Python PDF 抓取

Python PDF Scraping

任务:

PDF 是银行对帐单,包含列,即(日期、描述、存款、取款、余额),用各自的字段解析列,并以 CSV 格式导出该数据。PDF.

我的代码:

import pdftotext
import re
import csv

# open PDF file
with open('test.pdf', 'rb') as pdf_file:
pdf = pdftotext.PDF(pdf_file)

# extract tabular text
lines = pdf[2].split('\n')[4:]
# CSV table
table = []

# loop over lines in table
for line in lines:
# replace trailing spaces with comas
row = re.sub('   ', ',', line)

# reducing the number of comas to one
row = [cols.strip() for cols in re.sub(',+', ',', row).split(',')]

# handling missed separators
row = ','.join(row).replace('  ', ',').split(',')

# append row to table
table.append(row)

print(table)

# write CSV output
with open('test.csv', 'w') as csv_file:
writer = csv.writer(csv_file)
writer.writerows(table)

问题:

我没有得到想要的输出,即一半的描述显示在 table.I 日期下,我正在附加 csv 以供进一步参考 here

期望的输出:

例如

['04/02','CHRYSLER CAPITAL PAYMENT 0023582513','$469.88-','$51.15']

Example of output

你可以使用 pdfplumber 库,它非常有用,我在不到五分钟的时间内就得到了这个输出,它需要使用 table 参数

import pandas as pd
import pdfplumber
pdf = pdfplumber.open(r'C:\Users\Erkin\Downloads\test.pdf')
df = pd.DataFrame()
table_settings={"vertical_strategy": "text", 
    "horizontal_strategy": "lines","intersection_y_tolerance": 8}
df = pd.DataFrame(pdf.pages[3].extract_table(table_settings))
df.to_csv(r'C:\Users\Erkin\Downloads\test.csv')