堆栈内存溢出
  简体   繁体   English

从python中的迭代行和混合行提取字符串

[英]String extraction from iterating and mixed lines in python

 原文  2018-07-30 21:27:37  1   2    python/ regex/ string

问题描述

I have a dataset as below; 我有一个如下的数据集； 

"birth_date_1:25        birth_date_2:august     birth_date_3:1945    birth_place_1:france   death_date:<none>   "
"birth_date_1:14        birth_date_2:june       birth_date_3:1995   birth_place_1:dvůr     birth_place_2:králové     birth_place_3:nad       birth_place_4:labem     birth_place_5:,     birth_place_6:czech     birth_place_7:republic  "
"birth_date_1:21        birth_date_2:february       birth_date_3:1869   birth_place_1:blackburn     birth_place_2:,     birth_place_3:england   death_date_1:12     death_date_2:march      death_date_3:1917   "
"birth_date_1:07        birth_date_2:july       birth_date_3:1979   birth_place_1:ghana     birth_place_2:,     birth_place_3:accra "
"birth_date_1:27        birth_date_2:february       birth_date_3:1979   birth_place_1:durban        birth_place_2:,     birth_place_3:south     birth_place_4:africa    "
"birth_date_1:1989  birth_place_1:lima      birth_place_2:,     birth_place_3:peru  "
"birth_date_1:5     birth_date_2:september      birth_date_3:1980   birth_place_1:angola    death_date:<none>   "
"birth_date_1:1     birth_date_2:february       birth_date_3:1856   birth_place_1:hampstead     birth_place_2:,     birth_place_3:london    death_date_1:14     death_date_2:august     death_date_3:1905   "
"birth_date_1:28        birth_date_2:december       birth_date_3:1954   birth_place_1:hickory       birth_place_2:,     birth_place_3:north     birth_place_4:carolina  death_date:<none>   "
"birth_date:<none>  "
"birth_date:<none>  birth_place:<none>  death_date:<none>   "
"birth_date:<none>  birth_place_1:belfast       birth_place_2:,     birth_place_3:northern      birth_place_4:ireland   "
"birth_date:<none>  birth_place:<none>  death_date:<none>   "
"birth_date_1:28        birth_date_2:february       birth_date_3:1891   birth_place_1:carberry      birth_place_2:,     birth_place_3:manitoba  death_date_1:20     death_date_2:september      death_date_3:1968   "
"birth_date_1:4     birth_date_2:november       birth_date_3:1993   birth_place_1:portim√£o     birth_place_2:,     birth_place_3:portugal  "

Within these dataset I am trying to extract information as below; 在这些数据集中，我尝试按以下方式提取信息； 

25.08.1945 \t France \t NA
14.06.1995 \t Dvůr Králové nad Labem,Czech Republic \t 
21.02.1896 \t Blackburn,England \t 12.03.1917
.
.
.
1989 \t Lima,Peru \t NA
.
.
.
NA \t NA \t NA
NA \t NA \t NA
NA \t Belfast, Northern Ireland \t NA
.
.
04.11.1993 \t Portimeo,Portugal \t NA

I wrote below code to achieve this however because of the several scenarios I will encounter in my dataset such as birth_date_1 information can be null, a month name or a year, the loop below I came up with feels like is going to fail somewhere and won't be feasible. 我写了下面的代码来实现这一点，但是由于我将在数据集中遇到几种情况，例如birth_date_1信息可以为null，月名或年份，因此我想出的下面的循环感觉会在某处失败并赢得胜利。不可行。 

    outputfile = open('ornek_box_seperated_update.csv','w',encoding="utf-8")
    inputfile = open('ornek_box_seperated.csv','r',encoding="utf-8")
    import numpy as np

    birthDatePlace = [[ np.nan for i in range(9) ] for j in range(20000)]

    for line in inputfile:
        d = line.split(":")
        print(d)
        d = line.split(d)
        d = "\t".join(d)
        print(d)
        if(d[1]<40 and d[1]>0):
            birthDatePlace[line,1] = d[1]
        elif(d[1]<2020):
            birthDatePlace[line,3] = d[1]
        if(d[1]<40 and d[1]>0 and isinstance(d[3])==str):
            birthDatePlace[line,2] = d[3]
        elif(d[1]<2020 and isinstance(d[3])==int):
            birthDatePlace[line,4] = d[3]

        # this code planned to continue from here until cover the all birth place and death date information in required format

        outputfile.write(d)
        outputfile.write('\n')
    outputfile.close()

I appreciate any help you can provide. 感谢您提供的任何帮助。 I am kinda newbie in python and especially in regex or string extraction methodologies. 我是python的新手，尤其是正则表达式或字符串提取方法。

Thank you in advance for your kind support. 预先感谢您的支持。

2 个解决方案

解决方案1
 0 已采纳  2018-07-31 00:11:10

If you want to avoid code breakages, it would be better to have explicit checks. 如果要避免代码损坏，最好进行显式检查。 Please check the code below. 请检查下面的代码。 I have parsed the information and stored it in an a class object. 我已经解析了信息并将其存储在一个类对象中。 The class has some helper functions to modify the parsed data. 该类具有一些帮助程序功能来修改已解析的数据。 

# -*- coding: utf-8 -*-

# Class for storing parsed information
class Info(object):
    def __init__(self, birth_date_1, birth_date_2, birth_date_3, birth_place, death_date_1, death_date_2, death_date_3):
        if not (birth_date_1 or birth_date_2 or birth_date_3):
            self.birth_date = "NA"
        else:
            if birth_date_2 and birth_date_2.isalpha():
                birth_date_2 = self.month_string_to_number(birth_date_2)
            self.birth_date = '.'.join([birth_date_1, birth_date_2, birth_date_3]).strip(".")

        self.birth_place = birth_place if birth_place.strip(",") else "NA"

        if not (death_date_1 or death_date_2 or death_date_3):
            self.death_date = "NA"
        else:
            if death_date_2 and death_date_2.isalpha():
                death_date_2 = self.month_string_to_number(death_date_2)
            self.death_date = '.'.join([death_date_1, death_date_2, death_date_3]).strip(".")

        self.sanitize()

    def print_req_format(self):
        print '\t'.join([self.birth_date, self.birth_place, self.death_date])

    def sanitize(self):
        if "<none>" in self.birth_date:
            self.birth_date = "NA"
        if "<none>" in self.birth_place:
            self.birth_place = "NA"
        if "<none>" in self.death_date:
            self.death_date = "NA"

    def month_string_to_number(self, month):
        m = {
            'jan': 1,
            'feb': 2,
            'mar': 3,
            'apr': 4,
            'may': 5,
            'jun': 6,
            'jul': 7,
            'aug': 8,
            'sep': 9,
            'oct': 10,
            'nov': 11,
            'dec': 12
        }
        s = month.strip()[:3].lower()

        try:
            out = m[s]
            return str(out)
        except:
            return ""

dataset = [
"birth_date_1:25        birth_date_2:august     birth_date_3:1945    birth_place_1:france   death_date:<none>",
"birth_date_1:14        birth_date_2:june       birth_date_3:1995   birth_place_1:dvůr     birth_place_2:králové     birth_place_3:nad       birth_place_4:labem     birth_place_5:,     birth_place_6:czech     birth_place_7:republic",
"birth_date_1:21        birth_date_2:february       birth_date_3:1869   birth_place_1:blackburn     birth_place_2:,     birth_place_3:england   death_date_1:12     death_date_2:march      death_date_3:1917",
"birth_date_1:07        birth_date_2:july       birth_date_3:1979   birth_place_1:ghana     birth_place_2:,     birth_place_3:accra",
"birth_date_1:27        birth_date_2:february       birth_date_3:1979   birth_place_1:durban        birth_place_2:,     birth_place_3:south     birth_place_4:africa",
"birth_date_1:1989  birth_place_1:lima      birth_place_2:,     birth_place_3:peru",
"birth_date_1:5     birth_date_2:september      birth_date_3:1980   birth_place_1:angola    death_date:<none>",
"birth_date_1:1     birth_date_2:february       birth_date_3:1856   birth_place_1:hampstead     birth_place_2:,     birth_place_3:london    death_date_1:14     death_date_2:august     death_date_3:1905",
"birth_date_1:28        birth_date_2:december       birth_date_3:1954   birth_place_1:hickory       birth_place_2:,     birth_place_3:north     birth_place_4:carolina  death_date:<none>",
"birth_date:<none>",
"birth_date:<none>  birth_place:<none>  death_date:<none>",
"birth_date:<none>  birth_place_1:belfast       birth_place_2:,     birth_place_3:northern      birth_place_4:ireland",
"birth_date:<none>  birth_place:<none>  death_date:<none>",
"birth_date_1:28        birth_date_2:february       birth_date_3:1891   birth_place_1:carberry      birth_place_2:,     birth_place_3:manitoba  death_date_1:20     death_date_2:september      death_date_3:1968",
"birth_date_1:4     birth_date_2:november       birth_date_3:1993   birth_place_1:portim√£o     birth_place_2:,     birth_place_3:portugal",
]

for line in dataset:
    split_data_line = line.split()
    birth_date_1 = birth_date_2 = birth_date_3 = birth_place = death_date_1 = death_date_2 = death_date_3 = ""
    for data in split_data_line:
        split_data = data.split(":")
        if len(split_data) < 2:
            continue

        val = split_data[1]
        if data.startswith("birth_date_1"):
            birth_date_1 = val
        elif data.startswith("birth_date_2"):
            birth_date_2 = val
        elif data.startswith("birth_date_3"):
            birth_date_3 = val
        elif data.startswith("birth_place"):
            if not birth_place or val == ",":
                birth_place += val
            else:
                birth_place += " " + val
        elif data.startswith("death_date_1"):
            death_date_1 = val
        elif data.startswith("death_date_2"):
            death_date_2 = val
        elif data.startswith("death_date_3"):
            death_date_3 = val

    info = Info(birth_date_1, birth_date_2, birth_date_3, birth_place, death_date_1, death_date_2, death_date_3)
    info.print_req_format()

The output for this code according to your data provided is: 根据提供的数据，此代码的输出为： 

25.8.1945   france  NA
14.6.1995   dvůr králové nad labem, czech republic   NA
21.2.1869   blackburn, england  12.3.1917
07.7.1979   ghana, accra    NA
27.2.1979   durban, south africa    NA
1989    lima, peru  NA
5.9.1980    angola  NA
1.2.1856    hampstead, london   14.8.1905
28.12.1954  hickory, north carolina NA
NA  NA  NA
NA  NA  NA
NA  belfast, northern ireland   NA
NA  NA  NA
28.2.1891   carberry, manitoba  20.9.1968
4.11.1993   portim√£o, portugal NA

The code is simple enough to understand. 该代码非常简单易懂。 Hope this works for you. 希望这对您有用。 Cheers. 干杯。

解决方案2
 0  2018-07-31 00:12:32

import csv

FIELDNAMES = ('birth_date', 'birth_place', 'death_date')

with open('infile', 'r') as f:
    result = []
    for line in f:
        record = {k: '' for k in FIELDNAMES}
        for kv in line.strip('" \n').split():
            k, v = kv.split(':')
            if v == '<none>':
                continue
            key = k.rstrip('_0123456789')
            value = ' ' + v if record[key] and v != ',' else v
            record[key] += value
        result.append(record)

with open('outfile.csv', 'w') as f:
    writer = csv.DictWriter(f, fieldnames=FIELDNAMES)
    writer.writeheader()
    writer.writerows(result)

'outfile.csv' : 'outfile.csv' ： 

birth_date,birth_place,death_date
25 august 1945,france,
14 june 1995,"dvůr králové nad labem, czech republic",
21 february 1869,"blackburn, england",12 march 1917
07 july 1979,"ghana, accra",
27 february 1979,"durban, south africa",
1989,"lima, peru",
5 september 1980,angola,
1 february 1856,"hampstead, london",14 august 1905
28 december 1954,"hickory, north carolina",
,,
,,
,"belfast, northern ireland",
,,
28 february 1891,"carberry, manitoba",20 september 1968
4 november 1993,"portim√£o, portugal",

暂无
暂无

声明:本站的技术帖子网页，遵循CC BY-SA 4.0协议，如果您需要转载，请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.

相关问题 从python中提取多行 - Extraction from python over multiple lines 从子过程中提取Python字符串 - Python string extraction from Subprocess 从R或Python中的字符串中提取 - Extraction from a string in R or Python 从文本文件中提取Python字符串 - Python String Extraction from text file 从字符串 pandas python 中提取值？ - Values extraction from string pandas python? Python：使用for循环遍历字符串在打印时返回两行 - Python: Iterating through string with a for loop returns two lines when printing Python正则表达式字符串提取 - Python Regex String Extraction Python字符串提取和剥离 - Python string extraction and stripping Python字符串提取 - Python string extraction 从网站中提取特定行 - Extraction Specific Lines From A Website
相关标签
 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM