Question

我有以下代码从一堆PDF文件中提取一些数据。它适用于少数文件然后它给我一个索引超出范围错误。

__author__ = 'thavan'


import os
import sys

import convertor


def _get_file_list(root):
    """
    Get all the pdf files in a given directory.
    :param root:
    :return:
    """
    file_list = []
    for root, dir, files in os.walk(root):
        if len(files) > 0:
            for f in files:
                file_path = os.path.join(root, f)
                if file_path.endswith('.pdf'):
                    file_list.append(file_path)

    return file_list


def _match_key(key, match_list):
    for match in match_list:
        if match == key:
            return True
    return False


class SedaScraper(object):
    process_only = [] # Add here any PDF file with full path if you want to process only these files.

    def __init__(self):
        """
        Update this list whenever need to fetch a new value.
        :return:
        """
        self.total_spend_key = ['Total Spend', 'Total spend']
        self.total_spend_all_media_key = ['Total spend All Media']
        self.outlet_per_all_media_key = ['Press % All Media', 'Internet % All Media', 'Outdoor % All Media',
                                         'TV % All Media', 'Cinema % All Media']
        self.no_of_new_create_key = ['No of New Banners', 'No. of New Creatives']

    def _get_csv_values(self, pdf_file):
        """Extracts values from a given PDF file.
        """
        pdf_text = convertor.convert(pdf_file).get_text()
        lines = pdf_text.split('\n')
        data = []
        sub_data = []
        for line in lines:
            if line.strip() != '':
                sub_data.append(line.strip())
            else:
                data.append(sub_data)
                sub_data = []
        outlet = data[0][0]
        company_name = data[1][0]
        date = data[2][0]
        start_date = date.split(' to ')[0]
        end_date  = date.split(' to ')[1]
        for x in range(3, len(data)):
            try:
                if _match_key(data[x][0], self.no_of_new_create_key):
                    metric_data = dict(zip(data[x], data[x+1]))
                    break
            except IndexError:
                print "Some required text not found. Please check following data... {}".format(data)
                sys.exit(1)

        total_spend, total_spend_all_media, outlet_per_all_media, no_of_new_creatives = self._parse_metric(metric_data)
        print company_name, outlet, start_date, end_date, total_spend, total_spend_all_media, outlet_per_all_media, no_of_new_creatives
        # change below CSV separator as required.
        return '|'.join((company_name, outlet, start_date, end_date, total_spend, total_spend_all_media, outlet_per_all_media, no_of_new_creatives))

    def _parse_metric(self, metric_data):
        total_spend = None
        total_spend_all_media = None
        outlet_per_all_media = None
        no_of_new_creatives = None
        for key, value in metric_data.items():
            if _match_key(key, self.total_spend_key):
                total_spend = value
            elif _match_key(key, self.total_spend_all_media_key):
                total_spend_all_media = value
            elif _match_key(key, self.outlet_per_all_media_key):
                outlet_per_all_media = value
            elif _match_key(key, self.no_of_new_create_key):
                no_of_new_creatives = value

        return total_spend, total_spend_all_media, outlet_per_all_media, no_of_new_creatives

    def process(self, root):
        """Iteratively goes through every PDF file.
        :param root:
        :return:
        """
        pdf_list = _get_file_list(root)
        out_file = open(os.path.join(root, 'output.csv'), 'w')
        if self.process_only:
            pdf_list = self.process_only

        for pdf in pdf_list:
            print "Processing", pdf
            csv_line = self._get_csv_values(pdf)
            out_file.write(csv_line + '\n')
        print "Output file: {}".format(out_file.name)

if __name__ == '__main__':
    if len(sys.argv) == 1:
        print "Usage: pdf_scraper.py <path>"
        sys.exit(1)
    SedaScraper().process(sys.argv[1])

这是我得到的错误：

C:\Users\soz\Documents\Python\seda_pdf\src>python pdf_scraper.py C:\Users\soz\Do
cuments\Python\seda_pdf\2
Processing C:\Users\soz\Documents\Python\seda_pdf\2\2009\Internet\01.pdf
Aberdeen Asset Management Internet 01 January 2009 31 January 2009 ┬ú5,505 ┬ú166
,384 3.31% 5
Processing C:\Users\soz\Documents\Python\seda_pdf\2\2009\Internet\02.pdf
Aberdeen Asset Management Internet 01 February 2009 28 February 2009 ┬ú5,906 ┬ú2
26,575 2.61% 5
Processing C:\Users\soz\Documents\Python\seda_pdf\2\2009\Internet\03.pdf
Traceback (most recent call last):
  File "pdf_scraper.py", line 117, in <module>
    SedaScraper().process(sys.argv[1])
  File "pdf_scraper.py", line 109, in process
    csv_line = self._get_csv_values(pdf)
  File "pdf_scraper.py", line 65, in _get_csv_values
    end_date  = date.split(' to ')[1]
IndexError: list index out of range

我无法弄清楚错误是什么，因为它实际上正在为我的某些PDF文件工作。我也检查了我的文件，它们没有任何问题。

我的Python知识相当有限，所以如果你的答案可能是dummyproof，我会很感激。

Answer 1

to

如果数据变量中没有end_date = "" if ' to ' in date: end_date = date.split(' to ')[1]，该怎么办？因此不会发生拆分，那么你如何获取拆分部分的索引1。

所以将上面改为，

{{1}}

Answer 2

尝试：

if  not ' to ' in date:
   // raise exception - validate if string contains your condition

date_chunks = date.split(' to ')
start_date = date_chunks [0]
end_date  = date_chunks [1]

在Python中，就像在Java中一样，我们从零开始索引，而不是一个

Python sys.argv [1]索引超出范围

2 个答案: