前两天捣鼓了一下Github page + Hexo搭建博客，发现本地编写的Markdown博文发布以后图片链接还是用的本地路径，导致图片没法正常显示。网上搜了一下，解决办法大都是安装插件、手动上传图床再插入链接、用在线Markdown编辑器等。感觉不是很方便，于是写了一个Python脚本在文章发布以前自动把文章里的图片上传到SM.MS图床或者腾讯云对象存储并替换文章内的图片链接。
import re import requests import json from qcloud_cos import CosConfig from qcloud_cos import CosS3Client import sys #import logging import argparse import os import time import shutil from urllib import parse def get_artical_path(art_tittle): ''' 合成文章路径 ''' try: dir_path = os.getcwd() #如果是在博客根目录放置脚本，请修改testdir\_posts路径为source\_posts或其他放置文章的目录。 article_path = os.path.join(dir_path,r"source\_posts",art_tittle) return article_path print(article_path) except BaseException as err: print("error in get_artical_path\n{}".format(err)) def change_pic_path(article_path,upld_method): ''' article_path: 文章的本地路径 upld_method: 上传方法，sm.ms，腾讯cos，本地（github）。 ''' print("please wait a moment") try: with open (article_path,'r',encoding = 'utf-8') as md: article_content = md.read() pic_block = re.findall(r'\!.*?\)',article_content) #获取添加图片的Markdown文本 if upld_method == "smms": for i in range(len(pic_block)): pic_origin_url = re.findall(r'\((.*?)\)',pic_block[i]) #获取插入图片时图片的位置 pic_new_url = smms(pic_origin_url[0]) print("pic_new_url is {}".format(pic_new_url)) article_content = article_content.replace(pic_origin_url[0],pic_new_url) elif upld_method == "tx": for i in range(len(pic_block)): pic_origin_url = re.findall(r'\((.*?)\)',pic_block[i]) #获取插入图片时图片的位置 #print("pic_origin_url is {}".format(pic_origin_url)) pic_new_url = tx(pic_origin_url[0]) print("pic_new_url is {}".format(pic_new_url)) article_content = article_content.replace(pic_origin_url[0],pic_new_url) elif upld_method == "local": for i in range(len(pic_block)): pic_origin_url = re.findall(r'\((.*?)\)',pic_block[i]) #获取插入图片时图片的位置 pic_new_url = local(pic_origin_url[0]) print("pic_new_url is {}".format(pic_new_url)) article_content = article_content.replace(pic_origin_url[0],pic_new_url) else: print("part of get_pic_path error") with open (article_path,'w',encoding = 'utf-8') as md: md.write(article_content) print("job done") except BaseException as err: print("error in change_pic_path\n{}".format(err)) #上传至SM.MS def smms(pic_origin_url): try: smms_url = 'https://sm.ms/api/upload' #file_path = r"C:\Users\Root\AppData\Roaming\Typora\typora-user-images\1542107269017.png" #for test data = requests.post( smms_url, files={'smfile':open(pic_origin_url,'rb'), 'format':'json'} ) pic_new_url = json.loads(data.text) cloud_path = pic_new_url['data']['url'] return(cloud_path) except BaseException as err: print("error in smms\n{}".format(err)) def tx(pic_origin_url): ''' 上传至腾讯云 ''' try: secret_id = '' # 替换为用户的 secretId secret_key = '' # 替换为用户的 secretKey region = '' # 替换为用户的 Region Bucket = '' #替换为用户的Bucket token = None # 使用临时密钥需要传入 Token，默认为空，可不填 scheme = 'https' # 指定使用 http/https 协议来访问 COS，默认为 https，可不填 config = CosConfig(Region=region, SecretId=secret_id, SecretKey=secret_key, Token=token, Scheme=scheme) client = CosS3Client(config) pic_basename = os.path.basename(pic_origin_url) #获取文件名及后缀 time_for_dir = time.strftime("%Y-%m-%d") #获取当前日期用于命名文件夹 #默认上传后建立一个以当前日期命名的文件夹，如果不需要，注释掉这一行用下一行。 cloud_path = time_for_dir + r'/' + pic_basename #cloud_path = pic_basename key = cloud_path # 2. 获取客户端对象 with open(pic_origin_url, 'rb') as fp: response = client.put_object( Bucket=Bucket, Body=fp, Key=key, #StorageClass='STANDARD', #ContentType='text/html; #charset='utf-8' ) cloud_path = client._conf.uri(bucket=Bucket, path=key) return cloud_path except BaseException as err: print("error in tx\n{}".format(err)) if __name__ == '__main__': parser = argparse.ArgumentParser() subparsers = parser.add_subparsers(help="commands", dest="command") smms_parser = subparsers.add_parser("smms",help="use smms") smms_parser.add_argument("art_tittle",help="input filename") tx_parser = subparsers.add_parser("tx",help="use tx cos") tx_parser.add_argument("art_tittle",help="input filename") local_parser = subparsers.add_parser("local",help="use local") local_parser.add_argument("art_tittle",help="input filename") args = parser.parse_args() if args.command == "smms": article_path = get_artical_path(args.art_tittle) change_pic_path(article_path,'smms') if args.command == "tx": print ('use tx,file name is {}'.format(args.art_tittle)) article_path = get_artical_path(args.art_tittle) change_pic_path(article_path,'tx')
将脚本放在博客本地根目录，当前目录下命令行输入python uploadpic.py <上传方式> <文章名.md> 貌似通常情况下博客文章都是放在source/_posts文件夹的，所以脚本自动查找的是source/_posts目录，如果你的文章是放在其他目录中，请手动修改代码中的目录。 上传方式：smms - SM.MSts - 腾讯云COS，需要填写相应的secret_id、secret_key、region、Bucket 及安装腾讯云cos-python-sdk-v5库
关注公众号：Python爬虫数据分析挖掘，获取更多开源项目源码
