初用python-docx

事情的起因是这样的，女朋友的公司在做一个数字化管道的工作，需要在大量的word中提取想要的数据，手动输入实在太麻烦，就看到这个python-docx的库，肥肠的高效，脚本发出来给大家。

!/usr/bin/python

-- coding: UTF-8 --

author = 'Administrator'

读取docx中的文本代码示例

import docx
from docx import Document #导入库
import re
import glob
import os

第一种表格

path = "E:\test\test.docx" #文件路径

document = Document(path) #读入文件

tables = document.tables #获取文件中的表格集

for i in range(0,len(tables)):

table = tables[0]#获取文件中的第一个表格

print(table.cell(2,1).text)

print(table.cell(4,1).text)

print(table.cell(4,6).text)

datestr = table.cell(8,0).text.replace(" ","")

mat = re.search(r"(\d{4}年\d{1,2}月\d{1,2}日)",datestr)

print(mat.group(0))

for i in range(0,len(table.rows)):#从表格第二行开始循环读取表格数据

result = table.cell(i,0).text + "" +table.cell(i,1).text+table.cell(i,2).text + table.cell(i,3).text

#cell(i,0)表示第(i+1)行第1列数据，以此类推

print(result)

第二种表格

f = open("result.txt",'w',encoding='UTF-8')

path = "test02.docx" #文件路径

document = Document(path) #读入文件

tables = document.tables #获取文件中的表格集

table = tables[0]#获取文件中的第一个表格

print(len(tables))

for table in tables:

for i in range(0,len(table.rows)):#从表格第二行开始循环读取表格数据

if table.cell(i,0).text.isdigit():

mark = table.cell(i,10).text

num = mark

split = ""

m = '0'

if mark.endswith('m'):

mat = re.search(r"([\d,\.]*m)",mark)

m =mat.group(0)

num = mark.replace(m,"")

split = num[len(num)-1]

num=num[0:len(num)-1]

result = table.cell(i,0).text + "\t" +table.cell(i,1).text+"\t"+table.cell(i,7).text +"\t"+ table.cell(i,6).text+"\t"+ table.cell(i,2).text.replace("X:","")+"\t"+ table.cell(i,3).text.replace("Y:","")+"\t"+ num+"\t"+ split+"\t"+ m.replace('m','')

print(result)

f.write(result+"\n")

f.close()

第三种表格

os.chdir("E:\test\input\815\QAB管道焊缝检测报告\")

paths = []
paths.extend(glob.glob('*.%s' % 'docx'))
fs = open("射线.txt",'w',encoding='UTF-8')
fc = open("超声波.txt",'w',encoding='UTF-8')
for path in paths:
#path = "1.管道焊缝检测报告（QAB171-1～QAB189-1）.docx" #文件路径
document = Document(path) #读入文件
tables = document.tables #获取文件中的表格集
#table = tables[0]#获取文件中的第一个表格
#print(len(tables))
for table in tables:
title = (table.cell(0,2).text)
if("无损检测返修通知单" not in title and "超声波检测报告" not in title and "附页" in title):
a2 = (table.cell(1,2).text)
a3 = (table.cell(2,2).text.replace(" ",""))
try:
a6 = table.cell(24,0).text.split("\n")[0].replace("评定人员：","").replace("评定人员:","").split("级 别：")[0]
a7 = table.cell(24,11).text.split("\n")[0].replace("监理（签字）：","")
except IndexError as e:
a6 = table.cell(23,0).text.split("\n")[0].replace("评定人员：","").replace("评定人员:","").split("级 别：")[0]
a7 = table.cell(23,11).text.split("\n")[0].replace("监理（签字）：","")

        for i in range(0,len(table.rows)):#从表格第二行开始循环读取表格数据
            if table.cell(i,0).text.isdigit():
                result = table.cell(i,1).text+"\t"+a2+"\t"+a3+"\t"+table.cell(i,10).text+"\t"+table.cell(i,12).text+"\t"+a6+"\t"+a7
                result = result.replace('\n','')
                print(path+"==="+result)
                fs.write(result+"\n")
    elif("无损检测返修通知单" not in title and "射线检测报告" not in title and "附页" in title):
        a2 = (table.cell(1,2).text)
        a3 = (table.cell(2,2).text.replace(" ",""))
        a6 = table.cell(25,0).text.split("\n")[0].replace("检测人员：","").replace("检测人员:","").split("级  别：")[0]
        a7 = table.cell(25,11).text.split("\n")[0].replace("监理（签字）：","")
        for i in range(0,len(table.rows)):#从表格第二行开始循环读取表格数据
            if table.cell(i,0).text.isdigit():
                result = table.cell(i,1).text+"\t"+a2+"\t"+a3+"\t"+table.cell(i,6).text+" "+table.cell(i,7).text+" "+table.cell(i,9).text+"\t"+table.cell(i,12).text+"\t"+a6+"\t"+a7
                result = result.replace('\n','')
                print(path+"==="+result)
                fc.write(result+"\n")

fs.close()
fc.close()

第四种表格

f = open("result.txt",'w',encoding='UTF-8')

path = "E:\test\input\815\1.docx" #文件路径

document = Document(path) #读入文件

tables = document.tables #获取文件中的表格集

table = tables[0]#获取文件中的第一个表格

print(len(tables))

for table in tables:

a = table.cell(1,4).text

ass = a.split("-")

if len(ass) < 3:

a =ass[0]

else:

print(a)

b = table.cell(8,2).text

bs = b.replace('\n\n\n',"\n").split('\n')

b1 = bs[0].replace("监理代表：","")

b2 = bs[1].replace(" ","").replace("年","-").replace("月","-").replace("日","")

result = a.replace("\n","")+"\t"+b2.replace("\n","")+"\t"+b1.replace("\n","")

#print(result)

f.write(result+"\n")

f.close()