Docx Toolkit — Docx 工具kit

DOCX 工具kit

A complete 工具kit for processing Microsoft Word documents (.docx and legacy .doc 格式化s).

Capabilities

• Text + Table 提取ion (.docx)

python3 {baseDir}/scripts/提取_text.py 输入.docx 输出.txt

提取s all paragraphs and tables with structure preserved. Tables are 格式化ted as pipe-delimited rows for easy parsing.

• Text 提取ion (Legacy .doc)

python3 {baseDir}/scripts/提取_doc_text.py 输入.doc 输出.txt

Handles legacy OLE2 .doc 格式化 using olefile. 提取s Unicode text from the WordDocument 流.

• Image 提取ion (.docx)

python3 {baseDir}/scripts/提取_images.py 输入.docx 输出_dir/

提取s all embedded images with:

Automatic deduplication (MD5 哈希 comparison) Size 过滤器ing (skips tiny icons <5KB by default) Sequential renaming (img_001.png, img_002.jpg, etc.)

• Image 压缩ion

python3 {baseDir}/scripts/resize_images.py 输入_dir/ 输出_dir/ [--max-width 1024]

Batch resize/压缩 images for API processing (saves 50-70% on vision API costs).

Dependencies Python 3.6+ python-docx — for .docx processing olefile — for legacy .doc processing Pillow — for image resizing (optional, only needed for resize script)

安装:

pip3 安装 python-docx olefile Pillow

Use Cases Document analysis: 提取 text for AI review/summarization 迁移: Pull content from Word docs into other 格式化s Image 审计: 提取 and review all embedded images Cost optimization: 压缩 images before 发送ing to vision APIs Batch processing: Process multiple documents in a 流水线 Notes Large .doc files (>200MB) may require 签名ificant RAM for olefile processing Image 提取ion preserves original 格式化 (png/jpg/gif/etc.) Deduplication catches exact duplicates; near-duplicates still pass through CJK (Chinese/Japanese/Korean) text is fully supported in 机器人h 提取器s