数据修改（文件的读出与写入）

"""
dev数据集有错误；改变一下错误索引
"""
BEGIN_DOCUMENT_REGEX = re.compile(r"#begin document \((.*)\); part (\d+)")

def devtoture(gold_path):
    with open(gold_path, "r") as gold_file,open('users_new.txt', 'w') as output_file:
        for line in gold_file.readlines():
            row = line.split()
            if len(row) == 0:
                output_file.write("\n")
            elif row[0].startswith('#'):
                begin_match = re.match(BEGIN_DOCUMENT_REGEX, line)
                if begin_match:
                    partnum = row[4]
                    print(partnum)
                output_file.write(line)
            else:
                if row[1] != partnum:
                    print("不相等")
                    len0 = len(row[0])+3
                    lineout=line[:len0]+partnum[0]+line[len0+2:]
                    output_file.write(lineout)
                else:
                    output_file.write(line)

 

错误的数据集：part值和第二列对不上，修改

学到的方法：

python 修改字符串：转成list修改，或者用切片修改。用replace

lineout=line[:len0]+partnum[0]+line[len0+2:]

参考：https://www.cnblogs.com/yvonnes/p/10020911.html

re匹配：

line="#begin document (bc/cctv/00/cctv_0001); part 000"

BEGIN_DOCUMENT_REGEX=re.compile(r"#begin document \((.*)\); part (\d+)")

beginmatch = re.match(BEGIN_DOCUMENT_REGEX, line)