Question

我有两个名为df1，df2的数据框，但是当我尝试加入它时，它无法完成。让我为每个数据帧和每个样本输出设置我的模式。

df1
Out[160]: DataFrame[BibNum: string, CallNumber: string, CheckoutDateTime: string, ItemBarcode: string, ItemCollection: string, ItemType: string]

Row(BibNum=u'BibNum', CallNumber=u'CallNumber', CheckoutDateTime=u'CheckoutDateTime', ItemBarcode=u'ItemBarcode', ItemCollection=u'ItemCollection', ItemType=u'ItemType'),
 Row(BibNum=u'1842225', CallNumber=u'MYSTERY ELKINS1999', CheckoutDateTime=u'05/23/2005 03:20:00 PM', ItemBarcode=u'10035249209', ItemCollection=u'namys', ItemType=u'acbk')]



df2    
DataFrame[Author: string, BibNum: string, FloatingItem: string, ISBN: string, ItemCollection: string, ItemCount: string, ItemLocation: string, ItemType: string, PublicationDate: string, Publisher: string, ReportDate: string, Subjects: string, Title: string]

[Row(Author=u'Author', BibNum=u'BibNum', FloatingItem=u'FloatingItem', ISBN=u'ISBN', ItemCollection=u'ItemCollection', ItemCount=u'ItemCount', ItemLocation=u'ItemLocation', ItemType=u'ItemType', PublicationDate=u'PublicationYear', Publisher=u'Publisher', ReportDate=u'ReportDate', Subjects=u'Subjects', Title=u'Title'),
 Row(Author=u"O'Ryan| Ellie", BibNum=u'3011076', FloatingItem=u'Floating', ISBN=u'1481425730| 1481425749| 9781481425735| 9781481425742', ItemCollection=u'ncrdr', ItemCount=u'1', ItemLocation=u'qna', ItemType=u'jcbk', PublicationDate=u'2014', Publisher=u'Simon Spotlight|', ReportDate=u'09/01/2017', Subjects=u'Musicians Fiction| Bullfighters Fiction| Best friends Fiction| Friendship Fiction| Adventure and adventurers Fiction', Title=u"A tale of two friends / adapted by Ellie O'Ryan ; illustrated by Tom Caulfield| Frederick Gardner| Megan Petasky| and Allen Tam.")]

当我尝试使用此命令加入两个时：

df3=df1.join(df2, df1.BibNum==df2.BibNum)

，没有错误，但数据框看起来像是重叠的列：

DataFrame[BibNum: string, CallNumber: string, CheckoutDateTime: string, ItemBarcode: string, ItemCollection: string, ItemType: string, Author: string, BibNum: string, FloatingItem: string, ISBN: string, ItemCollection: string, ItemCount: string, ItemLocation: string, ItemType: string, PublicationDate: string, Publisher: string, ReportDate: string, Subjects: string, Title: string]

最后，在我获得df3（加入数据帧）之后，当我尝试df3.take（2）时，错误：list index out of range发生了。我想要的结果是我想通过计算哪些天（checkoutDateTime）被借用最多的书来找出哪些ItemLocation可用。

Answer 1

您需要在公共列上加入数据框，否则它将从2个不同的数据框生成2个相同名称的冲突列。

common_cols = [x for x in df1.columns if x in df2.columns]
df3 = df1.join(df2, on=common_cols, how='outer')

您可以根据您的要求使用外部联接或左联接。请不要就同一问题提出多个问题。您已在以下网址获得有效答案：when trying to join two tables, happening IndexError: list index out of range in pyspark

不能在pyspark加入两个RDD

1 个答案: