2017-12-08 11:27:42 +01:00
|
|
|
import os
|
2018-03-25 13:42:04 +02:00
|
|
|
import ujson
|
2017-12-08 11:27:42 +01:00
|
|
|
import shutil
|
|
|
|
import subprocess
|
2018-02-08 00:06:02 +01:00
|
|
|
import logging
|
2018-03-13 20:43:39 +01:00
|
|
|
import random
|
2018-02-01 00:56:57 +01:00
|
|
|
import requests
|
2017-12-08 11:27:42 +01:00
|
|
|
|
2018-12-05 04:13:47 +01:00
|
|
|
from collections import defaultdict
|
|
|
|
|
2018-02-21 10:02:44 +01:00
|
|
|
from django.conf import settings
|
2017-12-08 11:27:42 +01:00
|
|
|
from django.utils.timezone import now as timezone_now
|
2018-04-23 23:36:40 +02:00
|
|
|
from django.forms.models import model_to_dict
|
2018-12-05 04:13:47 +01:00
|
|
|
from typing import Any, Dict, List, Optional, Tuple, Set, Iterator
|
2019-02-02 23:53:34 +01:00
|
|
|
from zerver.models import Reaction, RealmEmoji, UserProfile, Recipient, \
|
2020-01-24 15:00:18 +01:00
|
|
|
CustomProfileField, CustomProfileFieldValue, Realm
|
2018-08-01 00:18:04 +02:00
|
|
|
from zerver.data_import.slack_message_conversion import convert_to_zulip_markdown, \
|
2018-01-11 15:52:31 +01:00
|
|
|
get_user_full_name
|
2018-08-01 01:10:55 +02:00
|
|
|
from zerver.data_import.import_util import ZerverFieldsT, build_zerver_realm, \
|
2018-08-03 21:11:47 +02:00
|
|
|
build_avatar, build_subscription, build_recipient, build_usermessages, \
|
2018-08-03 22:33:41 +02:00
|
|
|
build_defaultstream, build_attachment, process_avatars, process_uploads, \
|
2019-07-02 07:41:51 +02:00
|
|
|
process_emojis, build_realm, build_stream, build_huddle, build_message, \
|
2018-10-25 17:33:52 +02:00
|
|
|
create_converted_data_files, make_subscriber_map
|
2018-10-28 18:55:32 +01:00
|
|
|
from zerver.data_import.sequencer import NEXT_ID
|
2018-02-26 06:57:00 +01:00
|
|
|
from zerver.lib.upload import random_name, sanitize_name
|
2018-06-18 16:24:01 +02:00
|
|
|
from zerver.lib.export import MESSAGE_BATCH_CHUNK_SIZE
|
2020-02-06 07:07:10 +01:00
|
|
|
from zerver.lib.emoji import name_to_codepoint
|
2020-01-24 15:00:18 +01:00
|
|
|
from zerver.lib.upload import resize_logo
|
2019-08-07 13:58:16 +02:00
|
|
|
from urllib.parse import urlencode
|
2017-12-29 10:57:48 +01:00
|
|
|
|
2019-08-12 13:44:07 +02:00
|
|
|
SlackToZulipUserIDT = Dict[str, int]
|
2018-04-07 00:00:05 +02:00
|
|
|
AddedChannelsT = Dict[str, Tuple[str, int]]
|
2019-07-02 07:41:51 +02:00
|
|
|
AddedMPIMsT = Dict[str, Tuple[str, int]]
|
2019-06-28 10:34:14 +02:00
|
|
|
DMMembersT = Dict[str, Tuple[str, str]]
|
2019-08-13 10:02:03 +02:00
|
|
|
SlackToZulipRecipientT = Dict[str, int]
|
2017-12-08 11:27:42 +01:00
|
|
|
|
|
|
|
def rm_tree(path: str) -> None:
|
|
|
|
if os.path.exists(path):
|
|
|
|
shutil.rmtree(path)
|
|
|
|
|
2018-02-25 14:46:34 +01:00
|
|
|
def slack_workspace_to_realm(domain_name: str, realm_id: int, user_list: List[ZerverFieldsT],
|
2018-04-18 19:10:17 +02:00
|
|
|
realm_subdomain: str, slack_data_dir: str,
|
2019-08-12 13:44:07 +02:00
|
|
|
custom_emoji_list: ZerverFieldsT) -> Tuple[ZerverFieldsT,
|
|
|
|
SlackToZulipUserIDT,
|
2019-08-13 10:02:03 +02:00
|
|
|
SlackToZulipRecipientT,
|
2019-01-31 14:32:37 +01:00
|
|
|
AddedChannelsT,
|
2019-07-02 07:41:51 +02:00
|
|
|
AddedMPIMsT,
|
2019-06-28 10:34:14 +02:00
|
|
|
DMMembersT,
|
2019-01-31 14:32:37 +01:00
|
|
|
List[ZerverFieldsT],
|
|
|
|
ZerverFieldsT]:
|
2018-01-23 19:04:59 +01:00
|
|
|
"""
|
|
|
|
Returns:
|
|
|
|
1. realm, Converted Realm data
|
2019-08-12 13:44:07 +02:00
|
|
|
2. slack_user_id_to_zulip_user_id, which is a dictionary to map from slack user id to zulip user id
|
2019-08-13 10:02:03 +02:00
|
|
|
3. slack_recipient_name_to_zulip_recipient_id, which is a dictionary to map from slack recipient
|
|
|
|
name(channel names, mpim names, usernames, etc) to zulip recipient id
|
2018-04-07 00:00:05 +02:00
|
|
|
4. added_channels, which is a dictionary to map from channel name to channel id, zulip stream_id
|
2019-07-02 07:41:51 +02:00
|
|
|
5. added_mpims, which is a dictionary to map from MPIM name to MPIM id, zulip huddle_id
|
2019-06-28 10:34:14 +02:00
|
|
|
6. dm_members, which is a dictionary to map from DM id to tuple of DM participants.
|
|
|
|
7. avatars, which is list to map avatars to zulip avatar records.json
|
|
|
|
8. emoji_url_map, which is maps emoji name to its slack url
|
2018-01-23 19:04:59 +01:00
|
|
|
"""
|
|
|
|
NOW = float(timezone_now().timestamp())
|
|
|
|
|
2018-08-01 01:01:55 +02:00
|
|
|
zerver_realm = build_zerver_realm(realm_id, realm_subdomain, NOW, 'Slack') # type: List[ZerverFieldsT]
|
2018-08-10 02:27:17 +02:00
|
|
|
realm = build_realm(zerver_realm, realm_id, domain_name)
|
2018-01-23 19:04:59 +01:00
|
|
|
|
2019-08-12 13:44:07 +02:00
|
|
|
zerver_userprofile, avatars, slack_user_id_to_zulip_user_id, zerver_customprofilefield, \
|
2018-04-09 22:58:03 +02:00
|
|
|
zerver_customprofilefield_value = users_to_zerver_userprofile(slack_data_dir, user_list,
|
|
|
|
realm_id, int(NOW), domain_name)
|
2019-08-13 10:02:03 +02:00
|
|
|
realm, added_channels, added_mpims, dm_members, slack_recipient_name_to_zulip_recipient_id = \
|
|
|
|
channels_to_zerver_stream(slack_data_dir, realm_id, realm, slack_user_id_to_zulip_user_id,
|
|
|
|
zerver_userprofile)
|
2019-07-26 18:39:50 +02:00
|
|
|
|
2018-03-29 18:58:10 +02:00
|
|
|
zerver_realmemoji, emoji_url_map = build_realmemoji(custom_emoji_list, realm_id)
|
|
|
|
realm['zerver_realmemoji'] = zerver_realmemoji
|
|
|
|
|
2018-01-23 19:04:59 +01:00
|
|
|
# See https://zulipchat.com/help/set-default-streams-for-new-users
|
|
|
|
# for documentation on zerver_defaultstream
|
|
|
|
realm['zerver_userprofile'] = zerver_userprofile
|
|
|
|
|
2018-04-09 22:58:03 +02:00
|
|
|
realm['zerver_customprofilefield'] = zerver_customprofilefield
|
2018-05-31 19:17:55 +02:00
|
|
|
realm['zerver_customprofilefieldvalue'] = zerver_customprofilefield_value
|
2018-04-09 22:58:03 +02:00
|
|
|
|
2019-08-13 10:02:03 +02:00
|
|
|
return realm, slack_user_id_to_zulip_user_id, slack_recipient_name_to_zulip_recipient_id, \
|
|
|
|
added_channels, added_mpims, dm_members, avatars, emoji_url_map
|
2018-01-23 19:04:59 +01:00
|
|
|
|
2018-03-29 18:58:10 +02:00
|
|
|
def build_realmemoji(custom_emoji_list: ZerverFieldsT,
|
|
|
|
realm_id: int) -> Tuple[List[ZerverFieldsT],
|
|
|
|
ZerverFieldsT]:
|
|
|
|
zerver_realmemoji = []
|
|
|
|
emoji_url_map = {}
|
|
|
|
emoji_id = 0
|
|
|
|
for emoji_name, url in custom_emoji_list.items():
|
|
|
|
if 'emoji.slack-edge.com' in url:
|
|
|
|
# Some of the emojis we get from the api have invalid links
|
|
|
|
# this is to prevent errors related to them
|
2018-08-17 23:24:57 +02:00
|
|
|
realmemoji = RealmEmoji(
|
2018-03-29 18:58:10 +02:00
|
|
|
name=emoji_name,
|
|
|
|
id=emoji_id,
|
|
|
|
file_name=os.path.basename(url),
|
|
|
|
deactivated=False)
|
2018-08-17 23:24:57 +02:00
|
|
|
|
|
|
|
realmemoji_dict = model_to_dict(realmemoji, exclude=['realm', 'author'])
|
|
|
|
realmemoji_dict['author'] = None
|
|
|
|
realmemoji_dict['realm'] = realm_id
|
|
|
|
|
2018-03-29 18:58:10 +02:00
|
|
|
emoji_url_map[emoji_name] = url
|
2018-08-17 23:24:57 +02:00
|
|
|
zerver_realmemoji.append(realmemoji_dict)
|
2018-03-29 18:58:10 +02:00
|
|
|
emoji_id += 1
|
|
|
|
return zerver_realmemoji, emoji_url_map
|
|
|
|
|
2018-02-01 00:56:57 +01:00
|
|
|
def users_to_zerver_userprofile(slack_data_dir: str, users: List[ZerverFieldsT], realm_id: int,
|
|
|
|
timestamp: Any, domain_name: str) -> Tuple[List[ZerverFieldsT],
|
2018-02-17 00:42:59 +01:00
|
|
|
List[ZerverFieldsT],
|
2019-08-12 13:44:07 +02:00
|
|
|
SlackToZulipUserIDT,
|
2018-04-09 22:58:03 +02:00
|
|
|
List[ZerverFieldsT],
|
|
|
|
List[ZerverFieldsT]]:
|
2017-12-08 11:27:42 +01:00
|
|
|
"""
|
|
|
|
Returns:
|
|
|
|
1. zerver_userprofile, which is a list of user profile
|
2018-02-17 00:42:59 +01:00
|
|
|
2. avatar_list, which is list to map avatars to zulip avatard records.json
|
2019-08-12 13:44:07 +02:00
|
|
|
3. slack_user_id_to_zulip_user_id, which is a dictionary to map from slack user id to zulip
|
2017-12-08 11:27:42 +01:00
|
|
|
user id
|
2018-04-09 22:58:03 +02:00
|
|
|
4. zerver_customprofilefield, which is a list of all custom profile fields
|
|
|
|
5. zerver_customprofilefield_values, which is a list of user profile fields
|
2017-12-08 11:27:42 +01:00
|
|
|
"""
|
2018-02-08 00:06:02 +01:00
|
|
|
logging.info('######### IMPORTING USERS STARTED #########\n')
|
2017-12-08 11:27:42 +01:00
|
|
|
zerver_userprofile = []
|
2018-04-09 13:53:32 +02:00
|
|
|
zerver_customprofilefield = [] # type: List[ZerverFieldsT]
|
|
|
|
zerver_customprofilefield_values = [] # type: List[ZerverFieldsT]
|
2018-02-16 23:54:38 +01:00
|
|
|
avatar_list = [] # type: List[ZerverFieldsT]
|
2019-08-12 13:44:07 +02:00
|
|
|
slack_user_id_to_zulip_user_id = {}
|
2018-01-20 10:01:17 +01:00
|
|
|
|
2018-04-09 13:53:32 +02:00
|
|
|
# The user data we get from the slack api does not contain custom profile data
|
|
|
|
# Hence we get it from the slack zip file
|
|
|
|
slack_data_file_user_list = get_data_file(slack_data_dir + '/users.json')
|
|
|
|
|
2019-08-13 13:05:40 +02:00
|
|
|
slack_user_id_to_custom_profile_fields = {} # type: ZerverFieldsT
|
2019-08-13 13:16:38 +02:00
|
|
|
slack_custom_field_name_to_zulip_custom_field_id = {} # type: ZerverFieldsT
|
2018-04-09 13:53:32 +02:00
|
|
|
|
|
|
|
for user in slack_data_file_user_list:
|
2019-08-13 13:05:40 +02:00
|
|
|
process_slack_custom_fields(user, slack_user_id_to_custom_profile_fields)
|
2018-04-09 13:53:32 +02:00
|
|
|
|
2018-02-06 22:19:47 +01:00
|
|
|
# We have only one primary owner in slack, see link
|
|
|
|
# https://get.slack.help/hc/en-us/articles/201912948-Owners-and-Administrators
|
|
|
|
# This is to import the primary owner first from all the users
|
2019-08-13 14:03:02 +02:00
|
|
|
user_id_count = custom_profile_field_value_id_count = custom_profile_field_id_count = 0
|
2018-02-06 22:19:47 +01:00
|
|
|
primary_owner_id = user_id_count
|
|
|
|
user_id_count += 1
|
|
|
|
|
2017-12-08 11:27:42 +01:00
|
|
|
for user in users:
|
|
|
|
slack_user_id = user['id']
|
|
|
|
|
2018-02-06 22:19:47 +01:00
|
|
|
if user.get('is_primary_owner', False):
|
2018-03-17 12:15:57 +01:00
|
|
|
user_id = primary_owner_id
|
2018-02-06 22:19:47 +01:00
|
|
|
else:
|
2018-03-17 12:15:57 +01:00
|
|
|
user_id = user_id_count
|
2018-02-06 22:19:47 +01:00
|
|
|
|
2018-01-23 19:04:59 +01:00
|
|
|
email = get_user_email(user, domain_name)
|
2019-03-09 17:43:48 +01:00
|
|
|
# ref: https://chat.zulip.org/help/set-your-profile-picture
|
2018-02-16 23:54:38 +01:00
|
|
|
avatar_url = build_avatar_url(slack_user_id, user['team_id'],
|
|
|
|
user['profile']['avatar_hash'])
|
2018-03-07 13:53:35 +01:00
|
|
|
build_avatar(user_id, realm_id, email, avatar_url, timestamp, avatar_list)
|
2019-10-05 02:35:07 +02:00
|
|
|
role = UserProfile.ROLE_MEMBER
|
|
|
|
if get_admin(user):
|
|
|
|
role = UserProfile.ROLE_REALM_ADMINISTRATOR
|
2019-10-31 13:18:49 +01:00
|
|
|
if get_guest(user):
|
|
|
|
role = UserProfile.ROLE_GUEST
|
2018-01-23 19:04:59 +01:00
|
|
|
timezone = get_user_timezone(user)
|
2017-12-08 11:27:42 +01:00
|
|
|
|
2019-08-13 13:05:40 +02:00
|
|
|
if slack_user_id in slack_user_id_to_custom_profile_fields:
|
2019-08-13 14:03:02 +02:00
|
|
|
slack_custom_field_name_to_zulip_custom_field_id, custom_profile_field_id_count = \
|
2019-08-13 13:16:38 +02:00
|
|
|
build_customprofile_field(zerver_customprofilefield,
|
|
|
|
slack_user_id_to_custom_profile_fields[slack_user_id],
|
2019-08-13 14:03:02 +02:00
|
|
|
custom_profile_field_id_count, realm_id,
|
2019-08-13 13:16:38 +02:00
|
|
|
slack_custom_field_name_to_zulip_custom_field_id)
|
2019-08-13 13:58:37 +02:00
|
|
|
custom_profile_field_value_id_count = build_customprofilefields_values(
|
2019-08-13 13:16:38 +02:00
|
|
|
slack_custom_field_name_to_zulip_custom_field_id,
|
|
|
|
slack_user_id_to_custom_profile_fields[slack_user_id], user_id,
|
2019-08-13 13:58:37 +02:00
|
|
|
custom_profile_field_value_id_count, zerver_customprofilefield_values)
|
2018-04-09 13:53:32 +02:00
|
|
|
|
2018-05-01 18:16:12 +02:00
|
|
|
userprofile = UserProfile(
|
|
|
|
full_name=get_user_full_name(user),
|
|
|
|
short_name=user['name'],
|
2019-08-08 19:39:26 +02:00
|
|
|
is_active=not user.get('deleted', False) and not user["is_mirror_dummy"],
|
|
|
|
is_mirror_dummy=user["is_mirror_dummy"],
|
2018-05-01 18:16:12 +02:00
|
|
|
id=user_id,
|
|
|
|
email=email,
|
2018-08-17 16:16:49 +02:00
|
|
|
delivery_email=email,
|
2018-05-01 18:16:12 +02:00
|
|
|
avatar_source='U',
|
|
|
|
is_bot=user.get('is_bot', False),
|
|
|
|
pointer=-1,
|
2019-10-05 02:35:07 +02:00
|
|
|
role=role,
|
2018-05-01 18:16:12 +02:00
|
|
|
bot_type=1 if user.get('is_bot', False) else None,
|
|
|
|
date_joined=timestamp,
|
|
|
|
timezone=timezone,
|
|
|
|
last_login=timestamp)
|
2018-05-01 06:30:11 +02:00
|
|
|
userprofile_dict = model_to_dict(userprofile)
|
|
|
|
# Set realm id separately as the corresponding realm is not yet a Realm model instance
|
|
|
|
userprofile_dict['realm'] = realm_id
|
2017-12-08 11:27:42 +01:00
|
|
|
|
2018-05-01 06:30:11 +02:00
|
|
|
zerver_userprofile.append(userprofile_dict)
|
2019-08-12 13:44:07 +02:00
|
|
|
slack_user_id_to_zulip_user_id[slack_user_id] = user_id
|
2018-02-06 22:19:47 +01:00
|
|
|
if not user.get('is_primary_owner', False):
|
|
|
|
user_id_count += 1
|
|
|
|
|
2020-04-09 21:51:58 +02:00
|
|
|
logging.info("{} -> {}".format(user['name'], userprofile_dict['email']))
|
2018-04-09 13:53:32 +02:00
|
|
|
|
|
|
|
process_customprofilefields(zerver_customprofilefield, zerver_customprofilefield_values)
|
2018-02-08 00:06:02 +01:00
|
|
|
logging.info('######### IMPORTING USERS FINISHED #########\n')
|
2019-08-12 13:44:07 +02:00
|
|
|
return zerver_userprofile, avatar_list, slack_user_id_to_zulip_user_id, zerver_customprofilefield, \
|
2018-04-09 22:58:03 +02:00
|
|
|
zerver_customprofilefield_values
|
2017-12-08 11:27:42 +01:00
|
|
|
|
2018-04-09 13:53:32 +02:00
|
|
|
def build_customprofile_field(customprofile_field: List[ZerverFieldsT], fields: ZerverFieldsT,
|
2019-08-13 14:03:02 +02:00
|
|
|
custom_profile_field_id: int, realm_id: int,
|
2019-08-13 13:16:38 +02:00
|
|
|
slack_custom_field_name_to_zulip_custom_field_id: ZerverFieldsT) \
|
|
|
|
-> Tuple[ZerverFieldsT, int]:
|
2018-04-09 13:53:32 +02:00
|
|
|
# The name of the custom profile field is not provided in the slack data
|
|
|
|
# Hash keys of the fields are provided
|
|
|
|
# Reference: https://api.slack.com/methods/users.profile.set
|
|
|
|
for field, value in fields.items():
|
2019-08-13 13:16:38 +02:00
|
|
|
if field not in slack_custom_field_name_to_zulip_custom_field_id:
|
2018-06-09 13:44:47 +02:00
|
|
|
slack_custom_fields = ['phone', 'skype']
|
|
|
|
if field in slack_custom_fields:
|
|
|
|
field_name = field
|
|
|
|
else:
|
2019-08-13 14:03:02 +02:00
|
|
|
field_name = "slack custom field %s" % (str(custom_profile_field_id + 1),)
|
2018-08-17 23:24:57 +02:00
|
|
|
customprofilefield = CustomProfileField(
|
2019-08-13 14:03:02 +02:00
|
|
|
id=custom_profile_field_id,
|
2018-04-09 13:53:32 +02:00
|
|
|
name=field_name,
|
|
|
|
field_type=1 # For now this is defaulted to 'SHORT_TEXT'
|
|
|
|
# Processing is done in the function 'process_customprofilefields'
|
|
|
|
)
|
2018-08-17 23:24:57 +02:00
|
|
|
|
|
|
|
customprofilefield_dict = model_to_dict(customprofilefield,
|
|
|
|
exclude=['realm'])
|
|
|
|
customprofilefield_dict['realm'] = realm_id
|
|
|
|
|
2019-08-13 14:03:02 +02:00
|
|
|
slack_custom_field_name_to_zulip_custom_field_id[field] = custom_profile_field_id
|
|
|
|
custom_profile_field_id += 1
|
2018-08-17 23:24:57 +02:00
|
|
|
customprofile_field.append(customprofilefield_dict)
|
2019-08-13 14:03:02 +02:00
|
|
|
return slack_custom_field_name_to_zulip_custom_field_id, custom_profile_field_id
|
2018-04-09 13:53:32 +02:00
|
|
|
|
2018-06-09 13:44:47 +02:00
|
|
|
def process_slack_custom_fields(user: ZerverFieldsT,
|
2019-08-13 13:05:40 +02:00
|
|
|
slack_user_id_to_custom_profile_fields: ZerverFieldsT) -> None:
|
|
|
|
slack_user_id_to_custom_profile_fields[user['id']] = {}
|
2018-06-09 13:44:47 +02:00
|
|
|
if user['profile'].get('fields'):
|
2019-08-13 13:05:40 +02:00
|
|
|
slack_user_id_to_custom_profile_fields[user['id']] = user['profile']['fields']
|
2018-06-09 13:44:47 +02:00
|
|
|
|
|
|
|
slack_custom_fields = ['phone', 'skype']
|
|
|
|
for field in slack_custom_fields:
|
|
|
|
if field in user['profile']:
|
2019-08-13 13:05:40 +02:00
|
|
|
slack_user_id_to_custom_profile_fields[user['id']][field] = {'value': user['profile'][field]}
|
2018-06-09 13:44:47 +02:00
|
|
|
|
2019-08-13 13:16:38 +02:00
|
|
|
def build_customprofilefields_values(slack_custom_field_name_to_zulip_custom_field_id: ZerverFieldsT,
|
|
|
|
fields: ZerverFieldsT, user_id: int, custom_field_id: int,
|
2018-04-09 13:53:32 +02:00
|
|
|
custom_field_values: List[ZerverFieldsT]) -> int:
|
|
|
|
for field, value in fields.items():
|
2018-12-12 21:56:14 +01:00
|
|
|
if value['value'] == "":
|
|
|
|
continue
|
2018-08-17 23:24:57 +02:00
|
|
|
custom_field_value = CustomProfileFieldValue(
|
2018-04-09 13:53:32 +02:00
|
|
|
id=custom_field_id,
|
|
|
|
value=value['value'])
|
2018-08-17 23:24:57 +02:00
|
|
|
|
|
|
|
custom_field_value_dict = model_to_dict(custom_field_value,
|
|
|
|
exclude=['user_profile', 'field'])
|
|
|
|
custom_field_value_dict['user_profile'] = user_id
|
2019-08-13 13:16:38 +02:00
|
|
|
custom_field_value_dict['field'] = slack_custom_field_name_to_zulip_custom_field_id[field]
|
2018-08-17 23:24:57 +02:00
|
|
|
|
|
|
|
custom_field_values.append(custom_field_value_dict)
|
2018-04-09 13:53:32 +02:00
|
|
|
custom_field_id += 1
|
|
|
|
return custom_field_id
|
|
|
|
|
|
|
|
def process_customprofilefields(customprofilefield: List[ZerverFieldsT],
|
|
|
|
customprofilefield_value: List[ZerverFieldsT]) -> None:
|
|
|
|
for field in customprofilefield:
|
|
|
|
for field_value in customprofilefield_value:
|
|
|
|
if field_value['field'] == field['id'] and len(field_value['value']) > 50:
|
2018-05-06 11:40:31 +02:00
|
|
|
field['field_type'] = 2 # corresponding to Long text
|
2018-04-09 13:53:32 +02:00
|
|
|
break
|
|
|
|
|
2018-01-23 19:04:59 +01:00
|
|
|
def get_user_email(user: ZerverFieldsT, domain_name: str) -> str:
|
2018-03-13 00:21:54 +01:00
|
|
|
if 'email' in user['profile']:
|
|
|
|
return user['profile']['email']
|
2019-08-08 19:39:26 +02:00
|
|
|
if user['is_mirror_dummy']:
|
|
|
|
return "{}@{}.slack.com".format(user["name"], user["team_domain"])
|
2018-03-13 00:21:54 +01:00
|
|
|
if 'bot_id' in user['profile']:
|
|
|
|
if 'real_name_normalized' in user['profile']:
|
|
|
|
slack_bot_name = user['profile']['real_name_normalized']
|
|
|
|
elif 'first_name' in user['profile']:
|
|
|
|
slack_bot_name = user['profile']['first_name']
|
|
|
|
else:
|
|
|
|
raise AssertionError("Could not identify bot type")
|
|
|
|
return slack_bot_name.replace("Bot", "").replace(" ", "") + "-bot@%s" % (domain_name,)
|
2019-01-27 17:54:44 +01:00
|
|
|
if get_user_full_name(user).lower() == "slackbot":
|
2018-10-17 01:33:21 +02:00
|
|
|
return "imported-slackbot-bot@%s" % (domain_name,)
|
|
|
|
raise AssertionError("Could not find email address for Slack user %s" % (user,))
|
2018-01-23 19:04:59 +01:00
|
|
|
|
2018-02-16 23:54:38 +01:00
|
|
|
def build_avatar_url(slack_user_id: str, team_id: str, avatar_hash: str) -> str:
|
|
|
|
avatar_url = "https://ca.slack-edge.com/{}-{}-{}".format(team_id, slack_user_id,
|
|
|
|
avatar_hash)
|
|
|
|
return avatar_url
|
|
|
|
|
2018-02-06 21:02:23 +01:00
|
|
|
def get_admin(user: ZerverFieldsT) -> bool:
|
|
|
|
admin = user.get('is_admin', False)
|
|
|
|
owner = user.get('is_owner', False)
|
|
|
|
primary_owner = user.get('is_primary_owner', False)
|
|
|
|
|
|
|
|
if admin or owner or primary_owner:
|
|
|
|
return True
|
|
|
|
return False
|
|
|
|
|
2019-10-31 13:18:49 +01:00
|
|
|
def get_guest(user: ZerverFieldsT) -> bool:
|
|
|
|
restricted_user = user.get('is_restricted', False)
|
|
|
|
ultra_restricted_user = user.get('is_ultra_restricted', False)
|
|
|
|
|
|
|
|
# Slack's Single channel and multi channel guests both have
|
|
|
|
# is_restricted set to True. So assuming Slack doesn't change their
|
|
|
|
# data model, it would also be correct to just check whether
|
|
|
|
# is_restricted is set to True.
|
|
|
|
return restricted_user or ultra_restricted_user
|
|
|
|
|
2018-01-23 19:04:59 +01:00
|
|
|
def get_user_timezone(user: ZerverFieldsT) -> str:
|
|
|
|
_default_timezone = "America/New_York"
|
|
|
|
timezone = user.get("tz", _default_timezone)
|
|
|
|
if timezone is None or '/' not in timezone:
|
|
|
|
timezone = _default_timezone
|
|
|
|
return timezone
|
|
|
|
|
2019-07-26 18:39:50 +02:00
|
|
|
def channels_to_zerver_stream(slack_data_dir: str, realm_id: int,
|
2019-08-12 13:44:07 +02:00
|
|
|
realm: Dict[str, Any],
|
|
|
|
slack_user_id_to_zulip_user_id: SlackToZulipUserIDT,
|
2019-07-26 18:39:50 +02:00
|
|
|
zerver_userprofile: List[ZerverFieldsT]) \
|
2019-08-13 10:02:03 +02:00
|
|
|
-> Tuple[Dict[str, List[ZerverFieldsT]], AddedChannelsT, AddedMPIMsT,
|
|
|
|
DMMembersT, SlackToZulipRecipientT]:
|
2017-12-08 11:27:42 +01:00
|
|
|
"""
|
|
|
|
Returns:
|
2019-07-26 18:39:50 +02:00
|
|
|
1. realm, Converted Realm data
|
|
|
|
2. added_channels, which is a dictionary to map from channel name to channel id, zulip stream_id
|
|
|
|
3. added_mpims, which is a dictionary to map from MPIM(multiparty IM) name to MPIM id, zulip huddle_id
|
|
|
|
4. dm_members, which is a dictionary to map from DM id to tuple of DM participants.
|
2019-08-13 10:02:03 +02:00
|
|
|
5. slack_recipient_name_to_zulip_recipient_id, which is a dictionary to map from slack recipient
|
|
|
|
name(channel names, mpim names, usernames etc) to zulip recipient_id
|
2017-12-08 11:27:42 +01:00
|
|
|
"""
|
2018-02-08 00:06:02 +01:00
|
|
|
logging.info('######### IMPORTING CHANNELS STARTED #########\n')
|
2018-01-20 10:01:17 +01:00
|
|
|
|
2017-12-08 11:27:42 +01:00
|
|
|
added_channels = {}
|
2019-07-02 07:41:51 +02:00
|
|
|
added_mpims = {}
|
2019-06-28 10:34:14 +02:00
|
|
|
dm_members = {}
|
2019-08-13 10:02:03 +02:00
|
|
|
slack_recipient_name_to_zulip_recipient_id = {}
|
2017-12-08 11:27:42 +01:00
|
|
|
|
2019-07-26 18:39:50 +02:00
|
|
|
realm["zerver_stream"] = []
|
|
|
|
realm["zerver_huddle"] = []
|
|
|
|
realm["zerver_subscription"] = []
|
|
|
|
realm["zerver_recipient"] = []
|
|
|
|
realm["zerver_defaultstream"] = []
|
2017-12-08 11:27:42 +01:00
|
|
|
|
2019-07-02 07:41:51 +02:00
|
|
|
subscription_id_count = recipient_id_count = 0
|
|
|
|
stream_id_count = defaultstream_id = 0
|
|
|
|
huddle_id_count = 0
|
2017-12-08 11:27:42 +01:00
|
|
|
|
2019-06-28 11:08:45 +02:00
|
|
|
def process_channels(channels: List[Dict[str, Any]], invite_only: bool=False) -> None:
|
2019-06-28 11:07:03 +02:00
|
|
|
nonlocal stream_id_count
|
|
|
|
nonlocal recipient_id_count
|
|
|
|
nonlocal defaultstream_id
|
|
|
|
nonlocal subscription_id_count
|
|
|
|
|
|
|
|
for channel in channels:
|
|
|
|
# map Slack's topic and purpose content into Zulip's stream description.
|
|
|
|
# WARN This mapping is lossy since the topic.creator, topic.last_set,
|
|
|
|
# purpose.creator, purpose.last_set fields are not preserved.
|
|
|
|
description = channel["purpose"]["value"]
|
|
|
|
stream_id = stream_id_count
|
|
|
|
recipient_id = recipient_id_count
|
|
|
|
|
|
|
|
stream = build_stream(float(channel["created"]), realm_id, channel["name"],
|
2019-06-28 11:08:45 +02:00
|
|
|
description, stream_id, channel["is_archived"], invite_only)
|
2019-07-26 18:39:50 +02:00
|
|
|
realm["zerver_stream"].append(stream)
|
2019-06-28 11:07:03 +02:00
|
|
|
|
2019-08-13 16:07:36 +02:00
|
|
|
slack_default_channels = ['general', 'random']
|
|
|
|
if channel['name'] in slack_default_channels:
|
2019-06-28 11:07:03 +02:00
|
|
|
defaultstream = build_defaultstream(realm_id, stream_id,
|
|
|
|
defaultstream_id)
|
2019-07-26 18:39:50 +02:00
|
|
|
realm["zerver_defaultstream"].append(defaultstream)
|
2019-06-28 11:07:03 +02:00
|
|
|
defaultstream_id += 1
|
|
|
|
|
|
|
|
added_channels[stream['name']] = (channel['id'], stream_id)
|
|
|
|
|
|
|
|
recipient = build_recipient(stream_id, recipient_id, Recipient.STREAM)
|
2019-07-26 18:39:50 +02:00
|
|
|
realm["zerver_recipient"].append(recipient)
|
2019-08-13 10:02:03 +02:00
|
|
|
slack_recipient_name_to_zulip_recipient_id[stream['name']] = recipient_id
|
2019-06-28 11:07:03 +02:00
|
|
|
|
2019-07-26 18:39:50 +02:00
|
|
|
subscription_id_count = get_subscription(channel['members'], realm["zerver_subscription"],
|
2019-08-12 13:44:07 +02:00
|
|
|
recipient_id, slack_user_id_to_zulip_user_id,
|
2019-06-28 11:07:03 +02:00
|
|
|
subscription_id_count)
|
|
|
|
|
|
|
|
stream_id_count += 1
|
|
|
|
recipient_id_count += 1
|
2020-04-09 21:51:58 +02:00
|
|
|
logging.info("{} -> created".format(channel['name']))
|
2019-06-28 11:07:03 +02:00
|
|
|
|
|
|
|
# TODO map Slack's pins to Zulip's stars
|
|
|
|
# There is the security model that Slack's pins are known to the team owner
|
|
|
|
# as evident from where it is stored at (channels)
|
|
|
|
# "pins": [
|
|
|
|
# {
|
|
|
|
# "id": "1444755381.000003",
|
|
|
|
# "type": "C",
|
|
|
|
# "user": "U061A5N1G",
|
|
|
|
# "owner": "U061A5N1G",
|
|
|
|
# "created": "1444755463"
|
|
|
|
# }
|
|
|
|
# ],
|
|
|
|
|
2019-06-28 11:08:45 +02:00
|
|
|
public_channels = get_data_file(slack_data_dir + '/channels.json')
|
|
|
|
process_channels(public_channels)
|
|
|
|
|
|
|
|
try:
|
|
|
|
private_channels = get_data_file(slack_data_dir + '/groups.json')
|
|
|
|
except FileNotFoundError:
|
|
|
|
private_channels = []
|
|
|
|
process_channels(private_channels, True)
|
2017-12-08 11:27:42 +01:00
|
|
|
|
2019-07-02 07:41:51 +02:00
|
|
|
# mpim is the Slack equivalent of huddle.
|
|
|
|
def process_mpims(mpims: List[Dict[str, Any]]) -> None:
|
|
|
|
nonlocal huddle_id_count
|
|
|
|
nonlocal recipient_id_count
|
|
|
|
nonlocal subscription_id_count
|
|
|
|
|
|
|
|
for mpim in mpims:
|
|
|
|
huddle = build_huddle(huddle_id_count)
|
2019-07-26 18:39:50 +02:00
|
|
|
realm["zerver_huddle"].append(huddle)
|
2019-07-02 07:41:51 +02:00
|
|
|
|
|
|
|
added_mpims[mpim['name']] = (mpim['id'], huddle_id_count)
|
|
|
|
|
|
|
|
recipient = build_recipient(huddle_id_count, recipient_id_count, Recipient.HUDDLE)
|
2019-07-26 18:39:50 +02:00
|
|
|
realm["zerver_recipient"].append(recipient)
|
2019-08-13 10:02:03 +02:00
|
|
|
slack_recipient_name_to_zulip_recipient_id[mpim['name']] = recipient_id_count
|
2019-07-02 07:41:51 +02:00
|
|
|
|
2019-07-26 18:39:50 +02:00
|
|
|
subscription_id_count = get_subscription(mpim['members'], realm["zerver_subscription"],
|
2019-08-12 13:44:07 +02:00
|
|
|
recipient_id_count, slack_user_id_to_zulip_user_id,
|
2019-07-02 07:41:51 +02:00
|
|
|
subscription_id_count)
|
|
|
|
|
|
|
|
huddle_id_count += 1
|
|
|
|
recipient_id_count += 1
|
2020-04-09 21:51:58 +02:00
|
|
|
logging.info("{} -> created".format(mpim['name']))
|
2019-07-02 07:41:51 +02:00
|
|
|
|
|
|
|
try:
|
|
|
|
mpims = get_data_file(slack_data_dir + '/mpims.json')
|
|
|
|
except FileNotFoundError:
|
|
|
|
mpims = []
|
|
|
|
process_mpims(mpims)
|
|
|
|
|
2019-08-12 13:44:07 +02:00
|
|
|
for slack_user_id, zulip_user_id in slack_user_id_to_zulip_user_id.items():
|
2019-07-30 23:35:52 +02:00
|
|
|
recipient = build_recipient(zulip_user_id, recipient_id_count, Recipient.PERSONAL)
|
2019-08-13 10:02:03 +02:00
|
|
|
slack_recipient_name_to_zulip_recipient_id[slack_user_id] = recipient_id_count
|
2019-07-30 23:35:52 +02:00
|
|
|
sub = build_subscription(recipient_id_count, zulip_user_id, subscription_id_count)
|
2019-07-26 18:39:50 +02:00
|
|
|
realm["zerver_recipient"].append(recipient)
|
|
|
|
realm["zerver_subscription"].append(sub)
|
2018-01-23 19:04:59 +01:00
|
|
|
recipient_id_count += 1
|
2019-06-28 10:34:14 +02:00
|
|
|
subscription_id_count += 1
|
|
|
|
|
|
|
|
def process_dms(dms: List[Dict[str, Any]]) -> None:
|
|
|
|
for dm in dms:
|
|
|
|
user_a = dm["members"][0]
|
|
|
|
user_b = dm["members"][1]
|
|
|
|
dm_members[dm["id"]] = (user_a, user_b)
|
|
|
|
|
|
|
|
try:
|
|
|
|
dms = get_data_file(slack_data_dir + '/dms.json')
|
|
|
|
except FileNotFoundError:
|
|
|
|
dms = []
|
|
|
|
process_dms(dms)
|
2017-12-08 11:27:42 +01:00
|
|
|
|
2018-02-08 00:06:02 +01:00
|
|
|
logging.info('######### IMPORTING STREAMS FINISHED #########\n')
|
2019-08-13 10:02:03 +02:00
|
|
|
return realm, added_channels, added_mpims, dm_members, slack_recipient_name_to_zulip_recipient_id
|
2018-01-23 19:04:59 +01:00
|
|
|
|
2018-08-02 00:35:02 +02:00
|
|
|
def get_subscription(channel_members: List[str], zerver_subscription: List[ZerverFieldsT],
|
2019-08-12 13:44:07 +02:00
|
|
|
recipient_id: int, slack_user_id_to_zulip_user_id: SlackToZulipUserIDT,
|
2018-08-02 00:35:02 +02:00
|
|
|
subscription_id: int) -> int:
|
2019-08-12 13:44:07 +02:00
|
|
|
for slack_user_id in channel_members:
|
|
|
|
sub = build_subscription(recipient_id, slack_user_id_to_zulip_user_id[slack_user_id],
|
|
|
|
subscription_id)
|
2017-12-08 11:27:42 +01:00
|
|
|
zerver_subscription.append(sub)
|
2018-03-17 12:15:57 +01:00
|
|
|
subscription_id += 1
|
|
|
|
return subscription_id
|
2017-12-08 11:27:42 +01:00
|
|
|
|
2018-12-17 02:26:51 +01:00
|
|
|
def process_long_term_idle_users(slack_data_dir: str, users: List[ZerverFieldsT],
|
2019-08-12 13:44:07 +02:00
|
|
|
slack_user_id_to_zulip_user_id: SlackToZulipUserIDT,
|
|
|
|
added_channels: AddedChannelsT,
|
2019-06-28 10:34:14 +02:00
|
|
|
added_mpims: AddedMPIMsT, dm_members: DMMembersT,
|
2018-12-17 02:26:51 +01:00
|
|
|
zerver_userprofile: List[ZerverFieldsT]) -> Set[int]:
|
|
|
|
"""Algorithmically, we treat users who have sent at least 10 messages
|
|
|
|
or have sent a message within the last 60 days as active.
|
|
|
|
Everyone else is treated as long-term idle, which means they will
|
2020-03-28 01:25:56 +01:00
|
|
|
have a slightly slower first page load when coming back to
|
2018-12-17 02:26:51 +01:00
|
|
|
Zulip.
|
|
|
|
"""
|
2019-06-28 10:34:14 +02:00
|
|
|
all_messages = get_messages_iterator(slack_data_dir, added_channels, added_mpims, dm_members)
|
2018-12-17 02:26:51 +01:00
|
|
|
|
|
|
|
sender_counts = defaultdict(int) # type: Dict[str, int]
|
|
|
|
recent_senders = set() # type: Set[str]
|
|
|
|
NOW = float(timezone_now().timestamp())
|
|
|
|
for message in all_messages:
|
|
|
|
timestamp = float(message['ts'])
|
|
|
|
slack_user_id = get_message_sending_user(message)
|
|
|
|
if not slack_user_id:
|
|
|
|
continue
|
|
|
|
|
|
|
|
if slack_user_id in recent_senders:
|
|
|
|
continue
|
|
|
|
|
|
|
|
if NOW - timestamp < 60:
|
|
|
|
recent_senders.add(slack_user_id)
|
|
|
|
|
|
|
|
sender_counts[slack_user_id] += 1
|
|
|
|
for (slack_sender_id, count) in sender_counts.items():
|
|
|
|
if count > 10:
|
|
|
|
recent_senders.add(slack_sender_id)
|
|
|
|
|
|
|
|
long_term_idle = set()
|
|
|
|
|
|
|
|
for slack_user in users:
|
|
|
|
if slack_user["id"] in recent_senders:
|
|
|
|
continue
|
2019-08-12 13:44:07 +02:00
|
|
|
zulip_user_id = slack_user_id_to_zulip_user_id[slack_user['id']]
|
2018-12-17 02:26:51 +01:00
|
|
|
long_term_idle.add(zulip_user_id)
|
|
|
|
|
|
|
|
for user_profile_row in zerver_userprofile:
|
|
|
|
if user_profile_row['id'] in long_term_idle:
|
|
|
|
user_profile_row['long_term_idle'] = True
|
|
|
|
# Setting last_active_message_id to 1 means the user, if
|
|
|
|
# imported, will get the full message history for the
|
|
|
|
# streams they were on.
|
|
|
|
user_profile_row['last_active_message_id'] = 1
|
|
|
|
|
|
|
|
return long_term_idle
|
|
|
|
|
2018-03-07 14:14:08 +01:00
|
|
|
def convert_slack_workspace_messages(slack_data_dir: str, users: List[ZerverFieldsT], realm_id: int,
|
2019-08-12 13:44:07 +02:00
|
|
|
slack_user_id_to_zulip_user_id: SlackToZulipUserIDT,
|
2019-08-13 10:02:03 +02:00
|
|
|
slack_recipient_name_to_zulip_recipient_id: SlackToZulipRecipientT,
|
2019-07-02 07:41:51 +02:00
|
|
|
added_channels: AddedChannelsT,
|
|
|
|
added_mpims: AddedMPIMsT,
|
2019-06-28 10:34:14 +02:00
|
|
|
dm_members: DMMembersT,
|
2019-07-02 07:41:51 +02:00
|
|
|
realm: ZerverFieldsT,
|
2018-12-17 02:26:51 +01:00
|
|
|
zerver_userprofile: List[ZerverFieldsT],
|
2018-06-13 20:15:35 +02:00
|
|
|
zerver_realmemoji: List[ZerverFieldsT], domain_name: str,
|
2018-06-18 16:24:01 +02:00
|
|
|
output_dir: str,
|
|
|
|
chunk_size: int=MESSAGE_BATCH_CHUNK_SIZE) -> Tuple[List[ZerverFieldsT],
|
|
|
|
List[ZerverFieldsT],
|
|
|
|
List[ZerverFieldsT]]:
|
2018-01-23 19:04:59 +01:00
|
|
|
"""
|
|
|
|
Returns:
|
2018-06-13 20:15:35 +02:00
|
|
|
1. reactions, which is a list of the reactions
|
2018-02-26 10:03:48 +01:00
|
|
|
2. uploads, which is a list of uploads to be mapped in uploads records.json
|
2018-02-26 10:16:34 +01:00
|
|
|
3. attachment, which is a list of the attachments
|
2018-01-23 19:04:59 +01:00
|
|
|
"""
|
2018-02-25 10:12:30 +01:00
|
|
|
|
2019-08-12 13:44:07 +02:00
|
|
|
long_term_idle = process_long_term_idle_users(slack_data_dir, users, slack_user_id_to_zulip_user_id,
|
|
|
|
added_channels, added_mpims, dm_members,
|
|
|
|
zerver_userprofile)
|
2018-12-17 02:26:51 +01:00
|
|
|
|
2019-06-28 10:34:14 +02:00
|
|
|
all_messages = get_messages_iterator(slack_data_dir, added_channels, added_mpims, dm_members)
|
2018-02-08 00:06:02 +01:00
|
|
|
logging.info('######### IMPORTING MESSAGES STARTED #########\n')
|
2018-02-25 09:09:32 +01:00
|
|
|
|
2018-06-17 19:16:42 +02:00
|
|
|
total_reactions = [] # type: List[ZerverFieldsT]
|
|
|
|
total_attachments = [] # type: List[ZerverFieldsT]
|
|
|
|
total_uploads = [] # type: List[ZerverFieldsT]
|
|
|
|
|
|
|
|
dump_file_id = 1
|
|
|
|
|
2018-10-25 17:33:52 +02:00
|
|
|
subscriber_map = make_subscriber_map(
|
|
|
|
zerver_subscription=realm['zerver_subscription'],
|
|
|
|
)
|
|
|
|
|
2018-06-17 19:16:42 +02:00
|
|
|
while True:
|
2018-12-05 04:13:47 +01:00
|
|
|
message_data = []
|
|
|
|
_counter = 0
|
|
|
|
for msg in all_messages:
|
|
|
|
_counter += 1
|
|
|
|
message_data.append(msg)
|
|
|
|
if _counter == chunk_size:
|
|
|
|
break
|
2018-06-17 19:16:42 +02:00
|
|
|
if len(message_data) == 0:
|
|
|
|
break
|
2018-12-05 04:13:47 +01:00
|
|
|
|
2018-10-28 18:55:32 +01:00
|
|
|
zerver_message, zerver_usermessage, attachment, uploads, reactions = \
|
|
|
|
channel_message_to_zerver_message(
|
2019-08-13 10:02:03 +02:00
|
|
|
realm_id, users, slack_user_id_to_zulip_user_id, slack_recipient_name_to_zulip_recipient_id,
|
|
|
|
message_data, zerver_realmemoji, subscriber_map, added_channels, dm_members,
|
2018-12-17 02:26:51 +01:00
|
|
|
domain_name, long_term_idle)
|
2018-06-17 19:16:42 +02:00
|
|
|
|
|
|
|
message_json = dict(
|
|
|
|
zerver_message=zerver_message,
|
|
|
|
zerver_usermessage=zerver_usermessage)
|
|
|
|
|
|
|
|
message_file = "/messages-%06d.json" % (dump_file_id,)
|
2019-04-20 01:00:46 +02:00
|
|
|
logging.info("Writing Messages to %s\n" % (output_dir + message_file,))
|
2018-06-17 19:16:42 +02:00
|
|
|
create_converted_data_files(message_json, output_dir, message_file)
|
|
|
|
|
|
|
|
total_reactions += reactions
|
|
|
|
total_attachments += attachment
|
|
|
|
total_uploads += uploads
|
|
|
|
|
|
|
|
dump_file_id += 1
|
2018-01-23 19:04:59 +01:00
|
|
|
|
2018-06-13 20:15:35 +02:00
|
|
|
logging.info('######### IMPORTING MESSAGES FINISHED #########\n')
|
2018-06-17 19:16:42 +02:00
|
|
|
return total_reactions, total_uploads, total_attachments
|
2018-01-23 19:04:59 +01:00
|
|
|
|
2019-08-08 19:39:26 +02:00
|
|
|
def get_messages_iterator(slack_data_dir: str, added_channels: Dict[str, Any],
|
2019-06-28 10:34:14 +02:00
|
|
|
added_mpims: AddedMPIMsT, dm_members: DMMembersT) -> Iterator[ZerverFieldsT]:
|
2018-12-05 04:13:47 +01:00
|
|
|
"""This function is an iterator that returns all the messages across
|
|
|
|
all Slack channels, in order by timestamp. It's important to
|
|
|
|
not read all the messages into memory at once, because for
|
|
|
|
large imports that can OOM kill."""
|
2019-07-02 07:41:51 +02:00
|
|
|
|
2019-06-28 10:34:14 +02:00
|
|
|
dir_names = list(added_channels.keys()) + list(added_mpims.keys()) + list(dm_members.keys())
|
2018-12-05 04:13:47 +01:00
|
|
|
all_json_names = defaultdict(list) # type: Dict[str, List[str]]
|
2019-07-02 07:41:51 +02:00
|
|
|
for dir_name in dir_names:
|
|
|
|
dir_path = os.path.join(slack_data_dir, dir_name)
|
|
|
|
json_names = os.listdir(dir_path)
|
2018-02-25 07:08:28 +01:00
|
|
|
for json_name in json_names:
|
2019-07-02 07:41:51 +02:00
|
|
|
all_json_names[json_name].append(dir_path)
|
2018-12-05 04:13:47 +01:00
|
|
|
|
|
|
|
# Sort json_name by date
|
|
|
|
for json_name in sorted(all_json_names.keys()):
|
|
|
|
messages_for_one_day = [] # type: List[ZerverFieldsT]
|
2019-07-02 07:41:51 +02:00
|
|
|
for dir_path in all_json_names[json_name]:
|
|
|
|
message_dir = os.path.join(dir_path, json_name)
|
|
|
|
dir_name = os.path.basename(dir_path)
|
2020-01-26 07:47:21 +01:00
|
|
|
messages = []
|
|
|
|
for message in get_data_file(message_dir):
|
|
|
|
if message.get('user') == 'U00':
|
|
|
|
# Skip messages involving the the "U00" user,
|
|
|
|
# which is apparently used in some channel rename
|
|
|
|
# messages. It's likely just the result of some
|
|
|
|
# bug in Slack's export system. Arguably we could
|
|
|
|
# change this to point to slackbot instead, but
|
|
|
|
# skipping those messages is simpler.
|
|
|
|
continue
|
2019-08-12 12:58:54 +02:00
|
|
|
if dir_name in added_channels:
|
2019-07-02 07:41:51 +02:00
|
|
|
message['channel_name'] = dir_name
|
2019-08-12 12:58:54 +02:00
|
|
|
elif dir_name in added_mpims:
|
2019-07-02 07:41:51 +02:00
|
|
|
message['mpim_name'] = dir_name
|
2019-08-12 12:58:54 +02:00
|
|
|
elif dir_name in dm_members:
|
2019-06-28 10:34:14 +02:00
|
|
|
message['pm_name'] = dir_name
|
2020-01-26 07:47:21 +01:00
|
|
|
messages.append(message)
|
2018-12-05 04:13:47 +01:00
|
|
|
messages_for_one_day += messages
|
|
|
|
|
|
|
|
# we sort the messages according to the timestamp to show messages with
|
|
|
|
# the proper date order
|
2020-04-09 21:51:58 +02:00
|
|
|
yield from sorted(messages_for_one_day, key=lambda m: m['ts'])
|
2018-02-25 07:08:28 +01:00
|
|
|
|
2018-10-28 16:48:52 +01:00
|
|
|
def channel_message_to_zerver_message(realm_id: int,
|
|
|
|
users: List[ZerverFieldsT],
|
2019-08-12 13:44:07 +02:00
|
|
|
slack_user_id_to_zulip_user_id: SlackToZulipUserIDT,
|
2019-08-13 10:02:03 +02:00
|
|
|
slack_recipient_name_to_zulip_recipient_id: SlackToZulipRecipientT,
|
2018-02-25 07:08:28 +01:00
|
|
|
all_messages: List[ZerverFieldsT],
|
2018-03-30 12:38:03 +02:00
|
|
|
zerver_realmemoji: List[ZerverFieldsT],
|
2018-10-25 17:33:52 +02:00
|
|
|
subscriber_map: Dict[int, Set[int]],
|
2018-04-07 00:00:05 +02:00
|
|
|
added_channels: AddedChannelsT,
|
2019-06-28 10:34:14 +02:00
|
|
|
dm_members: DMMembersT,
|
2018-12-17 02:26:51 +01:00
|
|
|
domain_name: str,
|
|
|
|
long_term_idle: Set[int]) -> Tuple[List[ZerverFieldsT],
|
|
|
|
List[ZerverFieldsT],
|
|
|
|
List[ZerverFieldsT],
|
|
|
|
List[ZerverFieldsT],
|
|
|
|
List[ZerverFieldsT]]:
|
2017-12-29 10:57:48 +01:00
|
|
|
"""
|
|
|
|
Returns:
|
|
|
|
1. zerver_message, which is a list of the messages
|
|
|
|
2. zerver_usermessage, which is a list of the usermessages
|
2018-02-26 10:16:34 +01:00
|
|
|
3. zerver_attachment, which is a list of the attachments
|
|
|
|
4. uploads_list, which is a list of uploads to be mapped in uploads records.json
|
2018-03-25 13:42:04 +02:00
|
|
|
5. reaction_list, which is a list of all user reactions
|
2017-12-29 10:57:48 +01:00
|
|
|
"""
|
|
|
|
zerver_message = []
|
2018-01-23 19:04:59 +01:00
|
|
|
zerver_usermessage = [] # type: List[ZerverFieldsT]
|
2018-02-26 06:57:00 +01:00
|
|
|
uploads_list = [] # type: List[ZerverFieldsT]
|
2018-02-26 08:48:14 +01:00
|
|
|
zerver_attachment = [] # type: List[ZerverFieldsT]
|
2018-03-25 13:42:04 +02:00
|
|
|
reaction_list = [] # type: List[ZerverFieldsT]
|
|
|
|
|
2018-12-17 02:26:51 +01:00
|
|
|
total_user_messages = 0
|
|
|
|
total_skipped_user_messages = 0
|
2018-02-25 09:54:53 +01:00
|
|
|
for message in all_messages:
|
2019-08-12 13:44:07 +02:00
|
|
|
slack_user_id = get_message_sending_user(message)
|
|
|
|
if not slack_user_id:
|
|
|
|
# Ignore messages without slack_user_id
|
2018-02-25 09:54:53 +01:00
|
|
|
# These are Sometimes produced by slack
|
|
|
|
continue
|
2018-08-10 21:55:14 +02:00
|
|
|
|
|
|
|
subtype = message.get('subtype', False)
|
|
|
|
if subtype in [
|
2018-04-18 20:36:31 +02:00
|
|
|
# Zulip doesn't have a pinned_item concept
|
|
|
|
"pinned_item",
|
|
|
|
"unpinned_item",
|
|
|
|
# Slack's channel join/leave notices are spammy
|
2018-04-18 20:35:59 +02:00
|
|
|
"channel_join",
|
|
|
|
"channel_leave",
|
|
|
|
"channel_name"
|
|
|
|
]:
|
|
|
|
continue
|
2018-02-25 09:54:53 +01:00
|
|
|
|
2018-04-18 20:29:40 +02:00
|
|
|
try:
|
2018-10-25 17:33:52 +02:00
|
|
|
content, mentioned_user_ids, has_link = convert_to_zulip_markdown(
|
2019-08-12 13:44:07 +02:00
|
|
|
message['text'], users, added_channels, slack_user_id_to_zulip_user_id)
|
2018-04-18 20:29:40 +02:00
|
|
|
except Exception:
|
|
|
|
print("Slack message unexpectedly missing text representation:")
|
2018-10-17 01:14:09 +02:00
|
|
|
print(ujson.dumps(message, indent=4))
|
2018-04-18 20:29:40 +02:00
|
|
|
continue
|
2018-02-25 09:54:53 +01:00
|
|
|
rendered_content = None
|
2018-02-26 08:48:14 +01:00
|
|
|
|
2019-07-02 07:41:51 +02:00
|
|
|
if "channel_name" in message:
|
|
|
|
is_private = False
|
2019-08-13 10:02:03 +02:00
|
|
|
recipient_id = slack_recipient_name_to_zulip_recipient_id[message['channel_name']]
|
2019-07-02 07:41:51 +02:00
|
|
|
elif "mpim_name" in message:
|
|
|
|
is_private = True
|
2019-08-13 10:02:03 +02:00
|
|
|
recipient_id = slack_recipient_name_to_zulip_recipient_id[message['mpim_name']]
|
2019-06-28 10:34:14 +02:00
|
|
|
elif "pm_name" in message:
|
|
|
|
is_private = True
|
|
|
|
sender = get_message_sending_user(message)
|
|
|
|
members = dm_members[message['pm_name']]
|
|
|
|
if sender == members[0]:
|
2019-08-13 10:02:03 +02:00
|
|
|
recipient_id = slack_recipient_name_to_zulip_recipient_id[members[1]]
|
|
|
|
sender_recipient_id = slack_recipient_name_to_zulip_recipient_id[members[0]]
|
2019-06-28 10:34:14 +02:00
|
|
|
else:
|
2019-08-13 10:02:03 +02:00
|
|
|
recipient_id = slack_recipient_name_to_zulip_recipient_id[members[0]]
|
|
|
|
sender_recipient_id = slack_recipient_name_to_zulip_recipient_id[members[1]]
|
2019-07-02 07:41:51 +02:00
|
|
|
|
2018-10-28 18:55:32 +01:00
|
|
|
message_id = NEXT_ID('message')
|
2018-02-26 08:48:14 +01:00
|
|
|
|
2018-03-25 13:42:04 +02:00
|
|
|
if 'reactions' in message.keys():
|
2019-08-12 13:44:07 +02:00
|
|
|
build_reactions(reaction_list, message['reactions'], slack_user_id_to_zulip_user_id,
|
2020-02-06 07:07:10 +01:00
|
|
|
message_id, zerver_realmemoji)
|
2018-03-25 13:42:04 +02:00
|
|
|
|
2018-03-20 19:26:35 +01:00
|
|
|
# Process different subtypes of slack messages
|
|
|
|
|
2018-08-10 21:55:14 +02:00
|
|
|
# Subtypes which have only the action in the message should
|
|
|
|
# be rendered with '/me' in the content initially
|
|
|
|
# For example "sh_room_created" has the message 'started a call'
|
|
|
|
# which should be displayed as '/me started a call'
|
|
|
|
if subtype in ["bot_add", "sh_room_created", "me_message"]:
|
2019-04-20 01:00:46 +02:00
|
|
|
content = '/me %s' % (content,)
|
2018-11-06 22:14:23 +01:00
|
|
|
if subtype == 'file_comment':
|
|
|
|
# The file_comment message type only indicates the
|
|
|
|
# responsible user in a subfield.
|
|
|
|
message['user'] = message['comment']['user']
|
2018-08-10 21:55:14 +02:00
|
|
|
|
2018-10-28 16:48:52 +01:00
|
|
|
file_info = process_message_files(
|
|
|
|
message=message,
|
|
|
|
domain_name=domain_name,
|
|
|
|
realm_id=realm_id,
|
|
|
|
message_id=message_id,
|
2019-08-12 13:44:07 +02:00
|
|
|
slack_user_id=slack_user_id,
|
2018-10-28 16:48:52 +01:00
|
|
|
users=users,
|
2019-08-12 13:44:07 +02:00
|
|
|
slack_user_id_to_zulip_user_id=slack_user_id_to_zulip_user_id,
|
2018-10-28 16:48:52 +01:00
|
|
|
zerver_attachment=zerver_attachment,
|
|
|
|
uploads_list=uploads_list,
|
|
|
|
)
|
2018-02-26 06:57:00 +01:00
|
|
|
|
2018-10-28 16:48:52 +01:00
|
|
|
content += file_info['content']
|
|
|
|
has_link = has_link or file_info['has_link']
|
2018-02-26 06:57:00 +01:00
|
|
|
|
2018-10-28 16:48:52 +01:00
|
|
|
has_attachment = file_info['has_attachment']
|
|
|
|
has_image = file_info['has_image']
|
2018-03-15 14:12:38 +01:00
|
|
|
|
2018-11-10 17:10:45 +01:00
|
|
|
topic_name = 'imported from slack'
|
2018-08-17 23:47:37 +02:00
|
|
|
|
2018-11-10 17:10:45 +01:00
|
|
|
zulip_message = build_message(topic_name, float(message['ts']), message_id, content,
|
2019-08-12 13:44:07 +02:00
|
|
|
rendered_content, slack_user_id_to_zulip_user_id[slack_user_id],
|
|
|
|
recipient_id, has_image, has_link, has_attachment)
|
2018-08-17 23:47:37 +02:00
|
|
|
zerver_message.append(zulip_message)
|
2018-02-25 09:54:53 +01:00
|
|
|
|
2018-12-17 02:26:51 +01:00
|
|
|
(num_created, num_skipped) = build_usermessages(
|
2018-10-25 17:33:52 +02:00
|
|
|
zerver_usermessage=zerver_usermessage,
|
|
|
|
subscriber_map=subscriber_map,
|
|
|
|
recipient_id=recipient_id,
|
|
|
|
mentioned_user_ids=mentioned_user_ids,
|
|
|
|
message_id=message_id,
|
2019-07-02 07:41:51 +02:00
|
|
|
is_private=is_private,
|
2018-12-17 02:26:51 +01:00
|
|
|
long_term_idle=long_term_idle,
|
2018-10-25 17:33:52 +02:00
|
|
|
)
|
2018-12-17 02:26:51 +01:00
|
|
|
total_user_messages += num_created
|
|
|
|
total_skipped_user_messages += num_skipped
|
2018-02-25 09:54:53 +01:00
|
|
|
|
2019-06-28 10:34:14 +02:00
|
|
|
if "pm_name" in message and recipient_id != sender_recipient_id:
|
|
|
|
(num_created, num_skipped) = build_usermessages(
|
|
|
|
zerver_usermessage=zerver_usermessage,
|
|
|
|
subscriber_map=subscriber_map,
|
|
|
|
recipient_id=sender_recipient_id,
|
|
|
|
mentioned_user_ids=mentioned_user_ids,
|
|
|
|
message_id=message_id,
|
|
|
|
is_private=is_private,
|
|
|
|
long_term_idle=long_term_idle,
|
|
|
|
)
|
|
|
|
total_user_messages += num_created
|
|
|
|
total_skipped_user_messages += num_skipped
|
|
|
|
|
2018-12-17 02:26:51 +01:00
|
|
|
logging.debug("Created %s UserMessages; deferred %s due to long-term idle" % (
|
|
|
|
total_user_messages, total_skipped_user_messages))
|
2018-06-17 19:10:54 +02:00
|
|
|
return zerver_message, zerver_usermessage, zerver_attachment, uploads_list, \
|
2018-10-28 18:55:32 +01:00
|
|
|
reaction_list
|
2017-12-29 10:57:48 +01:00
|
|
|
|
2018-10-28 16:48:52 +01:00
|
|
|
def process_message_files(message: ZerverFieldsT,
|
|
|
|
domain_name: str,
|
|
|
|
realm_id: int,
|
|
|
|
message_id: int,
|
2019-08-12 13:44:07 +02:00
|
|
|
slack_user_id: str,
|
2018-10-28 16:48:52 +01:00
|
|
|
users: List[ZerverFieldsT],
|
2019-08-12 13:44:07 +02:00
|
|
|
slack_user_id_to_zulip_user_id: SlackToZulipUserIDT,
|
2018-10-28 16:48:52 +01:00
|
|
|
zerver_attachment: List[ZerverFieldsT],
|
|
|
|
uploads_list: List[ZerverFieldsT]) -> Dict[str, Any]:
|
|
|
|
has_attachment = False
|
|
|
|
has_image = False
|
|
|
|
has_link = False
|
|
|
|
|
|
|
|
files = message.get('files', [])
|
|
|
|
|
|
|
|
subtype = message.get('subtype')
|
|
|
|
|
|
|
|
if subtype == 'file_share':
|
|
|
|
# In Slack messages, uploads can either have the subtype as 'file_share' or
|
|
|
|
# have the upload information in 'files' keyword
|
|
|
|
files = [message['file']]
|
|
|
|
|
|
|
|
markdown_links = []
|
|
|
|
|
|
|
|
for fileinfo in files:
|
2019-05-30 12:47:12 +02:00
|
|
|
if fileinfo.get('mode', '') in ['tombstone', 'hidden_by_limit']:
|
2019-03-13 20:42:42 +01:00
|
|
|
# Slack sometimes includes tombstone mode files with no
|
|
|
|
# real data on the actual file (presumably in cases where
|
2019-05-30 12:47:12 +02:00
|
|
|
# the file was deleted). hidden_by_limit mode is for files
|
|
|
|
# that are hidden because of 10k cap in free plan.
|
2019-03-13 20:42:42 +01:00
|
|
|
continue
|
|
|
|
|
2018-10-28 16:48:52 +01:00
|
|
|
url = fileinfo['url_private']
|
|
|
|
|
|
|
|
if 'files.slack.com' in url:
|
|
|
|
# For attachments with slack download link
|
|
|
|
has_attachment = True
|
|
|
|
has_link = True
|
|
|
|
has_image = True if 'image' in fileinfo['mimetype'] else False
|
|
|
|
|
|
|
|
file_user = [iterate_user for iterate_user in users if message['user'] == iterate_user['id']]
|
|
|
|
file_user_email = get_user_email(file_user[0], domain_name)
|
|
|
|
|
|
|
|
s3_path, content_for_link = get_attachment_path_and_content(fileinfo, realm_id)
|
|
|
|
markdown_links.append(content_for_link)
|
|
|
|
|
2019-08-12 13:44:07 +02:00
|
|
|
build_uploads(slack_user_id_to_zulip_user_id[slack_user_id], realm_id, file_user_email,
|
|
|
|
fileinfo, s3_path, uploads_list)
|
2018-10-28 16:48:52 +01:00
|
|
|
|
2019-08-12 13:44:07 +02:00
|
|
|
build_attachment(realm_id, {message_id}, slack_user_id_to_zulip_user_id[slack_user_id],
|
2018-10-28 16:48:52 +01:00
|
|
|
fileinfo, s3_path, zerver_attachment)
|
|
|
|
else:
|
|
|
|
# For attachments with link not from slack
|
|
|
|
# Example: Google drive integration
|
|
|
|
has_link = True
|
|
|
|
if 'title' in fileinfo:
|
|
|
|
file_name = fileinfo['title']
|
|
|
|
else:
|
|
|
|
file_name = fileinfo['name']
|
|
|
|
markdown_links.append('[%s](%s)' % (file_name, fileinfo['url_private']))
|
|
|
|
|
|
|
|
content = '\n'.join(markdown_links)
|
|
|
|
|
|
|
|
return dict(
|
|
|
|
content=content,
|
|
|
|
has_attachment=has_attachment,
|
|
|
|
has_image=has_image,
|
|
|
|
has_link=has_link,
|
|
|
|
)
|
|
|
|
|
2018-02-26 06:57:00 +01:00
|
|
|
def get_attachment_path_and_content(fileinfo: ZerverFieldsT, realm_id: int) -> Tuple[str,
|
|
|
|
str]:
|
|
|
|
# Should be kept in sync with its equivalent in zerver/lib/uploads in the function
|
2018-03-28 18:14:17 +02:00
|
|
|
# 'upload_message_file'
|
2018-02-26 06:57:00 +01:00
|
|
|
s3_path = "/".join([
|
|
|
|
str(realm_id),
|
2018-04-03 20:52:16 +02:00
|
|
|
'SlackImportAttachment', # This is a special placeholder which should be kept
|
|
|
|
# in sync with 'exports.py' function 'import_message_data'
|
2018-02-26 06:57:00 +01:00
|
|
|
format(random.randint(0, 255), 'x'),
|
|
|
|
random_name(18),
|
|
|
|
sanitize_name(fileinfo['name'])
|
|
|
|
])
|
2019-04-20 01:00:46 +02:00
|
|
|
attachment_path = '/user_uploads/%s' % (s3_path,)
|
2018-03-13 00:33:42 +01:00
|
|
|
content = '[%s](%s)' % (fileinfo['title'], attachment_path)
|
2018-02-26 06:57:00 +01:00
|
|
|
|
|
|
|
return s3_path, content
|
|
|
|
|
2018-03-25 13:42:04 +02:00
|
|
|
def build_reactions(reaction_list: List[ZerverFieldsT], reactions: List[ZerverFieldsT],
|
2019-08-12 13:44:07 +02:00
|
|
|
slack_user_id_to_zulip_user_id: SlackToZulipUserIDT, message_id: int,
|
2018-10-28 18:55:32 +01:00
|
|
|
zerver_realmemoji: List[ZerverFieldsT]) -> None:
|
2018-03-30 12:38:03 +02:00
|
|
|
realmemoji = {}
|
|
|
|
for realm_emoji in zerver_realmemoji:
|
|
|
|
realmemoji[realm_emoji['name']] = realm_emoji['id']
|
|
|
|
|
2018-03-25 13:42:04 +02:00
|
|
|
# For the unicode emoji codes, we use equivalent of
|
|
|
|
# function 'emoji_name_to_emoji_code' in 'zerver/lib/emoji' here
|
|
|
|
for slack_reaction in reactions:
|
|
|
|
emoji_name = slack_reaction['name']
|
|
|
|
if emoji_name in name_to_codepoint:
|
2018-03-30 12:38:03 +02:00
|
|
|
emoji_code = name_to_codepoint[emoji_name]
|
|
|
|
reaction_type = Reaction.UNICODE_EMOJI
|
|
|
|
elif emoji_name in realmemoji:
|
|
|
|
emoji_code = realmemoji[emoji_name]
|
|
|
|
reaction_type = Reaction.REALM_EMOJI
|
2018-03-25 13:42:04 +02:00
|
|
|
else:
|
|
|
|
continue
|
2018-03-30 12:38:03 +02:00
|
|
|
|
2019-08-12 13:44:07 +02:00
|
|
|
for slack_user_id in slack_reaction['users']:
|
2018-10-28 18:55:32 +01:00
|
|
|
reaction_id = NEXT_ID('reaction')
|
2018-08-17 20:35:00 +02:00
|
|
|
reaction = Reaction(
|
2018-03-30 12:38:03 +02:00
|
|
|
id=reaction_id,
|
|
|
|
emoji_code=emoji_code,
|
|
|
|
emoji_name=emoji_name,
|
2018-08-17 20:35:00 +02:00
|
|
|
reaction_type=reaction_type)
|
|
|
|
|
|
|
|
reaction_dict = model_to_dict(reaction,
|
|
|
|
exclude=['message', 'user_profile'])
|
|
|
|
reaction_dict['message'] = message_id
|
2019-08-12 13:44:07 +02:00
|
|
|
reaction_dict['user_profile'] = slack_user_id_to_zulip_user_id[slack_user_id]
|
2018-08-17 20:35:00 +02:00
|
|
|
|
|
|
|
reaction_list.append(reaction_dict)
|
2018-03-25 13:42:04 +02:00
|
|
|
|
2018-02-26 06:57:00 +01:00
|
|
|
def build_uploads(user_id: int, realm_id: int, email: str, fileinfo: ZerverFieldsT, s3_path: str,
|
|
|
|
uploads_list: List[ZerverFieldsT]) -> None:
|
|
|
|
upload = dict(
|
|
|
|
path=fileinfo['url_private'], # Save slack's url here, which is used later while processing
|
|
|
|
realm_id=realm_id,
|
|
|
|
content_type=None,
|
|
|
|
user_profile_id=user_id,
|
|
|
|
last_modified=fileinfo['timestamp'],
|
|
|
|
user_profile_email=email,
|
|
|
|
s3_path=s3_path,
|
|
|
|
size=fileinfo['size'])
|
|
|
|
uploads_list.append(upload)
|
|
|
|
|
2018-04-25 19:18:32 +02:00
|
|
|
def get_message_sending_user(message: ZerverFieldsT) -> Optional[str]:
|
|
|
|
if 'user' in message:
|
|
|
|
return message['user']
|
|
|
|
if message.get('file'):
|
|
|
|
return message['file'].get('user')
|
|
|
|
return None
|
2018-01-23 19:04:59 +01:00
|
|
|
|
2019-08-08 19:39:26 +02:00
|
|
|
def fetch_shared_channel_users(user_list: List[ZerverFieldsT], slack_data_dir: str, token: str) -> None:
|
|
|
|
normal_user_ids = set()
|
|
|
|
mirror_dummy_user_ids = set()
|
|
|
|
added_channels = {}
|
|
|
|
team_id_to_domain = {} # type: Dict[str, str]
|
|
|
|
for user in user_list:
|
|
|
|
user["is_mirror_dummy"] = False
|
|
|
|
normal_user_ids.add(user["id"])
|
|
|
|
|
|
|
|
public_channels = get_data_file(slack_data_dir + '/channels.json')
|
|
|
|
try:
|
|
|
|
private_channels = get_data_file(slack_data_dir + '/groups.json')
|
|
|
|
except FileNotFoundError:
|
|
|
|
private_channels = []
|
|
|
|
for channel in public_channels + private_channels:
|
|
|
|
added_channels[channel["name"]] = True
|
|
|
|
for user_id in channel["members"]:
|
|
|
|
if user_id not in normal_user_ids:
|
|
|
|
mirror_dummy_user_ids.add(user_id)
|
|
|
|
|
|
|
|
all_messages = get_messages_iterator(slack_data_dir, added_channels, {}, {})
|
|
|
|
for message in all_messages:
|
|
|
|
user_id = get_message_sending_user(message)
|
|
|
|
if user_id is None or user_id in normal_user_ids:
|
|
|
|
continue
|
|
|
|
mirror_dummy_user_ids.add(user_id)
|
|
|
|
|
|
|
|
# Fetch data on the mirror_dummy_user_ids from the Slack API (it's
|
|
|
|
# not included in the data export file).
|
|
|
|
for user_id in mirror_dummy_user_ids:
|
|
|
|
user = get_slack_api_data("https://slack.com/api/users.info", "user", token=token, user=user_id)
|
|
|
|
team_id = user["team_id"]
|
|
|
|
if team_id not in team_id_to_domain:
|
|
|
|
team = get_slack_api_data("https://slack.com/api/team.info", "team", token=token, team=team_id)
|
|
|
|
team_id_to_domain[team_id] = team["domain"]
|
|
|
|
user["team_domain"] = team_id_to_domain[team_id]
|
|
|
|
user["is_mirror_dummy"] = True
|
|
|
|
user_list.append(user)
|
|
|
|
|
2020-01-24 15:00:18 +01:00
|
|
|
def fetch_team_icons(zerver_realm: Dict[str, Any], team_info_dict: Dict[str, Any],
|
|
|
|
output_dir: str) -> List[Dict[str, Any]]:
|
|
|
|
records = []
|
|
|
|
|
|
|
|
team_icons_dict = team_info_dict["icon"]
|
|
|
|
if "image_default" in team_icons_dict and team_icons_dict["image_default"]:
|
|
|
|
return []
|
|
|
|
|
|
|
|
icon_url = team_icons_dict.get("image_original", None) or team_icons_dict.get("image_230", None) or \
|
|
|
|
team_icons_dict.get("image_132", None) or team_icons_dict.get("image_102", None)
|
|
|
|
if icon_url is None:
|
|
|
|
return []
|
|
|
|
|
|
|
|
response = requests.get(icon_url, stream=True)
|
|
|
|
response_raw = response.raw
|
|
|
|
|
|
|
|
realm_id = zerver_realm["id"]
|
|
|
|
os.makedirs(os.path.join(output_dir, str(realm_id)), exist_ok=True)
|
|
|
|
|
|
|
|
original_icon_output_path = os.path.join(output_dir, str(realm_id), "icon.original")
|
|
|
|
with open(original_icon_output_path, 'wb') as output_file:
|
|
|
|
shutil.copyfileobj(response_raw, output_file)
|
|
|
|
records.append({
|
|
|
|
"realm_id": realm_id,
|
|
|
|
"path": os.path.join(str(realm_id), "icon.original"),
|
|
|
|
"s3_path": os.path.join(str(realm_id), "icon.original"),
|
|
|
|
})
|
|
|
|
|
|
|
|
resized_icon_output_path = os.path.join(output_dir, str(realm_id), "icon.png")
|
|
|
|
with open(resized_icon_output_path, 'wb') as output_file:
|
|
|
|
with open(original_icon_output_path, 'rb') as original_file:
|
|
|
|
resized_data = resize_logo(original_file.read())
|
|
|
|
output_file.write(resized_data)
|
|
|
|
records.append({
|
|
|
|
"realm_id": realm_id,
|
|
|
|
"path": os.path.join(str(realm_id), "icon.png"),
|
|
|
|
"s3_path": os.path.join(str(realm_id), "icon.png"),
|
|
|
|
})
|
|
|
|
|
|
|
|
zerver_realm["icon_source"] = Realm.ICON_UPLOADED
|
|
|
|
|
|
|
|
return records
|
|
|
|
|
2018-04-15 16:21:02 +02:00
|
|
|
def do_convert_data(slack_zip_file: str, output_dir: str, token: str, threads: int=6) -> None:
|
2018-04-06 02:21:16 +02:00
|
|
|
# Subdomain is set by the user while running the import command
|
|
|
|
realm_subdomain = ""
|
2018-04-18 19:10:17 +02:00
|
|
|
realm_id = 0
|
2018-02-25 14:46:34 +01:00
|
|
|
domain_name = settings.EXTERNAL_HOST
|
|
|
|
|
2018-01-06 19:42:18 +01:00
|
|
|
slack_data_dir = slack_zip_file.replace('.zip', '')
|
2018-01-20 14:49:40 +01:00
|
|
|
if not os.path.exists(slack_data_dir):
|
|
|
|
os.makedirs(slack_data_dir)
|
2018-02-08 21:38:14 +01:00
|
|
|
|
|
|
|
os.makedirs(output_dir, exist_ok=True)
|
|
|
|
if os.listdir(output_dir):
|
|
|
|
raise Exception('Output directory should be empty!')
|
|
|
|
|
2018-01-20 14:49:40 +01:00
|
|
|
subprocess.check_call(['unzip', '-q', slack_zip_file, '-d', slack_data_dir])
|
2017-12-08 11:27:42 +01:00
|
|
|
|
2018-03-29 14:38:11 +02:00
|
|
|
# We get the user data from the legacy token method of slack api, which is depreciated
|
|
|
|
# but we use it as the user email data is provided only in this method
|
2019-08-07 13:58:16 +02:00
|
|
|
user_list = get_slack_api_data("https://slack.com/api/users.list", "members", token=token)
|
2019-08-08 19:39:26 +02:00
|
|
|
fetch_shared_channel_users(user_list, slack_data_dir, token)
|
2019-08-07 13:58:16 +02:00
|
|
|
|
|
|
|
custom_emoji_list = get_slack_api_data("https://slack.com/api/emoji.list", "emoji", token=token)
|
2018-03-29 14:38:11 +02:00
|
|
|
|
2019-08-13 10:02:03 +02:00
|
|
|
realm, slack_user_id_to_zulip_user_id, slack_recipient_name_to_zulip_recipient_id, \
|
|
|
|
added_channels, added_mpims, dm_members, avatar_list, \
|
|
|
|
emoji_url_map = slack_workspace_to_realm(domain_name, realm_id, user_list,
|
|
|
|
realm_subdomain, slack_data_dir,
|
|
|
|
custom_emoji_list)
|
2018-02-17 00:42:59 +01:00
|
|
|
|
2018-06-13 20:15:35 +02:00
|
|
|
reactions, uploads_list, zerver_attachment = convert_slack_workspace_messages(
|
2019-08-12 13:44:07 +02:00
|
|
|
slack_data_dir, user_list, realm_id, slack_user_id_to_zulip_user_id,
|
2019-08-13 10:02:03 +02:00
|
|
|
slack_recipient_name_to_zulip_recipient_id, added_channels, added_mpims, dm_members, realm,
|
2019-08-12 13:44:07 +02:00
|
|
|
realm['zerver_userprofile'], realm['zerver_realmemoji'], domain_name, output_dir)
|
2017-12-08 11:27:42 +01:00
|
|
|
|
2018-05-24 13:56:15 +02:00
|
|
|
# Move zerver_reactions to realm.json file
|
2018-06-13 20:15:35 +02:00
|
|
|
realm['zerver_reaction'] = reactions
|
2018-05-24 13:56:15 +02:00
|
|
|
|
2018-03-30 00:09:29 +02:00
|
|
|
emoji_folder = os.path.join(output_dir, 'emoji')
|
|
|
|
os.makedirs(emoji_folder, exist_ok=True)
|
2018-04-15 16:21:02 +02:00
|
|
|
emoji_records = process_emojis(realm['zerver_realmemoji'], emoji_folder, emoji_url_map, threads)
|
2018-03-30 00:09:29 +02:00
|
|
|
|
2018-02-17 01:46:50 +01:00
|
|
|
avatar_folder = os.path.join(output_dir, 'avatars')
|
2018-03-07 14:14:08 +01:00
|
|
|
avatar_realm_folder = os.path.join(avatar_folder, str(realm_id))
|
2018-02-17 01:46:50 +01:00
|
|
|
os.makedirs(avatar_realm_folder, exist_ok=True)
|
2018-08-02 15:24:34 +02:00
|
|
|
avatar_records = process_avatars(avatar_list, avatar_folder, realm_id, threads, size_url_suffix='-512')
|
2018-02-17 01:46:50 +01:00
|
|
|
|
2018-02-26 11:04:13 +01:00
|
|
|
uploads_folder = os.path.join(output_dir, 'uploads')
|
|
|
|
os.makedirs(os.path.join(uploads_folder, str(realm_id)), exist_ok=True)
|
2018-04-15 16:21:02 +02:00
|
|
|
uploads_records = process_uploads(uploads_list, uploads_folder, threads)
|
2018-01-23 19:04:59 +01:00
|
|
|
attachment = {"zerver_attachment": zerver_attachment}
|
2017-12-08 11:27:42 +01:00
|
|
|
|
2020-01-24 15:00:18 +01:00
|
|
|
team_info_dict = get_slack_api_data("https://slack.com/api/team.info", "team", token=token)
|
|
|
|
realm_icons_folder = os.path.join(output_dir, 'realm_icons')
|
|
|
|
realm_icon_records = fetch_team_icons(realm["zerver_realm"][0], team_info_dict, realm_icons_folder)
|
|
|
|
|
2018-03-09 17:50:48 +01:00
|
|
|
create_converted_data_files(realm, output_dir, '/realm.json')
|
2018-03-30 00:09:29 +02:00
|
|
|
create_converted_data_files(emoji_records, output_dir, '/emoji/records.json')
|
2018-03-09 17:50:48 +01:00
|
|
|
create_converted_data_files(avatar_records, output_dir, '/avatars/records.json')
|
|
|
|
create_converted_data_files(uploads_records, output_dir, '/uploads/records.json')
|
|
|
|
create_converted_data_files(attachment, output_dir, '/attachment.json')
|
2020-01-24 15:00:18 +01:00
|
|
|
create_converted_data_files(realm_icon_records, output_dir, '/realm_icons/records.json')
|
2017-12-08 11:27:42 +01:00
|
|
|
|
2018-01-06 19:42:18 +01:00
|
|
|
rm_tree(slack_data_dir)
|
2017-12-08 11:27:42 +01:00
|
|
|
subprocess.check_call(["tar", "-czf", output_dir + '.tar.gz', output_dir, '-P'])
|
|
|
|
|
2018-02-08 00:06:02 +01:00
|
|
|
logging.info('######### DATA CONVERSION FINISHED #########\n')
|
2019-04-20 01:00:46 +02:00
|
|
|
logging.info("Zulip data dump created at %s" % (output_dir,))
|
2018-01-23 19:04:59 +01:00
|
|
|
|
|
|
|
def get_data_file(path: str) -> Any:
|
2020-04-09 21:51:58 +02:00
|
|
|
with open(path) as fp:
|
2018-10-17 01:14:09 +02:00
|
|
|
data = ujson.load(fp)
|
|
|
|
return data
|
2018-01-23 19:04:59 +01:00
|
|
|
|
2019-08-07 13:58:16 +02:00
|
|
|
def get_slack_api_data(slack_api_url: str, get_param: str, **kwargs: Any) -> Any:
|
|
|
|
if not kwargs.get("token"):
|
2020-01-22 23:46:39 +01:00
|
|
|
raise AssertionError("Slack token missing in kwargs")
|
|
|
|
token = kwargs["token"]
|
|
|
|
if not token.startswith("xoxp-"):
|
|
|
|
raise Exception('Invalid Slack legacy token.\n'
|
|
|
|
' You must pass a Slack "legacy token" starting with "xoxp-".\n'
|
|
|
|
' Create one at https://api.slack.com/custom-integrations/legacy-tokens')
|
|
|
|
|
|
|
|
data = requests.get("{}?{}".format(slack_api_url, urlencode(kwargs)))
|
2019-08-07 13:58:16 +02:00
|
|
|
|
2018-03-29 14:38:11 +02:00
|
|
|
if data.status_code == requests.codes.ok:
|
2020-01-22 23:46:39 +01:00
|
|
|
result = data.json()
|
|
|
|
if not result['ok']:
|
|
|
|
raise Exception('Error accessing Slack API: %s' % (result['error'],))
|
|
|
|
return result[get_param]
|
|
|
|
|
|
|
|
raise Exception('HTTP error accessing the Slack API.')
|