scrapy csv文件有统一的空行？

Question

这里是蜘蛛：

import scrapy
from danmurphys.items import DanmurphysItem

class MySpider(scrapy.Spider):
    name = 'danmurphys'
    allowed_domains = ['danmurphys.com.au']
    start_urls = ['https://www.danmurphys.com.au/dm/navigation/navigation_results_gallery.jsp?params=fh_location%3D%2F%2Fcatalog01%2Fen_AU%2Fcategories%3C%7Bcatalog01_2534374302084767_2534374302027742%7D%26fh_view_size%3D120%26fh_sort%3D-sales_value_30_days%26fh_modification%3D&resetnav=false&storeExclusivePage=false']


    def parse(self, response):        
        urls = response.xpath('//h2/a/@href').extract()
        for url in urls:            
            request = scrapy.Request(url , callback=self.parse_page)      
            yield request

    def parse_page(self , response):
        item = DanmurphysItem()
        item['brand'] = response.xpath('//span[@itemprop="brand"]/text()').extract_first().strip()
        item['name'] = response.xpath('//span[@itemprop="name"]/text()').extract_first().strip()
        item['url'] = response.url     
        return item

以下是项目：

import scrapy
class DanmurphysItem(scrapy.Item):  
    brand = scrapy.Field()
    name = scrapy.Field()
    url = scrapy.Field()

使用以下命令运行spider时：

scrapy crawl danmurphys -o output.csv

Answer 1

我通过以下步骤设法解决此问题：

C:.
|   scrapy.cfg
|
\---my_scraper
    |   exporters.py
    |   items.py
    |   middlewares.py
    |   pipelines.py
    |   settings.py
    |   __init__.py
    |
    +---spiders
    |   |   my_spider.py
    |   |   __init__.py
    |

# -*- coding: utf-8 -*-
import csv
import io
import os
import six

from scrapy.conf import settings
from scrapy.exporters import CsvItemExporter

from scrapy.extensions.feedexport import IFeedStorage
from w3lib.url import file_uri_to_path
from zope.interface import implementer

@implementer(IFeedStorage)
class FixedFileFeedStorage(object):

    def __init__(self, uri):
        self.path = file_uri_to_path(uri)

    def open(self, spider):
        dirname = os.path.dirname(self.path)
        if dirname and not os.path.exists(dirname):
            os.makedirs(dirname)
        return open(self.path, 'ab')

    def store(self, file):
        file.close()



class MyCsvItemExporter(CsvItemExporter):

    def __init__(self, file, include_headers_line=True, join_multivalued=',', **kwargs):

        # Custom delimiter
        delimiter = settings.get('CSV_DELIMITER', ';')
        kwargs['delimiter'] = delimiter

        super(MyCsvItemExporter, self).__init__(file, include_headers_line, join_multivalued, **kwargs)

        self._configure(kwargs, dont_fail=True)
        self.stream.close()
        storage = FixedFileFeedStorage(file.name)
        file = storage.open(file.name)
        self.stream = io.TextIOWrapper(
            file,
            line_buffering=False,
            write_through=True,
            encoding=self.encoding,
            newline="",
        ) if six.PY3 else file
        self.csv_writer = csv.writer(self.stream, **kwargs)

# ...

FEED_EXPORT_ENCODING = 'utf-8'

FEED_EXPORTERS = {
    'csv': 'my_scraper.exporters.MyCsvItemExporter',
}

CSV_DELIMITER = ';'

我希望这对你有帮助

Answer 2

此输出显示了在windows上使用“w”模式打开csv文件句柄的典型症状（可能是为了修复Python 3的兼容性），但忽略了换行符。

虽然这对基于Linux/Unix的系统没有影响，但在Windows上，会发出2个回车符，在每个数据行后面插入一个伪空行。

with open("output.csv","w") as f:
     cr = csv.writer(f)

正确的操作方法（python 3）：

with open("output.csv","w",newline='') as f:  # python 3
     cr = csv.writer(f)

（在Python2中，将“wb”设置为开放模式可以修复它）

如果文件是由无法或不想修改的程序创建的，则始终可以按如下方式对文件进行后期处理：

with open("output.csv","rb") as f:
   with open("output_fix.csv","w") as f2:
       f2.write(f.read().decode().replace("\r","")) # python 3
       f2.write(f.read().replace("\r","")) # python 2

Answer 3

要在Scrapy 1.3中解决这个问题，您可以通过将newline="作为参数添加到io来修补它。TextIOWrapper中的__init__方法的CsvItemExporter类在scrapy.exporters中。