image.png

# pip install python-docx

import docx
def Stack():
    def __init__(self):
        self.items = []

    def is_empty(self):
        return self.items == []

    def push(self, item):
        return self.items.append(item)

    def pop(self):
        return self.items.pop()

    def peek(self):
        return self.items[len(self.items) - 1]

    def size(self):
        return len(self.items)


for i in files:
    doc = docx.Document(i)
    stack = Stack()
    res = []
    tmp_content = []
    for i ,j in enumerate(doc.paragraphs):
        if j.text == '': continue
        item = {}
        record = {}
        style_name = j.style.name
        # 如果当前过来的是标题，则将当前标题之前的内容写入ES
        if style_name.startswith('Heading'):
            if style_name.startswith('Heading1'):
                item['style'] = 1
                item['text'] = j.text
            elif style_name.startswith('Heading2'):
                item['style'] = 2
                item['text'] = j.text
            elif style_name.startswith('Heading3'):
                item['style'] = 3
                item['text'] = j.text
            # 如果当前标题之前的正文内容不为空
            # 1、合并多个正文段落
            # 2、合并多级标题
            if tmp_content != []:
                # 写入ES
                content = ' '.join(tmp_content)
                title = ''
                for i in range(stack.size()):
                    title_text = stack.items[i]
                    text = title_text['text']
                    title = title + '-' + text
                record['title'] = title
                record['content'] = content
                tmp_content = []
            # 清空栈
            # 取出栈顶元素，与当前标题对比，如果栈顶标题的级别 >= 当前标题，则清除，直到小于当前标题
            while not stack.is_empty():
                top = stack.peek()
                top_style = top['style']
                if top_style >= item['style']:
                    stack.pop()
                else:
                    break
            # 压入当前标题
            stack.push(item)
        # 如果当前过来的是正文，则放到正文的临时列表
        elif style_name.startswith('Normal'):
            tmp_text = j.text.strip('\n')
            tmp_text = tmp_text.strip(' ')
            tmp_content.append(tmp_text)

    for i ,item in enumerate(res):
        if item['title'] != '' and len(item['content']) != 0:
            title_strip = item['title'].strip('-')
            insert_es() # 待续