在Python中从(字符串的)元组编写csv

Question

为了理解我目前的问题，这里有一些关于更广泛问题的背景信息：

我有一个由多个文档组成的大型文本文件。我需要找到一种方法来将这个文件组织成它的组成部分。不幸的是，所有单独的文档都有不同的格式，它们之间唯一的共同点是每个文档的头部都包含日期，每次都以相同的格式编写：dd MONTH yyyy。我使用日期作为书挡，以隔离它们之间的文本。

#the date pattern with positive lookbehind
bookend_1 = "(?<=\d{1,2}\sJANUARY\s\d{4}|\d{1,2}\sFEBRUARY\s\d{4}|\d{1,2}\sMARCH\s\d{4}|\d{1,2}\sAPRIL\s\d{4}|\d{1,2}\sMAY\s\d{4}|\d{1,2}\sJUNE\s\d{4}|\d{1,2}\sJULY\s\d{4}|\d{1,2}\sAUGUST\s\d{4}|\d{1,2}\sSEPTEMBER\s\d{4}|\d{1,2}\sOCTOBER\s\d{4}|\d{1,2}\sNOVEMBER\s\d{4}|\d{1,2}\sDECEMBER\s\d)"

#The date pattern with positive lookahead
bookend_2 = "(?=\d{1,2}\sJANUARY\s\d{4}|\d{1,2}\sFEBRUARY\s\d{4}|\d{1,2}\sMARCH\s\d{4}|\d{1,2}\sAPRIL\s\d{4}|\d{1,2}\sMAY\s\d{4}|\d{1,2}\sJUNE\s\d{4}|\d{1,2}\sJULY\s\d{4}|\d{1,2}\sAUGUST\s\d{4}|\d{1,2}\sSEPTEMBER\s\d{4}|\d{1,2}\sOCTOBER\s\d{4}|\d{1,2}\sNOVEMBER\s\d{4}|\d{1,2}\sDECEMBER\s\d)"

#using the bookends to find the text in between dates
docs = regex.findall(bookend_1+'(.*?)'+ bookend_2, psc_comm_raw, re.DOTALL|re.MULTILINE) 

使用正则表达式，我创建了两个列表:一个是所有日期，另一个是日期之间出现的所有文本段落。我把这些列表压缩成一个元组。我不能把它们压缩到字典里，因为有些日期是重复的。

psc_comm_tuple = list(zip(date, docs))

下面是几行psc_comm_tuple代码。

[('27 JULY 2004',
  ' ADDIS ABABA, ETHIOPIA\n\nPSC/PR/Comm.(XIII)\n\nCOMMUNIQUÉ\n\nPSC/PR/Comm.(XIII) Page l\n\nCOMMUNIQUÉ OF THE THIRTEENTH MEETING OF THE PEACE AND SECURITY COUNCIL\n\nThe Peace and Security Council (PSC) of the African Union (AU), at its thirteenth meeting, held on 27 July 2004, adopted the following communiqué on the crisis in the Darfur region of the Sudan:\n\nCouncil,\n\n1.\tReiterates its deep concern over the grave situation that still prevails in the Darfur region of the Sudan, in particular the continued attacks by the Janjaweed militia against the civilian population, as well as other human rights abuses and the humanitarian crisis;\n\n2.\tUnderlines the urgent need to implement decision AU/Dec.54(111) on Darfur, adopted by the 3rd Ordinary Session of the Assembly...'),
 ('29 JANUARY 2001',
  '\n\nThe Central Organ of the OAU Mechanism for Conflict Prevention, Management and Resolution held its seventy-third * ordinary session at the level of Ambassadors on 29 January 2001, in Addis Ababa. The session was chaired by Ambassador Kati Ohara Korga, Permanent Representative of Togo to the OAU.\n\nHaving considered the Report of the Secretary General on the Democratic Republic of the Congo (DRC) and the situation in that country, the Central Organ:\n\n1.\tstrongly condemns the assassination of Pre...'),
 ('20 MARCH 2001',
  "\n\nThe Central Organ of the OAU Mechanism for Conflict Prevention, Management and Resolution held its 74th ordinary session at ambassadorial level, in Addis Ababa, Ethiopia, on Tuesday March 20, 2001. The session was chaired by Ambassador Ohara Korga, Permanent representative of Togo to the OAU....'),
 ('22 AUGUST 2001',
  '\n\nThe Central Organ of the OAU Mechanism for Conflict Prevention, Management and Resolution held its 75th Ordinary Session at Ambassadorial level in Addis Ababa, Ethiopia, on Wednesday 22 August 2001....')...]

我的最终目标是创建一个包含两列的CSV :一列用于日期，另一列用于与该日期相关联的文本正文。

import csv
import os

with open('psc_comm.csv','w') as out:
    csv_out=csv.writer(out)
    csv_out.writerow(['date','text'])
    for row in psc_comm_tuple:
        csv_out.writerow(row)

当我将元组输出写入csv时，其中一些行完全没有问题。但一些输出结果却乱七八糟--文本被分成了看似随机的几个块，还有一排排空白的句子片段。有成百上千个这样的事件。当我回顾原始文档并找到句子断开的相应位置时，我看不到文本本身的任何特殊或独特之处。没有特殊字符。这只是纯文本。但是，它们看起来确实是特别长的文本部分，所以我想知道CSV文件中的单个单元格可以包含的信息是否有限制。

我的问题是:为什么CSV输出在一些地方如此时髦，而在另一些地方却没有？每个单元格中可以放入多少文本有限制吗？

​

​

Answer 1

您没有提供足够的信息来识别问题，但是Excel在读取带有嵌入换行符的CSV单元格时往往会出现问题，所以我的第一个猜测是它的错误所在:您有一个带有嵌入换行符的csv，csvwriter可能正在以可逆的方式编写它，但是Excel没有正确地解析它。

换句话说，您的CSV文件可能没有任何问题；只是读取到Excel中的方式有问题。你没有说你是如何确定存在问题的。

如果您的目标是生成Excel可以读取的内容，那么我将省去CSV格式，直接使用电子表格。模块openpyxl可以生成xlsx文档，而且效果很好。