Streamxmlwriter

Streamxmlwriter is a simple python library for writing XML files.

It allows you to generate XML documents without building the whole document tree in memory first, which means you can generate arbitrarily large documents with a very small memory footprint.

Its features include pretty-printing and custom ordering of element attributes.

Usage example

>>> from io import BytesIO
>>> output = BytesIO()

>>> from streamxmlwriter import XMLWriter
>>> writer = XMLWriter(output, pretty_print=True)

>>> writer.start_ns("a", "http://example.org/ns")
>>> writer.start("foo", {"{http://example.org/ns}one": "1"}, two="2")
>>> writer.start("bar")
>>> writer.data("something")
>>> writer.end("bar")
>>> writer.comment("hello")
>>> writer.element("{http://example.org/ns}baz", data="whatnot", x="y")
>>> writer.start("empty")
>>> writer.close()

>>> output.getvalue()
b'<foo xmlns:a="http://example.org/ns" two="2" a:one="1">\n  <bar>something</bar>\n  <!--hello-->\n  <a:baz x="y">whatnot</a:baz>\n  <empty />\n</foo>'

The API

writer = XMLWriter(file, encoding="utf-8", pretty_print=False, sort=True, abbrev_empty=True)

creates a new writer instance that writes its output to the file-like object you pass as the first argument. There are a few optional arguments as well:

• encoding specifies the character encoding for the XML output. Default: "utf-8".
• If pretty_print is True, the XML output will be pretty-printed (indented). Default: False.
• If sort is True, every element's attributes will be lexicographically sorted by name. See "Attribute ordering" below for more advanced options. Default: True.
• If abbrev_empty is False, empty elements are serialized as a start-end tag pair (<foo></foo>), instead of the shorter form (<foo />). Default: True.

writer.start(tag, attributes=None, nsmap=None, **kwargs)

opens an element whose tag is tag. To specify attributes, you can pass it a dictionary as the second argument. In most cases, it's easier to specify each attribute as a keyword argument.

nsmap is an optional dictionary, mapping namespace prefixes to URIs. This is used automatically if you serialize Element instances from lxml using the element method.

writer.end(tag)

closes the most recently opened element. If you pass it a tag that doesn't match the open element, the writer raises an XMLSyntaxError. If you don't pass any tag at all, the current element is closed.

writer.data(data)

writes character data to the output file, properly encoded.

writer.element(element, attributes=None, data=None, **kwargs)

writes a complete element. element("foo", bar="baz", data="hello!") is exactly the same as calling start("foo", bar="baz"), data("hello!") and end("foo").

If element is an Element instance, the whole element will be serialized, including children.

writer.declaration()

outputs an XML declaration. If the character encoding is not us-ascii or utf-8, it is called automatically by the constructor. Does nothing if a declaration has already been written. Raises XMLSyntaxError if XML element data has already been written.

writer.comment(data)

outputs an XML comment.

writer.pi(target, data)

outputs an XML processing instruction.

writer.start_ns(prefix, uri)

adds a namespace prefix mapping to be included in the next start tag.

writer.end_ns()

does nothing (namespace scope is handled automatically).

writer.iterwrite(events)

writes XML data based on (event, elem) tuples of the kind that you get from iterparse in ElementTree and lxml. start, end, start-ns, end-ns, comment and pi events are currently supported. Note that the events iterable must include start events, since the document structure can't be inferred from end elements alone.

writer.close()

Closes all open elements.

Attribute ordering

For more control over attribute ordering, you can pass a dictionary to the constructor's sort argument. In that dictionary:

• each key is a tag name, and
• each corresponding value is a list of attribute names in the order you want them to occur in the XML data for that tag.

If None appears in the list, it acts as a wildcard for all attributes not explicitly named in the list. (By default, they will be placed last.)

Example::

attrib_order = {
    "person": ["id", "first_name", "last_name"],
    "foo": ["id", None, "put_me_last"],
}

License

Copyright (c) 2009-2020 Filip Salomonsson

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

(That's the MIT license.)