Working with DOCX Reader

Working with DOCX Reader

The DocumentReader class reads Office Open XML (.docx) files and produces an abstracted data structure — the Light Document Model (LDM). It handles paragraphs, tables, shapes, images, and formatting, making the content available for manipulation or conversion.

Reading a DOCX File

The simplest approach is using the Document constructor, which automatically selects DocumentReader for .docx files:

import aspose.words_foss as aw

doc = aw.Document("input.docx")
print(f"Sections: {len(doc.sections)}")
for para in doc.all_paragraphs:
    print(para.text)

DocumentReader Architecture

DocumentReader uses two mixins to split responsibilities:

All three work together internally — you interact with the result through the Document API.

Extracting Text

Use get_text() on a loaded document to extract plain text without saving to a file:

import aspose.words_foss as aw

doc = aw.Document("contract.docx")
text = doc.get_text()
print(text[:500])

Working with Paragraphs and Formatting

Access paragraph-level properties through the LDM:

import aspose.words_foss as aw

doc = aw.Document("styled.docx")
for para in doc.all_paragraphs:
    for run in para.runs:
        if run.font.bold:
            print(f"Bold: {run.text}")

Each Run carries font properties (bold, italic, size, color) that were parsed from the DOCX XML.

Handling Shapes and Images

The ShapeParserMixin extracts drawing elements during parsing. Shapes appear in the LDM as ShapeNode objects within paragraph content:

import aspose.words_foss as aw

doc = aw.Document("with-images.docx")
for para in doc.all_paragraphs:
    for node in para.content_sequence:
        if hasattr(node, 'shape_type'):
            print(f"Shape: {node.shape_type}")

Tips and Best Practices

• Use Document("file.docx") for standard workflows — it handles reader selection, package validation, and LDM construction automatically.
• For stream-based input (web uploads), pass a file-like object to the Document constructor.
• For files with tables, access them via doc.tables or doc.all_tables.
• For batch processing, reuse the same Python process to avoid repeated import overhead.

Common Issues

FAQ

What is the Light Document Model (LDM)?

The LDM is the internal representation shared across all readers. It abstracts away format-specific details (OPC XML, OLE2 binary) into a common structure of sections, paragraphs, runs, tables, and shapes.

Can DocumentReader handle DOCM (macro-enabled) files?

Macro-enabled .docm files use the same OPC structure. The reader can parse the document content, but macros are not preserved or executed.

How does the reader handle large documents?

The reader processes the XML parts sequentially. Memory usage scales with document size. For very large documents, consider processing sections individually.

API Reference Summary