畅言导出评论转换成导入评论格式

changyan.py

import json
import time

cmt = open('export.json').read()
cmt = json.loads(cmt)

print(len(cmt['comments']))
blk = set((
    '=======',
    '~~~~~~~~~~~~'
))

articles = {}

for cmt in cmt['comments']:
    ctime = int(time.mktime(time.strptime(cmt['ctime'], '%Y-%m-%d %H:%M:%S')) * 1000)
    remove = False
    for b in blk:
        if b in cmt['content']:
            remove = True
            break
    if remove:
        continue

    if cmt['topicSourceId'] not in articles:
        url = cmt['topicUrl']
        articles[cmt['topicSourceId']] = {
                "title": cmt['topicTitle'],
                "url": url,
                "ttime": ctime, # use cmt time as ttime
                "sourceid": cmt['topicSourceId'],
                "comments": [],
        }
    articles[cmt['topicSourceId']]['comments'].append({
        "cmtid": cmt['id'],
        "ctime": ctime,
        "content": cmt['content'],
        "replyid": cmt['replyId'],
        "user": {
            "userid": cmt['referUserId'],
            "nickname": cmt['nickname'],
            "usericon": cmt['iconUrl'],
            "userurl": cmt['userProfileUrl'],
        },
        "ip": cmt['ip'],
        "channeltype": "1",
        "spcount": cmt['sp'] if 'sp' in cmt else '0',
        "opcount": cmt['op'] if 'op' in cmt else '0',
        "attachment": [] if not cmt['attachs'] else cmt['attachs'],
    })

with open('import.json', 'wb') as f:
    for a in articles:
        f.write(json.dumps(articles[a]))
        f.write('\n')