统计各类商品的发贴量，画出柱状图。

从url中获取商品类别

table = mongoset('ganji', 'itemurls') # 访问数据表itemurls
dict = {}
datadict = []
for i in table.find():
    # 有部分url 指向zhuanzhuan商品，把它们都筛出去
    itemclass = i['itemurl'].split('/')[3] # itemclass = 'sh.ganji.com'
    # 下面判断 itemclass 是否为'sh.ganji.com'， 是则保留，否则丢弃
    if len(itemclass)<20 and itemclass != 'Mzhuanzhuan':
        data = {
            'iclass' : itemclass,
            'url': i['itemurl']   
        }
        datadict.append(data)
classlist = []
for i in datadict:
    #print(i)
    classlist.append(i['iclass'])
classidx = list(set(classlist))
print(classidx)

classidx打印如下

['fushixiaobaxuemao', 'jiadian', 'shouji', 'xuniwupin', 'zixingchemaimai', 'ruanjiantushu', 'bangong', 'shoucangpin', 'jiaju', 'xianzhilipin', 'nongyongpin', 'yingyouyunfu', 'meironghuazhuang', 'shuma', 'diannao', 'motuoche', 'baojianpin', 'laonianyongpin', 'rirongbaihuo']

[itemurls]表中的部分数据

{'_id': ObjectId('57786d9084a5fd53c0c0b2a6'), 'itemurl': 'http://sh.ganji.com/shouji/2170757686x.htm'}
{'_id': ObjectId('57786d9084a5fd53c0c0b2a7'), 'itemurl': 'http://sh.ganji.com/shouji/2091775260x.htm'}
{'_id': ObjectId('57786d9084a5fd53c0c0b2a8'), 'itemurl': 'http://sh.ganji.com/shouji/2154806461x.htm'}
{'_id': ObjectId('57786d9084a5fd53c0c0b2ac'), 'itemurl': 'http://sh.ganji.com/shouji/2156209920x.htm'}

统计各类别商品的发贴数量

classamount = []
for i in classidx:
    #print(i)
    classamount.append(classlist.count(i))
    #print(classamount)
print(classamount)

classamount打印如下

[1744, 2858, 1773, 867, 1664, 2729, 4569, 3276, 1656, 3490, 468, 1638, 1377, 907, 2157, 270, 982, 1377, 1770]

将数据转换成用于highcharts 绘图的格式

def chartformat(name, data, typestr):
    # 将数据转换成highcharts 格式的标准数据
    series = []
    for n,d in zip(name, data):
        item = {
            'name': n,
            'data': [int(d)],
            'type': typestr
        }
        series.append(item)
    print(series)
    return series

data = chartformat(classidx, classamount, 'column')
charts.plot(data, show='inline')

各类目发贴量统计柱状图

看上去发贴量最大的是家具类目

Screen Shot 2016-07-05 at 10.35.04 PM.png

总结

• 学习jupyter 与higncharts的使用，很赞的绘图工具