"""
将文章中的关键字加粗（第一行中关键字不加粗, 第一行后面_下划线删除）,去除标题中的日期
"""
import os
from docx import Document
import pythoncom
import re
from docx.opc.exceptions import PackageNotFoundError
from docx.enum.text import WD_PARAGRAPH_ALIGNMENT


# 读取文件夹下的docx文件名列表
def docx_file_name(file_dir):
    file_list = []
    for root, dirs, files in os.walk(file_dir):
        for file in files:
            if os.path.splitext(file)[1] == '.docx':
                file_list.append(os.path.join(root, file))
    return file_list


# 取关键词
def keyword_find(doc):
    paragraph1 = doc.paragraphs[0].text
    key_word = paragraph1.split("_")[-1]
    return key_word


# 去除标题中的_
def replace_(doc):
    para = doc.paragraphs[0]
    for i in range(len(para.runs)):
        para.runs[i].text = para.runs[i].text.replace("_", "")
    return doc


# 删除标题中的日期
def replace_date(doc):
    para = doc.paragraphs[0]
    # 删除“-2018-09-20”或“-2018-2-13”
    para.text = re.sub(r"(-\d{4}-\d{1,2}-\d{1,2})", "", para.text)
    # 删除“2018-09-20”或“2018-2-13”
    para.text = re.sub(r"(\d{4}-\d{1,2}-\d{1,2})", "", para.text)
    # 删除“2018年09月20日”或“2018年2月13日”
    para.text = re.sub(r"(\d{4}\u5E74\d{1,2}\u6708\d{1,2}\u65E5)", "", para.text)
    # 删除“20180324”
    para.text = re.sub(r"(\d{8})", "", para.text)
    return doc

# 删除正文中的一句话
def replace_text(doc):
    for i in range(1, len(doc.paragraphs)):
        para = doc.paragraphs[i]
        # 删除“这*发布的”  “这*发布”
        para.text = re.sub(r"(\u8fd9)([\u4e00-\u9fa5a-zA-Z]*)(\u53d1\u5e03)([\u7684]*)", "", para.text)
        # 删除“此*发布的”  “此*发布”
        para.text = re.sub(r"(\u6b64)([\u4e00-\u9fa5a-zA-Z]*)(\u53d1\u5e03)([\u7684]*)", "", para.text)
        # 删除“本*发布的”  “本*发布”
        para.text = re.sub(r"(\u672c)([\u4e00-\u9fa5a-zA-Z]*)(\u53d1\u5e03)([\u7684]*)", "", para.text)
        # 删除“战况报告由*发布的”  “战况报告由*发布”
        para.text = re.sub(r"(\u6218\u51b5\u62a5\u544a\u7531)([\u4e00-\u9fa5a-zA-Z]*)(\u53d1\u5e03)([\u7684]*)", "", para.text)
        # 删除“比赛报告由*发布的”  “比赛报告由*发布”
        para.text = re.sub(r"(\u6bd4\u8d5b\u62a5\u544a\u7531)([\u4e00-\u9fa5a-zA-Z]*)(\u53d1\u5e03)([\u7684]*)", "", para.text)
        # 删除“由*发布的”  “由*发布”
        para.text = re.sub(r"(\u7531)([\u4e00-\u9fa5a-zA-Z]*)(\u53d1\u5e03)([\u7684]*)", "", para.text)

        # [来*在等着你
        para.text = re.sub(r"(\[\u6765)([\s\S]*)(\u5728\u7b49\u7740\u4f60)", "", para.text)
        # [来*等着您！
        para.text = re.sub(r"(\[\u6765)([\s\S]*)(\u7b49\u7740\u60a8\uff01)", "", para.text)
        # 【足球*应用
        para.text = re.sub(r"(\u3010\u8db3\u7403)([\s\S]*)(\u5e94\u7528)", "", para.text)
        # 【*等你来拿。
        para.text = re.sub(r"(\u3010)([\s\S]*)(\u7b49\u4f60\u6765\u62ff\u3002)", "", para.text)
        # 【来*等你领！
        para.text = re.sub(r"(\u3010\u6765)([\s\S]*)(\u7b49\u4f60\u9886\uff01)", "", para.text)
        # 【来*等您来
        para.text = re.sub(r"(\u3010\u6765)([\s\S]*)(\u7b49\u60a8\u6765)", "", para.text)
        # [如果*应用中找到。
        para.text = re.sub(r"(\[\u5982\u679c)([\s\S]*)(\u5e94\u7528\u4e2d\u627e\u5230\u3002)", "", para.text)
        # [，*查看！
        para.text = re.sub(r"(\[\uff0c)([\s\S]*)(\u67e5\u770b\uff01)", "", para.text)
        # 新用户*等着你去取。
        para.text = re.sub(r"(\u65b0\u7528\u6237)([\s\S]*)(\u7b49\u7740\u4f60\u53bb\u53d6\u3002)", "", para.text)
        # 来-掷-赌！
        para.text = re.sub(r"(\u6765\u002d\u63b7\u002d\u8d4c\uff01)", "", para.text)

    return doc

# 删除正文中的日期
def replace_date_1(doc):
    for i in range(1, len(doc.paragraphs)):
        para = doc.paragraphs[i]
        # 删除“2018年09月20日，”或“2018年2月13日，”
        para.text = re.sub(r"(\d{4}\u5E74\d{1,2}\u6708\d{1,2}\u65E5\uFF0C)", "", para.text)
        # 删除“2018年09月20日”或“2018年2月13日”
        para.text = re.sub(r"(\d{4}\u5E74\d{1,2}\u6708\d{1,2}\u65E5)", "", para.text)
    return doc

# 加粗关键词
def key_word_bold(doc, keyword):
    for i in range(1, len(doc.paragraphs)):
        p = doc.paragraphs[i]
        # p.alignment = WD_PARAGRAPH_ALIGNMENT.LEFT
        for r in p.runs:
            if keyword not in r.text:
                pass
            rest = r.text.split(keyword)
            r.style = r.style
            r.text = ''
            for text in rest[:-1]:
                run = p.add_run(text=text)
                run.bold = False
                run = p.add_run(keyword)
                run.bold = True
            run = p.add_run(rest[-1])
            run.bold = False
    return doc


# 替换换行符（即向下箭头）
def replace_n_n(doc):
    for para in doc.paragraphs:
        for i in range(len(para.runs)):
            para.runs[i].text = para.runs[i].text.replace("\n\n", "\n")
    return doc


# 设置段落左对齐
def align_left(doc):
    for para in doc.paragraphs:
        # for left, 1 for center, 2 right, 3 justify ....
        para.alignment = WD_PARAGRAPH_ALIGNMENT.LEFT
    return doc


def main():
    file_list = docx_file_name("D:\\文件处理\\2020.8.12-350篇足球")
    print(len(file_list))

    for file in file_list:
        pythoncom.CoInitialize()
        try:
            doc = Document(file)
            print(len(doc.paragraphs))

            # 根据标题中的_查找关键词
            keyword = keyword_find(doc)
            # 去除标题中的_
            replace_(doc)
            # 去除标题中的日期
            replace_date(doc)
            # 去除正文中的日期
            replace_date_1(doc)
            # 去除正文中的一句话
            replace_text(doc)
            # 替换 \n\n 为 \n, 避免增加空行
            replace_n_n(doc)

            # 加粗关键词
            if keyword == "":
                pass
            elif "2020" in keyword:
                key_word_bold(doc, "2020")
                key_word_bold(doc, keyword.replace("2020", ""))
            else:
                key_word_bold(doc, keyword)

            doc.save(file)
        except PackageNotFoundError:
            pass
        finally:
            # 释放资源
            pythoncom.CoUninitialize()


if __name__ == '__main__':
    main()