"""
修改文档格式
"""
import os
import re

import pythoncom
import win32com.client as wc
from docx import Document
from docx.shared import Inches


def doc_file_name(file_dir):
    fileList = []
    for root, dirs, files in os.walk(file_dir):
        print(root)
        print(dirs)
        for file in files:
            if os.path.splitext(file)[1] == '.doc':
                fileList.append(os.path.join(root, file))
                docx_dir = root + '\\docx'
                if not os.path.exists(docx_dir):
                    os.makedirs(docx_dir)
    return fileList



# 从最后开始替换某字符串几次
def rreplace(s, old, new, occurrence):
    li = s.rsplit(old, occurrence)
    return new.join(li)


# doc文件另存为docx
def doc_to_docx(docName):
    pythoncom.CoInitialize()
    try:
        word = wc.Dispatch("Word.Application")
        doc = word.Documents.Open(docName)
        # 上面的地方只能使用完整绝对地址，相对地址找不到文件，且，只能用“\\”，不能用“/”，哪怕加了 r 也不行，涉及到将反斜杠看成转义字符。
        docName = rreplace(docName, "\\", "\\docx\\", 1)
        doc.SaveAs(docName.replace(".doc", ".docx"), FileFormat=12, Encoding=65001)
        # print(len(doc.paragraphs))
        # 内联图片个数
        # print(len(doc.InlineShapes))
        # 转换后的文件,12代表转换后为docx文件
        doc.Close
    except Exception as e:
        print(e.message)
    finally:
        # 对com操作，一定要确保退出word应用
        if word:
            word.Quit
            del word
        # 释放资源
        pythoncom.CoUninitialize()


# 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


# 替换换行符
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


# 正文插入两幅图片
def replace_n_picture(doc, picture1, picture2):
    para = doc.paragraphs[1]
    print(len(para.runs))
    for i in range(len(para.runs)):
        print(i)
        if '\n' in para.runs[i].text:
            #para.runs[i+1].add_break()  # 添加一个折行
            para.runs[i+1].add_picture(picture1, width=Inches(5.25))  # 在runs的最后一段文字后添加图片
            para.runs[i+1].add_break()
            break

    return doc



# 删除日期
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


# 插入空行
def add_null_line(doc):
    run = doc.paragraphs[0].add_run()
    run.add_break()
    return doc


def replace(string, new_string):
    word = wc.DispatchEx('Word.Application')
    word.Visible = 1
    word.DisplayAlerts = 0
    word.Selection.Find.Execute(string, False, False, False, False, False, True, 1, True, new_string, 2)


# 标题行之后插入两张图片
def add_picture(doc, picture1, picture2):
    run = doc.paragraphs[0].add_run()
    run.add_break()
    run.add_picture(picture1, width=Inches(5.25))
    run.add_break()
    run.add_picture(picture2, width=Inches(5.25))


# 标题行之后插入一段文字及一幅图片
def add_para_picture(doc, para, picture):
    run = doc.paragraphs[0].add_run("\n" + para)
    run.add_break()
    run.add_picture(picture, width=Inches(5.25))


# 随机取列表中的一个值返回
def pick_random_one(list):
    import random
    return random.choice(list)

# 加粗关键词
def key_word_bold(doc, keyword):
    for i in range(1, len(doc.paragraphs)):
        p = doc.paragraphs[i]
        for r in p.runs:
            if keyword not in r.text:
                pass
            rest = r.text.split(keyword)
            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


# 替换文本
def add_pic_after_text(doc, expect_text, picture_name):
    for i in range(1, len(doc.paragraphs)):
        p = doc.paragraphs[i]
        for r in p.runs:
            if expect_text not in r.text:
                pass
            rest = r.text.split(expect_text)
            r.text = ''
            for text in rest[0]:
                run = p.add_run(text=text)
                run.bold = False
                run = p.add_run(expect_text)
                run.add_picture(picture_name)
            run = p.add_run(rest[-1])
            run.bold = False


def find_char_pos(doc):
    p2 = doc.paragraphs[1].text
    brace_list = []
    for i, ch in enumerate(p2):
        if ch == "。":
            brace_list.append(i)
    p_new = p2.replace(p2[brace_list[1]], "\r\n[", 1)


# 读取文件夹下的特定前缀的文件名列表
def file_name(file_dir, file_prefix):
    fileList = []
    for root, dirs, files in os.walk(file_dir):
        for file in files:
            if os.path.splitext(file)[0].startswith(file_prefix):
                fileList.append(os.path.join(root, file))
    return fileList


def main():
    # doc_to_docx("D:\\workspace\\python-test\\N标题81.doc")
    doc_file_name("D:\\文件处理\\2020.6.29-200篇足球文章")
    # doc = Document("D:\\workspace\\python-test\\N标题8.docx")
    # print(len(doc.paragraphs))
    #
    # for parg in doc.paragraphs:
    #     # 仅修改文本
    #     # runt = []
    #     # for run in parg.runs:
    #     #     if run.text:
    #     #         runt.append(run.text)
    #     #         run.text = ''
    #     # parg.add_run('test***'+''.join(runt))
    #     # 仅保留文本
    #     if parg.text:
    #         parg.text = parg.text
    # doc.save("D:\\workspace\\python-test\\new-N标题8.docx")

    # replace_date(doc)
    # doc.save("D:\\workspace\\python-test\\docx\\2016年11月5日 第11轮格拉纳达VS拉科鲁尼亚全战巴塞罗那.doc")

    # replace_n_n(doc)
    # replace_n_picture(doc, "D:\\workspace\\python-test\\足球图片\\002807uf4yb2dccedfojmi.jpg", "D:\\workspace\\python-test\\足球图片\\002807uf4yb2dccedfojmi.jpg")

    # para = ['test1', 'test2', 'test3']
    # picture = "D:\\workspace\\python-test\\足球图片\\002807uf4yb2dccedfojmi.jpg"
    # add_para_picture(doc, pick_random_one(para), picture)
    # doc.save("D:\\workspace\\python-test\\N标题8-addpara.doc")

    # keyWord = keyword_find(doc)
    # replace_(doc)
    # 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)

    #find_char_pos(doc)
    #add_pic_after_text(doc, "。", "003019z4a1swz4ya2agava_wps图片.jpg")
    # add_null_line(doc)
    # print(len(doc.paragraphs))
    #add_picture(doc, "003019z4a1swz4ya2agava_wps图片.jpg", "002730v6pxgdd2462fe78z_wps图片.jpg")

    # replace_date(doc)
    #doc.save("replace_date.doc")
    # doc.save("D标题44_temp.docx".replace(".docx", ".doc"))


if __name__ == '__main__':
    main()