. Once created, an Element object may be manipulated by directly changing its fields, as well as adding new children (for example with Element.insert()).

Let’s say we want to find all box nodes in a gui and change its layers.

for element in root.iter_elements("nodes")
    if element.get_attribute("type") == "TYPE_BOX":
        element.set_attribute("layer", 'new_layer')

We can also add new attributes and elements all together.

new_element = root.add_element("layers")
new_element.add_attribute("name", 'new_layer')

DefTree Attribute of number types supports basic math functions directly

new_element = root.get_element("position")
attribute = new_element.get_attribute("x")
attribute += 10

We will probably then overwrite the file

tree.write(tree.get_document_path())

Changelog

1.0.2

Changed

  • How DefTree determines if a string is a string, int or float, fix for bigger numbers with science annotation

1.0.1

Added

  • Added Element.add_element(name)
  • Added Element.add_attribute(name, value)
  • Added Element.set_attribute(name, value)
  • Added Element.elements() - for getting top level elements of Element
  • Added Element.attribute() - for getting top level attribute of Element
  • Exposed deftree.dump and deftree.validate in the documentation
  • Added DefTree.get_document_path() to get the path of the document that was parsed
  • Attribute are now sub classed into different types this to make it easier when editing values as Defold is picky

Changed

  • Element.iter_all() is now Element.iter()
  • Element.iter_find_elements(name) is now Element.iter_elements(name)
  • Changed how attributes reports their value. They should now be easier to work with, without any need add quotationmarks and such.

Removed

  • Removed SubElement() factory, now use element.add_element()
  • Removed Element.iter_attributes()
  • Removed Element.iter_find_attributes()
  • Removed NaiveDefParser as it was obsolete and inferior
  • Removed Example folder

0.2.0

Added

  • Raises ParseError when reading invalid documents

Changed

  • Updated docstrings to be easier to read.
  • Refactored internal usage of a level variable to track how deep the item were in the tree

Removed

  • Removed Element.add(), use Element.append() Element.insert()
  • Removed Element.items(), use Element.iter_all()

0.1.1

Added

  • Licence to github repository
  • Setup files for PyPi to github repository
  • Example usage
  • Unittesting with unittest
  • Coverage exclusion for usage with Coverage.py
  • Using __all__ to define public api, in case of wild import

Changed

  • Elements __setitem__ raises exception on invalid types
  • Elements __next__ implementation was broken
  • serialize() is now a class method

0.1.0

Added

  • First release of DefTree

Contributing to DefTree

Bug fixes, feature additions, tests, documentation and more can be contributed via issues and/or pull requests. All contributions are welcome.

Bug fixes, feature additions, etc.

Please send a pull request to the master branch. Please include documentation and tests for new or changed features. Tests or documentation without bug fixes or feature additions are welcome too.

  • Fork the DefTree repository.
  • Create a branch from master.
  • Develop bug fixes, features, tests, etc.
  • Run the test suite.
  • Create a pull request to pull the changes from your branch to the DefTree master.

Guidelines

  • Separate code commits from reformatting commits.
  • Provide tests for any newly added code.
  • Follow PEP8.

Reporting Issues

When reporting issues, please include code that reproduces the issue. The best reproductions are self-contained scripts with minimal dependencies.

DefTree

DefTree is a python module modify Defold documents, it’s first implementation was inspired by the xml.ElementTree library.

It reads any Defold document into a object tree hierarchy, the module have three main concepts

  1. DefTree represents the whole Defold document as a tree and
  2. Element represents a single node or block in this tree and
  3. Attribute represent a name value pair

Installation

Note

DefTree is only supported by python >= 3.3.0

DefTree is a native python implementation and thus should work under the most common platforms that supports python. The package is distributed in the wheel format and is easily installed with pip.

pip install deftree

Old Versions

You can download old distributions from PyPI.