epicyon/posts.py

274 lines
11 KiB
Python
Raw Normal View History

2019-06-28 18:55:29 +00:00
__filename__ = "posts.py"
__author__ = "Bob Mottram"
__license__ = "AGPL3+"
__version__ = "0.0.1"
__maintainer__ = "Bob Mottram"
__email__ = "bob@freedombone.net"
__status__ = "Production"
import requests
import json
2019-06-29 10:08:59 +00:00
import commentjson
2019-06-28 18:55:29 +00:00
import html
2019-06-29 10:08:59 +00:00
import datetime
2019-06-29 11:47:33 +00:00
import os, shutil
2019-06-29 10:08:59 +00:00
from pprint import pprint
2019-06-28 18:55:29 +00:00
from random import randint
from session import getJson
try:
from BeautifulSoup import BeautifulSoup
except ImportError:
from bs4 import BeautifulSoup
2019-06-28 20:43:37 +00:00
def permitted(url: str,federationList) -> bool:
2019-06-28 18:55:29 +00:00
"""Is a url from one of the permitted domains?
"""
2019-06-28 20:43:37 +00:00
for domain in federationList:
2019-06-28 18:55:29 +00:00
if domain in url:
return True
return False
def cleanHtml(rawHtml: str) -> str:
text = BeautifulSoup(rawHtml, 'html.parser').get_text()
return html.unescape(text)
def getUserUrl(wfRequest) -> str:
if wfRequest.get('links'):
for link in wfRequest['links']:
if link.get('type') and link.get('href'):
if link['type'] == 'application/activity+json':
return link['href']
return None
2019-06-28 19:36:39 +00:00
def parseUserFeed(session,feedUrl,asHeader) -> None:
2019-06-29 10:08:59 +00:00
feedJson = getJson(session,feedUrl,asHeader,None)
pprint(feedJson)
2019-06-28 18:55:29 +00:00
2019-06-29 10:08:59 +00:00
if 'orderedItems' in feedJson:
2019-06-29 10:59:16 +00:00
for item in feedJson['orderedItems']:
2019-06-28 18:55:29 +00:00
yield item
nextUrl = None
2019-06-29 10:08:59 +00:00
if 'first' in feedJson:
2019-06-29 10:59:16 +00:00
nextUrl = feedJson['first']
2019-06-29 10:08:59 +00:00
elif 'next' in feedJson:
nextUrl = feedJson['next']
2019-06-28 18:55:29 +00:00
if nextUrl:
2019-06-28 19:36:39 +00:00
for item in parseUserFeed(session,nextUrl,asHeader):
2019-06-28 18:55:29 +00:00
yield item
2019-06-28 20:43:37 +00:00
def getUserPosts(session,wfRequest,maxPosts,maxMentions,maxEmoji,maxAttachments,federationList) -> {}:
2019-06-28 18:55:29 +00:00
userPosts={}
asHeader = {'Accept': 'application/ld+json; profile="https://www.w3.org/ns/activitystreams"'}
userUrl = getUserUrl(wfRequest)
if not userUrl:
return userPosts
userJson = getJson(session,userUrl,asHeader,None)
if not userJson.get('outbox'):
return userPosts
feedUrl = userJson['outbox']
i = 0
2019-06-28 19:36:39 +00:00
for item in parseUserFeed(session,feedUrl,asHeader):
2019-06-28 18:55:29 +00:00
if not item.get('type'):
continue
if item['type'] != 'Create':
continue
if not item.get('object'):
continue
published = item['object']['published']
if not userPosts.get(published):
content = item['object']['content']
mentions=[]
emoji={}
if item['object'].get('tag'):
for tagItem in item['object']['tag']:
tagType=tagItem['type'].lower()
if tagType=='emoji':
if tagItem.get('name') and tagItem.get('icon'):
if tagItem['icon'].get('url'):
# No emoji from non-permitted domains
2019-06-28 20:43:37 +00:00
if permitted(tagItem['icon']['url'],federationList):
2019-06-28 18:55:29 +00:00
emojiName=tagItem['name']
emojiIcon=tagItem['icon']['url']
emoji[emojiName]=emojiIcon
if tagType=='mention':
if tagItem.get('name'):
if tagItem['name'] not in mentions:
mentions.append(tagItem['name'])
if len(mentions)>maxMentions:
continue
if len(emoji)>maxEmoji:
continue
summary = ''
if item['object'].get('summary'):
if item['object']['summary']:
summary = item['object']['summary']
inReplyTo = ''
if item['object'].get('inReplyTo'):
if item['object']['inReplyTo']:
# No replies to non-permitted domains
2019-06-28 20:43:37 +00:00
if not permitted(item['object']['inReplyTo'],federationList):
2019-06-28 18:55:29 +00:00
continue
inReplyTo = item['object']['inReplyTo']
conversation = ''
if item['object'].get('conversation'):
if item['object']['conversation']:
# no conversations originated in non-permitted domains
2019-06-28 20:43:37 +00:00
if permitted(item['object']['conversation'],federationList):
2019-06-28 18:55:29 +00:00
conversation = item['object']['conversation']
attachment = []
if item['object'].get('attachment'):
if item['object']['attachment']:
for attach in item['object']['attachment']:
if attach.get('name') and attach.get('url'):
# no attachments from non-permitted domains
2019-06-28 20:43:37 +00:00
if permitted(attach['url'],federationList):
2019-06-28 18:55:29 +00:00
attachment.append([attach['name'],attach['url']])
sensitive = False
if item['object'].get('sensitive'):
sensitive = item['object']['sensitive']
userPosts[published] = {
"sensitive": sensitive,
"inreplyto": inReplyTo,
"summary": summary,
"html": content,
"plaintext": cleanHtml(content),
"attachment": attachment,
"mentions": mentions,
"emoji": emoji,
"conversation": conversation
}
#print(str(item)+'\n')
i += 1
if i == maxPosts:
break
return userPosts
2019-06-29 10:08:59 +00:00
def createOutboxDir(username: str,domain: str) -> (str,str):
"""Create an outbox for a person and returns the feed filename and directory
"""
handle=username.lower()+'@'+domain.lower()
baseDir=os.getcwd()
if not os.path.isdir(baseDir+'/accounts/'+handle):
os.mkdir(baseDir+'/accounts/'+handle)
outboxDir=baseDir+'/accounts/'+handle+'/outbox'
if not os.path.isdir(outboxDir):
os.mkdir(outboxDir)
outboxJsonFilename=baseDir+'/accounts/'+handle+'/outbox.json'
return outboxJsonFilename,outboxDir
2019-06-29 11:47:33 +00:00
def deleteAllPosts(username: str, domain: str) -> None:
"""Deletes all posts for a person
"""
outboxJsonFilename,outboxDir = createOutboxDir(username,domain)
for deleteFilename in os.listdir(outboxDir):
filePath = os.path.join(outboxDir, deleteFilename)
try:
if os.path.isfile(filePath):
os.unlink(filePath)
elif os.path.isdir(filePath): shutil.rmtree(filePath)
except Exception as e:
print(e)
# TODO update output feed
2019-06-29 10:41:22 +00:00
def createPublicPost(username: str, domain: str, https: bool, content: str, followersOnly: bool, saveToFile: bool, inReplyTo=None, inReplyToAtomUri=None, subject=None) -> {}:
2019-06-29 11:47:33 +00:00
"""Creates a public post
2019-06-29 10:08:59 +00:00
"""
2019-06-28 18:55:29 +00:00
prefix='https'
if not https:
prefix='http'
currTime=datetime.datetime.utcnow()
2019-06-29 10:08:59 +00:00
daysSinceEpoch=(currTime - datetime.datetime(1970,1,1)).days
2019-06-29 10:23:40 +00:00
# status is the number of seconds since epoch
2019-06-29 10:08:59 +00:00
statusNumber=str((daysSinceEpoch*24*60*60) + (currTime.hour*60*60) + (currTime.minute*60) + currTime.second)
2019-06-28 18:55:29 +00:00
published=currTime.strftime("%Y-%m-%dT%H:%M:%SZ")
conversationDate=currTime.strftime("%Y-%m-%d")
2019-06-29 10:08:59 +00:00
conversationId=statusNumber
2019-06-28 18:55:29 +00:00
postTo='https://www.w3.org/ns/activitystreams#Public'
postCC=prefix+'://'+domain+'/users/'+username+'/followers'
if followersOnly:
postTo=postCC
postCC=''
2019-06-29 10:08:59 +00:00
newPostId=prefix+'://'+domain+'/users/'+username+'/statuses/'+statusNumber
2019-06-29 10:23:40 +00:00
sensitive=False
if subject:
summary=subject
sensitive=True
2019-06-28 18:55:29 +00:00
newPost = {
2019-06-29 10:08:59 +00:00
'id': newPostId+'/activity',
2019-06-28 18:55:29 +00:00
'type': 'Create',
'actor': prefix+'://'+domain+'/users/'+username,
'published': published,
'to': ['https://www.w3.org/ns/activitystreams#Public'],
'cc': [prefix+'://'+domain+'/users/'+username+'/followers'],
2019-06-29 10:08:59 +00:00
'object': {'id': newPostId,
2019-06-28 18:55:29 +00:00
'type': 'Note',
2019-06-29 10:23:40 +00:00
'summary': summary,
2019-06-29 10:41:22 +00:00
'inReplyTo': inReplyTo,
2019-06-28 18:55:29 +00:00
'published': published,
'url': prefix+'://'+domain+'/@'+username+'/'+statusNumber,
'attributedTo': prefix+'://'+domain+'/users/'+username,
'to': ['https://www.w3.org/ns/activitystreams#Public'],
'cc': [prefix+'://'+domain+'/users/'+username+'/followers'],
2019-06-29 10:23:40 +00:00
'sensitive': sensitive,
2019-06-28 18:55:29 +00:00
'atomUri': prefix+'://'+domain+'/users/'+username+'/statuses/'+statusNumber,
2019-06-29 10:41:22 +00:00
'inReplyToAtomUri': inReplyToAtomUri,
2019-06-28 18:55:29 +00:00
'conversation': 'tag:'+domain+','+conversationDate+':objectId='+conversationId+':objectType=Conversation',
'content': content,
'contentMap': {
'en': content
},
'attachment': [],
'tag': [],
'replies': {}
# 'id': 'https://'+domain+'/users/'+username+'/statuses/'+statusNumber+'/replies',
# 'type': 'Collection',
# 'first': {
# 'type': 'CollectionPage',
# 'partOf': 'https://'+domain+'/users/'+username+'/statuses/'+statusNumber+'/replies',
# 'items': []
# }
#}
}
}
2019-06-29 10:08:59 +00:00
if saveToFile:
outboxJsonFilename,outboxDir = createOutboxDir(username,domain)
filename=outboxDir+'/'+newPostId.replace('/','#')+'.json'
with open(filename, 'w') as fp:
commentjson.dump(newPost, fp, indent=4, sort_keys=False)
2019-06-29 11:47:33 +00:00
# TODO update output feed
2019-06-28 18:55:29 +00:00
return newPost
2019-06-29 10:08:59 +00:00
def createOutbox(username: str,domain: str,https: bool,noOfItems: int):
prefix='https'
if not https:
prefix='http'
outboxJsonFilename,outboxDir = createOutboxDir(username,domain)
outboxItems=0
outboxHeader = {'@context': 'https://www.w3.org/ns/activitystreams',
'first': prefix+'://'+domain+'/users/'+username+'/outbox?page=true',
'id': prefix+'://'+domain+'/users/'+username+'/outbox',
'last': prefix+'://'+domain+'/users/'+username+'/outbox?min_id=0&page=true',
'totalItems': str(outboxItems),
'type': 'OrderedCollection'}
maxMessageId=100000000000000000
minMessageId=100000000000000000
outboxItems = {'@context': 'https://www.w3.org/ns/activitystreams',
'id': prefix+'://'+domain+'/users/'+username+'/outbox?page=true',
'next': prefix+'://'+domain+'/users/'+username+'/outbox?max_id='+str(maxMessageId)+'&page=true',
'orderedItems': [
],
'partOf': prefix+'://'+domain+'/users/'+username+'/outbox',
'prev': prefix+'://'+domain+'/users/'+username+'/outbox?min_id='+str(minMessageId)+'&page=true',
'type': 'OrderedCollectionPage'}