Python实现对比Word文档差异的三种方式小结

pip install spire.doc

from spire.doc import *


# 加载原始文档
original = Document()
original.LoadFromFile("原始.docx")

# 加载修订文档
revised = Document()
revised.LoadFromFile("修订.docx")

# 执行对比
original.Compare(revised, "Reviewer")

# 保存对比结果（包含修订标记）
original.SaveToFile("对比结果.docx", FileFormat.Docx2013)

from spire.doc import *

original = Document()
original.LoadFromFile("原始.docx")
revised = Document()
revised.LoadFromFile("修订.docx")

# 设置对比选项
options = CompareOptions()
options.IgnoreFormatting = True   # 忽略字体、颜色和样式变化

# 执行对比
original.Compare(revised, "Reviewer", DateTime.get_Now(), options)

# 保存结果
original.SaveToFile("忽略格式差异.docx", FileFormat.Docx2013)

from spire.doc import *

def save_text(fname: str, content: str):
    """将字符串内容保存到文本文件"""
    with open(fname, "w", encoding="utf-8") as fp:
        fp.write(content)

def format_revision(index: int, revision, content: str) -> str:
    """格式化修订信息，包含序号、类型、作者和修订内容"""
    return (
        f"序号: {index}\n"
        f"修订类型: {revision.Type.name}\n"
        f"作者: {revision.Author}\n"
        f"修订内容: {content}\n"
    )

# 输入的对比结果文档
input_file = "文档比对结果.docx"
# 输出文件：插入修订
output_insert = "插入修订.txt"
# 输出文件：删除修订
output_delete = "删除修订.txt"

# 创建 Document 对象并加载 Word 文件
document = Document()
document.LoadFromFile(input_file)

# 用于收集插入和删除修订结果
insert_revisions = ["【插入修订列表】"]
delete_revisions = ["【删除修订列表】"]

# 插入和删除修订的计数器
index_insert = 0
index_delete = 0

# 遍历文档中的所有节
for k in range(document.Sections.Count):
    sec = document.Sections.get_Item(k)

    # 遍历节中的所有子对象
    for m in range(sec.Body.ChildObjects.Count):
        doc_item = sec.Body.ChildObjects.get_Item(m)

        # 判断对象是否为段落
        if isinstance(doc_item, Paragraph):
            para = doc_item

            # 段落级别的插入修订
            if para.IsInsertRevision:
                index_insert += 1
                insert_revisions.append(format_revision(index_insert, para.InsertRevision, para.Text))

            # 段落级别的删除修订
            elif para.IsDeleteRevision:
                index_delete += 1
                delete_revisions.append(format_revision(index_delete, para.DeleteRevision, para.Text))

            # 遍历段落中的所有子对象
            for j in range(para.ChildObjects.Count):
                obj = para.ChildObjects.get_Item(j)

                # 判断对象是否为文本范围
                if isinstance(obj, TextRange):
                    text_range = obj

                    # 文本级别的插入修订
                    if text_range.IsInsertRevision:
                        index_insert += 1
                        insert_revisions.append(format_revision(index_insert, text_range.InsertRevision, text_range.Text))

                    # 文本级别的删除修订
                    elif text_range.IsDeleteRevision:
                        index_delete += 1
                        delete_revisions.append(format_revision(index_delete, text_range.DeleteRevision, text_range.Text))

# 将结果保存到文本文件
save_text(output_insert, "\n".join(insert_revisions))
save_text(output_delete, "\n".join(delete_revisions))