grouped similar filenames

grouped_similar_filenames.py

def groupedSimilarFilenames(filenames):
    '''
    将相似文件名分组\n
    输出类似格式\n
        filenames preffix suffix  size \n
    0   cover.jpg     NaN    NaN     0 \n
    1     top.png     NaN    NaN     0 \n
    2       9.jpg     NaN   .jpg     0 \n
    3    015a.jpg       0  a.jpg     1 \n
    4     008.jpg       0   .jpg     9 \n
    5     010.jpg       0   .jpg     9 \n
    6     011.jpg       0   .jpg     9 \n
    7     012.jpg       0   .jpg     9 \n
    8     013.jpg       0   .jpg     9 \n
    9     014.jpg       0   .jpg     9 \n
    10    016.jpg       0   .jpg     9 \n
    11    017.jpg       0   .jpg     9 \n
    12    018.jpg       0   .jpg     9 \n
    \n
    size 为该项所在分组的成员个数\n
    依据条件可筛选特别的文件名\n
    '''
    df = pd.DataFrame(filenames, columns=['filenames']) # 即使一维 filenames 也能直接变成竖向
    pattern = r'^(?P<preffix>.+\D)?(?P<number>\d+)(?P<suffix>.+)?'
    # pattern = r'^(?P<preffix>.+?\D)?(?P<number>\d+)(?P<suffix>.+)?'
    df1 = pd.concat([df, df['filenames'].str.extract(pattern, flags=re.IGNORECASE)], axis=1)
    gf = ['preffix','suffix']
    if df1['preffix'].isna().all():
        gf = ['suffix']
    g = df1.groupby(gf, as_index=False)
    df1['size'] = g.transform('size').fillna(0).astype(int)
    return df1.sort_values('size', ascending=False, ignore_index=True)