Python 使用 pandas：如何忽略“”中的分隔符？

Question

我的 CSV 文件包含一个包含 16 列的标题。数据行包含用“，”分隔的 16 个值。

刚刚发现有些行包含"" 中包含, 的值。这使解析器感到困惑。它不是期望 15 个逗号，而是找到 18 个。下面的一个示例：

"23210","Cosmetic","Lancome","Eyes Virtuose Palette Makeup","**7,2g**","W","Decorative range","5x**1,2**g Eye Shadow + **1,2**g Powder","http://image.jpg","","3660732000104","","No","","1","1"

如何让解析器忽略""中的逗号？

我的代码如下所示：

import pandas

csv1 = pandas.read_csv('Produktlista.csv', quoting=3)
csv2 = pandas.read_csv('Prislista.csv', quoting= 3)
merged = csv1.merge(csv2, on='id')
merged.to_csv("output.csv", index=False, quoting=3)

Answer 1

传递参数quotechar='"'。来自Pandas Documentation：

quotechar：str（长度为1），可选

用于表示引用项的开始和结束的字符。引用的项目可以包含分隔符，它将被忽略。

例如：

In [9]:

t='''"23210","Cosmetic","Lancome","Eyes Virtuose Palette Makeup","7,2g","W","Decorative range","5x1,2g Eye Shadow + 1,2g Powder","http://image.jpg","","3660732000104","","No","","1","1"'''
df = pd.read_csv(io.StringIO(t), quotechar='"', header=None)
df
Out[9]:
      0         1        2                             3     4  5   \
0  23210  Cosmetic  Lancome  Eyes Virtuose Palette Makeup  7,2g  W   

                 6                                7                 8   9   \
0  Decorative range  5x1,2g Eye Shadow + 1,2g Powder  http://image.jpg NaN   

              10  11  12  13  14  15  
0  3660732000104 NaN  No NaN   1   1