Das deutsche Python-Forum

Code: Alles auswählen

<rss>
<channel>
<item>
<title>Oversized-Shirt</title>
<link>
http://www.otto.de/artikel.up?artinr=34634234
</link>
<description>
Oversized-Shirt mit dekorativ verziertem Peace-Zeichen vorne
</description>
<g-core:price>29.95 EUR</g-core:price> -> preis 1
<g-core:shipping>
<g-core:country>de</g-core:country>
<g-core:service>Standard</g-core:service>
<g-core:price>5.95</g-core:price> -> preis 2
</g-core:shipping>
</item>
</channel>
</rss>

Code: Alles auswählen

types = ["title",
         "link",
         "description",
         "guid",
         "g-core:apparel_type",
         "g-core:brand",
         "g-core:color",
         "g-core:condition",
         "g-core:currency",
         "g-core:expisation_date",
         "g-core:id",
         "g-core:image_link",
         "g-core:manufacturer",
         "g-core:payment_accepted",
         "g-core:price",
         "g-core:product_type",
         "g-custom:language type",
         "g-core:product_type",
         "g-core:country",
         "g-core:service",
         "shipping_price"] # hier soll der preis fürs shipping eingetragen werden

class FeedHandler(ContentHandler):
	def __init__(self):
		self.data = []
		self.item = None
		self.type = None
		self.shipping = False

	def startElement(self, name, attrs):
		if name == "g-core:shipping":
			self.shipping = True

		if name == 'item':
			self.item = {} # mein dict
			for i in types: self.item[i] = u""

		if name in types:
			self.type = name

 	def characters (self, ch):
		ch = unicode(ch)
		if self.shipping:
                    # ---> wie bekomme ich hier die tags innerhalb von shipping in mein dict item
		if self.type and self.item != None:
			self.item[self.type] +=ch

 	def endElement(self, name):
		if name == "g-core:shipping":
			self.shipping = False
		if name in types: self.type = None
	 	if name == 'item':
			self.data.append(self.item)
			self.item = None

data = open("RSS.xml")
parser = make_parser()
handler = FeedHandler()
parser.setContentHandler(handler)
parser.parse(data)

Code: Alles auswählen

	def characters (self, ch):
		ch = unicode(ch)
		if self.type and self.item != None:
                    if self.shipping:
                        if self.item["g-core:price"]:
                            print self.item["g-core:price"]
                        else:
                            self.item[self.type] += ch
                    else:
                        self.item[self.type] +=ch                           

Code: Alles auswählen

from lxml import html

TYPES = [
    ('title', 'title'),
    ('link', 'link'),
    ('description', 'description'),
    ('price', 'price'),
    ('country', 'shipping/country'),
    ('service', 'shipping/service'),
    ('shipping_price', 'shipping/price')
]


def process_item_node(node):
    result = dict()
    for name, path in TYPES:
        values = node.xpath('./%s/text()' % path)
        result[name] = values[0] if values else ''
    return result


def main():
    doc = html.parse('test.xml').getroot()
    result = map(process_item_node, doc.xpath('//item'))
    print result


if __name__ == '__main__':
    main()