Ich versuche, eine Scrapy Spinne mit Item Loaders Refactoring. Ich habe einen Teil erreicht, in dem ich mehrere Felder aus Daten entpacke, die aus einem einzelnen XPath-Ausdruck extrahiert wurden, und der Kürze halber möchte ich Nested Loaders verwenden. HierIn Scrapy, wie verschachtelte Artikel Loader ohne weitere 'add_xpath' Aufrufe
ist die Spinne so weit:
from scrapy.spiders import SitemapSpider
from apkmirror_scraper.items import ApkmirrorScraperItem, ApkmirrorItemLoader
class ApkmirrorSitemapSpider(SitemapSpider):
name = 'apkmirror-spider'
sitemap_urls = ['http://www.apkmirror.com/sitemap_index.xml']
sitemap_rules = [(r'.*-android-apk-download/$', 'parse')]
def parse(self, response):
loader = ApkmirrorItemLoader(item=ApkmirrorScraperItem(), response=response)
loader.add_value('url', response.url)
loader.add_xpath(field_name='title', xpath='//h1[@title]/text()')
loader.add_xpath(field_name='developer', xpath='//h3[@title]/a/text()')
loader.add_xpath(field_name='app', xpath='//*[contains(@data-channel-name, "App Updates")]/@data-channel-name')
apk_details_loader = loader.nested_xpath('//*[@title="APK details"]/following-sibling::*[@class="appspec-value"]')
apk_details_loader.add_xpath(field_name="version_name", xpath=".//text()")
apk_details_loader.add_xpath(field_name="version_code", xpath=".//text()")
return loader.load_item()
wo items.py
ist
import re
import scrapy
import scrapy.loader
from scrapy.loader.processors import Compose, MapCompose, TakeFirst
class ApkmirrorScraperItem(scrapy.Item):
url = scrapy.Field()
title = scrapy.Field()
developer = scrapy.Field()
app = scrapy.Field()
version_name = scrapy.Field()
version_code = scrapy.Field()
architectures = scrapy.Field()
package = scrapy.Field()
apk_file_size = scrapy.Field()
android_min_version = scrapy.Field()
android_target_version = scrapy.Field()
supported_dpis = scrapy.Field()
md5_signature = scrapy.Field()
time_uploaded = scrapy.Field()
time_scraped = scrapy.Field()
download_link = scrapy.Field()
def parse_app(data_channel_name):
'''Parse the name of the app from the "data-channel-name" attribute of the button named "Follow [app_name] Updates".'''
pattern = re.compile(r'(?P<app>.+) App Updates')
return pattern.search(data_channel_name).groupdict().get("app")
def get_version_line(apk_details):
'''Get the line containing the version from the 'APK details' section.'''
return next(line for line in apk_details if line.startswith("Version:"))
def get_architectures_line(apk_details):
'''Get the line containing the supported architectures (e.g. "arm", "x64") from the 'APK details' section, if present.'''
return apk_details[1] if not apk_details[1].startswith("Package:") else None # The line does not contain any keywords and may not be present, in which case None is returned
def get_package_line(apk_details):
return next(line for line in apk_details if line.startswith("Package:")) # The 'package line' is always present and starts with "Package:"
def parse_version_line(version_line):
'''Parse the 'versionName' and 'versionCode' from the relevant line in 'APK details'.'''
PATTERN = r"^Version: (?P<version_name>.+) \((?P<version_code>\d+)\)\s*$" # Note that the pattern includes the end-of-line character ($). This is necessary because some package names (e.g. Google Play) themselves contain brackets.
return re.match(PATTERN, version_line).groupdict()
class ApkmirrorItemLoader(scrapy.loader.ItemLoader):
url_out = TakeFirst()
title_in = MapCompose(unicode.strip)
title_out = TakeFirst()
developer_in = MapCompose(unicode.strip)
developer_out = TakeFirst()
app_in = MapCompose(parse_app)
app_out = TakeFirst()
version_name_in = Compose(get_version_line, parse_version_line, lambda d: d.get("version_name"))
version_name_out = TakeFirst()
version_code_in = Compose(get_version_line, parse_version_line, lambda d: d.get("version_code"))
version_code_out = TakeFirst()
Wenn ich die Spinne laufen, zum Beispiel mit dem Befehl
scrapy parse --spider=apkmirror-spider http://www.apkmirror.com/apk/google-inc/sheets/sheets-1-7-152-06-release/google-sheets-1-7-152-06-30-android-apk-download/
Es ist die Extrakte Felder wie erwartet:
# Scraped Items ------------------------------------------------------------
[{'app': u'Google Sheets',
'developer': u'Google Inc.',
'title': u'Google Sheets 1.7.152.06.30 (arm) (nodpi)',
'url': 'http://www.apkmirror.com/apk/google-inc/sheets/sheets-1-7-152-06-release/google-sheets-1-7-152-06-30-android-apk-download/',
'version_code': u'71520630',
'version_name': u'1.7.152.06.30'}]
Ich möchte jedoch die apk_details_loader
weiter refaktorieren. Beachten Sie, wie die beiden Zeilen mit apk_details_load.add_xpath
das gleiche xpath
Argument haben, ".//text()"
. Um diesen Gutscheincode Wiederholungen zu vermeiden, würde ich stattdessen gerne definieren apk_details_loader
als
apk_details_loader = loader.nested_xpath('//*[@title="APK details"]/following-sibling::*[@class="appspec-value"]//text()')
einschließlich der //text()
am Ende, und statt add_xpath
etwas tun, wie nur add
, die statt weiter den XPath-Ausdruck zu verfeinern und dann ruft die Eingabe-Prozessoren auf ihnen, ruft einfach die Eingabe-Prozessoren direkt. Ist so etwas möglich?