new_tiku_structure_v3_sci/configs.py

#!/usr/bin/env/python
# -*- coding:utf-8 -*-

import logging
import time
import os
import sys
import datetime


class myLog(object):
    '''
    封装后的logging
    '''

    def __init__(self, logger=None, log_cate='my_log'):
        '''
            指定保存日志的文件路径，日志级别，以及调用文件
            将日志存入到指定的文件中
        '''
        # 创建一个logger
        self.logger = logging.getLogger(logger)
        self.logger.setLevel(logging.INFO)  # DEBUG
        # 创建一个handler，用于写入日志文件
        # self.log_time = time.strftime("%Y_%m_%d")

        # file_dir = os.getcwd() + '/../log'
        # if not os.path.exists(file_dir):
        #     os.mkdir(file_dir)
        # self.log_path = file_dir
        # self.log_name = self.log_path + "/" + log_cate + "." + self.log_time + '.log'
        # self.log_name = os.path.join(log_dir, 'parse_log.log')  # 日志地址
        self.log_name = os.path.join(parse_log_dir, '{}.log'.format(log_cate))  # 日志地址
        if os.path.exists(self.log_name):  # 设置日志定长自动新建
            logsize = os.path.getsize(self.log_name)
            if logsize > 180000000:  # 180M
                os.rename(self.log_name, os.path.join(parse_log_dir, '{}_{}.log'.format(log_cate,
                                                      datetime.datetime.now().strftime('%m_%d'))))

        # fh = logging.FileHandler(self.log_name, 'a')  # 追加模式  这个是python2的
        fh = logging.FileHandler(self.log_name, mode='a', encoding='utf8', delay=True)
        # fh = logging.FileHandler(self.log_name, 'a', encoding='utf-8')  # 这个是python3的
        fh.setLevel(logging.INFO)

        # 再创建一个handler，用于输出到控制台
        # ch = logging.StreamHandler()
        # ch.setLevel(logging.INFO)

        # 定义handler的输出格式
        # formatter = logging.Formatter(
        #     '[%(asctime)s] %(filename)s->%(funcName)s line:%(lineno)d [%(levelname)s]%(message)s')
        # formatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s')
        # formatter = logging.Formatter('{"[%(asctime)s-%(levelname)s-%(filename)s-%(lineno)s]": %(message)s}')
        formatter = logging.Formatter('{"host-ip": ' + '"{}"'.format(external_ip)
                                      + ', "log-msg": %(message)s, "other-msg": "%(filename)s-%(lineno)s-%(asctime)s"}')

        fh.setFormatter(formatter)
        # ch.setFormatter(formatter)

        # 给logger添加handler
        self.logger.addHandler(fh)
        # self.logger.addHandler(ch)

        #  添加下面一句，在记录日志之后移除句柄
        # self.logger.removeHandler(ch)
        # self.logger.removeHandler(fh)
        # 关闭打开的文件
        fh.close()
        # ch.close()

    def getlog(self):
        return self.logger


class TestingCfg:  # testing
    internal_ip = '192.168.1.140'  # internal
    external_ip = '192.168.1.140'  # external
    server_port = 11066
    public_bucket = 'zxhx-1302712961'  # 桶名称
    region = "ap-shanghai"  # 存储桶地域
    public_bucket_addr = 'zxhx-1302712961.cos.ap-shanghai.myqcloud.com'
    FAIL_FOLDER = "F:/zwj/Text_Structure/fail_files"
    IMG_FOLDER = "F:/zwj/Text_Structure/img_folder"
    RES_FOLDER = "F:/zwj/Text_Structure/res_folder"
    parse_log_dir = "F:/zwj/Text_Structure/logs"
    raw_img_upload_folder = 'F:/zwj/word_folder'
    old_img_ip = "http://192.168.1.140:8800/ser_static"
    mathpix_ip = "http://192.168.1.208:8001/segment/formula/"
    kps_phy_ip = "http://192.168.1.86:11088/phy_mark_and_connect"
    kps_Hmath_ip = "http://192.168.1.192:13356/auto_labels"
    repeat_ip = "http://192.168.1.192:8866/api/repeat/subject"
    # repeat_ip = "http://82.156.68.22:8888/repeat/subject"  # 全学科查重
    # repeat_ip = "http://82.156.68.22:8888/api/repeat/subject"  # 保存入库查重
    # callback_url_taskcheck = "http://zsytk3api.dev.xueping.com/v1/interior-api/record"
    callback_url_taskcheck = "http://zsytk3api.testing.xueping.com/v1/interior-api/record"


class ProductionCfg:  # production
    internal_ip = '0.0.0.0'  # internal
    external_ip = '82.156.255.225'  # external
    server_port = 11088
    public_bucket = 'zxhx-pro-1302712961'  # 桶名称
    region = "ap-beijing"  # 存储桶地域
    public_bucket_addr = 'zxhx-pro-1302712961.cos.ap-beijing.myqcloud.com'  # 桶地址
    FAIL_FOLDER = "E:/Text_Structure/fail_files"
    IMG_FOLDER = "E:/Text_Structure/img_folder"
    RES_FOLDER = "E:/Text_Structure/res_folder"
    parse_log_dir = "E:/Text_Structure/logs"
    raw_img_upload_folder = 'E:/word_uploads'
    old_img_ip = "http://82.156.255.225:11086/ser_static"
    mathpix_ip = "http://10.19.1.11:7080/segment/formula/"
    kps_phy_ip = "http:/49.232.72.198:11088/phy_mark_and_connect"
    kps_Hmath_ip = ""
    repeat_ip = "http://10.19.1.18:8866/api/repeat/subject"
    callback_url_taskcheck = "http://api.tk.zhixinhuixue.com/v1/interior-api/record"


# sys.argv：从控制台窗台运行程序，程序后加参数，以空格隔开，sys.argv[0]即程序本身
config_class = TestingCfg   # 没有参数时，默认按测试环境
if len(sys.argv) > 1:
    print(sys.argv, sys.argv[0])
    print(sys.argv[1])
    if sys.argv[1] == 'test':
        config_class = TestingCfg
    elif sys.argv[1] == 'product':
        config_class = ProductionCfg
    else:
        print('cmd should be: python server.py test')
        print('or: python server.py product')
        raise ValueError("命令不正确")


server_ip = config_class.internal_ip
external_ip = config_class.external_ip
server_port = config_class.server_port

public_bucket_addr = config_class.public_bucket_addr
public_bucket = config_class.public_bucket
region = config_class.region

parse_log_dir = config_class.parse_log_dir
# 定义结构化失败文件保存路径
FAIL_FOLDER = config_class.FAIL_FOLDER
if not os.path.isdir(FAIL_FOLDER):
    os.makedirs(FAIL_FOLDER)

# 定义再结构化文件中新图片（base64）的保存路径及访问目录
IMG_FOLDER = config_class.IMG_FOLDER
if not os.path.isdir(IMG_FOLDER):
    os.makedirs(IMG_FOLDER)

# 回调结果保存路径
RES_FOLDER = config_class.RES_FOLDER
if not os.path.isdir(RES_FOLDER):
    os.makedirs(RES_FOLDER)

new_img_ip = "http://{0}:{1}/ser_static".format(external_ip, server_port)
old_img_ip = config_class.old_img_ip
RawImg_UploadFolder = config_class.raw_img_upload_folder
mathpix_ip = config_class.mathpix_ip
kps_phy_ip = config_class.kps_phy_ip
kps_Hmath_ip = config_class.kps_Hmath_ip
callback_url_taskcheck = config_class.callback_url_taskcheck
repeat_ip = config_class.repeat_ip

# 注意：
# 单题解析中，线上css_conflict_deal与线下不一样

#  作答类型
answer_type = {"选择题": 1, "填空题": 2, "解答题": 3, "判断题": 4,
               "英语作文": 5, "语文作文": 6}