1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
|
# frozen_string_literal: true
require_relative '../../config/boot'
require_relative '../../config/environment'
require_relative 'cli_helper'
module Mastodon
class MediaCLI < Thor
include ActionView::Helpers::NumberHelper
include CLIHelper
def self.exit_on_failure?
true
end
option :days, type: :numeric, default: 7, aliases: [:d]
option :concurrency, type: :numeric, default: 5, aliases: [:c]
option :verbose, type: :boolean, default: false, aliases: [:v]
option :dry_run, type: :boolean, default: false
desc 'remove', 'Remove remote media files'
long_desc <<-DESC
Removes locally cached copies of media attachments from other servers.
The --days option specifies how old media attachments have to be before
they are removed. It defaults to 7 days.
DESC
def remove
time_ago = options[:days].days.ago
dry_run = options[:dry_run] ? '(DRY RUN)' : ''
processed, aggregate = parallelize_with_progress(MediaAttachment.cached.where.not(remote_url: '').where('created_at < ?', time_ago)) do |media_attachment|
next if media_attachment.file.blank?
size = media_attachment.file_file_size
unless options[:dry_run]
media_attachment.file.destroy
media_attachment.save
end
size
end
say("Removed #{processed} media attachments (approx. #{number_to_human_size(aggregate)}) #{dry_run}", :green, true)
end
option :start_after
option :prefix
option :dry_run, type: :boolean, default: false
desc 'remove-orphans', 'Scan storage and check for files that do not belong to existing media attachments'
long_desc <<~LONG_DESC
Scans file storage for files that do not belong to existing media attachments. Because this operation
requires iterating over every single file individually, it will be slow.
Please mind that some storage providers charge for the necessary API requests to list objects.
LONG_DESC
def remove_orphans
progress = create_progress_bar(nil)
reclaimed_bytes = 0
removed = 0
dry_run = options[:dry_run] ? ' (DRY RUN)' : ''
prefix = options[:prefix]
case Paperclip::Attachment.default_options[:storage]
when :s3
paperclip_instance = MediaAttachment.new.file
s3_interface = paperclip_instance.s3_interface
bucket = s3_interface.bucket(Paperclip::Attachment.default_options[:s3_credentials][:bucket])
last_key = options[:start_after]
loop do
objects = begin
begin
bucket.objects(start_after: last_key, prefix: prefix).limit(1000).map { |x| x }
rescue => e
progress.log(pastel.red("Error fetching list of files: #{e}"))
progress.log("If you want to continue from this point, add --start-after=#{last_key} to your command") if last_key
break
end
end
break if objects.empty?
last_key = objects.last.key
record_map = preload_records_from_mixed_objects(objects)
objects.each do |object|
path_segments = object.key.split('/')
path_segments.delete('cache')
model_name = path_segments.first.classify
attachment_name = path_segments[1].singularize
record_id = path_segments[2..-2].join.to_i
file_name = path_segments.last
record = record_map.dig(model_name, record_id)
attachment = record&.public_send(attachment_name)
progress.increment
next unless attachment.blank? || !attachment.variant?(file_name)
begin
object.delete unless options[:dry_run]
reclaimed_bytes += object.size
removed += 1
progress.log("Found and removed orphan: #{object.key}")
rescue => e
progress.log(pastel.red("Error processing #{object.key}: #{e}"))
end
end
end
when :fog
say('The fog storage driver is not supported for this operation at this time', :red)
exit(1)
when :filesystem
require 'find'
root_path = ENV.fetch('PAPERCLIP_ROOT_PATH', File.join(':rails_root', 'public', 'system')).gsub(':rails_root', Rails.root.to_s)
Find.find(File.join(*[root_path, prefix].compact)) do |path|
next if File.directory?(path)
key = path.gsub("#{root_path}#{File::SEPARATOR}", '')
path_segments = key.split(File::SEPARATOR)
path_segments.delete('cache')
model_name = path_segments.first.classify
record_id = path_segments[2..-2].join.to_i
attachment_name = path_segments[1].singularize
file_name = path_segments.last
next unless PRELOAD_MODEL_WHITELIST.include?(model_name)
record = model_name.constantize.find_by(id: record_id)
attachment = record&.public_send(attachment_name)
progress.increment
next unless attachment.blank? || !attachment.variant?(file_name)
begin
size = File.size(path)
File.delete(path) unless options[:dry_run]
reclaimed_bytes += size
removed += 1
progress.log("Found and removed orphan: #{key}")
rescue => e
progress.log(pastel.red("Error processing #{key}: #{e}"))
end
end
end
progress.total = progress.progress
progress.finish
say("Removed #{removed} orphans (approx. #{number_to_human_size(reclaimed_bytes)})#{dry_run}", :green, true)
end
option :account, type: :string
option :domain, type: :string
option :status, type: :numeric
option :concurrency, type: :numeric, default: 5, aliases: [:c]
option :verbose, type: :boolean, default: false, aliases: [:v]
option :dry_run, type: :boolean, default: false
option :force, type: :boolean, default: false
desc 'refresh', 'Fetch remote media files'
long_desc <<-DESC
Re-downloads media attachments from other servers. You must specify the
source of media attachments with one of the following options:
Use the --status option to download attachments from a specific status,
using the status local numeric ID.
Use the --account option to download attachments from a specific account,
using username@domain handle of the account.
Use the --domain option to download attachments from a specific domain.
By default, attachments that are believed to be already downloaded will
not be re-downloaded. To force re-download of every URL, use --force.
DESC
def refresh
dry_run = options[:dry_run] ? ' (DRY RUN)' : ''
if options[:status]
scope = MediaAttachment.where(status_id: options[:status])
elsif options[:account]
username, domain = username.split('@')
account = Account.find_remote(username, domain)
if account.nil?
say('No such account', :red)
exit(1)
end
scope = MediaAttachment.where(account_id: account.id)
elsif options[:domain]
scope = MediaAttachment.joins(:account).merge(Account.by_domain_and_subdomains(options[:domain]))
else
exit(1)
end
processed, aggregate = parallelize_with_progress(scope) do |media_attachment|
next if media_attachment.remote_url.blank? || (!options[:force] && media_attachment.file_file_name.present?)
unless options[:dry_run]
media_attachment.reset_file!
media_attachment.save
end
media_attachment.file_file_size
end
say("Downloaded #{processed} media attachments (approx. #{number_to_human_size(aggregate)})#{dry_run}", :green, true)
end
desc 'usage', 'Calculate disk space consumed by Mastodon'
def usage
say("Attachments:\t#{number_to_human_size(MediaAttachment.sum(:file_file_size))} (#{number_to_human_size(MediaAttachment.where(account: Account.local).sum(:file_file_size))} local)")
say("Custom emoji:\t#{number_to_human_size(CustomEmoji.sum(:image_file_size))} (#{number_to_human_size(CustomEmoji.local.sum(:image_file_size))} local)")
say("Preview cards:\t#{number_to_human_size(PreviewCard.sum(:image_file_size))}")
say("Avatars:\t#{number_to_human_size(Account.sum(:avatar_file_size))} (#{number_to_human_size(Account.local.sum(:avatar_file_size))} local)")
say("Headers:\t#{number_to_human_size(Account.sum(:header_file_size))} (#{number_to_human_size(Account.local.sum(:header_file_size))} local)")
say("Backups:\t#{number_to_human_size(Backup.sum(:dump_file_size))}")
say("Imports:\t#{number_to_human_size(Import.sum(:data_file_size))}")
say("Settings:\t#{number_to_human_size(SiteUpload.sum(:file_file_size))}")
end
desc 'lookup URL', 'Lookup where media is displayed by passing a media URL'
def lookup(url)
path = Addressable::URI.parse(url).path
path_segments = path.split('/')[2..-1]
path_segments.delete('cache')
model_name = path_segments.first.classify
record_id = path_segments[2..-2].join.to_i
unless PRELOAD_MODEL_WHITELIST.include?(model_name)
say("Cannot find corresponding model: #{model_name}", :red)
exit(1)
end
record = model_name.constantize.find_by(id: record_id)
record = record.status if record.respond_to?(:status)
unless record
say('Cannot find corresponding record', :red)
exit(1)
end
display_url = ActivityPub::TagManager.instance.url_for(record)
if display_url.blank?
say('No public URL for this type of record', :red)
exit(1)
end
say(display_url, :blue)
rescue Addressable::URI::InvalidURIError
say('Invalid URL', :red)
exit(1)
end
private
PRELOAD_MODEL_WHITELIST = %w(
Account
Backup
CustomEmoji
Import
MediaAttachment
PreviewCard
SiteUpload
).freeze
def preload_records_from_mixed_objects(objects)
preload_map = Hash.new { |hash, key| hash[key] = [] }
objects.map do |object|
segments = object.key.split('/')
segments.delete('cache')
model_name = segments.first.classify
record_id = segments[2..-2].join.to_i
next unless PRELOAD_MODEL_WHITELIST.include?(model_name)
preload_map[model_name] << record_id
end
preload_map.each_with_object({}) do |(model_name, record_ids), model_map|
model_map[model_name] = model_name.constantize.where(id: record_ids).each_with_object({}) { |record, record_map| record_map[record.id] = record }
end
end
end
end
|