energy-data-trans/utils/file/trans_methods.py

# -*- coding: utf-8 -*-
# @Time    : 2024/5/16
# @Author  : 魏志亮
import datetime
import os
import shutil
import warnings

import chardet
import pandas as pd

from utils.log.trans_log import trans_print

warnings.filterwarnings("ignore")


# 获取文件编码
def detect_file_encoding(filename):
    # 读取文件的前1000个字节（足够用于大多数编码检测）
    with open(filename, 'rb') as f:
        rawdata = f.read(1000)
    result = chardet.detect(rawdata)
    encoding = result['encoding']

    trans_print("文件类型:", filename, encoding)

    if encoding is None:
        encoding = 'gb18030'

    if encoding.lower() in ['utf-8', 'ascii', 'utf8']:
        return 'utf-8'

    return 'gb18030'


def del_blank(df=pd.DataFrame(), cols=list()):
    for col in cols:
        if df[col].dtype == object:
            df[col] = df[col].str.strip()
    return df


# 切割数组到多个数组
def split_array(array, num):
    return [array[i:i + num] for i in range(0, len(array), num)]


def find_read_header(file_path, trans_cols):
    df = read_file_to_df(file_path, nrows=20)
    df.reset_index(inplace=True)
    count = 0
    header = None
    for col in trans_cols:
        if col in df.columns:
            count = count + 1
            if count >= 2:
                header = 0
                break

    count = 0

    values = list()
    for index, row in df.iterrows():
        for col in trans_cols:
            if col in row.values:
                count = count + 1
                if count > 2:
                    header = index + 1
                    break

    read_cols = []
    for col in values:
        if col in trans_cols:
            read_cols.append(col)

    return header, read_cols


# 读取数据到df
def read_file_to_df(file_path, read_cols=list(), trans_cols=None, nrows=None):
    begin = datetime.datetime.now()
    trans_print('开始读取文件', file_path)
    header = 0
    find_cols = list()
    if trans_cols:
        header, find_cols = find_read_header(file_path, trans_cols)
        trans_print(os.path.basename(file_path), "读取第", header, "行")
        if header is None:
            message = '未匹配到开始行，请检查并重新指定'
            trans_print(message)
            raise Exception(message)

    read_cols.extend(find_cols)

    try:
        df = pd.DataFrame()
        if str(file_path).lower().endswith("csv") or str(file_path).lower().endswith("gz"):
            encoding = detect_file_encoding(file_path)
            end_with_gz = str(file_path).lower().endswith("gz")
            if read_cols:
                if end_with_gz:
                    df = pd.read_csv(file_path, encoding=encoding, usecols=read_cols, compression='gzip', header=header,
                                     nrows=nrows)
                else:
                    df = pd.read_csv(file_path, encoding=encoding, usecols=read_cols, header=header,
                                     on_bad_lines='warn', nrows=nrows)
            else:

                if end_with_gz:
                    df = pd.read_csv(file_path, encoding=encoding, compression='gzip', header=header, nrows=nrows)
                else:
                    df = pd.read_csv(file_path, encoding=encoding, header=header, on_bad_lines='warn', nrows=nrows)

        else:
            xls = pd.ExcelFile(file_path, engine="calamine")
            # 获取所有的sheet名称
            sheet_names = xls.sheet_names
            for sheet_name in sheet_names:
                if read_cols:
                    now_df = pd.read_excel(xls, sheet_name=sheet_name, header=header, usecols=read_cols, nrows=nrows)
                else:
                    now_df = pd.read_excel(xls, sheet_name=sheet_name, header=header, nrows=nrows)

                now_df['sheet_name'] = sheet_name
                df = pd.concat([df, now_df])
            xls.close()
        trans_print('文件读取成功:', file_path, '数据数量:', df.shape, '耗时:', datetime.datetime.now() - begin)
    except Exception as e:
        trans_print('读取文件出错', file_path, str(e))
        message = '文件:' + os.path.basename(file_path) + ',' + str(e)
        raise ValueError(message)

    return df


def __build_directory_dict(directory_dict, path, filter_types=None):
    # 遍历目录下的所有项
    for item in os.listdir(path):
        item_path = os.path.join(path, item)
        if os.path.isdir(item_path):
            __build_directory_dict(directory_dict, item_path, filter_types=filter_types)
        elif os.path.isfile(item_path):
            if path not in directory_dict:
                directory_dict[path] = []

            if filter_types is None or len(filter_types) == 0:
                directory_dict[path].append(item_path)
            elif str(item_path).split(".")[-1] in filter_types:
                if str(item_path).count("~$") == 0:
                    directory_dict[path].append(item_path)


# 读取路径下所有的excel文件
def read_excel_files(read_path):
    if os.path.isfile(read_path):
        return [read_path]

    directory_dict = {}
    __build_directory_dict(directory_dict, read_path, filter_types=['xls', 'xlsx', 'csv', 'gz'])

    return [path for paths in directory_dict.values() for path in paths if path]


# 读取路径下所有的文件
def read_files(read_path):
    directory_dict = {}
    __build_directory_dict(directory_dict, read_path, filter_types=['xls', 'xlsx', 'csv', 'gz', 'zip', 'rar'])

    return [path for paths in directory_dict.values() for path in paths if path]


def copy_to_new(from_path, to_path):
    is_file = False
    if to_path.count('.') > 0:
        is_file = True

    create_file_path(to_path, is_file_path=is_file)

    shutil.copy(from_path, to_path)


# 创建路径
def create_file_path(path, is_file_path=False):
    if is_file_path:
        path = os.path.dirname(path)

    if not os.path.exists(path):
        os.makedirs(path, exist_ok=True)


if __name__ == '__main__':
    datas = read_excel_files(r"D:\data\清理数据\招远风电场\WOF053600062-WOB000009_ZYFDC000012\minute")
    for data in datas:
        print(data)

    print("*" * 20)

    datas = read_excel_files(r"D:\data\清理数据\招远风电场\WOF053600062-WOB000009_ZYFDC000012\minute\WOG00066.csv.gz")
    for data in datas:
        print(data)