Python 爬虫实战计划：第三周作业

最近一直忙于工作，作业现在才交，实在不好意思。
根据赶集网北京二手市场数据源，统计两个数据项
1. 北京是各城区发帖量最多的TOP3商品类目
2. 各大类目中各成色对应的平均价位

代码如下：

#coding=utf-8
import pymongo
import charts

client = pymongo.MongoClient('localhost', 27017)
db = client['ganji']
items_collection = db['item_info']

def get_areas():
    """获得所有的城区的名字"""
    pipeline = [
        {'$group': {'_id': {'$slice': ['$area', 1]}}}
    ]
    areas = [info['_id'][0] for info in items_collection.aggregate((pipeline))]
    return areas

def get_top3(area):
    """获得指定城区的发帖量前三的类目，以及发帖量"""
    pipeline2 = [
        {'$match': {'area': area}},
        {'$group': {'_id': {'$slice': ['$cates', 2, 1]}, 'counts': {'$sum': 1}}},
        {'$sort': {'counts': -1}},
        {'$limit': 3}
    ]
    info = [{'name':info['_id'][0],'data':[info['counts']],'type':'column'}
                for info in items_collection.aggregate(pipeline2)]
    return info
#选择一个城区，在notebook中展示发帖量统计
areas = get_areas()
series = get_top3(areas[0])
options = {
    'title': {'text': '发帖量统计'},
    'subtitle': {'text': str(areas[0].encode('utf-8')) + '城区发帖量TOP3'}
}
charts.plot(series,options,show='inline')

效果如下：

image.png

各大类目中各成色对应的平均价位

def get_cates():
    """获得所有的类目名称"""
    pipeline = [
        {'$group':{'_id':{'$slice':['$cates',2,1]}}}
    ]
    return [cate['_id'][0] for cate in items_collection.aggregate(pipeline)]

def get_avg_price(cate):
    """ 在这里需要对成色进行一下转换，因为数据库里面成色信息是字符串，不好进行排序，因为我们需要按照 成色信息按照从全新到报废进行排序"""
    def zhuanhuan_chengse(chengse,info):
        """成色转换"""
        data = {
            u'全新':{'chengse':100,'info':info['counts'],'cate':info['_id']},
            u'99成新':{'chengse':99,'info':info['counts'],'cate':info['_id']},
            u'95成新':{'chengse':95,'info':info['counts'],'cate':info['_id']},
            u'9成新':{'chengse':90,'info':info['counts'],'cate':info['_id']},
            u'8成新':{'chengse':80,'info':info['counts'],'cate':info['_id']},
            u'7成新及以下':{'chengse':70,'info':info['counts'],'cate':info['_id']},
            u'报废机/尸体':{'chengse':0,'info':info['counts'],'cate':info['_id']},
        }
        return data.get(chengse,'None')

    pipeline = [
        {'$match': {'look': {'$ne': '-'},'cates':cate}},
        {'$group': {'_id': '$look', 'counts': {'$avg': '$price'}}},
    ]

    infos = sorted([zhuanhuan_chengse(info['_id'],info)
                for info in items_collection.aggregate(pipeline)],key=lambda x:x['chengse'],reverse=True)
    return [info['info'] for info in infos],[info['cate'] for info in infos]

#选取其中第一个类目
cates = get_cates()
cate = cates[0]
date = get_avg_price(cate)
serice = [{
    'name':cate,
    'data':date[0]
}]
options = {
    'title': {'text': '成色平均价格统计'},
    'subtitle': {'text': str(cate.encode('utf-8')) + '类目成色价格平均统计'},
    'xAxis':{'categories':date[1],'title':{'text':'价格'}}
}
charts.plot(serice,options,show='inline')

效果如下：

image.png