discourse/app/models/upload.rb

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

509 lines
14 KiB
Ruby
Raw Normal View History

# frozen_string_literal: true
2013-11-06 02:04:47 +08:00
require "digest/sha1"
2013-02-06 03:16:51 +08:00
class Upload < ActiveRecord::Base
self.ignored_columns = [
"verified" # TODO(2020-12-10): remove
]
include ActionView::Helpers::NumberHelper
include HasUrl
SHA1_LENGTH = 40
SEEDED_ID_THRESHOLD = 0
URL_REGEX ||= /(\/original\/\dX[\/\.\w]*\/(\h+)[\.\w]*)/
MAX_IDENTIFY_SECONDS = 5
2013-02-06 03:16:51 +08:00
belongs_to :user
FEATURE: Secure media allowing duplicated uploads with category-level privacy and post-based access rules (#8664) ### General Changes and Duplication * We now consider a post `with_secure_media?` if it is in a read-restricted category. * When uploading we now set an upload's secure status straight away. * When uploading if `SiteSetting.secure_media` is enabled, we do not check to see if the upload already exists using the `sha1` digest of the upload. The `sha1` column of the upload is filled with a `SecureRandom.hex(20)` value which is the same length as `Upload::SHA1_LENGTH`. The `original_sha1` column is filled with the _real_ sha1 digest of the file. * Whether an upload `should_be_secure?` is now determined by whether the `access_control_post` is `with_secure_media?` (if there is no access control post then we leave the secure status as is). * When serializing the upload, we now cook the URL if the upload is secure. This is so it shows up correctly in the composer preview, because we set secure status on upload. ### Viewing Secure Media * The secure-media-upload URL will take the post that the upload is attached to into account via `Guardian.can_see?` for access permissions * If there is no `access_control_post` then we just deliver the media. This should be a rare occurrance and shouldn't cause issues as the `access_control_post` is set when `link_post_uploads` is called via `CookedPostProcessor` ### Removed We no longer do any of these because we do not reuse uploads by sha1 if secure media is enabled. * We no longer have a way to prevent cross-posting of a secure upload from a private context to a public context. * We no longer have to set `secure: false` for uploads when uploading for a theme component.
2020-01-16 11:50:27 +08:00
belongs_to :access_control_post, class_name: 'Post'
2013-02-06 03:16:51 +08:00
# when we access this post we don't care if the post
# is deleted
def access_control_post
Post.unscoped { super }
end
2013-11-06 02:04:47 +08:00
has_many :post_uploads, dependent: :destroy
2013-06-14 05:44:24 +08:00
has_many :posts, through: :post_uploads
has_many :optimized_images, dependent: :destroy
has_many :user_uploads, dependent: :destroy
FEATURE: Include optimized thumbnails for topics (#9215) This introduces new APIs for obtaining optimized thumbnails for topics. There are a few building blocks required for this: - Introduces new `image_upload_id` columns on the `posts` and `topics` table. This replaces the old `image_url` column, which means that thumbnails are now restricted to uploads. Hotlinked thumbnails are no longer possible. In normal use (with pull_hotlinked_images enabled), this has no noticeable impact - A migration attempts to match existing urls to upload records. If a match cannot be found then the posts will be queued for rebake - Optimized thumbnails are generated during post_process_cooked. If thumbnails are missing when serializing a topic list, then a sidekiq job is queued - Topic lists and topics now include a `thumbnails` key, which includes all the available images: ``` "thumbnails": [ { "max_width": null, "max_height": null, "url": "//example.com/original-image.png", "width": 1380, "height": 1840 }, { "max_width": 1024, "max_height": 1024, "url": "//example.com/optimized-image.png", "width": 768, "height": 1024 } ] ``` - Themes can request additional thumbnail sizes by using a modifier in their `about.json` file: ``` "modifiers": { "topic_thumbnail_sizes": [ [200, 200], [800, 800] ], ... ``` Remember that these are generated asynchronously, so your theme should include logic to fallback to other available thumbnails if your requested size has not yet been generated - Two new raw plugin outlets are introduced, to improve the customisability of the topic list. `topic-list-before-columns` and `topic-list-before-link`
2020-05-05 16:07:50 +08:00
has_many :topic_thumbnails
2013-06-16 16:39:48 +08:00
attr_accessor :for_group_message
attr_accessor :for_theme
attr_accessor :for_private_message
attr_accessor :for_export
2018-11-14 15:03:02 +08:00
attr_accessor :for_site_setting
attr_accessor :for_gravatar
2013-02-06 03:16:51 +08:00
validates_presence_of :filesize
validates_presence_of :original_filename
validates_with UploadValidator
2014-04-15 04:55:57 +08:00
before_destroy do
UserProfile.where(card_background_upload_id: self.id).update_all(card_background_upload_id: nil)
UserProfile.where(profile_background_upload_id: self.id).update_all(profile_background_upload_id: nil)
end
after_destroy do
User.where(uploaded_avatar_id: self.id).update_all(uploaded_avatar_id: nil)
UserAvatar.where(gravatar_upload_id: self.id).update_all(gravatar_upload_id: nil)
UserAvatar.where(custom_upload_id: self.id).update_all(custom_upload_id: nil)
end
scope :by_users, -> { where("uploads.id > ?", SEEDED_ID_THRESHOLD) }
def self.verification_statuses
@verification_statuses ||= Enum.new(
unchecked: 1,
verified: 2,
invalid_etag: 3
)
end
2018-11-14 15:03:02 +08:00
def to_s
self.url
end
def thumbnail(width = self.thumbnail_width, height = self.thumbnail_height)
optimized_images.find_by(width: width, height: height)
2013-06-17 07:00:25 +08:00
end
2013-11-06 02:04:47 +08:00
def has_thumbnail?(width, height)
2013-09-27 16:55:50 +08:00
thumbnail(width, height).present?
2013-06-17 07:00:25 +08:00
end
def create_thumbnail!(width, height, opts = nil)
2013-06-17 07:00:25 +08:00
return unless SiteSetting.create_thumbnails?
opts ||= {}
if get_optimized_image(width, height, opts)
save(validate: false)
2013-09-27 16:55:50 +08:00
end
2013-06-17 07:00:25 +08:00
end
# this method attempts to correct old incorrect extensions
def get_optimized_image(width, height, opts = nil)
opts ||= {}
if (!extension || extension.length == 0)
fix_image_extension
end
opts = opts.merge(raise_on_error: true)
begin
OptimizedImage.create_for(self, width, height, opts)
rescue => ex
Rails.logger.info ex if Rails.env.development?
opts = opts.merge(raise_on_error: false)
if fix_image_extension
OptimizedImage.create_for(self, width, height, opts)
else
nil
end
end
end
def fix_image_extension
return false if extension == "unknown"
begin
# this is relatively cheap once cached
original_path = Discourse.store.path_for(self)
if original_path.blank?
external_copy = Discourse.store.download(self) rescue nil
original_path = external_copy.try(:path)
end
image_info = FastImage.new(original_path) rescue nil
new_extension = image_info&.type&.to_s || "unknown"
if new_extension != self.extension
self.update_columns(extension: new_extension)
true
end
rescue
self.update_columns(extension: "unknown")
true
end
end
def destroy
Upload.transaction do
2013-08-14 04:08:29 +08:00
Discourse.store.remove_upload(self)
super
end
end
def short_url
"upload://#{short_url_basename}"
end
def uploaded_before_secure_media_enabled?
original_sha1.blank?
end
def matching_access_control_post?(post)
access_control_post_id == post.id
end
def copied_from_other_post?(post)
return false if access_control_post_id.blank?
!matching_access_control_post?(post)
end
def short_path
self.class.short_path(sha1: self.sha1, extension: self.extension)
end
def self.consider_for_reuse(upload, post)
return upload if !SiteSetting.secure_media? || upload.blank? || post.blank?
return nil if !upload.matching_access_control_post?(post) || upload.uploaded_before_secure_media_enabled?
upload
end
def self.secure_media_url?(url)
# we do not want to exclude topic links that for whatever reason
# have secure-media-uploads in the URL e.g. /t/secure-media-uploads-are-cool/223452
route = UrlHelper.rails_route_from_url(url)
return false if route.blank?
route[:action] == "show_secure" && route[:controller] == "uploads" && FileHelper.is_supported_media?(url)
rescue ActionController::RoutingError
false
end
def self.signed_url_from_secure_media_url(url)
route = UrlHelper.rails_route_from_url(url)
url = Rails.application.routes.url_for(route.merge(only_path: true))
secure_upload_s3_path = url[url.index(route[:path])..-1]
Discourse.store.signed_url_for_path(secure_upload_s3_path)
end
def self.secure_media_url_from_upload_url(url)
return url if !url.include?(SiteSetting.Upload.absolute_base_url)
uri = URI.parse(url)
Rails.application.routes.url_for(
controller: "uploads",
action: "show_secure",
path: uri.path[1..-1],
only_path: true
)
end
def self.short_path(sha1:, extension:)
@url_helpers ||= Rails.application.routes.url_helpers
@url_helpers.upload_short_path(
base62: self.base62_sha1(sha1),
extension: extension
)
end
def self.base62_sha1(sha1)
Base62.encode(sha1.hex)
end
def base62_sha1
2019-06-04 14:10:46 +08:00
Upload.base62_sha1(self.sha1)
end
def local?
!(url =~ /^(https?:)?\/\//)
end
def fix_dimensions!
return if !FileHelper.is_supported_image?("image.#{extension}")
path =
if local?
Discourse.store.path_for(self)
else
Discourse.store.download(self).path
end
begin
if extension == 'svg'
w, h = Discourse::Utils.execute_command("identify", "-format", "%w %h", path, timeout: MAX_IDENTIFY_SECONDS).split(' ') rescue [0, 0]
else
w, h = FastImage.new(path, raise_on_failure: true).size
end
self.width = w || 0
self.height = h || 0
self.thumbnail_width, self.thumbnail_height = ImageSizer.resize(w, h)
self.update_columns(
width: width,
height: height,
thumbnail_width: thumbnail_width,
thumbnail_height: thumbnail_height
)
rescue => e
Discourse.warn_exception(e, message: "Error getting image dimensions")
end
nil
end
# on demand image size calculation, this allows us to null out image sizes
# and still handle as needed
def get_dimension(key)
if v = read_attribute(key)
return v
end
fix_dimensions!
read_attribute(key)
end
def width
get_dimension(:width)
end
def height
get_dimension(:height)
end
def thumbnail_width
get_dimension(:thumbnail_width)
end
def thumbnail_height
get_dimension(:thumbnail_height)
end
def target_image_quality(local_path, test_quality)
@file_quality ||= Discourse::Utils.execute_command("identify", "-format", "%Q", local_path, timeout: MAX_IDENTIFY_SECONDS).to_i rescue 0
if @file_quality == 0 || @file_quality > test_quality
test_quality
end
end
def self.sha1_from_short_path(path)
if path =~ /(\/uploads\/short-url\/)([a-zA-Z0-9]+)(\..*)?/
self.sha1_from_base62_encoded($2)
end
end
def self.sha1_from_short_url(url)
if url =~ /(upload:\/\/)?([a-zA-Z0-9]+)(\..*)?/
self.sha1_from_base62_encoded($2)
end
end
def self.sha1_from_base62_encoded(encoded_sha1)
sha1 = Base62.decode(encoded_sha1).to_s(16)
if sha1.length > SHA1_LENGTH
nil
else
sha1.rjust(SHA1_LENGTH, '0')
end
end
def self.generate_digest(path)
Digest::SHA1.file(path).hexdigest
end
def human_filesize
number_to_human_size(self.filesize)
end
def rebake_posts_on_old_scheme
self.posts.where("cooked LIKE '%/_optimized/%'").find_each(&:rebake!)
end
def update_secure_status(source: "unknown", override: nil)
if override.nil?
mark_secure, reason = UploadSecurity.new(self).should_be_secure_with_reason
else
mark_secure = override
reason = "manually overridden"
end
secure_status_did_change = self.secure? != mark_secure
self.update(secure_params(mark_secure, reason, source))
Discourse.store.update_upload_ACL(self) if Discourse.store.external?
secure_status_did_change
end
def secure_params(secure, reason, source = "unknown")
{
secure: secure,
security_last_changed_reason: reason + " | source: #{source}",
security_last_changed_at: Time.zone.now
}
end
def self.migrate_to_new_scheme(limit: nil)
problems = []
DistributedMutex.synchronize("migrate_upload_to_new_scheme") do
if SiteSetting.migrate_to_new_scheme
max_file_size_kb = [
SiteSetting.max_image_size_kb,
SiteSetting.max_attachment_size_kb
].max.kilobytes
local_store = FileStore::LocalStore.new
db = RailsMultisite::ConnectionManagement.current_db
scope = Upload.by_users
.where("url NOT LIKE '%/original/_X/%' AND url LIKE '%/uploads/#{db}%'")
.order(id: :desc)
scope = scope.limit(limit) if limit
if scope.count == 0
SiteSetting.migrate_to_new_scheme = false
return problems
end
remap_scope = nil
scope.each do |upload|
begin
# keep track of the url
previous_url = upload.url.dup
# where is the file currently stored?
external = previous_url =~ /^\/\//
# download if external
if external
url = SiteSetting.scheme + ":" + previous_url
begin
retries ||= 0
file = FileHelper.download(
url,
max_file_size: max_file_size_kb,
tmp_file_name: "discourse",
follow_redirect: true
)
rescue OpenURI::HTTPError
retry if (retries += 1) < 1
next
end
path = file.path
else
path = local_store.path_for(upload)
end
# compute SHA if missing
if upload.sha1.blank?
upload.sha1 = Upload.generate_digest(path)
end
# store to new location & update the filesize
File.open(path) do |f|
upload.url = Discourse.store.store_upload(f, upload)
upload.filesize = f.size
upload.save!(validate: false)
end
# remap the URLs
DbHelper.remap(UrlHelper.absolute(previous_url), upload.url) unless external
DbHelper.remap(
previous_url,
upload.url,
excluded_tables: %w{
posts
post_search_data
incoming_emails
notifications
single_sign_on_records
stylesheet_cache
topic_search_data
users
user_emails
draft_sequences
optimized_images
}
)
remap_scope ||= begin
Post.with_deleted
.where("raw ~ '/uploads/#{db}/\\d+/' OR raw ~ '/uploads/#{db}/original/(\\d|[a-z])/'")
.select(:id, :raw, :cooked)
.all
end
remap_scope.each do |post|
post.raw.gsub!(previous_url, upload.url)
post.cooked.gsub!(previous_url, upload.url)
Post.with_deleted.where(id: post.id).update_all(raw: post.raw, cooked: post.cooked) if post.changed?
end
upload.optimized_images.find_each(&:destroy!)
upload.rebake_posts_on_old_scheme
# remove the old file (when local)
unless external
FileUtils.rm(path, force: true)
end
rescue => e
problems << { upload: upload, ex: e }
ensure
file&.unlink
file&.close
end
end
end
end
problems
end
private
def short_url_basename
"#{Upload.base62_sha1(sha1)}#{extension.present? ? ".#{extension}" : ""}"
end
2013-02-06 03:16:51 +08:00
end
# == Schema Information
#
# Table name: uploads
#
# id :integer not null, primary key
# user_id :integer not null
# original_filename :string not null
# filesize :integer not null
# width :integer
# height :integer
# url :string not null
# created_at :datetime not null
# updated_at :datetime not null
# sha1 :string(40)
# origin :string(1000)
# retain_hours :integer
# extension :string(10)
# thumbnail_width :integer
# thumbnail_height :integer
# etag :string
# secure :boolean default(FALSE), not null
# access_control_post_id :bigint
# original_sha1 :string
# verification_status :integer default(1), not null
# animated :boolean
# security_last_changed_at :datetime
# security_last_changed_reason :string
#
# Indexes
#
# idx_uploads_on_verification_status (verification_status)
FEATURE: Secure media allowing duplicated uploads with category-level privacy and post-based access rules (#8664) ### General Changes and Duplication * We now consider a post `with_secure_media?` if it is in a read-restricted category. * When uploading we now set an upload's secure status straight away. * When uploading if `SiteSetting.secure_media` is enabled, we do not check to see if the upload already exists using the `sha1` digest of the upload. The `sha1` column of the upload is filled with a `SecureRandom.hex(20)` value which is the same length as `Upload::SHA1_LENGTH`. The `original_sha1` column is filled with the _real_ sha1 digest of the file. * Whether an upload `should_be_secure?` is now determined by whether the `access_control_post` is `with_secure_media?` (if there is no access control post then we leave the secure status as is). * When serializing the upload, we now cook the URL if the upload is secure. This is so it shows up correctly in the composer preview, because we set secure status on upload. ### Viewing Secure Media * The secure-media-upload URL will take the post that the upload is attached to into account via `Guardian.can_see?` for access permissions * If there is no `access_control_post` then we just deliver the media. This should be a rare occurrance and shouldn't cause issues as the `access_control_post` is set when `link_post_uploads` is called via `CookedPostProcessor` ### Removed We no longer do any of these because we do not reuse uploads by sha1 if secure media is enabled. * We no longer have a way to prevent cross-posting of a secure upload from a private context to a public context. * We no longer have to set `secure: false` for uploads when uploading for a theme component.
2020-01-16 11:50:27 +08:00
# index_uploads_on_access_control_post_id (access_control_post_id)
# index_uploads_on_etag (etag)
# index_uploads_on_extension (lower((extension)::text))
# index_uploads_on_id_and_url (id,url)
# index_uploads_on_original_sha1 (original_sha1)
# index_uploads_on_sha1 (sha1) UNIQUE
# index_uploads_on_url (url)
# index_uploads_on_user_id (user_id)
#