批量分析word文档中的数据

1 读取文件名字

import docx
import re, os
word_name_list = os.listdir(r'C:\Users\xue\Desktop\各种表\2019级研究生培养计划\all')
word_name_list = [i for i in word_name_list if i.endswith('.doc') and '$' not in i]
print(len(word_name_list),'\n',word_name_list[:10])

输出

2 尝试采用py-win32

py-win32可以同时处理doc和docx

import docx
import re, os
from win32com.client import Dispatch

word = Dispatch('Word.Application')     # 打开word应用程序
# word = DispatchEx('Word.Application') # 启动独立的进程
word.Visible = 0        # 后台运行,不显示
word.DisplayAlerts = 0  # 不警告

path = r"C:\Users\xue\Desktop\各种表\2019级研究生培养计划\all\畅檀应用统计QSZ20190070.doc"
doc = word.Documents.Open(FileName=path, Encoding='gbk')

# for para in doc.paragraphs: # 会出现表格数据读取不全
#     print(para.Range.Text)

for t in doc.Tables: 3 # 依然容易出错!!并且出现表格数据读取不全
    for row in t.Rows:
        for cell in row.Cells:
            print(cell.Range.Text)


doc.Close()
word.Quit

输出出错！

3 改用docx模块

docx模块只能处理docx后缀文件！

import pythoncom
import os
import win32com.client as wc

4 .doc文件转化为.docx文件

4.1 先将所有文件转化为docx

root = r'C:\Users\xue\Desktop\各种表\2019级研究生培养计划\all'

#　找出doc文件名字
doc_name_list = os.listdir(root)
doc_name_list = [i for i in doc_name_list if i.endswith('.doc') and '$' not in i]
print(len(doc_name_list),'\n',doc_name_list[:3])

4.2 找出docx文件名字

docx_name_list = os.listdir(root)
docx_name_list = [i for i in docx_name_list if i.endswith('.docx') and '$' not in i]
print(len(docx_name_list),'\n',docx_name_list[:])

4.3 先创建一个放docx的文件夹

docx_dir = root + '\\docx'
if not os.path.exists(docx_dir):
    os.makedirs(docx_dir)

4.4 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)
    

4.5 doc转化为docx

def doc_to_docx(doc_name):
    pythoncom.CoInitialize()
    try:
        word = wc.Dispatch("Word.Application") # 打开word应用程序
        doc = word.Documents.Open(doc_name, Encoding='utf-8')
        # 上面的地方只能使用完整绝对地址，相对地址找不到文件，
        # 并且，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。
        doc_name = rreplace(doc_name, "\\", "\\docx\\", 1)
        doc.SaveAs(doc_name.replace(".doc", ".docx"), 12, False, "", True, "", False, False, False, False)
        # 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()

4.6 改变doc文件名字为docx，并存到docx文件夹下面

for i in doc_name_list:
    doc_to_docx(root + '\\' + i)

4.7 移动docx文件到docx文件夹

import shutil
for i in docx_name_list:
    shutil.move(root + '\\' + i, root + "\\docx")

5 读取所有的docx列表（就是所有已经转化为docx的文件）

import os

root = r"C:\Users\xue\Desktop\各种表\2019级研究生培养计划\all\docx" # 改变根路径
word_name_list = os.listdir(root)
word_name_list = [i for i in word_name_list if '.docx' in i and '$' not in i]
print(len(word_name_list),'\n',word_name_list[:5])

6 读取docx文件中的所有字符串

import docx
def read_docx(fn):
    s = ''
    doc = docx.Document(fn)

    # # 按段落读取全部数据
    # for paragraph in doc.paragraphs:
    #     print(paragraph.text)
    # doc = docx.Document(fn)

    # # 按段落读取全部数据
    # for paragraph in doc.paragraphs:
    #     print(paragraph.text)

    # 按表格读取全部数据
    text_list = []
    for table in doc.tables:
        for row in table.rows:
            for cell in row.cells:
                if cell.text not in text_list:
                    text_list.append(cell.text.strip())
    s = ''.join(text_list)
    return s

    # table_num = len(doc.tables)
    # # 获取文档的表格个数
    # print(table_num)

    # table_0 = doc.tables[0]
    # # 选取第一个表
    # table_rows = len(table_0.rows)
    # # 获取第一个表的行数
    # print(table_rows)

    # tab = doc.tables[0].rows[0].cells[0]
    # # 获取第一张表第一行第一列数据
    # print(tab.text)

    # par = doc.paragraphs[2]
    # # 读取第三段数据
    # print(par.text)

7 读取和匹配信息

import re
information = []
for i, item in enumerate(word_name_list):
    text = read_docx(root + '\\' + item)
#     print(text[:1000])
    pattern = re.compile(r'名(.*?)入')
    names = pattern.findall(text)[0]
    names = names.replace(' ', '')
    pattern = re.compile(r'职务(.*?)专')
    school = max(pattern.findall(text)[0].split(), key = len)
    school = school.replace(' ', '')
    years = re.compile(r'学制(.*?)入学').findall(text)[0]
    kind = re.compile(r'类别(.*?)入学').findall(text)[0]
    major = re.compile(r'称(.*?)研').findall(text)[0]
    field = re.compile(r'向(.*?)等').findall(text)[0]

#     print('-'* 30)
    info = (i, names, school, years, kind, major, field)
    information.append(info)
    print(info)
print(len(information), information[:3], information[-3:], sep = '\n')

输出

8 测试代码（分析数据，编写正则）

# 读取和匹配信息
import re
information = []

text = read_docx(root + '\\' + word_name_list[27])
# print(text[:1000])
pattern = re.compile(r'名(.*?)入')
names = pattern.findall(text)[0]
names = names.replace(' ', '')
pattern = re.compile(r'职务(.*?)专')
school = pattern.findall(text)
print(school)
# print(max(school, key=len))
years = re.compile(r'学制(.*?)入学').findall(text)[0]
kind = re.compile(r'类别(.*?)入学').findall(text)[0]
major = re.compile(r'称(.*?)研').findall(text)[0]
field = re.compile(r'向(.*?)等').findall(text)[0]

#     print('-'* 30)
info = (str(i), names, school, years, kind, major, field)
information.append(info)
print(info)
# print(len(information), information[:3], information[-3:], sep = '\n')

输出

9 存入数据到磁盘

9.1 存入txt

path = r"C:\Users\xue\Desktop\各种表\2019级研究生培养计划\2019所有研究生信息汇总.txt"
with open(path, 'w') as fp:
    fp.write('\n'.join('%s %s %s %s %s %s %s' % x for x in information))

9.2 转化为pandas并存入csv

import pandas as pd

path = r"C:\Users\xue\Desktop\各种表\2019级研究生培养计划\2019所有研究生信息汇总.csv"
data = information
print(data[:5])

df = pd.DataFrame(data=data, columns=['序号','姓名','本科学校','学制','学位类别','专业','研究方向'])
df = df.reindex(index = df['序号']) # 重新设置索引为第一列
df = df.drop(columns=['序号'], axis=1) # 删除“序号”这一列
df = df.rename_axis("序号", axis=0).rename_axis(None, axis=1)  # 设置行索引的名称
df.to_csv(path)

输出

9.3 存入excel

path = r"C:\Users\xue\Desktop\各种表\2019级研究生培养计划\2019所有研究生信息汇总.xls"
df.to_excel(path)