apache-spark - 分析异常 : u'Cannot resolve column name

Question

我是 spark 的新手。我已将 csv 文件解析为数据框。我无法从数据框中选择某些列。以下是我的表的架构:根

 |-- _c0: integer (nullable = true)
 |-- id: integer (nullable = true)
 |-- track_name: string (nullable = true)
 |-- size_bytes: long (nullable = true)
 |-- currency: string (nullable = true)
 |-- price: double (nullable = true)
 |-- rating_count_tot: integer (nullable = true)
 |-- rating_count_ver: integer (nullable = true)
 |-- user_rating: double (nullable = true)
 |-- user_rating_ver: double (nullable = true)
 |-- ver: string (nullable = true)
 |-- cont_rating: string (nullable = true)
 |-- prime_genre: string (nullable = true)
 |-- sup_devices.num: integer (nullable = true)
 |-- ipadSc_urls.num: integer (nullable = true)
 |-- lang.num: integer (nullable = true)
 |-- vpp_lic: integer (nullable = true)
 |-- Gigabytes: double (nullable = true)

虽然我能够选择一些列，但是当我尝试选择列“ipadSc_urls.num”(数据框中的列之一)时，出现错误。

Answer 1

尝试将列名括在反引号(`)中。

my_df.select('`ipadSc_urls.num`').show()