在 Python 中将 XML 文件转换为 CSV 文件

Question

我正在将 XML 文件转换为 CSV 文件，但在此过程中遇到了问题。

我收到一个 AttributeError: 'NoneType' 对象没有属性 'tag'。

我已经为解决这个问题做了很多研究，不幸的是我没有成功地想出一个合适的结果，我看不出我哪里出错了。

XML 文件片段

<?xml version="1.0" encoding="UTF-8"?>
<filebooks xmlns="http://www.something.com/xml/xxx/filebook/2006-10-31"\>
  <filebook>
    <header filebook-id="Some Title">
      <currency>GBP</currency>
      <display-name xml:lang="x-default">Some Name</display-name>
    </header>
    <file-tables>
      <file_table product-id="111">
        <amount quantity="1">21.5000</amount>
        <file-info>xxx 01/06/2020:Test</file-info>
        <date-from>2020-06-01</date-from>
        <date-to>2020-06-02</date-to>
      </file_table>
      <file_table product-id="222">
        <amount quantity="1">18.3000</amount>
        <file-info>xxx 01/07/2020: Txt</file-info>
        <date-from>2020-07-02</date-from>
        <date-to>2020-07-02</date-to>
      </file_table>
    </file-tables>
  </filebook>
</filebooks>

我希望在 CSV 文件中以这种格式输出：

product_id; currency; amount; quantity; file_info; date_from; date_to
111;GBP;21.500;1;xxx 01/06/2020:Test;2020-06-01;2020-06-02
222;GBP;18.300;1;xxx 01/07/2020: Txt;2020-07-02;2020-07-02

这是我到目前为止所做的。

import xml.etree.ElementTree as ET
import csv


tree = ET.parse("sample.xml")   
root = tree.getroot()     
ns = {'nspace': 'http://www.something.com/xml/xxx/filebook/2006-10-31'}

price_data = open('data.csv', 'w')

csvwriter = csv.writer(price_data)
price_head = []
count = 0
print (root.findall('nspace:filebook', ns)) #debugging


for member in root.findall('nspace:filebook', ns):
    price = []
    if count == 0:
        currency = member.find('nspace:currency', ns).tag
        price_head.append(currency)
        product_id = member.find('nspace:product_id').tag
        price_head.append(product_id)
        amount = member.find('nspace:amount').tag
        price_head.append(amount)
        quantity = member.find('nspace:quantity').tag
        price_head.append(quantity)
        file_info = member.find('nspace:file-info').tag
        price_head.append(file_info)
        date_from = member.find('nspace:date-from').tag
        price_head.append(date_from)
        date_to = member.find('nspace:date-to').tag
        price_head.append(date_to)

        count = count + 1

    currency = member.find('nspace:currency', ns).text
    price.append(currency)
    product_id = member.find('nspace:product_id').text
    price.append(product_id)
    amount = member.find('nspace:amount').text
    price.append(amount)
    quantity = member.find('nspace:quantity').text
    price.append(quantity)
    file_info = member.find('nspace:file-info').text
    price.append(file_info)
    date_from = member.find('nspace:date-from').text
    price.append(date_from)
    date_to = member.find('nspace:date-to').text
    price.append(date_to)

price_data.close()

Answer 1

另一种方法是使用 xml 解析器，因为这是 xml 格式。这忽略了 csv 方面；这显然不是这里的问题：

cash = """[your xml above - validated as xml]"""
from lxml import etree
doc = etree.XML(cash.encode())

namespaces = {'nspace': 'http://www.something.com/xml/xxx/filebook/2006-10-31'}
rows = []
for member in doc.findall('nspace:filebook',namespaces):    
    cur = member.find('.//nspace:currency',namespaces).text
    for m in member.findall('.//nspace:file_table',namespaces):
        row = []
        id = m.attrib['product-id']
        amount = m.find('./nspace:amount',namespaces).text
        file_i = m.find('./nspace:file-info',namespaces).text
        date_f = m.find('./nspace:date-from',namespaces).text
        date_t = m.find('./nspace:date-to',namespaces).text        
        row.extend([cur,id,amount,file_i,date_f,date_t])
        rows.append(row)
for r in rows:
    print(r)  

输出：

['GBP', '111', '21.5000', 'xxx 01/06/2020:Test', '2020-06-01', '2020-06-02']
['GBP', '222', '18.3000', 'xxx 01/07/2020: Txt', '2020-07-02', '2020-07-02']

显然，您可以对字段、名称标题等重新排序。

Answer 2

另一种方法。

from simplified_scrapy import SimplifiedDoc, utils, req
html = '''<?xml version="1.0" encoding="UTF-8"?>
<filebooks xmlns=\"http://www.something.com/xml/xxx/filebook/2006-10-31"\>
  <filebook>
    <header filebook-id="Some Title">
      <currency>GBP</currency>
      <display-name xml:lang="x-default">Some Name</display-name>
    </header>
    <file-tables>
      <file_table product-id="111">
        <amount quantity="1">21.5000</amount>
        <file-info>xxx 01/06/2020:Test</file-info>
        <date-from>2020-06-01</date-from>
        <date-to>2020-06-02</date-to>
      </file_table>
      <file_table product-id="222">
        <amount quantity="1">18.3000</amount>
        <file-info>xxx 01/07/2020: Txt</file-info>
        <date-from>2020-07-02</date-from>
        <date-to>2020-07-02</date-to>
      </file_table>
    </file-tables>
  </filebook>
</filebooks>'''

doc = SimplifiedDoc(html)
rows = []
header = ['product_id', 'currency', 'amount', 'quantity', 'file_info', 'date_from','date_to']
rows.append(header)
filebooks = doc.selects('filebook')
for filebook in filebooks:
    currency = filebook.currency.text
    file_tables = filebook.selects('file_table')
    for file_table in file_tables:
        amount = file_table.amount
        row = [
            file_table['product-id'], currency, amount.text,
            amount['quantity'],
            file_table.select('file-info>text()'),
            file_table.select('date-from>text()'),
            file_table.select('date-to>text()')
        ]
        rows.append(row)
utils.save2csv('data.csv', rows)

结果：

product_id,currency,amount,quantity,file_info,date_from,date_to
111,GBP,21.5000,1,xxx 01/06/2020:Test,2020-06-01,2020-06-02
222,GBP,18.3000,1,xxx 01/07/2020: Txt,2020-07-02,2020-07-02

这里有更多示例：https://github.com/yiyedata/simplified-scrapy-demo/tree/master/doc_examples