core/migration/tables/content_items.py

197 lines
6.7 KiB
Python
Raw Normal View History

2021-08-20 09:27:19 +00:00
from dateutil.parser import parse
2021-08-25 21:20:53 +00:00
from orm import User
2021-08-20 09:27:19 +00:00
from os.path import abspath
2021-08-20 15:10:15 +00:00
import frontmatter
2021-08-20 09:27:19 +00:00
import json
from orm import Shout
2021-08-23 08:44:46 +00:00
from bs4 import BeautifulSoup
from migration.html2text import html2text
2021-08-25 21:20:53 +00:00
from transliterate import translit
from datetime import datetime
from sqlalchemy.exc import IntegrityError
from orm.base import local_session
2021-08-20 09:27:19 +00:00
users_dict = json.loads(open(abspath('migration/data/users.dict.json')).read())
2021-08-25 21:20:53 +00:00
users_dict['0'] = {
'id': 9999999,
'slug': 'discours',
2021-10-03 16:19:48 +00:00
'name': 'Дискурс',
2021-08-25 21:20:53 +00:00
'userpic': 'https://discours.io/images/logo-mini.svg'
}
2021-08-20 09:27:19 +00:00
2021-08-25 21:20:53 +00:00
ts = datetime.now()
2021-08-20 09:27:19 +00:00
type2layout = {
'Article': 'article',
'Literature': 'prose',
'Music': 'music',
'Video': 'video',
'Image': 'image'
}
2021-08-25 21:20:53 +00:00
def migrate(entry, limit=3626, start=0):
2021-08-20 09:27:19 +00:00
'''
type Shout {
slug: String!
author: Int!
body: String!
createdAt: DateTime!
updatedAt: DateTime!
deletedAt: DateTime
deletedBy: Int
rating: Int
ratigns: [Rating]
published: Bool!
publishedAt: DateTime # if there is no published field - it is not published
replyTo: String # another shout
tags: [String] # actual values
topics: [String] # topic-slugs, order has matter
title: String
versionOf: String
visibleForRoles: [String] # role ids are strings
visibleForUsers: [Int]
views: Int
}
'''
r = {
'layout': type2layout[entry['type']],
'title': entry['title'],
2021-08-25 21:20:53 +00:00
'authors': [],
2021-08-20 09:27:19 +00:00
'topics': [],
2021-08-25 21:20:53 +00:00
'published': entry.get('published', False),
'views': entry.get('views', 0),
'rating': entry.get('rating', 0),
2021-08-20 09:27:19 +00:00
'ratings': []
}
2021-08-23 08:44:46 +00:00
r['slug'] = entry.get('slug', '')
body_orig = entry.get('body', '')
2021-08-20 09:27:19 +00:00
if not r['slug'] and entry.get('friendlySlugs') is not None:
r['slug'] = entry['friendlySlugs']['slug'][0]['slug']
if(r['slug'] is None):
r['slug'] = entry['friendlySlugs'][0]['slug']
2021-08-23 08:44:46 +00:00
if not r['slug']:
print('NO SLUG ERROR')
# print(entry)
raise Exception
2021-08-20 09:27:19 +00:00
if entry.get('image') is not None:
r['cover'] = entry['image']['url']
2021-08-23 08:44:46 +00:00
if entry.get('thumborId') is not None:
r['cover'] = 'https://assets.discours.io/unsafe/1600x/' + entry['thumborId']
2021-08-20 09:27:19 +00:00
if entry.get('updatedAt') is not None:
2021-08-25 21:20:53 +00:00
r['updatedAt'] = parse(entry['updatedAt'])
2021-08-20 09:27:19 +00:00
if entry.get('type') == 'Literature':
2021-08-23 08:44:46 +00:00
media = entry.get('media', '')
# print(media[0]['literatureBody'])
if type(media) == list:
body_orig = media[0].get('literatureBody', '')
if body_orig == '':
print('EMPTY BODY!')
else:
# body_html = str(BeautifulSoup(body_orig, features="html.parser"))
2021-08-25 21:20:53 +00:00
# body = html2text(body_orig).replace('****', '**')
r['old_id'] = entry.get('_id')
r['body'] = body_orig
2021-08-23 08:44:46 +00:00
else:
print(r['slug'] + ': literature has no media')
2021-08-20 09:27:19 +00:00
elif entry.get('type') == 'Video':
2021-08-23 08:44:46 +00:00
m = entry['media'][0]
yt = m.get('youtubeId', '')
vm = m.get('vimeoId', '')
videoUrl = 'https://www.youtube.com/watch?v=' + yt if yt else '#'
if videoUrl == '#':
videoUrl = 'https://vimeo.com/' + vm if vm else '#'
if videoUrl == '#':
2021-08-25 21:20:53 +00:00
print(entry.get('media', 'NO MEDIA!'))
2021-08-23 08:44:46 +00:00
# raise Exception
r['body'] = '<ShoutVideo src=\"' + videoUrl + '\" />' + html2text(m.get('body', '')) # FIXME
2021-08-20 09:27:19 +00:00
elif entry.get('type') == 'Music':
2021-08-23 08:44:46 +00:00
r['body'] = '<ShoutMusic media={\"' + json.dumps(entry['media']) +'\"} />' # FIXME
if r.get('body') is None:
body_orig = entry.get('body', '')
2021-08-25 21:20:53 +00:00
body_html = BeautifulSoup(body_orig, features="html.parser")
r['body'] = body_html # html2text(body_orig).replace('****', '**')
r['old_id'] = entry.get('_id')
body = r.get('body')
user = None
try:
userdata = users_dict[entry['createdBy']]
slug = userdata['slug']
2021-10-03 16:19:48 +00:00
name = userdata['name']
2021-08-25 21:20:53 +00:00
userpic = userdata['userpic']
except KeyError:
app = entry.get('application')
if app is not None:
authordata = {
'username': app['email'],
'email': app['email'],
2021-10-03 16:19:48 +00:00
'name': app['name'],
2021-08-25 21:20:53 +00:00
'bio': app.get('bio', ''),
'emailConfirmed': False,
'slug': translit(app['name'], 'ru', reversed=True).replace(' ', '-').lower(),
'createdAt': ts,
'wasOnlineAt': ts
}
try:
user = User.create(**authordata)
except IntegrityError:
with local_session() as session:
user = session.query(User).filter(User.email == authordata['email']).first()
if user is None:
user = session.query(User).filter(User.slug == authordata['slug']).first()
slug = user.slug
2021-10-03 16:19:48 +00:00
name = user.name
2021-08-25 21:20:53 +00:00
userpic = user.userpic
else:
# no application, no author!
slug = 'discours'
name = 'Дискурс'
userpic = 'https://discours.io/images/logo-mini.svg'
with local_session() as session:
user = session.query(User).filter(User.slug == slug).first()
r['authors'].append({
'slug': slug,
'name': name,
'pic': userpic
})
2021-08-23 08:44:46 +00:00
metadata = {}
metadata['title'] = r.get('title')
metadata['authors'] = r.get('authors')
if r.get('cover', False):
metadata['cover'] = r.get('cover')
post = frontmatter.Post(body, **metadata)
dumped = frontmatter.dumps(post)
2021-08-25 21:20:53 +00:00
if entry['published']:
if r.get('old_id', None):
ext = 'html'
content = str(body).replace('<p></p>', '').replace('<p> </p>', '')
else:
ext = 'md'
content = dumped
open('migration/content/' + entry['type'].lower() + '/' + r['slug'] + '.' + ext, 'w').write(content)
try:
shout_dict = r.copy()
shout_dict['authors'] = [user, ]
if entry.get('createdAt') is not None:
shout_dict['createdAt'] = parse(entry.get('createdAt'))
else:
shout_dict['createdAt'] = ts
if entry.get('published'):
if entry.get('publishedAt') is not None:
shout_dict['publishedAt'] = parse(entry.get('publishedAt'))
else:
shout_dict['publishedAt'] = ts
s = Shout.create(**shout_dict)
r['id'] = s.id
except:
r['body'] = 'body moved'
print(r)
# print(s)
raise Exception
2021-08-20 09:27:19 +00:00
return r