WrangLing DataFrame Test Command

>>> import pandas as pd


# 读取测试数据
>>> df = pd.read_csv('test.csv')

>>> from wrangling_core import wrangling_df

# 根据pandas的dataframe生成wrangling dataframe，初始化会判断数据类型，数据质量
>>> wl_df = wrangling_df.WrangLingDF(df)
num     float64
date     object
dtype: object

# 打印wrangling dataframe数据
>>> wl_df.dataframe
   num        date
0  NaN       sdfsf
1  NaN       sdfsf
2  1.0  01/03/1987
3  2.0  01/03/1987
4  3.0         NaN
5  4.0       sdfsf
6  5.0      sdfsdf
# 打印wrangling dataframe数据类型
>>> wl_df.dtypes
{'date': ('object', 0.86), 'num': ('float64', 0.71)}
# 重新关联wrangling dataframe指定数据列的数据质量统计参考模型并重新统计数据质量
>>> wl_df.relate_deal_module('date', 'data_type_num')
# 获取重新关联之后的数据类型
>>> wl_df.dtypes
{'date': ('int', 0.0), 'num': ('float64', 0.71)}
>>> from wrangling_core import wl_operation

# 声明一个数据处理操作，测试操作为取前n行，是一个callable属性
>>> wl_op = wl_operation.WlOperation('head',True)
# 获取操作的属性值
>>> wl_op.attr_name
'head'
# 为该callable操作添加参数，参数值为3
>>> wl_op.op_param_append(3)

>>> wl_op.get_checked_param()
(3,)
# 在wrangling dataframe上执行该操作
>>> wl_df.extract_dataframe(wl_op)
# 打印执行后的输出结果
>>> wl_df.dataframe
   num        date
0  NaN       sdfsf
1  NaN       sdfsf
2  1.0  01/03/1987

# 声明一个数据处理操作，测试操作为转置表格，该属性不是一个callable属性
>>> wl_op2 = wl_operation.WlOperation('T',False)
# 获取操作的属性值
>>> wl_op.attr_name
'head'
# 获取该操作的参数 非callable，无参数
>>> wl_op.get_checked_param()
(3,)
# 在wrangling dataframe上执行该操作
>>> wl_df.extract_dataframe(wl_op2)
# 打印执行后的输出结果
>>> wl_df.dataframe
          0      1           2           3    4      5       6
num     NaN    NaN           1           2    3      4       5
date  sdfsf  sdfsf  01/03/1987  01/03/1987  NaN  sdfsf  sdfsdf

>>> import pandas as pd
>>> df = pd.read_csv('test.csv')
>>> df
   num        date  float
0  NaN       sdfsf  0.230
1  NaN       sdfsf  0.324
2  1.0  01/03/1987  0.230
3  2.0  01/03/1987  0.345
4  3.0         NaN    NaN
5  4.0       sdfsf  1.230
6  5.0      sdfsdf  4.230
>>> from wrangling_core import wrangling_df
>>> from wrangling_core import rule_loader
>>> rule = rule_loader.RuleLoader('conf/')
>>> wl_df = wrangling_df.WrangLingDF(rule,df)
num      float64
date      object
float    float64
dtype: object
wrangling_core/wrangling_df.py:67: UserWarning: This pattern has match groups. To actually get the groups, use str.extract.
  right_value = self._dataframe[item].str.contains(regex, regex=True)

>>> ops = wl_df.get_recommend_operation((2,0))
当前选择的操作对象索引为(2,0)，该对象为行级操作
>>> for item in ops:
...     print item
...
drop--|--drop the checked row
dropna--|--drop the row has nan value
>>> wl_df.extract_dataframe(ops[1])
num      float64
date      object
float    float64
dtype: object
>>> wl_df.dataframe
   num        date  float
2  1.0  01/03/1987  0.230
3  2.0  01/03/1987  0.345
5  4.0       sdfsf  1.230
6  5.0      sdfsdf  4.230

>>> ops = wl_df.get_recommend_operation((0,1))
当前选择的操作对象索引为(0,1)，该对象为列级操作
>>> for item in ops:
...     print item
...
drop--|--drop the checked column
dropna--|--drop the column has nan value
set_index--|--set the checked column as index
groupby--|--group by the checked column
sort_values--|--sort by the checked column desc
>>> wl_df.extract_dataframe(ops[2])
date      object
float    float64
dtype: object
>>> wl_df.dataframe
           date  float
num
1.0  01/03/1987  0.230
2.0  01/03/1987  0.345
4.0       sdfsf  1.230
5.0      sdfsdf  4.230
>>> ops = wl_df.get_recommend_operation((0,0))
当前选择的操作对象索引为(0,0)，该对象为表级操作
>>> for item in ops:
...     print item

>>> wl_df.extract_dataframe(ops[1])
>>> wl_df.dataframe
num           1.0         2.0    4.0     5.0
date   01/03/1987  01/03/1987  sdfsf  sdfsdf
float        0.23       0.345   1.23    4.23
>>> wl_df.get_code()
dropna--|--drop the row has nan value
set_index--|--set the checked column as index
T--|--convert table
>>>