如果有不同模式的文件，我不知道这是否有效(请注意，我使用的是python 2.7.11)：# -*- coding: utf-8 -*-

from docx import Document

import sys

import os

import re

reload(sys)

sys.setdefaultencoding('utf8')

for root, dirs, files in os.walk("."):

for name in files:

doc_file = os.path.join(root, name)

if doc_file.endswith('docx'):

main_file = Document(doc_file)

table = main_file.tables[1] # this is same for every document

data = []

keys = None

for i, row in enumerate(table.rows):

text = (cell.text for cell in row.cells)

if i == 0:

keys = tuple(text)

continue

row_data = tuple(text)

data.append(row_data)

regexReference = re.compile("(C.-[0-9-]+)")

regexCoordinate = re.compile(r'(N-(.{,12})([0-9]|\')|[0-9].{,12}N)[;, ]+(E-(.{,12})([0-9]|\')|[0-9].{,12}E)')

result = []

for item in data:

tmp = dict()

matchReference = regexReference.search(item[1])

matchCoordinate = regexCoordinate.search(unicode(item[2]))

if matchReference:

tmp['reference'] = matchReference.group()

if matchCoordinate:

tmp['x'] = matchCoordinate.group(1)

tmp['y'] = matchCoordinate.group(4)

tmp['description'] = unicode(item[2])

tmp['location'] = unicode(item[3])

result.append(tmp)

for rs in result:

if 'reference' in rs:

for k, v in rs.iteritems():

print('{} = {}'.format(k, v))

print

# Output:

#

# y = 91°38'28.2"E

# x = 22°40'34.3"N

# description = The existing CMC Office at Bariyodhala (22°40'34.3"N; 91°38'28.2"E) requires some repair/maintenance works including electrical wiring and electrical lights and appliances like ceiling fans supplies. Detail specification of the works are attached.

# reference = C1-20701-17-1

# location = xxxxx Site, c Region