mackuba.eu
+1
-1
Gemfile
+1
-1
Gemfile
+39
-34
Gemfile.lock
+39
-34
Gemfile.lock
···
7
7
GEM
8
8
remote: https://rubygems.org/
9
9
specs:
10
-
activemodel (7.2.2.2)
11
-
activesupport (= 7.2.2.2)
12
-
activerecord (7.2.2.2)
13
-
activemodel (= 7.2.2.2)
14
-
activesupport (= 7.2.2.2)
10
+
activemodel (7.2.3)
11
+
activesupport (= 7.2.3)
12
+
activerecord (7.2.3)
13
+
activemodel (= 7.2.3)
14
+
activesupport (= 7.2.3)
15
15
timeout (>= 0.4.0)
16
-
activesupport (7.2.2.2)
16
+
activesupport (7.2.3)
17
17
base64
18
18
benchmark (>= 0.3)
19
19
bigdecimal
···
29
29
base58 (0.2.3)
30
30
base64 (0.3.0)
31
31
bcrypt_pbkdf (1.1.1)
32
-
benchmark (0.4.1)
33
-
bigdecimal (3.2.2)
32
+
benchmark (0.5.0)
33
+
bigdecimal (3.3.1)
34
34
capistrano (2.15.11)
35
35
highline
36
36
net-scp (>= 1.0.0)
···
39
39
net-ssh-gateway (>= 1.1.0)
40
40
cbor (0.5.10.1)
41
41
concurrent-ruby (1.3.5)
42
-
connection_pool (2.5.3)
43
-
date (3.4.1)
42
+
connection_pool (2.5.4)
43
+
daemons (1.4.1)
44
+
date (3.5.0)
44
45
dotenv (3.1.8)
45
46
drb (2.2.3)
46
47
ed25519 (1.4.0)
47
-
erb (5.0.2)
48
+
erb (6.0.0)
48
49
eventmachine (1.2.7)
49
50
faye-websocket (0.12.0)
50
51
eventmachine (>= 0.12.0)
···
54
55
i18n (1.14.7)
55
56
concurrent-ruby (~> 1.0)
56
57
io-console (0.8.1)
57
-
irb (1.15.2)
58
+
irb (1.15.3)
58
59
pp (>= 0.6.0)
59
60
rdoc (>= 4.0.0)
60
61
reline (>= 0.4.2)
···
63
64
logger (1.7.0)
64
65
minisky (0.5.0)
65
66
base64 (~> 0.1)
66
-
minitest (5.25.5)
67
+
minitest (5.26.1)
67
68
mustermann (3.0.4)
68
69
ruby2_keywords (~> 0.0.1)
69
70
net-scp (4.1.0)
···
73
74
net-ssh (7.3.0)
74
75
net-ssh-gateway (2.0.0)
75
76
net-ssh (>= 4.0.0)
76
-
nio4r (2.7.4)
77
-
pg (1.6.1)
78
-
pg (1.6.1-aarch64-linux)
79
-
pg (1.6.1-aarch64-linux-musl)
80
-
pg (1.6.1-arm64-darwin)
81
-
pg (1.6.1-x86_64-darwin)
82
-
pg (1.6.1-x86_64-linux)
83
-
pg (1.6.1-x86_64-linux-musl)
84
-
pp (0.6.2)
77
+
pg (1.6.2)
78
+
pg (1.6.2-aarch64-linux)
79
+
pg (1.6.2-aarch64-linux-musl)
80
+
pg (1.6.2-arm64-darwin)
81
+
pg (1.6.2-x86_64-darwin)
82
+
pg (1.6.2-x86_64-linux)
83
+
pg (1.6.2-x86_64-linux-musl)
84
+
pp (0.6.3)
85
85
prettyprint
86
86
prettyprint (0.2.0)
87
87
psych (5.2.6)
88
88
date
89
89
stringio
90
-
puma (6.6.1)
91
-
nio4r (~> 2.0)
92
-
rack (3.2.0)
93
-
rack-protection (4.1.1)
90
+
rack (3.2.4)
91
+
rack-protection (4.2.1)
94
92
base64 (>= 0.1.0)
95
93
logger (>= 1.6.0)
96
94
rack (>= 3.0.0, < 4)
···
100
98
rackup (2.2.1)
101
99
rack (>= 3)
102
100
rainbow (3.1.1)
103
-
rake (13.3.0)
104
-
rdoc (6.14.2)
101
+
rake (13.3.1)
102
+
rdoc (6.15.1)
105
103
erb
106
104
psych (>= 4.0.0)
107
-
reline (0.6.2)
105
+
tsort
106
+
reline (0.6.3)
108
107
io-console (~> 0.5)
109
108
ruby2_keywords (0.0.5)
110
109
securerandom (0.4.1)
111
-
sinatra (4.1.1)
110
+
sinatra (4.2.1)
112
111
logger (>= 1.6.0)
113
112
mustermann (~> 3.0)
114
113
rack (>= 3.0.0, < 4)
115
-
rack-protection (= 4.1.1)
114
+
rack-protection (= 4.2.1)
116
115
rack-session (>= 2.0.0, < 3)
117
116
tilt (~> 2.0)
118
117
sinatra-activerecord (2.0.28)
···
124
123
cbor (~> 0.5, >= 0.5.9.6)
125
124
eventmachine (~> 1.2, >= 1.2.7)
126
125
faye-websocket (~> 0.12)
127
-
stringio (3.1.7)
126
+
stringio (3.1.8)
127
+
thin (2.0.1)
128
+
daemons (~> 1.0, >= 1.0.9)
129
+
eventmachine (~> 1.0, >= 1.0.4)
130
+
logger
131
+
rack (>= 1, < 4)
128
132
tilt (2.6.1)
129
-
timeout (0.4.3)
133
+
timeout (0.4.4)
134
+
tsort (0.2.0)
130
135
tzinfo (2.0.6)
131
136
concurrent-ruby (~> 1.0)
132
137
websocket-driver (0.8.0)
···
155
160
jwt
156
161
minisky (~> 0.5)
157
162
pg
158
-
puma
159
163
rackup
160
164
rainbow
161
165
rake
162
166
sinatra
163
167
sinatra-activerecord (~> 2.0)
164
168
skyfall (~> 0.6)
169
+
thin
165
170
166
171
BUNDLED WITH
167
172
2.7.0
+110
README.md
+110
README.md
···
1
+
# Lycan 🐺
2
+
3
+
A service which downloads and indexes the Bluesky posts you've liked, reposted, quoted or bookmarked, and allows you to search in that archive.
4
+
5
+
6
+
## How it works
7
+
8
+
Lycan is kind of like a tiny specialized AppView, which only indexes some specific things from some specific people. To avoid having to keep a full-network AppView, it only indexes posts and likes on demand from people who request to use it. So the first time you want to use it, you need to ask it to run an import process, which can take anything between a few minutes and an hour, depending on how much data there is to download. After that, new likes are being indexed live from the firehose.
9
+
10
+
At the moment, Lycan indexes four types of content:
11
+
12
+
- posts you've liked
13
+
- posts you've reposted
14
+
- posts you've quoted
15
+
- your old-style bookmarks (using the 📌 emoji method)
16
+
17
+
New bookmarks are private data, so at the moment they can't be imported until support for OAuth is added.
18
+
19
+
Lycan is written in Ruby, using Sinatra and ActiveRecord, with Postgres as the database. The official instance runs at [lycan.feeds.blue](https://lycan.feeds.blue) (this service only implements an XRPC API – the UI is implemented as part of [Skythread](https://skythread.mackuba.eu)).
20
+
21
+
The service consists of three separate components:
22
+
23
+
- a **firehose client**, which streams events from a relay/Jetstream and saves new data for the users whose data is/has been imported
24
+
- a **background worker**, which runs the import process
25
+
- an **HTTP server**, which serves the XRPC endpoints (currently there are 3: `startImport`, `getImportStatus` and `searchPosts`, plus a `did.json`); all the endpoints require service authentication through PDS proxying
26
+
27
+
28
+
## Setting up on localhost
29
+
30
+
This app should run on any somewhat recent version of Ruby, though of course it's recommended to run one that's still getting maintenance updates, ideally the latest one. It's also recommended to install it with [YJIT support](https://shopify.engineering/ruby-yjit-is-production-ready), and on Linux also with [jemalloc](https://scalingo.com/blog/improve-ruby-application-memory-jemalloc). You will probably need to have some familiarity with the Ruby ecosystem in order to set it up and run it.
31
+
32
+
A Postgres database is also required (again, any non-ancient version should work).
33
+
34
+
Download or clone the repository, then install the dependencies:
35
+
36
+
```
37
+
bundle install
38
+
```
39
+
40
+
Next, create the database – the configuration is defined in [`config/database.yml`](config/database.yml), for development it's `lycan_development`. Create it either manually, or with a rake task:
41
+
42
+
```
43
+
bundle exec rake db:create
44
+
```
45
+
46
+
Then, run the migrations:
47
+
48
+
```
49
+
bundle exec rake db:migrate
50
+
```
51
+
52
+
To run an import, you will need to run three separate processes, probably in separate terminal tabs:
53
+
54
+
1) the firehose client, [`bin/firehose`](bin/firehose)
55
+
2) the background worker, [`bin/worker`](bin/worker)
56
+
3) the Sinatra HTTP server, [`bin/server`](bin/server)
57
+
58
+
The UI can be accessed through Skythread, either on the official site on [skythread.mackuba.eu](https://skythread.mackuba.eu), or a copy you can download [from the repo](https://tangled.org/mackuba.eu/skythread). Log in and open "[Archive search](https://skythread.mackuba.eu/?page=search&mode=likes)" from the account menu – but importantly, to use the `localhost` Lycan instance, add `&lycan=local` to the URL.
59
+
60
+
You should then be able to start an import from there, and see the worker process printing some logs as it starts to download the data. (The firehose process needs to be running too, because the import job needs to pass through it first.)
61
+
62
+
63
+
## Configuration
64
+
65
+
There's a few things you can configure through ENV variables:
66
+
67
+
- `RELAY_HOST` – hostname of the relay to use for the firehose (default: `bsky.network`)
68
+
- `JETSTREAM_HOST` – alternatively, instead of `RELAY_HOST`, set this to a hostname of a [Jetstream](https://github.com/bluesky-social/jetstream) instance
69
+
- `FIREHOSE_USER_AGENT` – when running in production, it's recommended that you set this to some name that identifies who is running the service
70
+
- `APPVIEW_HOST` – hostname of the AppView used to download posts (default: `public.api.bsky.app`)
71
+
- `SERVER_HOSTNAME` – hostname of the server on which you're running the service in production
72
+
73
+
74
+
## Rake tasks
75
+
76
+
Some Rake tasks that might be useful:
77
+
78
+
```
79
+
bundle exec rake enqueue_user DID=did:plc:qweqwe
80
+
```
81
+
82
+
- request an import of the given account (to be handled by firehose + worker)
83
+
84
+
```
85
+
bundle exec rake import_user DID=did:plc:qweqwe COLLECTION=likes/reposts/posts/all
86
+
```
87
+
88
+
- run a complete import synchronously
89
+
90
+
```
91
+
bundle exec rake process_posts
92
+
```
93
+
94
+
- process all previously queued and unfinished or failed items
95
+
96
+
97
+
## Running in production
98
+
99
+
This will probably heavily depend on where and how you prefer to run it, I'm using a Capistrano deploy config in [`config/deploy.rb`](config/deploy.rb) to deploy to a VPS at [lycan.feeds.blue](https://lycan.feeds.blue). To use something like Docker or a service like Fly or Railway, you'll need to adapt the config for your specific setup.
100
+
101
+
On the server, you need to make sure that the firehose & worker processes are always running and are restarted if necessary. One option to do this (which I'm using) may be writing a `systemd` service config file and adding it to `/etc/systemd/system`. To run the HTTP server, you need Nginx/Caddy/Apache and a Ruby app server – my recommendation is Nginx with either Passenger (runs your app automatically from Nginx) or something like Puma (needs to be started by e.g. systemd like the firehose).
102
+
103
+
104
+
## Credits
105
+
106
+
Copyright © 2025 Kuba Suder ([@mackuba.eu](https://bsky.app/profile/did:plc:oio4hkxaop4ao4wz2pp3f4cr)).
107
+
108
+
The code is available under the terms of the [zlib license](https://choosealicense.com/licenses/zlib/) (permissive, similar to MIT).
109
+
110
+
Bug reports and pull requests are welcome 😎
+19
-1
app/firehose_client.rb
+19
-1
app/firehose_client.rb
···
151
151
end
152
152
end
153
153
end
154
+
rescue CBOR::UnpackError
155
+
# ignore invalid records
154
156
end
155
157
156
158
def process_account_event(msg)
157
159
if msg.status == :deleted
158
160
if user = User.find_by(did: msg.repo)
159
161
user.destroy
160
-
@active_users.delete_if { |u| u.id == user.id }
162
+
@active_users.delete_if { |k, u| u.id == user.id }
161
163
end
162
164
end
163
165
end
···
166
168
return unless @current_user
167
169
168
170
if op.action == :create
171
+
return if op.raw_record.nil?
169
172
@current_user.likes.import_from_record(op.uri, op.raw_record, queue: :firehose)
170
173
elsif op.action == :delete
171
174
@current_user.likes.where(rkey: op.rkey).delete_all
172
175
end
176
+
rescue StandardError => e
177
+
log "Error in #process_like (#{msg.seq}, #{op.uri}): #{e}"
178
+
log e.backtrace.reject { |x| x.include?('/ruby/') }
179
+
sleep 5 if e.is_a?(ActiveRecord::ConnectionFailed)
173
180
end
174
181
175
182
def process_repost(msg, op)
176
183
return unless @current_user
177
184
178
185
if op.action == :create
186
+
return if op.raw_record.nil?
179
187
@current_user.reposts.import_from_record(op.uri, op.raw_record, queue: :firehose)
180
188
elsif op.action == :delete
181
189
@current_user.reposts.where(rkey: op.rkey).delete_all
182
190
end
191
+
rescue StandardError => e
192
+
log "Error in #process_repost (#{msg.seq}, #{op.uri}): #{e}"
193
+
log e.backtrace.reject { |x| x.include?('/ruby/') }
194
+
sleep 5 if e.is_a?(ActiveRecord::ConnectionFailed)
183
195
end
184
196
185
197
def process_post(msg, op)
186
198
if op.action == :create
199
+
return if op.raw_record.nil?
200
+
187
201
if @current_user
188
202
@current_user.quotes.import_from_record(op.uri, op.raw_record, queue: :firehose)
189
203
@current_user.pins.import_from_record(op.uri, op.raw_record, queue: :firehose)
···
198
212
post.destroy
199
213
end
200
214
end
215
+
rescue StandardError => e
216
+
log "Error in #process_post (#{msg.seq}, #{op.uri}): #{e}"
217
+
log e.backtrace.reject { |x| x.include?('/ruby/') }
218
+
sleep 5 if e.is_a?(ActiveRecord::ConnectionFailed)
201
219
end
202
220
203
221
def log(text)
+19
-3
app/import_manager.rb
+19
-3
app/import_manager.rb
···
5
5
require_relative 'post_downloader'
6
6
7
7
class ImportManager
8
-
attr_accessor :report, :time_limit
8
+
attr_accessor :report, :time_limit, :logger, :log_status_updates
9
9
10
10
def initialize(user)
11
11
@user = user
···
39
39
40
40
downloader = PostDownloader.new
41
41
downloader.report = @report
42
+
downloader.logger = @logger
42
43
43
-
download_thread = Thread.new { downloader.import_from_queue(queue) }
44
+
download_thread = Thread.new do
45
+
@logger&.info "Starting downloader thread for #{@user}" if @log_status_updates
46
+
47
+
downloader.import_from_queue(queue)
48
+
49
+
@logger&.info "Ended downloader thread for #{@user}" if @log_status_updates
50
+
end
44
51
45
52
import_threads = importers.map do |import|
46
53
import.item_queue = queue
47
54
import.report = @report
55
+
import.logger = @logger
48
56
49
-
Thread.new { import.run_import(@time_limit) }
57
+
Thread.new do
58
+
@logger&.info "Starting #{import.class} thread for #{@user}" if @log_status_updates
59
+
60
+
import.run_import(@time_limit)
61
+
62
+
@logger&.info "Ended #{import.class} thread for #{@user}" if @log_status_updates
63
+
end
50
64
end
51
65
52
66
import_threads.each { |i| i.join }
67
+
68
+
@logger&.info "Finished all importer threads for #{@user}, waiting for downloader" if @log_status_updates
53
69
54
70
downloader.stop_when_empty = true
55
71
download_thread.join
+19
-5
app/import_worker.rb
+19
-5
app/import_worker.rb
···
9
9
require_relative 'reports/basic_report'
10
10
11
11
class ImportWorker
12
-
attr_accessor :verbose
12
+
attr_accessor :verbose, :logger
13
13
14
14
class UserThread < Thread
15
-
def initialize(user, collections, verbose = false)
15
+
def initialize(user, collections, logger, verbose = false)
16
16
@user = user
17
17
@verbose = verbose
18
+
@logger = logger
18
19
19
20
super { run(collections) }
20
21
end
···
24
25
end
25
26
26
27
def run(collections)
28
+
@logger&.info "Starting import thread for #{@user}"
29
+
27
30
if @user.registered_at.nil?
28
31
registration_time = get_registration_time(@user)
29
32
@user.update!(registered_at: registration_time)
30
33
end
31
34
32
35
import = ImportManager.new(@user)
33
-
import.report = BasicReport.new if @verbose
36
+
37
+
if @logger
38
+
import.report = BasicReport.new(@logger) if @verbose
39
+
import.logger = @logger
40
+
import.log_status_updates = true
41
+
end
42
+
34
43
import.start(collections)
44
+
45
+
@logger&.info "Ended import thread for #{@user}"
35
46
end
36
47
37
48
def get_registration_time(user)
···
47
58
48
59
@firehose_thread = Thread.new { process_firehose_items }
49
60
@downloader = PostDownloader.new
61
+
@downloader.logger = @logger
50
62
51
63
loop do
52
64
@user_threads.delete_if { |t| !t.alive? }
53
65
54
-
if user = User.with_unfinished_import.where.not(id: @user_threads.map(&:user_id)).first
66
+
users = User.with_unfinished_import.where.not(id: @user_threads.map(&:user_id)).to_a
67
+
68
+
users.each do |user|
55
69
collections = user.imports.unfinished.map(&:collection)
56
-
thread = UserThread.new(user, collections, @verbose)
70
+
thread = UserThread.new(user, collections, @logger, @verbose)
57
71
@user_threads << thread
58
72
end
59
73
+11
-3
app/importers/base_importer.rb
+11
-3
app/importers/base_importer.rb
···
1
1
require 'didkit'
2
2
require 'minisky'
3
+
require 'time'
3
4
4
5
require_relative '../at_uri'
6
+
require_relative '../errors'
5
7
require_relative '../models/post'
6
8
require_relative '../models/user'
7
9
8
10
class BaseImporter
9
-
attr_accessor :item_queue, :report
11
+
attr_accessor :item_queue, :report, :logger
10
12
11
13
def initialize(user)
12
14
@did = DID.new(user.did)
···
31
33
end
32
34
33
35
@time_limit = requested_time_limit || @import.last_completed
34
-
puts "Fetching until: #{@time_limit}" if @time_limit
36
+
@logger&.info "Fetching until: #{@time_limit}" if @time_limit
35
37
36
38
import_items
37
39
38
40
@import.update!(last_completed: @import.started_from) unless requested_time_limit
39
-
@import.update!(cursor: nil, started_from: nil)
41
+
@import.update!(cursor: nil, started_from: nil, fetched_until: nil)
40
42
@report&.update(importers: { importer_name => { :finished => true }})
41
43
end
42
44
43
45
def import_items
44
46
raise NotImplementedError
47
+
end
48
+
49
+
def created_at(record)
50
+
Time.parse(record['createdAt'])
51
+
rescue StandardError
52
+
raise InvalidRecordError
45
53
end
46
54
end
+12
-9
app/importers/likes_importer.rb
+12
-9
app/importers/likes_importer.rb
···
1
-
require 'time'
2
1
require_relative 'base_importer'
3
2
4
3
class LikesImporter < BaseImporter
···
11
10
12
11
records = response['records']
13
12
cursor = response['cursor']
14
-
15
-
@imported_count += records.length
16
-
@report&.update(importers: { importer_name => { :imported_items => @imported_count }})
17
-
@report&.update(importers: { importer_name => { :oldest_date => Time.parse(records.last['value']['createdAt']) }}) unless records.empty?
13
+
oldest_date = nil
18
14
19
15
records.each do |record|
20
16
begin
21
17
like = @user.likes.import_from_record(record['uri'], record['value'], queue: :import)
22
18
19
+
record_date = like&.time || created_at(record['value'])
20
+
oldest_date = [oldest_date, record_date].compact.min
21
+
23
22
if like && like.pending? && @item_queue
24
23
@item_queue.push(like)
25
24
@report&.update(queue: { length: @item_queue.length })
26
25
end
27
-
rescue StandardError => e
28
-
puts "Error in LikesImporter: #{record['uri']}: #{e}"
26
+
rescue InvalidRecordError => e
27
+
@logger&.warn "Error in LikesImporter: #{record['uri']}: #{e}"
29
28
end
30
29
end
31
30
31
+
@imported_count += records.length
32
+
@report&.update(importers: { importer_name => { :imported_items => @imported_count }})
33
+
@report&.update(importers: { importer_name => { :oldest_date => oldest_date }}) if oldest_date
34
+
32
35
params[:cursor] = cursor
33
-
@import.update!(cursor: cursor)
36
+
@import.update!(cursor: cursor, fetched_until: oldest_date)
34
37
35
38
break if !cursor
36
-
break if @time_limit && records.any? { |x| Time.parse(x['value']['createdAt']) < @time_limit }
39
+
break if @time_limit && oldest_date && oldest_date < @time_limit
37
40
end
38
41
end
39
42
end
+12
-9
app/importers/posts_importer.rb
+12
-9
app/importers/posts_importer.rb
···
1
-
require 'time'
2
1
require_relative 'base_importer'
3
2
4
3
class PostsImporter < BaseImporter
···
11
10
12
11
records = response['records']
13
12
cursor = response['cursor']
14
-
15
-
@imported_count += records.length
16
-
@report&.update(importers: { importer_name => { :imported_items => @imported_count }})
17
-
@report&.update(importers: { importer_name => { :oldest_date => Time.parse(records.last['value']['createdAt']) }}) unless records.empty?
13
+
oldest_date = nil
18
14
19
15
records.each do |record|
20
16
begin
21
17
quote = @user.quotes.import_from_record(record['uri'], record['value'], queue: :import)
22
18
pin = @user.pins.import_from_record(record['uri'], record['value'], queue: :import)
23
19
20
+
record_date = quote&.time || pin&.time || created_at(record['value'])
21
+
oldest_date = [oldest_date, record_date].compact.min
22
+
24
23
if @item_queue
25
24
if quote && quote.pending?
26
25
@item_queue.push(quote)
···
32
31
33
32
@report&.update(queue: { length: @item_queue.length })
34
33
end
35
-
rescue StandardError => e
36
-
puts "Error in PostsImporter: #{record['uri']}: #{e}"
34
+
rescue InvalidRecordError => e
35
+
@logger&.warn "Error in PostsImporter: #{record['uri']}: #{e}"
37
36
end
38
37
end
39
38
39
+
@imported_count += records.length
40
+
@report&.update(importers: { importer_name => { :imported_items => @imported_count }})
41
+
@report&.update(importers: { importer_name => { :oldest_date => oldest_date }}) if oldest_date
42
+
40
43
params[:cursor] = cursor
41
-
@import.update!(cursor: cursor)
44
+
@import.update!(cursor: cursor, fetched_until: oldest_date)
42
45
43
46
break if !cursor
44
-
break if @time_limit && records.any? { |x| Time.parse(x['value']['createdAt']) < @time_limit }
47
+
break if @time_limit && oldest_date && oldest_date < @time_limit
45
48
end
46
49
end
47
50
end
+12
-9
app/importers/reposts_importer.rb
+12
-9
app/importers/reposts_importer.rb
···
1
-
require 'time'
2
1
require_relative 'base_importer'
3
2
4
3
class RepostsImporter < BaseImporter
···
11
10
12
11
records = response['records']
13
12
cursor = response['cursor']
14
-
15
-
@imported_count += records.length
16
-
@report&.update(importers: { importer_name => { :imported_items => @imported_count }})
17
-
@report&.update(importers: { importer_name => { :oldest_date => Time.parse(records.last['value']['createdAt']) }}) unless records.empty?
13
+
oldest_date = nil
18
14
19
15
records.each do |record|
20
16
begin
21
17
repost = @user.reposts.import_from_record(record['uri'], record['value'], queue: :import)
22
18
19
+
record_date = repost&.time || created_at(record['value'])
20
+
oldest_date = [oldest_date, record_date].compact.min
21
+
23
22
if repost && repost.pending? && @item_queue
24
23
@item_queue.push(repost)
25
24
@report&.update(queue: { length: @item_queue.length })
26
25
end
27
-
rescue StandardError => e
28
-
puts "Error in RepostsImporter: #{record['uri']}: #{e}"
26
+
rescue InvalidRecordError => e
27
+
@logger&.warn "Error in RepostsImporter: #{record['uri']}: #{e}"
29
28
end
30
29
end
31
30
31
+
@imported_count += records.length
32
+
@report&.update(importers: { importer_name => { :imported_items => @imported_count }})
33
+
@report&.update(importers: { importer_name => { :oldest_date => oldest_date }}) if oldest_date
34
+
32
35
params[:cursor] = cursor
33
-
@import.update!(cursor: cursor)
36
+
@import.update!(cursor: cursor, fetched_until: oldest_date)
34
37
35
38
break if !cursor
36
-
break if @time_limit && records.any? { |x| Time.parse(x['value']['createdAt']) < @time_limit }
39
+
break if @time_limit && oldest_date && oldest_date < @time_limit
37
40
end
38
41
end
39
42
end
+26
app/models/import.rb
+26
app/models/import.rb
···
9
9
validates_uniqueness_of :collection, scope: :user_id
10
10
11
11
scope :unfinished, -> { where('(started_from IS NOT NULL) OR (last_completed IS NULL)') }
12
+
13
+
IMPORT_END = Time.at(0)
14
+
15
+
def imported_until
16
+
return nil if cursor.nil? && last_completed.nil?
17
+
18
+
groups = case collection
19
+
when 'likes'
20
+
[:likes]
21
+
when 'reposts'
22
+
[:reposts]
23
+
when 'posts'
24
+
[:pins, :quotes]
25
+
end
26
+
27
+
newest_queued_items = groups.map { |g| user.send(g).where(queue: :import).order(:time).last }
28
+
newest_queued = newest_queued_items.compact.sort_by(&:time).last
29
+
30
+
if newest_queued
31
+
newest_queued.time
32
+
elsif fetched_until
33
+
fetched_until
34
+
else
35
+
IMPORT_END
36
+
end
37
+
end
12
38
end
+7
app/models/post.rb
+7
app/models/post.rb
···
10
10
validates_length_of :text, maximum: 1000
11
11
validates_length_of :data, maximum: 10000
12
12
13
+
validate :check_null_bytes
14
+
13
15
belongs_to :user
14
16
15
17
has_many :likes, dependent: :delete_all
···
26
28
27
29
def at_uri
28
30
"at://#{user.did}/app.bsky.feed.post/#{rkey}"
31
+
end
32
+
33
+
def check_null_bytes
34
+
# Postgres doesn't allow null bytes in strings
35
+
errors.add(:text, 'must not contain a null byte') if text.include?("\u0000")
29
36
end
30
37
end
+19
-51
app/models/user.rb
+19
-51
app/models/user.rb
···
7
7
require_relative 'pin'
8
8
require_relative 'post'
9
9
require_relative 'repost'
10
+
require_relative 'user_importable'
10
11
11
12
class User < ActiveRecord::Base
12
13
validates_presence_of :did
···
19
20
20
21
before_destroy :delete_posts_cascading
21
22
22
-
has_many :likes, foreign_key: 'actor_id', dependent: :delete_all do
23
-
def import_from_record(like_uri, record, **args)
24
-
like = self.new_from_record(like_uri, record)
25
-
return nil if like.nil? || self.where(rkey: like.rkey).exists?
26
-
27
-
like.import_item!(args)
28
-
end
29
-
end
30
-
31
-
has_many :reposts, foreign_key: 'actor_id', dependent: :delete_all do
32
-
def import_from_record(repost_uri, record, **args)
33
-
repost = self.new_from_record(repost_uri, record)
34
-
return nil if repost.nil? || self.where(rkey: repost.rkey).exists?
35
-
36
-
repost.import_item!(args)
37
-
end
38
-
end
39
-
40
-
has_many :quotes, foreign_key: 'actor_id', dependent: :delete_all do
41
-
def import_from_record(post_uri, record, **args)
42
-
quote = self.new_from_record(post_uri, record)
43
-
return nil if quote.nil? || self.where(rkey: quote.rkey).exists?
44
-
45
-
quote.import_item!(args)
46
-
end
47
-
end
48
-
49
-
has_many :pins, foreign_key: 'actor_id', dependent: :delete_all do
50
-
def import_from_record(post_uri, record, **args)
51
-
pin = self.new_from_record(post_uri, record)
52
-
return nil if pin.nil? || self.where(rkey: pin.rkey).exists?
53
-
54
-
pin.import_item!(args)
55
-
end
56
-
end
23
+
has_many :likes, foreign_key: 'actor_id', dependent: :delete_all, extend: UserImportable
24
+
has_many :reposts, foreign_key: 'actor_id', dependent: :delete_all, extend: UserImportable
25
+
has_many :quotes, foreign_key: 'actor_id', dependent: :delete_all, extend: UserImportable
26
+
has_many :pins, foreign_key: 'actor_id', dependent: :delete_all, extend: UserImportable
57
27
58
28
def self.active
59
29
self.joins(:imports).distinct
···
80
50
end
81
51
82
52
def imported_until
83
-
return nil unless self.imports.exists?
53
+
import_positions = self.imports.map(&:imported_until)
84
54
85
-
oldest_imported_items = []
86
-
started = false
87
-
88
-
[:likes, :reposts, :pins, :quotes].each do |group|
89
-
if self.send(group).where(queue: :import).exists?
90
-
oldest_imported_items << self.send(group).where(queue: nil).order(:time).first
91
-
end
55
+
if import_positions.empty? || import_positions.any? { |x| x.nil? }
56
+
nil
57
+
else
58
+
import_positions.sort.last
92
59
end
60
+
end
93
61
94
-
earliest_oldest = oldest_imported_items.compact.sort_by(&:time).last
62
+
def erase_imports!
63
+
[:likes, :reposts, :quotes, :pins].map { |x| self.send(x).delete_all }
95
64
96
-
if earliest_oldest
97
-
earliest_oldest.time
98
-
elsif self.imports.merge(Import.unfinished).exists?
99
-
nil
100
-
else
101
-
:end
102
-
end
65
+
self.import_job&.destroy
66
+
self.imports.delete_all
103
67
end
104
68
105
69
def delete_posts_cascading
···
111
75
Quote.where(post_id: posts_subquery).delete_all
112
76
113
77
Post.where(user: self).delete_all
78
+
end
79
+
80
+
def to_s
81
+
%(<User id: #{id}, did: "#{did}">)
114
82
end
115
83
end
+20
app/models/user_importable.rb
+20
app/models/user_importable.rb
···
1
+
require_relative '../errors'
2
+
3
+
module UserImportable
4
+
def import_from_record(item_uri, record, **args)
5
+
item = try_build_from_record(item_uri, record)
6
+
return nil if item.nil? || already_imported?(item)
7
+
8
+
item.import_item!(args)
9
+
end
10
+
11
+
def try_build_from_record(item_uri, record)
12
+
self.new_from_record(item_uri, record)
13
+
rescue StandardError
14
+
raise InvalidRecordError
15
+
end
16
+
17
+
def already_imported?(item)
18
+
self.where(rkey: item.rkey).exists?
19
+
end
20
+
end
+34
-19
app/post_downloader.rb
+34
-19
app/post_downloader.rb
···
6
6
require_relative 'models/user'
7
7
8
8
class PostDownloader
9
-
attr_accessor :report, :stop_when_empty
9
+
attr_accessor :report, :logger, :stop_when_empty
10
10
11
11
def initialize
12
12
@sky = Minisky.new(ENV['APPVIEW_HOST'] || 'public.api.bsky.app', nil)
···
51
51
response = @sky.get_request('app.bsky.feed.getPosts', { uris: items.map(&:post_uri).uniq })
52
52
53
53
response['posts'].each do |data|
54
+
current_items = items.select { |x| x.post_uri == data['uri'] }
55
+
items -= current_items
56
+
54
57
begin
55
-
item = items.detect { |x| x.post_uri == data['uri'] }
56
-
items.delete(item)
57
-
58
58
post = save_post(data['uri'], data['record'])
59
59
60
60
if post.valid?
61
-
update_item(item, post)
61
+
current_items.each { |i| update_item(i, post) }
62
62
else
63
-
puts "Invalid post #{item.post_uri}: #{post.errors.full_messages.join("; ")}"
64
-
invalidate_item(item)
63
+
@logger&.warn "Invalid post #{data['uri']}: #{post.errors.full_messages.join("; ")}"
64
+
current_items.each { |i| invalidate_item(i) }
65
65
end
66
-
rescue StandardError => e
67
-
puts "Error in PostDownloader: #{item.post_uri}: #{e.class}: #{e}"
66
+
rescue InvalidRecordError => e
67
+
@logger&.warn "Error in PostDownloader: #{data['uri']}: #{e.class}: #{e}"
68
+
current_items.each { |i| i.update!(queue: nil) }
68
69
end
69
70
end
70
71
71
72
check_missing_items(items)
72
73
rescue StandardError => e
73
-
puts "Error in PostDownloader: #{e.class}: #{e}"
74
+
@logger&.warn "Error in PostDownloader: #{e.class}: #{e}"
74
75
end
75
76
end
76
77
77
78
def save_post(post_uri, record)
78
79
did, _, rkey = AT_URI(post_uri)
79
80
80
-
text = record.delete('text')
81
-
created = record.delete('createdAt')
82
-
83
-
author = User.find_or_create_by!(did: did)
81
+
begin
82
+
author = User.find_or_create_by!(did: did)
83
+
rescue ActiveRecord::RecordInvalid => e
84
+
raise InvalidRecordError
85
+
end
84
86
85
87
if post = Post.find_by(user: author, rkey: rkey)
86
88
return post
89
+
else
90
+
post = build_post(author, rkey, record)
91
+
post.save
92
+
post
87
93
end
94
+
end
88
95
89
-
Post.create(
96
+
def build_post(author, rkey, record)
97
+
text = record.delete('text')
98
+
created = record.delete('createdAt')
99
+
100
+
record.delete('$type')
101
+
102
+
Post.new(
90
103
user: author,
91
104
rkey: rkey,
92
105
time: Time.parse(created),
93
106
text: text,
94
107
data: JSON.generate(record)
95
108
)
109
+
rescue StandardError
110
+
raise InvalidRecordError
96
111
end
97
112
98
113
def update_item(item, post)
···
138
153
case status
139
154
when :active
140
155
# account is active but wasn't returned in getProfiles, probably was suspended on the AppView
141
-
puts "#{item.post_uri}: account #{did} exists on the PDS, account must have been taken down"
156
+
# puts "#{item.post_uri}: account #{did} exists on the PDS, account must have been taken down"
142
157
item.destroy
143
158
when nil
144
159
# account was deleted, so all posts were deleted too
145
-
puts "#{item.post_uri}: account #{did} doesn't exist on the PDS, post must have been deleted"
160
+
# puts "#{item.post_uri}: account #{did} doesn't exist on the PDS, post must have been deleted"
146
161
item.destroy
147
162
else
148
163
# account is inactive/suspended, but could come back, so leave it for now
149
-
puts "#{item.post_uri}: account #{did} is inactive: #{status}"
164
+
# puts "#{item.post_uri}: account #{did} is inactive: #{status}"
150
165
end
151
166
rescue StandardError => e
152
167
hostname = did_obj.document.pds_host rescue "???"
153
-
puts "#{item.post_uri}: couldn't check account status for #{did} on #{hostname}: #{e.class}: #{e}"
168
+
@logger&.warn "#{item.post_uri}: couldn't check account status for #{did} on #{hostname}: #{e.class}: #{e}"
154
169
155
170
# delete reference if the account's PDS is the old bsky.social (so it must have been deleted pre Nov 2023)
156
171
item.destroy if hostname == 'bsky.social'
+6
-2
app/reports/basic_report.rb
+6
-2
app/reports/basic_report.rb
···
1
1
class BasicReport
2
+
def initialize(logger)
3
+
@logger = logger
4
+
end
5
+
2
6
def update(data)
3
7
data.each do |k, v|
4
8
if k == :downloader
5
-
p ({ k => v }) if v[:downloaded_posts] && v[:downloaded_posts] % 100 == 0
9
+
@logger.info({ k => v }.inspect) if v[:downloaded_posts] && v[:downloaded_posts] % 100 == 0
6
10
elsif k == :queue
7
11
next
8
12
else
9
-
p ({ k => v})
13
+
@logger.info({ k => v}.inspect)
10
14
end
11
15
end
12
16
end
+9
app/reports/simple_logger.rb
+9
app/reports/simple_logger.rb
+11
-3
app/server.rb
+11
-3
app/server.rb
···
8
8
9
9
class Server < Sinatra::Application
10
10
register Sinatra::ActiveRecordExtension
11
-
set :port, 3000
11
+
set :port, ENV['PORT'] || 3000
12
12
13
13
PAGE_LIMIT = 25
14
14
HOSTNAME = ENV['SERVER_HOSTNAME'] || 'lycan.feeds.blue'
···
77
77
if settings.development?
78
78
headers['Access-Control-Allow-Origin'] = '*'
79
79
end
80
+
end
81
+
82
+
get '/' do
83
+
"Awoo 🐺"
80
84
end
81
85
82
86
get '/xrpc/blue.feeds.lycan.searchPosts' do
···
153
157
154
158
case until_date
155
159
when nil
156
-
json_response(status: 'not_started')
157
-
when :end
160
+
if user.import_job || user.imports.exists?
161
+
json_response(status: 'scheduled')
162
+
else
163
+
json_response(status: 'not_started')
164
+
end
165
+
when Import::IMPORT_END
158
166
json_response(status: 'finished')
159
167
else
160
168
progress = 1 - (until_date - user.registered_at) / (Time.now - user.registered_at)
+1
bin/server
+1
bin/server
+7
-2
bin/worker
+7
-2
bin/worker
···
2
2
3
3
require 'bundler/setup'
4
4
require_relative '../app/import_worker'
5
+
require_relative '../app/reports/simple_logger'
5
6
6
7
$stdout.sync = true
7
8
···
17
18
puts " -v = verbose"
18
19
end
19
20
21
+
logger = SimpleLogger.new
20
22
worker = ImportWorker.new
23
+
worker.logger = logger
21
24
22
25
args = ARGV.dup
23
26
···
36
39
end
37
40
38
41
trap("SIGINT") {
39
-
puts "Stopping..."
42
+
puts
43
+
puts "[#{Time.now}] Stopping..."
40
44
exit
41
45
}
42
46
43
47
trap("SIGTERM") {
44
-
puts "Shutting down the service..."
48
+
puts "[#{Time.now}] Shutting down the service..."
45
49
exit
46
50
}
47
51
52
+
puts "[#{Time.now}] Starting background worker..."
48
53
worker.run
+5
config/deploy.rb
+5
config/deploy.rb
···
22
22
23
23
after 'deploy', 'deploy:cleanup'
24
24
after 'deploy:migrations', 'deploy:cleanup'
25
+
after 'deploy:update_code', 'deploy:link_shared'
25
26
26
27
namespace :deploy do
27
28
task :restart, :roles => :web do
···
32
33
run "cd #{release_path} && bundle config set --local deployment 'true'"
33
34
run "cd #{release_path} && bundle config set --local path '#{shared_path}/bundle'"
34
35
run "cd #{release_path} && bundle config set --local without 'development test'"
36
+
end
37
+
38
+
task :link_shared do
39
+
run "ln -s #{shared_path}/env #{release_path}/.env"
35
40
end
36
41
end
+5
db/migrate/20251027134657_add_fetched_until_to_imports.rb
+5
db/migrate/20251027134657_add_fetched_until_to_imports.rb
+2
-1
db/schema.rb
+2
-1
db/schema.rb
···
10
10
#
11
11
# It's strongly recommended that you check this file into your version control system.
12
12
13
-
ActiveRecord::Schema[7.2].define(version: 2025_09_23_180153) do
13
+
ActiveRecord::Schema[7.2].define(version: 2025_10_27_134657) do
14
14
# These are extensions that must be enabled in order to support this database
15
15
enable_extension "plpgsql"
16
16
···
25
25
t.datetime "started_from"
26
26
t.datetime "last_completed"
27
27
t.string "collection", limit: 20, null: false
28
+
t.datetime "fetched_until"
28
29
t.index ["user_id", "collection"], name: "index_imports_on_user_id_and_collection", unique: true
29
30
end
30
31
+2
lib/tasks/import.rake
+2
lib/tasks/import.rake
···
3
3
require_relative '../../app/models/user'
4
4
require_relative '../../app/post_downloader'
5
5
require_relative '../../app/reports/console_report'
6
+
require_relative '../../app/reports/simple_logger'
6
7
7
8
task :enqueue_user do
8
9
unless ENV['DID']
···
34
35
35
36
import = ImportManager.new(user)
36
37
import.report = ConsoleReport.new
38
+
import.logger = SimpleLogger.new
37
39
import.time_limit = ENV['UNTIL']
38
40
39
41
trap("SIGINT") {