将文本文档中的字符串值分离到熊猫数据框中

Question

我有一个 6000 行的文本文件。每行由代表不同列的束字符组成。 例子： 'AS202003402092020MF1003 EXESTBOPF 01163500116000 000120200381R000540000116000WC05 Watawala Tea Ceylon Ltd. 1M'

上面的表达式是字符串，需要将每一列分别提取出来，如下：

Borkername = AS
Sale year = 2020
Saleno = 0340
sale_dte = 20/9/2020 # date need to be format
Factoryno = MF1003
Catalogu code= EXEST
Grade =BOPF
Gross weight =01163.50 #decimal point needed
Net Weight = 01160.50 #decimal point needed
Lot_No = 0001
invoice_year = 2020
invoice_no = WC05
price = 000540.00 #decimal point needed
Netweight = 01160.00 #decimal point needed
Buyer = 'Watawala Tea Ceylon Ltd.'
Buyer_code = '1M'

我用正则表达式编写了一个代码，用于在python中将每个字段分隔为熊猫数据框的列。

import re            
import csv
### headings of the dataframe

headings = [
"Borkername", "Sale year", "Saleno", "sale_dte", "Factoryno", "Catalogu code", "Grade", "Gross 
 weight",  "Net Weight", "Lot_No", "invoice_year", "invoice_no", "price", "Netweight", "Buyer", "Buyer_code"]

re_fields = re.compile(r'(.{2})(.{4})(.{3})(.{8})(.{6})(.{5})(.{4})(.{7})(.{7}) (.{4})(.{4})(.{5})(.{8})(.{7}).(.*?) (.{2})$')

with open('input.txt') as f_input, open('output.csv', 'w', newline='') as f_output:
  csv_writer = csv.writer(f_output)
  csv_writer.writerow(headings)

 for line in f_input:
    fields = list(re_fields.match(line).groups())

    fields[3] = "{}.{}.{}".format(fields[3][:2], fields[3][2:4], fields[3][4:])
    fields[7] = float("{}.{}".format(fields[7][:5], fields[7][5:]))
    fields[8] = float("{}.{}".format(fields[8][:5], fields[8][5:]))
    fields[12] = float("{}.{}".format(fields[12][:6], fields[12][6:]))
    fields[13] = float("{}.{}".format(fields[13][:5], fields[13][5:]))

    csv_writer.writerow(fields)

不幸的是，这段代码在尝试时出错

fields = list(re_fields.match(line).groups())

AttributeError: 'NoneType' 对象没有属性 'groups'

希望技术专家可以建议我以正确的方式执行此操作，同时删除现有的错误 附上文本文件示例

AS202003402092020MF1003 EXESTBOPF      01163500116000 000120200381R000540000116000WC05 Watawala Tea 
Ceylon Ltd.                          1M
AS202003402092020MF0663 EXESTBOPF      01123500112000 000420200165R000550000112000WC05 Watawala Tea 
Ceylon Ltd.                          1M
AS202003402092020MF0069 EXESTBOP       00963500096000 000520200278R000570000096000CM01 Ceylon Tea 
Marketing Ltd.                         1M
AS202003402092020MF0069 EXESTBOPF      01103500110000 000620200282R000580000110000CM01 Ceylon Tea 
Marketing Ltd.                         1M
AS202003402092020MF0348 EXESTBOPF      01163500116000 000720200259R000570000116000CM01 Ceylon Tea 
Marketing Ltd.                         1M
AS202003402092020MF0348 EXESTBOPF      01163500116000 000820200264R000560000116000TT01 Tea Tang (Pvt) 
Ltd                                0M
AS202003402092020MF0703 EXESTBOPF      01123500112000 000920200193R000540000112000AB01 Akbar Brothers 
(Pvt) Ltd                          1M
AS202003402092020MF0552 EXESTBOPF      01123500112000 001120200266 000520000112000AB01 Akbar Brothers 
(Pvt) Ltd                          1M
AS202003402092020MF0294 EXESTBOP       01003500100000 001220200097R000560000100000UL01 Unilever 
Lipton Ceylon Ltd, Tea Division          1M

Answer 1

您的正则表达式不适合您的输入字符串。

当您的代码尝试匹配时，它会返回 None。换句话说，你的字符串不适合你的模式。

我不是正则表达式专家，但是根据您的示例数据，这种模式已经奏效了：

r'(.{2})(.{4})(.{3})(.{8})(.{6})(\s*.{5})(.{4})(\s*.{7})(.{7})(\s*.{4})(.{4})(\s*.{5})(.{8})(.{7})(.{4})(.*?)(.{2})$'

我还更改了将字符串拆分为格式并包含小数点的方式。请检查这是否适合您：

import re            
import csv
### headings of the dataframe

headings = [
"Borkername", "Sale year", "Saleno", "sale_dte", "Factoryno", "Catalogu code", "Grade", "Gross weight",  "Net Weight", "Lot_No", "invoice_year", "invoice_no", "price", "Netweight", "Buyer", "Buyer_code"]

re_fields = re.compile(r'(.{2})(.{4})(.{3})(.{8})(.{6})(\s*.{5})(.{4})(\s*.{7})(.{7})(\s*.{4})(.{4})(\s*.{5})(.{8})(.{7})(.{4})(.*?)(.{2})$')


with open('input.txt') as f_input, open('output.csv', 'w', newline='') as f_output:
    csv_writer = csv.writer(f_output)
    csv_writer.writerow(headings)
    
    for line in f_input:
        fields = list(re_fields.match(line).groups())

        fields[3] = "{}.{}.{}".format(fields[3][:2], fields[3][2:4], fields[3][4:])
        fields[7] = float("{}.{}".format(fields[7][:-2], fields[7][-2:]))
        fields[8] = float("{}.{}".format(fields[8][:-2], fields[8][-2:]))
        fields[12] = float("{}.{}".format(fields[12][:6], fields[12][6:]))
        fields[13] = float("{}.{}".format(fields[13][:-2], fields[13][-2:]))

        csv_writer.writerow(fields)