主页 / 问题 / 784584
Accepted
Дмитрий Владимирович
Asked:2020-02-13 21:10:02 +0000 UTC2020-02-13 21:10:02 +0000 UTC

如何使用 python 从两个 csv 文件中收集信息？

  • 772

任务是这样的：在文件 1.csv 中有一个具有“OR-SDR-28HD-OLEC”类型的唯一值的列。此列中的每个单元格对应于相邻列中单元格的值。有一个 2.csv 文件，它的值来自 1.csv 文件的第一列。2.csv文件中第一列的值需要从第二列中拉出对应的值。

1.csv 看起来像这样：

Overwrite=No,,"* Changing the setting to ?Overwrite=Yes? will have the added effect of deactivating all of your items from the website except for those listed on this datafeed. If this is not intended, keep ?Overwrite=No?.",,,,,,,,
Part #,Item #,Currency,MSRP,MAP,Checkout MAP,Selling Price,Inventory,Fulfillment Option,Shipping,Activation Mark
PS2-BEC-5780,9SIAFDG6V86915,USD,,0.00,False,26.98,37,Seller,free,True

2.csv 应在第二列中包含来自 1.csv 的项目编号

Overwrite = Yes,,* Changing the setting to ?Overwrite=Yes? will have the added effect of deactivating all of your items from the website except for those listed on this datafeed. If this is not intended, keep ?Overwrite=No?.
Part #,Item #,Currency,MSRP,MAP,Checkout MAP,Selling Price,Inventory,Fulfillment Option,Shipping,Activation Mark
PS2-BEC-5780,,USD,,0.00,False,26.98,37,Seller,free,True
python

2 个回答

  1. 唯一值可以用作索引pandas.DataFrame，然后您可以简单地使用以下方法获得所需的值df.loc[]：

    #!/usr/bin/env python3
import pandas as pd

# use unique keys as index
df = pd.read_csv('1.csv', index_col=0, names=['key', 'value'], header=None)
print(df)

# read as series
keys = pd.read_csv('2.csv', squeeze=True, names=['key'])
print('\nKeys:', *keys)
print('\n', df.loc[keys])

    输出示例：

         value
key       
a        1
b        2
c        3
d        4

Keys: b d b

      value
key       
b        2
d        4
b        2
    • 1
  2. Best Answer

    如果我正确理解了任务：

    import pandas as pd

d1 = pd.read_csv(filename1, skiprows=1, usecols=['Part #','Item #'])
d2 = pd.read_csv(filename2, skiprows=1).drop('Item #',axis=1)

res = d2.merge(d1, on='Part #', how='left')

    如果您只想从另一个 CSV 文件（DataFrame）添加一列，那么使用该方法会更有效.map()：

    d1 = pd.read_csv(filename1, skiprows=1, usecols=['Part #','Item #'])
d2 = pd.read_csv(filename2, skiprows=1)

d2['Item #'] = d2['Part #'].map(d1.set_index('Part #')['Item #'])

    例子：

    In [106]: d2['Item #'] = d2['Part #'].map(d1.set_index('Part #')['Item #'])

In [107]: d2
Out[107]:
         Part #          Item # Currency  MSRP  MAP  Checkout MAP  Selling Price  Inventory Fulfillment Option Shipping  \
0  PS2-BEC-5780  9SIAFDG6V86915      USD   NaN  0.0         False          26.98         37             Seller     free

   Activation Mark
0             True

    更新： pandas.DataFrame 是内存中的“表”结构。要将 DataFrame 保存为 CSV，可以使用DataFrame.to_csv()方法

    d2.to_csv(r'/path/to/file.csv', index=False)
    • 1

相关问题