📢作者： 小小明-代码实体
📢博客主页：https://blog.csdn.net/as604049322
📢欢迎点赞 👍 收藏 ⭐留言 📝 欢迎讨论！
昨天我们处理Word文档的自动编号，详见《Python解析Word文档的自动编号》
链接：https://xxmdmst.blog.csdn.net/article/details/139638262
今天我们处理一下Word文档的目录和文本框，例如：
如果我们按照传统方式读取，是无法读取目录和文本框的。
这时，我们只需要想办法目录和文本框中的P节点与普通P节点一起被读取即可。
查看其xml结构后知道，w:sdt是目录节点，文本框节点存在于p里面的v:textbox节点下。
可以写出如下代码：
from docx import Document
from docx.oxml import ns
from docx.text.paragraph import Paragraph
doc = Document('目录测试.docx')
ns.nsmap.update(doc.element.nsmap)
body = doc.element.body
paragraphs = []
for p in body.xpath('w:p | w:sdt/w:sdtContent/w:p | w:p//v:textbox//w:p'):
paragraphs.append(Paragraph(p, body))
注意上面的代码使用Paragraph(p, body)封装是为了让paragraphs的结果类型与doc.paragraphs保存一致，可以看到其源码为：
如果不需要Paragraph的特殊功能，仅做基本的数据读取，也可以不封装。
然后就能将普通段落和目录内的段落以及文本框内的段落，都按顺序读取：
for paragraph in paragraphs:
print(paragraph.text)
如果这时，我们需要将自动编号也读取进来，需要注意文本框内的段落是单独计数的。
最后我们将该功能整合到上次的代码中：
import re
from docx import Document
from docx.oxml.ns import qn, nsmap
from docx.text.paragraph import Paragraph
class WithNumberDocxReader:
ideographTraditional = "甲乙丙丁戊己庚辛壬癸"
ideographZodiac = "子丑寅卯辰巳午未申酉戌亥"
def __init__(self, docx, gap_text="\t"):
self.docx = Document(docx)
nsmap.update(self.docx.element.nsmap)
self.numId2style = self.get_style_data()
self.gap_text = gap_text
self.cnt = {}
self.cache = {}
self.result = []
@property
def texts(self):
if self.result:
return self.result.copy()
self.clear()
for paragraph in self.paragraphs:
number_text = self.get_number_text(paragraph)
self.result.append(number_text + paragraph.text)
return self.result.copy()
def clear(self):
self.result.clear()
self.cnt.clear()
self.cache.clear()
@property
def paragraphs(self):
body = self.docx.element.body
result = []
for p in body.xpath('w:p | w:sdt/w:sdtContent/w:p | w:p//v:textbox//w:p'):
result.append(Paragraph(p, body))
return result
def get_style_data(self):
numbering_part = self.docx.part.numbering_part._element
abstractId2numId = {num.abstractNumId.val: num.numId for num in numbering_part.num_lst}
numId2style = {}
for abstractNumIdTag in numbering_part.findall(qn("w:abstractNum")):
abstractNumId = abstractNumIdTag.get(qn("w:abstractNumId"))
numId = abstractId2numId[int(abstractNumId)]
for lvlTag in abstractNumIdTag.findall(qn("w:lvl")):
ilvl = lvlTag.get(qn("w:ilvl"))
style = {tag.tag[tag.tag.rfind("}") + 1:]: tag.get(qn("w:val"))
for tag in lvlTag.xpath("./*[@w:val]", namespaces=nsmap)}
if "numFmt" not in style:
numFmtVal = lvlTag.xpath("./mc:AlternateContent/mc:Fallback/w:numFmt/@w:val",
namespaces=nsmap)
if numFmtVal and numFmtVal[0] == "decimal":
numFmt_format = lvlTag.xpath("./mc:AlternateContent/mc:Choice/w:numFmt/@w:format",
namespaces=nsmap)
if numFmt_format:
style["numFmt"] = "decimal" + numFmt_format[0].split(",")[0]
if style.get("numFmt") == "decimalZero":
style["numFmt"] = "decimal01"
numId2style[(numId, int(ilvl))] = style
return numId2style
@staticmethod
def int2upperLetter(num):
result = []
while num > 0:
num -= 1
remainder = num % 26
result.append(chr(remainder + ord('A')))
num //= 26
return "".join(reversed(result))
@staticmethod
def int2upperRoman(num):
t = [
(1000, 'M'), (900, 'CM'), (500, 'D'),
(400, 'CD'), (100, 'C'), (90, 'XC'),
(50, 'L'), (40, 'XL'), (10, 'X'),
(9, 'IX'), (5, 'V'), (4, 'IV'), (1, 'I')
]
roman_num = ''
i = 0
while num > 0:
val, syb = t[i]
for _ in range(num // val):
roman_num += syb
num -= val
i += 1
return roman_num
@staticmethod
def int2cardinalText(num):
if not isinstance(num, int) or num < 0 or num > 999999999:
raise ValueError(
"Invalid number: must be a positive integer within four digits")
base = ["Zero", "One", "Two", "Three", "Four", "Five", "Six",
"Seven", "Eight", "Nine", "Ten", "Eleven", "Twelve", "Thirteen", "Fourteen",
"Fifteen", "Sixteen", "Seventeen", "Eighteen", "Nineteen"]
tens = ["", "", "Twenty", "Thirty", "Fourty",
"Fifty", "Sixty", "Seventy", "Eighty", "Ninety"]
thousands = ["", "Thousand", "Million", "Billion"]
def two_digits(n):
if n < 20:
return base[n]
ten, unit = divmod(n, 10)
if unit == 0:
return f"{tens[ten]}"
else:
return f"{tens[ten]}-{base[unit]}"
def three_digits(n):
hundred, rest = divmod(n, 100)
if hundred == 0:
return two_digits(rest)
result = f"{base[hundred]} hundred "
if rest > 0:
result += two_digits(rest)
return result.strip()
if num < 99:
return two_digits(num)
chunks = []
while num > 0:
num, remainder = divmod(num, 1000)
chunks.append(remainder)
words = []
for i in range(len(chunks) - 1, -1, -1):
if chunks[i] == 0:
continue
chunk_word = three_digits(chunks[i])
if thousands[i]:
chunk_word += f" {thousands[i]}"
words.append(chunk_word)
words = " ".join(words).lower()
return words[0].upper() + words[1:]
@staticmethod
def int2ordinalText(num):
if not isinstance(num, int) or num < 0 or num > 999999:
raise ValueError(
"Invalid number: must be a positive integer within four digits")
base = ["Zero", "One", "Two", "Three", "Four", "Five", "Six",
"Seven", "Eight", "Nine", "Ten", "Eleven", "Twelve", "Thirteen", "Fourteen",
"Fifteen", "Sixteen", "Seventeen", "Eighteen", "Nineteen"]
baseth = ['Zeroth', 'First', 'Second', 'Third', 'Fourth', 'Fifth', 'Sixth', 'Seventh',
'Eighth', 'Ninth', 'Tenth', 'Eleventh', 'Twelfth', 'Thirteenth', 'Fourteenth',
'Fifteenth', 'Sixteenth', 'Seventeenth', 'Eighteenth', 'Nineteenth', 'Twentieth']
tens = ["", "", "Twenty", "Thirty", "Fourty",
"Fifty", "Sixty", "Seventy", "Eighty", "Ninety"]
tensth = ["", "", "Twentieth", "Thirtieth", "Fortieth",
"Fiftieth", "Sixtieth", "Seventieth", "Eightieth", "Ninetieth"]
def two_digits(n):
if n <= 20:
return baseth[n]
ten, unit = divmod(n, 10)
result = tensth[ten]
if unit != 0:
result = f"{tens[ten]}-{baseth[unit]}"
return result
thousand, num = divmod(num, 1000)
result = []
if thousand > 0:
if num == 0:
return f"{WithNumberDocxReader.int2cardinalText(thousand)} thousandth"
result.append(f"{WithNumberDocxReader.int2cardinalText(thousand)} thousand")
hundred, num = divmod(num, 100)
if hundred > 0:
if num == 0:
result.append(f"{base[hundred]} hundredth")
return " ".join(result)
result.append(f"{base[hundred]} hundred")
result.append(two_digits(num))
result = " ".join(result).lower()
return result[0].upper() + result[1:]
@staticmethod
def int2Chinese(num, ch_num, units):
if not (0 <= num <= 99999999):
raise ValueError("仅支持小于一亿以内的正整数")
def int2Chinese_in(num, ch_num, units):
if not (0 <= num <= 9999):
raise ValueError("仅支持小于一万以内的正整数")
result = [ch_num[int(i)] + unit for i, unit in zip(reversed(str(num).zfill(4)), units)]
result = "".join(reversed(result))
zero_char = ch_num[0]
result = re.sub(f"(?:{zero_char}[{units}])+", zero_char, result)
result = result.rstrip(units[0])
if result != zero_char:
result = result.rstrip(zero_char)
if result.lstrip(zero_char).startswith("一十"):
result = result.replace("一", "")
return result
if num < 10000:
result = int2Chinese_in(num, ch_num, units)
else:
left = num // 10000
right = num % 10000
result = int2Chinese_in(left, ch_num, units) + "万" + int2Chinese_in(right, ch_num, units)
if result != ch_num[0]:
result = result.strip(ch_num[0])
return result
@staticmethod
def int2ChineseCounting(num):
return WithNumberDocxReader.int2Chinese(num, ch_num='〇一二三四五六七八九', units='个十百千')
@staticmethod
def int2ChineseLegalSimplified(num):
return WithNumberDocxReader.int2Chinese(num, ch_num='零壹贰叁肆伍陆柒捌玖', units='个拾佰仟')
def get_number_text(self, paragraph):
numpr = paragraph._element.pPr.numPr
# paragraph.getparent().tag
if numpr is None or numpr.numId.val == 0:
return ""
numId = numpr.numId.val
ilvl = numpr.ilvl.val
style = self.numId2style[(numId, ilvl)]
numFmt: str = style.get("numFmt")
lvlText = style.get("lvlText")
isTxbxContent = paragraph._element.getparent().tag.endswith("txbxContent")
pos_key = (numId, ilvl, isTxbxContent)
if pos_key in self.cnt:
self.cnt[pos_key] += 1
else:
self.cnt[pos_key] = int(style["start"])
pos = self.cnt[pos_key]
num_text = str(pos)
if numFmt.startswith('decimal'):
num_text = num_text.zfill(numFmt.count("0") + 1)
elif numFmt == 'upperRoman':
num_text = self.int2upperRoman(pos)
elif numFmt == 'lowerRoman':
num_text = self.int2upperRoman(pos).lower()
elif numFmt == 'upperLetter':
num_text = self.int2upperLetter(pos)
elif numFmt == 'lowerLetter':
num_text = self.int2upperLetter(pos).lower()
elif numFmt == 'ordinal':
num_text = f"{pos}{'th' if 11 <= pos <= 13 else {1: 'st', 2: 'nd', 3: 'rd'}.get(pos % 10, 'th')}"
elif numFmt == 'cardinalText':
num_text = self.int2cardinalText(pos)
elif numFmt == 'ordinalText':
num_text = self.int2ordinalText(pos)
elif numFmt == 'ideographTraditional':
if 1 <= pos <= 10:
num_text = self.ideographTraditional[pos - 1]
elif numFmt == 'ideographZodiac':
if 1 <= pos <= 12:
num_text = self.ideographZodiac[pos - 1]
elif numFmt == 'chineseCounting':
num_text = self.int2ChineseCounting(pos)
elif numFmt == 'chineseLegalSimplified':
num_text = self.int2ChineseLegalSimplified(pos)
elif numFmt == 'decimalEnclosedCircleChinese':
pass
self.cache[pos_key] = num_text
for i in range(0, ilvl + 1):
lvlText = lvlText.replace(f'%{i + 1}', self.cache.get((numId, i, isTxbxContent), ""))
suff_text = {"space": " ", "nothing": ""}.get(style.get("suff"), self.gap_text)
lvlText += suff_text
return lvlText
测试一下：
if __name__ == '__main__':
doc = WithNumberDocxReader(r"目录测试.docx", "")
for text in doc.texts:
print(text)
结果顺利正确打印：
35.甲公司为增值税一般纳税人，委托外单位加工一批应交消费税的商品，以银行存款支付 加工费200万元、增值税税额26万元、消费税税额30万元(由受托方代收代缴),该批 商品收回后将直接用于销售。甲公司支付上述相关款项时，应编制的会计分录是( )。
A. 借：委托加工物资 256
贷：银行存款 256
B.借：委托加工物资 230
应交税费——应交增值税(进项税额) 26
贷：银行存款 256
C. 借：委托加工物资 200
应交税费 应交增值税(进项税额) 26
——应交消费税 30
贷：银行存款 256
D. 借：委托加工物资 256
贷：银行存款 200
应交税费——应交增值税(销项税额) 26
——应交消费税 30
A.已确认销售收入但尚未发出商品
B.结转完工入库产品成本
C.已收到材料但尚未收到发票账单
D.已收到发票账单并付款但尚未收到材料
36.下列各项中，不会引起企业期末存货账面价值发生变动的是( )。
A.已确认销售收入但尚未发出商品
B.结转完工入库产品成本
C.已收到材料但尚未收到发票账单
D.已收到发票账单并付款但尚未收到材料
以上是文本框的内容。
再次编号：
E.已确认销售收入但尚未发出商品
F.结转完工入库产品成本
G.已收到材料但尚未收到发票账单
H.已收到发票账单并付款但尚未收到材料
