使用Python处理CSV文件的一些代码示例

笔记：使用Python处理CSV文件的一些代码示例，来自于《Python数据分析基础》一书，有删改

# 读写CSV文件，不使用CSV模块，仅使用基础Python 
# 20181110 wangml

#!/usr/bin/env python3

input_file = 'D:wangmDocumentslearningcodepythonsupplier_data.csv'
output_file = 'D:wangmDocumentslearningcodepythonsupplier_data_out.csv'

# 分别以读、写方式打开input_file、output_file，当以 w 方式打开的文件不存在，则创建它
with open(input_file, 'r', newline='') as filereader:
    with open(output_file, 'w', newline='') as filewriter:
        # 读取一行文件内容
        header = filereader.readline()
        header = header.strip()
        header_list = header.split(',')
        print(header_list)
        filewriter.write(','.join(map(str, header_list))+'
')
        for row in filereader:
            row = row.strip()
            row_list = row.split(',')
            print(row_list)
            filewriter.write(','.join(map(str, row_list))+'
')

# 使用CSV模块读写CSV文件
# 20181112 wangml
# csv_pandas_1
#!/usr/bin/env python3
# 导入CSV库
import csv
input_file = 'D:wangmDocumentslearningcodepythonsupplier_data.csv'
output_file = 'D:wangmDocumentslearningcodepythonsupplier_data_out.csv'
with open(input_file, 'r', newline='') as csv_in_file:
    with open(output_file, 'w', newline='') as csv_out_file:
        # 使用CVS模块中csv.reader()、csv.writer()函数，创建一个读取对象、一个写入对象
        # delimiter指定CSV文件的分隔符，默认为 , 逗号
        filereader = csv.reader(csv_in_file, delimiter=',')
        filewriter = csv.writer(csv_out_file, delimiter=',')
        header = next(filereader)
        filewriter.writerow(header)
        # 循环，每次从CSV读取文件中读取一行数据，并将其打印出来，然后写入CSV写入对象
        for row_list in filereader:
            print(row_list)
            filewriter.writerow(row_list)
        # 筛选符合条件的行
        for row_list in filereader:
            #print(row_list[1])
            name = str(row_list[0]).strip()
            #print(row_list[3])
            cost = str(row_list[3]).strip('$').replace(',', '')
            #print(cost)
            #print(type(cost))
            # 选择name为z或者cost大于600的row，此处使用float()函数将cost由str类型转换为flost
            if name == 'z' or float(cost) > 600.0:
                filewriter.writerow(row_list)

# # csv_pandas_1
#!/usr/bin/env python3
import pandas as pd
input_file = 'D:wangmDocumentslearningcodepythonsupplier_data.csv'
output_file = 'D:wangmDocumentslearningcodepythonsupplier_data_out.csv'
# 使用pandas库函数pandas.read_csv()读取一个CSV文件，并由此创建一个数据框对象
data_frame = pd.read_csv(input_file)
# 通过列名作为index选取该数据框中的指定列
data_frame['Cost'] = data_frame['Cost'].str.strip('$').astype(float)
#print(type(data_frame['Cost']))
data_frame_value_meets_condition = data_frame.loc[(data_frame['Name'].str.contains('Z')) | (data_frame['Cost'] > 600.0), :]
# 此处导致CSV文件的Cost列的$消失了
# 下面的语句并没有将$加上去，暂时不知道怎么弄
data_frame['Cost'] = '$' + str(data_frame['Cost'])
# 将data_frame_value_meets_condition写入输出文件
data_frame_value_meets_condition.to_csv(output_file, index=False)

# 20181113
# csv_pandas_2
#!/usr/bin/env python3
# 导入CSV库
import csv
input_file = 'D:wangmDocumentslearningcodepythonsupplier_data.csv'
output_file = 'D:wangmDocumentslearningcodepythonsupplier_data_out.csv'
important_dates = ['1/1/2018', '2/1/2018']
with open(input_file, 'r', newline='') as csv_in_file:
    with open(output_file, 'w', newline='') as csv_out_file:
        filereader = csv.reader(csv_in_file)
        filewriter = csv.writer(csv_out_file)
        header = next(filereader)
        filewriter.writerow(header)
        for row_list in filereader:
            a_date = row_list[4]
            # 选取date值在important_dates中的行
            if a_date in important_dates:
                filewriter.writerow(row_list)

# # csv_pandas_2
#!/usr/bin/env python3
import pandas as pd
input_file = 'D:wangmDocumentslearningcodepythonsupplier_data.csv'
output_file = 'D:wangmDocumentslearningcodepythonsupplier_data_out.csv'
# 使用pandas库函数pandas.read_csv()读取一个CSV文件，并由此创建一个数据框对象
data_frame = pd.read_csv(input_file)
important_dates = ['1/1/2018', '2/1/2018']
# 选取date值在important_dates中的行
data_frame_value_set = data_frame.loc[data_frame['Date'].isin(important_dates), :]
data_frame_value_set.to_csv(output_file, index=False)

# 20181113
# csv_pandas_3
#!/usr/bin/env python3
# 导入CSV库、正则表达式库
import csv
import re
input_file = 'D:wangmDocumentslearningcodepythonsupplier_data.csv'
output_file = 'D:wangmDocumentslearningcodepythonsupplier_data_out.csv'
# 使用re.compile(正则表达式)，创建一个正则表达式变量
# 元字符?P<my_pattern_group>捕获了名为<my_pattern_group>的组中匹配了的字符串
# pattern表示满足以：'001-'开头，后面可跟除任意字串的字符串
# re.I表示大小写敏感
pattern = re.compile(r'(?P<my_pattern_group>^001-.*)', re.I)
with open(input_file, 'r', newline='') as csv_in_file:
    with open(output_file, 'w', newline='') as csv_out_file:
        filereader = csv.reader(csv_in_file)
        filewriter = csv.writer(csv_out_file)
        header = next(filereader)
        filewriter.writerow(header)
        for row_list in filereader:
            id_number = row_list[1]
            if pattern.search(id_number):
                filewriter.writerow(row_list)

# 20181113
# csv_pandas_3
#!/usr/bin/env python3
import pandas as pd
input_file = 'D:wangmDocumentslearningcodepythonsupplier_data.csv'
output_file = 'D:wangmDocumentslearningcodepythonsupplier_data_out.csv'
# 使用pandas库函数pandas.read_csv()读取一个CSV文件，并由此创建一个数据框对象
data_frame = pd.read_csv(input_file)
# 筛选出ID值以001-开头的行
data_frame_value_matches_pattern = data_frame.loc[data_frame['ID'].str.startswith("001-"), :]
data_frame_value_matches_pattern.to_csv(output_file, index=False)

# 选取CSV文件中符合条件的列

# 20181113
# csv_pandas_4
# 通过列索引值选取特定列
# 在只知道需要选取的列名称时，我们可以通过列名称取得相应的索引值，在进行选取
# 具体方法是判断相应标题行每个元素是否在已知列名称中，若是，记下该item的index
#!/usr/bin/env python3
import csv
input_file = 'D:wangmDocumentslearningcodepythonsupplier_data.csv'
output_file = 'D:wangmDocumentslearningcodepythonsupplier_data_out.csv'
my_columns = [0, 3]
with open(input_file, 'r', newline='') as csv_in_file:
    with open(output_file, 'w', newline='') as csv_out_file:
        filereader = csv.reader(csv_in_file)
        filewriter = csv.writer(csv_out_file)
        for row_list in filereader:
            # 每次向输出文件中写入的一行值
            row_list_output = []
            for index_value in my_columns:
                row_list_output.append(row_list[index_value])
            filewriter.writerow(row_list_output)

# 选取CSV文件中符合条件的列

# 20181113
# csv_pandas_4
# 通过列索引值选取特定列
# 在只知道需要选取的列名称时，不需要像基本Python一样处理标题行，pandas可以将列名称当做index一样处理
#!/usr/bin/env python3
import pandas as pd
input_file = 'D:wangmDocumentslearningcodepythonsupplier_data.csv'
output_file = 'D:wangmDocumentslearningcodepythonsupplier_data_out.csv'
# 使用pandas库函数pandas.read_csv()读取一个CSV文件，并由此创建一个数据框对象
data_frame = pd.read_csv(input_file)
# 选取data_frame数据框对象中的所有行的列索引值为0,3的列
# iloc(行,列)函数可以选取数据框中选定的行、列
data_frame_value_column_by_value = data_frame.iloc[:, [0, 3]]
# data_frame_value_column_by_value = data_frame.iloc[:, [‘Name’, 'Cost']]
data_frame_value_column_by_value.to_csv(output_file, index=False)

# 给一个CSV文件添加标题行，在基础Python中，可能是将标题行通过csv库的writerow()函数写入
# 而pandas库提供了更加简单的方法
# title = [‘One’, 'Two'...]
# data_frame = pd.read_csv(input_file, header=None, names=title)

# 读取多个CSV文件，输出读取了多少个CSV文件
#!/usr/bin/env python3
import csv
import glob
import os

input_path = 'D:wangmDocumentslearningcodepython'
file_counter = 0
for input_file in glob.glob(os.path.join(input_path, '*.csv')):
    file_counter = file_counter + 1
    #row_counter = 1
    #with open(input_file, 'r', newline='') as csv_input_file:
        #filereader = csv.reader(csv_input_file)
        #... 
print(file_counter)

# 20181114
# 合并多个CSV文件
#!/usv/bin/env python3
import pandas as pd
import os
import glob
input_path = 'D:wangmDocumentslearningcodepython'
output_file = 'D:wangmDocumentslearningcodepythonsupplier_data_out.csv'
#all_files = glob.glob(os.path.join(input_path, 'supplier_data_副本*'))
# OSError: Initializing from file failed上面这句出现错误，因为文件名含有中文，改成下面这句就行了
all_files = glob.glob(os.path.join(input_path, 'supplier_data_copy*'))
all_data_frame = []
for file in all_files:
    data_frame = pd.read_csv(file, index_col=None)
    all_data_frame.append(data_frame)
# pandas.concat()函数将数据框数据垂直堆叠(axis=0), 当水平连接数据时(asis=1)
data_frame_concat = pd.concat(all_data_frame, axis=0, ignore_index=True)
data_frame_concat.to_csv(output_file, index=False)

# 分别计算多个CSV文件中的某项数据的和、平均值等
# 在基本python中，可以读取多个CSV文件，然后要被计算的项的值一个一个取出来，然后计算
# 这里展示了使用pandas提供的方法 
#!/usv/bin/env python3
import pandas as pd
import os
import glob
input_path = 'D:wangmDocumentslearningcodepython'
output_file = 'D:wangmDocumentslearningcodepythonsupplier_data_out.csv'
all_files = glob.glob(os.path.join(input_path, 'supplier_data_copy*'))
all_data_frame = []
for file in all_files:
    data_frame = pd.read_csv(file, index_col=None)
    # 和
    total_cost = pd.DataFrame([float(str(value).strip('$').replace(',', '')) 
                               for value in data_frame.loc[:, 'Cost']]).sum()
    # 平均值
    average_cost = pd.DataFrame([float(str(value).strip('$').replace(',', '')) 
                               for value in data_frame.loc[:, 'Cost']]).mean()
    data = {'file_name': os.path.basename(file),
            'total_cost': total_cost,
            'average_cost': average_cost}
    all_data_frame.append(pd.DataFrame(data, columns=['file_name', 'total_cost', 'average_cost']))
data_frames_concat = pd.concat(all_data_frame, axis=0, ignore_index=True)
data_frames_concat.to_csv(output_file, index=False)

代码示例中使用的CSV文件：

上述代码分别使用CSV库、pandas库来对CSV文件进行相同的操作

上述代码运行在Python 3.6版本下，在Win10、Spyder中

有关Python的csv库的详细介绍：https://docs.python.org/2/library/csv.html