Working with XML in Python is a practical skill for anyone handling structured data transfers between systems. Unlike JSON, XML carries its own schema through Document Type Definitions and XSD files, making validation a first-class concern. This guide walks through parsing, editing, and generating XML using battle-tested libraries that fit into production workflows.
Choosing the Right XML Library for Python
The standard library supplies multiple XML solutions, and picking the right one depends on your constraints. For most tasks, ElementTree offers a lightweight API with decent performance and broad compatibility. If you need stricter validation against a schema, lxml adds XPath, XSLT, and DTD support while remaining familiar to ElementTree users. The built-in xml package also includes expat for event-driven parsing when memory usage must stay low.
Parsing XML with ElementTree
ElementTree reads an XML file or byte stream into a tree of elements that you can search and modify. The parse function handles files, while fromstring works with text already in memory. Once you have an Element object, methods like find and findall let you locate nodes by tag name or a simple XPath expression.
Example: Reading and Querying a Document
Use ElementTree.parse to load and access the root node.
Iterate over child elements to inspect repeating structures.
Retrieve attributes and text content with element.get and element.text.
Safe Handling of External XML Sources
XML features like external entities and document type declarations can open the door to billion laughs attacks and other exploits. The default parser in ElementTree disables external DTDs, but you should still avoid passing untrusted data to legacy XML functions. When you need more security, lxml provides parser options to block external entities and inline DTDs while preserving compatibility with ElementTree code.
Modifying and Writing XML
After locating elements, you can assign new text, update attributes, or insert and remove nodes. The ElementTree module writes the tree back to disk with write, where you control encoding and whether to include an XML declaration. For more control over formatting, use lxml with its pretty_print option to generate clean, human-readable output suitable for configuration files and API payloads.
XPath and Namespaces in Practice
Many real-world XML documents use namespaces, and ignoring them leads to confusing errors where find returns None. You can pass a namespace map to ElementTree methods or register prefixes so that XPath expressions resolve correctly. This approach keeps queries concise when you work with SOAP responses, SVG, or enterprise message formats that rely heavily on namespace-qualified tags.
Transforming XML with XSLT
XSLT shines when you need to reshape XML into another format, such as HTML for reporting or a simplified schema for downstream processing. The lxml library exposes a clean transform interface where you load an XSLT stylesheet once and apply it to multiple source documents. This pattern is efficient for batch conversions and keeps presentation logic separate from application code.
