discourse/lib/file_store/s3_store.rb

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

383 lines
12 KiB
Ruby
Raw Normal View History

# frozen_string_literal: true
2016-06-30 22:55:01 +08:00
require "uri"
require "mini_mime"
require_dependency "file_store/base_store"
require_dependency "s3_helper"
2014-04-15 19:04:14 +08:00
require_dependency "file_helper"
2013-11-06 02:04:47 +08:00
module FileStore
2013-11-06 02:04:47 +08:00
class S3Store < BaseStore
TOMBSTONE_PREFIX ||= "tombstone/"
def initialize(s3_helper = nil)
@s3_helper = s3_helper
end
def s3_helper
@s3_helper ||= S3Helper.new(s3_bucket,
Rails.configuration.multisite ? multisite_tombstone_prefix : TOMBSTONE_PREFIX
)
end
2015-05-30 00:39:47 +08:00
def store_upload(file, upload, content_type = nil)
upload.url = nil
path = get_path_for_upload(upload)
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
url, upload.etag = store_file(
file,
path,
filename: upload.original_filename,
content_type: content_type,
cache_locally: true,
private_acl: upload.secure?
)
url
2013-11-06 02:04:47 +08:00
end
FEATURE: Initial implementation of direct S3 uploads with uppy and stubs (#13787) This adds a few different things to allow for direct S3 uploads using uppy. **These changes are still not the default.** There are hidden `enable_experimental_image_uploader` and `enable_direct_s3_uploads` settings that must be turned on for any of this code to be used, and even if they are turned on only the User Card Background for the user profile actually uses uppy-image-uploader. A new `ExternalUploadStub` model and database table is introduced in this pull request. This is used to keep track of uploads that are uploaded to a temporary location in S3 with the direct to S3 code, and they are eventually deleted a) when the direct upload is completed and b) after a certain time period of not being used. ### Starting a direct S3 upload When an S3 direct upload is initiated with uppy, we first request a presigned PUT URL from the new `generate-presigned-put` endpoint in `UploadsController`. This generates an S3 key in the `temp` folder inside the correct bucket path, along with any metadata from the clientside (e.g. the SHA1 checksum described below). This will also create an `ExternalUploadStub` and store the details of the temp object key and the file being uploaded. Once the clientside has this URL, uppy will upload the file direct to S3 using the presigned URL. Once the upload is complete we go to the next stage. ### Completing a direct S3 upload Once the upload to S3 is done we call the new `complete-external-upload` route with the unique identifier of the `ExternalUploadStub` created earlier. Only the user who made the stub can complete the external upload. One of two paths is followed via the `ExternalUploadManager`. 1. If the object in S3 is too large (currently 100mb defined by `ExternalUploadManager::DOWNLOAD_LIMIT`) we do not download and generate the SHA1 for that file. Instead we create the `Upload` record via `UploadCreator` and simply copy it to its final destination on S3 then delete the initial temp file. Several modifications to `UploadCreator` have been made to accommodate this. 2. If the object in S3 is small enough, we download it. When the temporary S3 file is downloaded, we compare the SHA1 checksum generated by the browser with the actual SHA1 checksum of the file generated by ruby. The browser SHA1 checksum is stored on the object in S3 with metadata, and is generated via the `UppyChecksum` plugin. Keep in mind that some browsers will not generate this due to compatibility or other issues. We then follow the normal `UploadCreator` path with one exception. To cut down on having to re-upload the file again, if there are no changes (such as resizing etc) to the file in `UploadCreator` we follow the same copy + delete temp path that we do for files that are too large. 3. Finally we return the serialized upload record back to the client There are several errors that could happen that are handled by `UploadsController` as well. Also in this PR is some refactoring of `displayErrorForUpload` to handle both uppy and jquery file uploader errors.
2021-07-28 06:42:25 +08:00
def move_existing_stored_upload(existing_external_upload_key, upload, content_type = nil)
upload.url = nil
path = get_path_for_upload(upload)
url, upload.etag = store_file(
nil,
path,
filename: upload.original_filename,
content_type: content_type,
cache_locally: false,
private_acl: upload.secure?,
move_existing: true,
existing_external_upload_key: existing_external_upload_key
)
url
end
def store_optimized_image(file, optimized_image, content_type = nil, secure: false)
optimized_image.url = nil
path = get_path_for_optimized_image(optimized_image)
url, optimized_image.etag = store_file(file, path, content_type: content_type, private_acl: secure)
url
end
FEATURE: Initial implementation of direct S3 uploads with uppy and stubs (#13787) This adds a few different things to allow for direct S3 uploads using uppy. **These changes are still not the default.** There are hidden `enable_experimental_image_uploader` and `enable_direct_s3_uploads` settings that must be turned on for any of this code to be used, and even if they are turned on only the User Card Background for the user profile actually uses uppy-image-uploader. A new `ExternalUploadStub` model and database table is introduced in this pull request. This is used to keep track of uploads that are uploaded to a temporary location in S3 with the direct to S3 code, and they are eventually deleted a) when the direct upload is completed and b) after a certain time period of not being used. ### Starting a direct S3 upload When an S3 direct upload is initiated with uppy, we first request a presigned PUT URL from the new `generate-presigned-put` endpoint in `UploadsController`. This generates an S3 key in the `temp` folder inside the correct bucket path, along with any metadata from the clientside (e.g. the SHA1 checksum described below). This will also create an `ExternalUploadStub` and store the details of the temp object key and the file being uploaded. Once the clientside has this URL, uppy will upload the file direct to S3 using the presigned URL. Once the upload is complete we go to the next stage. ### Completing a direct S3 upload Once the upload to S3 is done we call the new `complete-external-upload` route with the unique identifier of the `ExternalUploadStub` created earlier. Only the user who made the stub can complete the external upload. One of two paths is followed via the `ExternalUploadManager`. 1. If the object in S3 is too large (currently 100mb defined by `ExternalUploadManager::DOWNLOAD_LIMIT`) we do not download and generate the SHA1 for that file. Instead we create the `Upload` record via `UploadCreator` and simply copy it to its final destination on S3 then delete the initial temp file. Several modifications to `UploadCreator` have been made to accommodate this. 2. If the object in S3 is small enough, we download it. When the temporary S3 file is downloaded, we compare the SHA1 checksum generated by the browser with the actual SHA1 checksum of the file generated by ruby. The browser SHA1 checksum is stored on the object in S3 with metadata, and is generated via the `UppyChecksum` plugin. Keep in mind that some browsers will not generate this due to compatibility or other issues. We then follow the normal `UploadCreator` path with one exception. To cut down on having to re-upload the file again, if there are no changes (such as resizing etc) to the file in `UploadCreator` we follow the same copy + delete temp path that we do for files that are too large. 3. Finally we return the serialized upload record back to the client There are several errors that could happen that are handled by `UploadsController` as well. Also in this PR is some refactoring of `displayErrorForUpload` to handle both uppy and jquery file uploader errors.
2021-07-28 06:42:25 +08:00
# File is an actual Tempfile on disk
#
# An existing_external_upload_key is given for cases where move_existing is specified.
# This is an object already uploaded directly to S3 that we are now moving
# to its final resting place with the correct sha and key.
#
2015-05-30 00:39:47 +08:00
# options
# - filename
# - content_type
# - cache_locally
FEATURE: Initial implementation of direct S3 uploads with uppy and stubs (#13787) This adds a few different things to allow for direct S3 uploads using uppy. **These changes are still not the default.** There are hidden `enable_experimental_image_uploader` and `enable_direct_s3_uploads` settings that must be turned on for any of this code to be used, and even if they are turned on only the User Card Background for the user profile actually uses uppy-image-uploader. A new `ExternalUploadStub` model and database table is introduced in this pull request. This is used to keep track of uploads that are uploaded to a temporary location in S3 with the direct to S3 code, and they are eventually deleted a) when the direct upload is completed and b) after a certain time period of not being used. ### Starting a direct S3 upload When an S3 direct upload is initiated with uppy, we first request a presigned PUT URL from the new `generate-presigned-put` endpoint in `UploadsController`. This generates an S3 key in the `temp` folder inside the correct bucket path, along with any metadata from the clientside (e.g. the SHA1 checksum described below). This will also create an `ExternalUploadStub` and store the details of the temp object key and the file being uploaded. Once the clientside has this URL, uppy will upload the file direct to S3 using the presigned URL. Once the upload is complete we go to the next stage. ### Completing a direct S3 upload Once the upload to S3 is done we call the new `complete-external-upload` route with the unique identifier of the `ExternalUploadStub` created earlier. Only the user who made the stub can complete the external upload. One of two paths is followed via the `ExternalUploadManager`. 1. If the object in S3 is too large (currently 100mb defined by `ExternalUploadManager::DOWNLOAD_LIMIT`) we do not download and generate the SHA1 for that file. Instead we create the `Upload` record via `UploadCreator` and simply copy it to its final destination on S3 then delete the initial temp file. Several modifications to `UploadCreator` have been made to accommodate this. 2. If the object in S3 is small enough, we download it. When the temporary S3 file is downloaded, we compare the SHA1 checksum generated by the browser with the actual SHA1 checksum of the file generated by ruby. The browser SHA1 checksum is stored on the object in S3 with metadata, and is generated via the `UppyChecksum` plugin. Keep in mind that some browsers will not generate this due to compatibility or other issues. We then follow the normal `UploadCreator` path with one exception. To cut down on having to re-upload the file again, if there are no changes (such as resizing etc) to the file in `UploadCreator` we follow the same copy + delete temp path that we do for files that are too large. 3. Finally we return the serialized upload record back to the client There are several errors that could happen that are handled by `UploadsController` as well. Also in this PR is some refactoring of `displayErrorForUpload` to handle both uppy and jquery file uploader errors.
2021-07-28 06:42:25 +08:00
# - move_existing
# - existing_external_upload_key
2015-05-30 00:39:47 +08:00
def store_file(file, path, opts = {})
path = path.dup
filename = opts[:filename].presence || File.basename(path)
2015-05-30 00:39:47 +08:00
# cache file locally when needed
cache_file(file, File.basename(path)) if opts[:cache_locally]
options = {
acl: opts[:private_acl] ? "private" : "public-read",
cache_control: 'max-age=31556952, public, immutable',
content_type: opts[:content_type].presence || MiniMime.lookup_by_filename(filename)&.content_type
}
# add a "content disposition: attachment" header with the original
# filename for everything but safe images (not SVG). audio and video will
# still stream correctly in HTML players, and when a direct link is
# provided to any file but an image it will download correctly in the
# browser.
if !FileHelper.is_inline_image?(filename)
options[:content_disposition] = ActionDispatch::Http::ContentDisposition.format(
disposition: "attachment", filename: filename
)
end
path.prepend(File.join(upload_path, "/")) if Rails.configuration.multisite
# if this fails, it will throw an exception
FEATURE: Initial implementation of direct S3 uploads with uppy and stubs (#13787) This adds a few different things to allow for direct S3 uploads using uppy. **These changes are still not the default.** There are hidden `enable_experimental_image_uploader` and `enable_direct_s3_uploads` settings that must be turned on for any of this code to be used, and even if they are turned on only the User Card Background for the user profile actually uses uppy-image-uploader. A new `ExternalUploadStub` model and database table is introduced in this pull request. This is used to keep track of uploads that are uploaded to a temporary location in S3 with the direct to S3 code, and they are eventually deleted a) when the direct upload is completed and b) after a certain time period of not being used. ### Starting a direct S3 upload When an S3 direct upload is initiated with uppy, we first request a presigned PUT URL from the new `generate-presigned-put` endpoint in `UploadsController`. This generates an S3 key in the `temp` folder inside the correct bucket path, along with any metadata from the clientside (e.g. the SHA1 checksum described below). This will also create an `ExternalUploadStub` and store the details of the temp object key and the file being uploaded. Once the clientside has this URL, uppy will upload the file direct to S3 using the presigned URL. Once the upload is complete we go to the next stage. ### Completing a direct S3 upload Once the upload to S3 is done we call the new `complete-external-upload` route with the unique identifier of the `ExternalUploadStub` created earlier. Only the user who made the stub can complete the external upload. One of two paths is followed via the `ExternalUploadManager`. 1. If the object in S3 is too large (currently 100mb defined by `ExternalUploadManager::DOWNLOAD_LIMIT`) we do not download and generate the SHA1 for that file. Instead we create the `Upload` record via `UploadCreator` and simply copy it to its final destination on S3 then delete the initial temp file. Several modifications to `UploadCreator` have been made to accommodate this. 2. If the object in S3 is small enough, we download it. When the temporary S3 file is downloaded, we compare the SHA1 checksum generated by the browser with the actual SHA1 checksum of the file generated by ruby. The browser SHA1 checksum is stored on the object in S3 with metadata, and is generated via the `UppyChecksum` plugin. Keep in mind that some browsers will not generate this due to compatibility or other issues. We then follow the normal `UploadCreator` path with one exception. To cut down on having to re-upload the file again, if there are no changes (such as resizing etc) to the file in `UploadCreator` we follow the same copy + delete temp path that we do for files that are too large. 3. Finally we return the serialized upload record back to the client There are several errors that could happen that are handled by `UploadsController` as well. Also in this PR is some refactoring of `displayErrorForUpload` to handle both uppy and jquery file uploader errors.
2021-07-28 06:42:25 +08:00
if opts[:move_existing] && opts[:existing_external_upload_key]
path, etag = s3_helper.copy(
opts[:existing_external_upload_key],
path,
options: options
)
s3_helper.delete_object(opts[:existing_external_upload_key])
else
path, etag = s3_helper.upload(file, path, options)
end
# return the upload url and etag
[File.join(absolute_base_url, path), etag]
2013-11-06 02:04:47 +08:00
end
def remove_file(url, path)
2015-05-30 00:39:47 +08:00
return unless has_been_uploaded?(url)
# copy the removed file to tombstone
s3_helper.remove(path, true)
2013-11-06 02:04:47 +08:00
end
2013-08-14 04:08:29 +08:00
2018-08-08 11:26:05 +08:00
def copy_file(url, source, destination)
return unless has_been_uploaded?(url)
s3_helper.copy(source, destination)
2018-08-08 11:26:05 +08:00
end
2013-11-06 02:04:47 +08:00
def has_been_uploaded?(url)
return false if url.blank?
2016-06-30 22:55:01 +08:00
begin
2020-06-17 15:47:05 +08:00
parsed_url = URI.parse(UrlHelper.encode(url))
rescue
# There are many exceptions possible here including Addressable::URI:: exceptions
# and URI:: exceptions, catch all may seem wide, but it makes no sense to raise ever
# on an invalid url here
return false
end
2016-06-30 22:55:01 +08:00
base_hostname = URI.parse(absolute_base_url).hostname
if url[base_hostname]
# if the hostnames match it means the upload is in the same
# bucket on s3. however, the bucket folder path may differ in
# some cases, and we do not want to assume the url is uploaded
# here. e.g. the path of the current site could be /prod and the
# other site could be /staging
if s3_bucket_folder_path.present?
return parsed_url.path.starts_with?("/#{s3_bucket_folder_path}")
else
return true
end
return false
end
2016-06-30 22:55:01 +08:00
return false if SiteSetting.Upload.s3_cdn_url.blank?
cdn_hostname = URI.parse(SiteSetting.Upload.s3_cdn_url || "").hostname
return true if cdn_hostname.presence && url[cdn_hostname]
false
end
def s3_bucket_folder_path
S3Helper.get_bucket_and_folder_path(s3_bucket)[1]
2013-11-06 02:04:47 +08:00
end
2013-08-14 04:08:29 +08:00
def s3_bucket_name
S3Helper.get_bucket_and_folder_path(s3_bucket)[0]
end
2013-11-06 02:04:47 +08:00
def absolute_base_url
@absolute_base_url ||= SiteSetting.Upload.absolute_base_url
2013-11-06 02:04:47 +08:00
end
2013-08-14 04:08:29 +08:00
def s3_upload_host
SiteSetting.Upload.s3_cdn_url.present? ? SiteSetting.Upload.s3_cdn_url : "https:#{absolute_base_url}"
end
2013-11-06 02:04:47 +08:00
def external?
true
end
2013-08-14 04:08:29 +08:00
2013-11-28 05:01:41 +08:00
def purge_tombstone(grace_period)
s3_helper.update_tombstone_lifecycle(grace_period)
2013-11-28 05:01:41 +08:00
end
def multisite_tombstone_prefix
File.join("uploads", "tombstone", RailsMultisite::ConnectionManagement.current_db, "/")
end
def download_url(upload)
return unless upload
"#{upload.short_path}?dl=1"
end
def path_for(upload)
url = upload&.url
FileStore::LocalStore.new.path_for(upload) if url && url[/^\/[^\/]/]
end
def url_for(upload, force_download: false)
upload.secure? || force_download ?
FEATURE: Initial implementation of direct S3 uploads with uppy and stubs (#13787) This adds a few different things to allow for direct S3 uploads using uppy. **These changes are still not the default.** There are hidden `enable_experimental_image_uploader` and `enable_direct_s3_uploads` settings that must be turned on for any of this code to be used, and even if they are turned on only the User Card Background for the user profile actually uses uppy-image-uploader. A new `ExternalUploadStub` model and database table is introduced in this pull request. This is used to keep track of uploads that are uploaded to a temporary location in S3 with the direct to S3 code, and they are eventually deleted a) when the direct upload is completed and b) after a certain time period of not being used. ### Starting a direct S3 upload When an S3 direct upload is initiated with uppy, we first request a presigned PUT URL from the new `generate-presigned-put` endpoint in `UploadsController`. This generates an S3 key in the `temp` folder inside the correct bucket path, along with any metadata from the clientside (e.g. the SHA1 checksum described below). This will also create an `ExternalUploadStub` and store the details of the temp object key and the file being uploaded. Once the clientside has this URL, uppy will upload the file direct to S3 using the presigned URL. Once the upload is complete we go to the next stage. ### Completing a direct S3 upload Once the upload to S3 is done we call the new `complete-external-upload` route with the unique identifier of the `ExternalUploadStub` created earlier. Only the user who made the stub can complete the external upload. One of two paths is followed via the `ExternalUploadManager`. 1. If the object in S3 is too large (currently 100mb defined by `ExternalUploadManager::DOWNLOAD_LIMIT`) we do not download and generate the SHA1 for that file. Instead we create the `Upload` record via `UploadCreator` and simply copy it to its final destination on S3 then delete the initial temp file. Several modifications to `UploadCreator` have been made to accommodate this. 2. If the object in S3 is small enough, we download it. When the temporary S3 file is downloaded, we compare the SHA1 checksum generated by the browser with the actual SHA1 checksum of the file generated by ruby. The browser SHA1 checksum is stored on the object in S3 with metadata, and is generated via the `UppyChecksum` plugin. Keep in mind that some browsers will not generate this due to compatibility or other issues. We then follow the normal `UploadCreator` path with one exception. To cut down on having to re-upload the file again, if there are no changes (such as resizing etc) to the file in `UploadCreator` we follow the same copy + delete temp path that we do for files that are too large. 3. Finally we return the serialized upload record back to the client There are several errors that could happen that are handled by `UploadsController` as well. Also in this PR is some refactoring of `displayErrorForUpload` to handle both uppy and jquery file uploader errors.
2021-07-28 06:42:25 +08:00
presigned_get_url(get_upload_key(upload), force_download: force_download, filename: upload.original_filename) :
upload.url
end
FEATURE: Initial implementation of direct S3 uploads with uppy and stubs (#13787) This adds a few different things to allow for direct S3 uploads using uppy. **These changes are still not the default.** There are hidden `enable_experimental_image_uploader` and `enable_direct_s3_uploads` settings that must be turned on for any of this code to be used, and even if they are turned on only the User Card Background for the user profile actually uses uppy-image-uploader. A new `ExternalUploadStub` model and database table is introduced in this pull request. This is used to keep track of uploads that are uploaded to a temporary location in S3 with the direct to S3 code, and they are eventually deleted a) when the direct upload is completed and b) after a certain time period of not being used. ### Starting a direct S3 upload When an S3 direct upload is initiated with uppy, we first request a presigned PUT URL from the new `generate-presigned-put` endpoint in `UploadsController`. This generates an S3 key in the `temp` folder inside the correct bucket path, along with any metadata from the clientside (e.g. the SHA1 checksum described below). This will also create an `ExternalUploadStub` and store the details of the temp object key and the file being uploaded. Once the clientside has this URL, uppy will upload the file direct to S3 using the presigned URL. Once the upload is complete we go to the next stage. ### Completing a direct S3 upload Once the upload to S3 is done we call the new `complete-external-upload` route with the unique identifier of the `ExternalUploadStub` created earlier. Only the user who made the stub can complete the external upload. One of two paths is followed via the `ExternalUploadManager`. 1. If the object in S3 is too large (currently 100mb defined by `ExternalUploadManager::DOWNLOAD_LIMIT`) we do not download and generate the SHA1 for that file. Instead we create the `Upload` record via `UploadCreator` and simply copy it to its final destination on S3 then delete the initial temp file. Several modifications to `UploadCreator` have been made to accommodate this. 2. If the object in S3 is small enough, we download it. When the temporary S3 file is downloaded, we compare the SHA1 checksum generated by the browser with the actual SHA1 checksum of the file generated by ruby. The browser SHA1 checksum is stored on the object in S3 with metadata, and is generated via the `UppyChecksum` plugin. Keep in mind that some browsers will not generate this due to compatibility or other issues. We then follow the normal `UploadCreator` path with one exception. To cut down on having to re-upload the file again, if there are no changes (such as resizing etc) to the file in `UploadCreator` we follow the same copy + delete temp path that we do for files that are too large. 3. Finally we return the serialized upload record back to the client There are several errors that could happen that are handled by `UploadsController` as well. Also in this PR is some refactoring of `displayErrorForUpload` to handle both uppy and jquery file uploader errors.
2021-07-28 06:42:25 +08:00
def path_from_url(url)
URI.parse(url).path.delete_prefix("/")
end
def cdn_url(url)
return url if SiteSetting.Upload.s3_cdn_url.blank?
2016-06-30 22:55:01 +08:00
schema = url[/^(https?:)?\/\//, 1]
folder = s3_bucket_folder_path.nil? ? "" : "#{s3_bucket_folder_path}/"
url.sub(File.join("#{schema}#{absolute_base_url}", folder), File.join(SiteSetting.Upload.s3_cdn_url, "/"))
end
def signed_url_for_path(path, expires_in: S3Helper::DOWNLOAD_URL_EXPIRES_AFTER_SECONDS, force_download: false)
key = path.sub(absolute_base_url + "/", "")
FEATURE: Initial implementation of direct S3 uploads with uppy and stubs (#13787) This adds a few different things to allow for direct S3 uploads using uppy. **These changes are still not the default.** There are hidden `enable_experimental_image_uploader` and `enable_direct_s3_uploads` settings that must be turned on for any of this code to be used, and even if they are turned on only the User Card Background for the user profile actually uses uppy-image-uploader. A new `ExternalUploadStub` model and database table is introduced in this pull request. This is used to keep track of uploads that are uploaded to a temporary location in S3 with the direct to S3 code, and they are eventually deleted a) when the direct upload is completed and b) after a certain time period of not being used. ### Starting a direct S3 upload When an S3 direct upload is initiated with uppy, we first request a presigned PUT URL from the new `generate-presigned-put` endpoint in `UploadsController`. This generates an S3 key in the `temp` folder inside the correct bucket path, along with any metadata from the clientside (e.g. the SHA1 checksum described below). This will also create an `ExternalUploadStub` and store the details of the temp object key and the file being uploaded. Once the clientside has this URL, uppy will upload the file direct to S3 using the presigned URL. Once the upload is complete we go to the next stage. ### Completing a direct S3 upload Once the upload to S3 is done we call the new `complete-external-upload` route with the unique identifier of the `ExternalUploadStub` created earlier. Only the user who made the stub can complete the external upload. One of two paths is followed via the `ExternalUploadManager`. 1. If the object in S3 is too large (currently 100mb defined by `ExternalUploadManager::DOWNLOAD_LIMIT`) we do not download and generate the SHA1 for that file. Instead we create the `Upload` record via `UploadCreator` and simply copy it to its final destination on S3 then delete the initial temp file. Several modifications to `UploadCreator` have been made to accommodate this. 2. If the object in S3 is small enough, we download it. When the temporary S3 file is downloaded, we compare the SHA1 checksum generated by the browser with the actual SHA1 checksum of the file generated by ruby. The browser SHA1 checksum is stored on the object in S3 with metadata, and is generated via the `UppyChecksum` plugin. Keep in mind that some browsers will not generate this due to compatibility or other issues. We then follow the normal `UploadCreator` path with one exception. To cut down on having to re-upload the file again, if there are no changes (such as resizing etc) to the file in `UploadCreator` we follow the same copy + delete temp path that we do for files that are too large. 3. Finally we return the serialized upload record back to the client There are several errors that could happen that are handled by `UploadsController` as well. Also in this PR is some refactoring of `displayErrorForUpload` to handle both uppy and jquery file uploader errors.
2021-07-28 06:42:25 +08:00
presigned_get_url(key, expires_in: expires_in, force_download: force_download)
end
def signed_url_for_temporary_upload(file_name, expires_in: S3Helper::UPLOAD_URL_EXPIRES_AFTER_SECONDS, metadata: {})
key = temporary_upload_path(file_name)
presigned_put_url(key, expires_in: expires_in, metadata: metadata)
end
def temporary_upload_path(file_name)
path = super(file_name)
s3_bucket_folder_path.nil? ? path : File.join(s3_bucket_folder_path, path)
end
def object_from_path(path)
s3_helper.object(path)
end
2015-05-30 00:39:47 +08:00
def cache_avatar(avatar, user_id)
source = avatar.url.sub(absolute_base_url + "/", "")
destination = avatar_template(avatar, user_id).sub(absolute_base_url + "/", "")
s3_helper.copy(source, destination)
2015-05-30 00:39:47 +08:00
end
2013-11-28 05:01:41 +08:00
2015-05-30 00:39:47 +08:00
def avatar_template(avatar, user_id)
UserAvatar.external_avatar_url(user_id, avatar.upload_id, avatar.width)
end
def s3_bucket
raise Discourse::SiteSettingMissing.new("s3_upload_bucket") if SiteSetting.Upload.s3_upload_bucket.blank?
SiteSetting.Upload.s3_upload_bucket.downcase
2015-05-30 00:39:47 +08:00
end
def list_missing_uploads(skip_optimized: false)
if SiteSetting.enable_s3_inventory
require 's3_inventory'
S3Inventory.new(s3_helper, :upload).backfill_etags_and_list_missing
S3Inventory.new(s3_helper, :optimized).backfill_etags_and_list_missing unless skip_optimized
else
2019-03-13 17:39:07 +08:00
list_missing(Upload.by_users, "original/")
list_missing(OptimizedImage, "optimized/") unless skip_optimized
end
end
def update_upload_ACL(upload)
key = get_upload_key(upload)
update_ACL(key, upload.secure?)
upload.optimized_images.find_each do |optimized_image|
optimized_image_key = get_path_for_optimized_image(optimized_image)
update_ACL(optimized_image_key, upload.secure?)
end
true
end
def download_file(upload, destination_path)
s3_helper.download_file(get_upload_key(upload), destination_path)
end
def copy_from(source_path)
local_store = FileStore::LocalStore.new
public_upload_path = File.join(local_store.public_dir, local_store.upload_path)
# The migration to S3 and lots of other code expects files to exist in public/uploads,
# so lets move them there before executing the migration.
if public_upload_path != source_path
if Dir.exist?(public_upload_path)
old_upload_path = "#{public_upload_path}_#{SecureRandom.hex}"
FileUtils.mv(public_upload_path, old_upload_path)
end
end
FileUtils.mkdir_p(File.expand_path("..", public_upload_path))
FileUtils.symlink(source_path, public_upload_path)
FileStore::ToS3Migration.new(
s3_options: FileStore::ToS3Migration.s3_options_from_site_settings,
migrate_to_multisite: Rails.configuration.multisite,
).migrate
ensure
FileUtils.rm(public_upload_path) if File.symlink?(public_upload_path)
FileUtils.mv(old_upload_path, public_upload_path) if old_upload_path
end
private
FEATURE: Initial implementation of direct S3 uploads with uppy and stubs (#13787) This adds a few different things to allow for direct S3 uploads using uppy. **These changes are still not the default.** There are hidden `enable_experimental_image_uploader` and `enable_direct_s3_uploads` settings that must be turned on for any of this code to be used, and even if they are turned on only the User Card Background for the user profile actually uses uppy-image-uploader. A new `ExternalUploadStub` model and database table is introduced in this pull request. This is used to keep track of uploads that are uploaded to a temporary location in S3 with the direct to S3 code, and they are eventually deleted a) when the direct upload is completed and b) after a certain time period of not being used. ### Starting a direct S3 upload When an S3 direct upload is initiated with uppy, we first request a presigned PUT URL from the new `generate-presigned-put` endpoint in `UploadsController`. This generates an S3 key in the `temp` folder inside the correct bucket path, along with any metadata from the clientside (e.g. the SHA1 checksum described below). This will also create an `ExternalUploadStub` and store the details of the temp object key and the file being uploaded. Once the clientside has this URL, uppy will upload the file direct to S3 using the presigned URL. Once the upload is complete we go to the next stage. ### Completing a direct S3 upload Once the upload to S3 is done we call the new `complete-external-upload` route with the unique identifier of the `ExternalUploadStub` created earlier. Only the user who made the stub can complete the external upload. One of two paths is followed via the `ExternalUploadManager`. 1. If the object in S3 is too large (currently 100mb defined by `ExternalUploadManager::DOWNLOAD_LIMIT`) we do not download and generate the SHA1 for that file. Instead we create the `Upload` record via `UploadCreator` and simply copy it to its final destination on S3 then delete the initial temp file. Several modifications to `UploadCreator` have been made to accommodate this. 2. If the object in S3 is small enough, we download it. When the temporary S3 file is downloaded, we compare the SHA1 checksum generated by the browser with the actual SHA1 checksum of the file generated by ruby. The browser SHA1 checksum is stored on the object in S3 with metadata, and is generated via the `UppyChecksum` plugin. Keep in mind that some browsers will not generate this due to compatibility or other issues. We then follow the normal `UploadCreator` path with one exception. To cut down on having to re-upload the file again, if there are no changes (such as resizing etc) to the file in `UploadCreator` we follow the same copy + delete temp path that we do for files that are too large. 3. Finally we return the serialized upload record back to the client There are several errors that could happen that are handled by `UploadsController` as well. Also in this PR is some refactoring of `displayErrorForUpload` to handle both uppy and jquery file uploader errors.
2021-07-28 06:42:25 +08:00
def presigned_put_url(key, expires_in: S3Helper::UPLOAD_URL_EXPIRES_AFTER_SECONDS, metadata: {})
signer = Aws::S3::Presigner.new(client: s3_helper.s3_client)
signer.presigned_url(
:put_object,
bucket: s3_bucket_name,
key: key,
acl: "private",
expires_in: expires_in,
metadata: metadata
)
end
def presigned_get_url(
url,
force_download: false,
filename: false,
expires_in: S3Helper::DOWNLOAD_URL_EXPIRES_AFTER_SECONDS
)
opts = { expires_in: expires_in }
if force_download && filename
opts[:response_content_disposition] = ActionDispatch::Http::ContentDisposition.format(
disposition: "attachment", filename: filename
)
end
FEATURE: Initial implementation of direct S3 uploads with uppy and stubs (#13787) This adds a few different things to allow for direct S3 uploads using uppy. **These changes are still not the default.** There are hidden `enable_experimental_image_uploader` and `enable_direct_s3_uploads` settings that must be turned on for any of this code to be used, and even if they are turned on only the User Card Background for the user profile actually uses uppy-image-uploader. A new `ExternalUploadStub` model and database table is introduced in this pull request. This is used to keep track of uploads that are uploaded to a temporary location in S3 with the direct to S3 code, and they are eventually deleted a) when the direct upload is completed and b) after a certain time period of not being used. ### Starting a direct S3 upload When an S3 direct upload is initiated with uppy, we first request a presigned PUT URL from the new `generate-presigned-put` endpoint in `UploadsController`. This generates an S3 key in the `temp` folder inside the correct bucket path, along with any metadata from the clientside (e.g. the SHA1 checksum described below). This will also create an `ExternalUploadStub` and store the details of the temp object key and the file being uploaded. Once the clientside has this URL, uppy will upload the file direct to S3 using the presigned URL. Once the upload is complete we go to the next stage. ### Completing a direct S3 upload Once the upload to S3 is done we call the new `complete-external-upload` route with the unique identifier of the `ExternalUploadStub` created earlier. Only the user who made the stub can complete the external upload. One of two paths is followed via the `ExternalUploadManager`. 1. If the object in S3 is too large (currently 100mb defined by `ExternalUploadManager::DOWNLOAD_LIMIT`) we do not download and generate the SHA1 for that file. Instead we create the `Upload` record via `UploadCreator` and simply copy it to its final destination on S3 then delete the initial temp file. Several modifications to `UploadCreator` have been made to accommodate this. 2. If the object in S3 is small enough, we download it. When the temporary S3 file is downloaded, we compare the SHA1 checksum generated by the browser with the actual SHA1 checksum of the file generated by ruby. The browser SHA1 checksum is stored on the object in S3 with metadata, and is generated via the `UppyChecksum` plugin. Keep in mind that some browsers will not generate this due to compatibility or other issues. We then follow the normal `UploadCreator` path with one exception. To cut down on having to re-upload the file again, if there are no changes (such as resizing etc) to the file in `UploadCreator` we follow the same copy + delete temp path that we do for files that are too large. 3. Finally we return the serialized upload record back to the client There are several errors that could happen that are handled by `UploadsController` as well. Also in this PR is some refactoring of `displayErrorForUpload` to handle both uppy and jquery file uploader errors.
2021-07-28 06:42:25 +08:00
obj = object_from_path(url)
obj.presigned_url(:get, opts)
end
def get_upload_key(upload)
if Rails.configuration.multisite
File.join(upload_path, "/", get_path_for_upload(upload))
else
get_path_for_upload(upload)
end
end
def update_ACL(key, secure)
begin
FEATURE: Initial implementation of direct S3 uploads with uppy and stubs (#13787) This adds a few different things to allow for direct S3 uploads using uppy. **These changes are still not the default.** There are hidden `enable_experimental_image_uploader` and `enable_direct_s3_uploads` settings that must be turned on for any of this code to be used, and even if they are turned on only the User Card Background for the user profile actually uses uppy-image-uploader. A new `ExternalUploadStub` model and database table is introduced in this pull request. This is used to keep track of uploads that are uploaded to a temporary location in S3 with the direct to S3 code, and they are eventually deleted a) when the direct upload is completed and b) after a certain time period of not being used. ### Starting a direct S3 upload When an S3 direct upload is initiated with uppy, we first request a presigned PUT URL from the new `generate-presigned-put` endpoint in `UploadsController`. This generates an S3 key in the `temp` folder inside the correct bucket path, along with any metadata from the clientside (e.g. the SHA1 checksum described below). This will also create an `ExternalUploadStub` and store the details of the temp object key and the file being uploaded. Once the clientside has this URL, uppy will upload the file direct to S3 using the presigned URL. Once the upload is complete we go to the next stage. ### Completing a direct S3 upload Once the upload to S3 is done we call the new `complete-external-upload` route with the unique identifier of the `ExternalUploadStub` created earlier. Only the user who made the stub can complete the external upload. One of two paths is followed via the `ExternalUploadManager`. 1. If the object in S3 is too large (currently 100mb defined by `ExternalUploadManager::DOWNLOAD_LIMIT`) we do not download and generate the SHA1 for that file. Instead we create the `Upload` record via `UploadCreator` and simply copy it to its final destination on S3 then delete the initial temp file. Several modifications to `UploadCreator` have been made to accommodate this. 2. If the object in S3 is small enough, we download it. When the temporary S3 file is downloaded, we compare the SHA1 checksum generated by the browser with the actual SHA1 checksum of the file generated by ruby. The browser SHA1 checksum is stored on the object in S3 with metadata, and is generated via the `UppyChecksum` plugin. Keep in mind that some browsers will not generate this due to compatibility or other issues. We then follow the normal `UploadCreator` path with one exception. To cut down on having to re-upload the file again, if there are no changes (such as resizing etc) to the file in `UploadCreator` we follow the same copy + delete temp path that we do for files that are too large. 3. Finally we return the serialized upload record back to the client There are several errors that could happen that are handled by `UploadsController` as well. Also in this PR is some refactoring of `displayErrorForUpload` to handle both uppy and jquery file uploader errors.
2021-07-28 06:42:25 +08:00
object_from_path(key).acl.put(acl: secure ? "private" : "public-read")
rescue Aws::S3::Errors::NoSuchKey
Rails.logger.warn("Could not update ACL on upload with key: '#{key}'. Upload is missing.")
end
end
def list_missing(model, prefix)
connection = ActiveRecord::Base.connection.raw_connection
connection.exec('CREATE TEMP TABLE verified_ids(val integer PRIMARY KEY)')
marker = nil
files = s3_helper.list(prefix, marker)
while files.count > 0 do
verified_ids = []
files.each do |f|
id = model.where("url LIKE '%#{f.key}' AND etag = '#{f.etag}'").pluck_first(:id)
verified_ids << id if id.present?
marker = f.key
end
verified_id_clause = verified_ids.map { |id| "('#{PG::Connection.escape_string(id.to_s)}')" }.join(",")
connection.exec("INSERT INTO verified_ids VALUES #{verified_id_clause}")
files = s3_helper.list(prefix, marker)
end
missing_uploads = model.joins('LEFT JOIN verified_ids ON verified_ids.val = id').where("verified_ids.val IS NULL")
missing_count = missing_uploads.count
if missing_count > 0
missing_uploads.find_each do |upload|
puts upload.url
end
puts "#{missing_count} of #{model.count} #{model.name.underscore.pluralize} are missing"
end
ensure
2018-11-27 03:45:29 +08:00
connection.exec('DROP TABLE verified_ids') unless connection.nil?
end
end
end