discourse/spec/multisite/s3_store_spec.rb

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

361 lines
14 KiB
Ruby
Raw Normal View History

# frozen_string_literal: true
require 'file_store/s3_store'
RSpec.describe 'Multisite s3 uploads', type: :multisite do
let(:original_filename) { "smallest.png" }
let(:uploaded_file) { file_from_fixtures(original_filename) }
let(:upload_sha1) { Digest::SHA1.hexdigest(File.read(uploaded_file)) }
let(:upload_path) { Discourse.store.upload_path }
def build_upload
Fabricate.build(:upload, sha1: upload_sha1, id: 1, original_filename: original_filename)
end
context 'uploading to s3' do
before(:each) do
setup_s3
end
describe "#store_upload" do
let(:s3_client) { Aws::S3::Client.new(stub_responses: true) }
let(:s3_helper) { S3Helper.new(SiteSetting.s3_upload_bucket, '', client: s3_client) }
let(:store) { FileStore::S3Store.new(s3_helper) }
let(:upload_opts) do
{
acl: "public-read",
cache_control: "max-age=31556952, public, immutable",
content_type: "image/png"
}
end
it "does not provide a content_disposition for images" do
s3_helper.expects(:upload).with(uploaded_file, kind_of(String), upload_opts).returns(["path", "etag"])
upload = build_upload
store.store_upload(uploaded_file, upload)
end
context "when the file is a PDF" do
let(:original_filename) { "small.pdf" }
let(:uploaded_file) { file_from_fixtures("small.pdf", "pdf") }
it "adds an attachment content-disposition with the original filename" do
disp_opts = { content_disposition: "attachment; filename=\"#{original_filename}\"; filename*=UTF-8''#{original_filename}", content_type: "application/pdf" }
s3_helper.expects(:upload).with(uploaded_file, kind_of(String), upload_opts.merge(disp_opts)).returns(["path", "etag"])
upload = build_upload
store.store_upload(uploaded_file, upload)
end
end
context "when the file is a video" do
let(:original_filename) { "small.mp4" }
let(:uploaded_file) { file_from_fixtures("small.mp4", "media") }
it "adds an attachment content-disposition with the original filename" do
disp_opts = { content_disposition: "attachment; filename=\"#{original_filename}\"; filename*=UTF-8''#{original_filename}", content_type: "application/mp4" }
s3_helper.expects(:upload).with(uploaded_file, kind_of(String), upload_opts.merge(disp_opts)).returns(["path", "etag"])
upload = build_upload
store.store_upload(uploaded_file, upload)
end
end
context "when the file is audio" do
let(:original_filename) { "small.mp3" }
let(:uploaded_file) { file_from_fixtures("small.mp3", "media") }
it "adds an attachment content-disposition with the original filename" do
disp_opts = { content_disposition: "attachment; filename=\"#{original_filename}\"; filename*=UTF-8''#{original_filename}", content_type: "audio/mpeg" }
s3_helper.expects(:upload).with(uploaded_file, kind_of(String), upload_opts.merge(disp_opts)).returns(["path", "etag"])
upload = build_upload
store.store_upload(uploaded_file, upload)
end
end
it "returns the correct url for default and second multisite db" do
test_multisite_connection('default') do
upload = build_upload
expect(store.store_upload(uploaded_file, upload)).to eq(
"//#{SiteSetting.s3_upload_bucket}.s3.dualstack.us-west-1.amazonaws.com/#{upload_path}/original/1X/c530c06cf89c410c0355d7852644a73fc3ec8c04.png"
)
expect(upload.etag).to eq("ETag")
end
test_multisite_connection('second') do
upload_path = Discourse.store.upload_path
upload = build_upload
expect(store.store_upload(uploaded_file, upload)).to eq(
"//#{SiteSetting.s3_upload_bucket}.s3.dualstack.us-west-1.amazonaws.com/#{upload_path}/original/1X/c530c06cf89c410c0355d7852644a73fc3ec8c04.png"
)
expect(upload.etag).to eq("ETag")
end
end
end
end
context 'removal from s3' do
before do
setup_s3
end
describe "#remove_upload" do
let(:store) { FileStore::S3Store.new }
let(:upload) { build_upload }
let(:upload_key) { "#{upload_path}/original/1X/#{upload.sha1}.png" }
def prepare_fake_s3
@fake_s3 = FakeS3.create
bucket = @fake_s3.bucket(SiteSetting.s3_upload_bucket)
bucket.put_object(
key: upload_key,
size: upload.filesize,
last_modified: upload.created_at
)
bucket
end
it "removes the file from s3 on multisite" do
test_multisite_connection('default') do
upload.update!(url: "//s3-upload-bucket.s3.dualstack.us-west-1.amazonaws.com/#{upload_path}/original/1X/#{upload.sha1}.png")
tombstone_key = "uploads/tombstone/default/original/1X/#{upload.sha1}.png"
bucket = prepare_fake_s3
expect(bucket.find_object(upload_key)).to be_present
expect(bucket.find_object(tombstone_key)).to be_nil
store.remove_upload(upload)
expect(bucket.find_object(upload_key)).to be_nil
expect(bucket.find_object(tombstone_key)).to be_present
end
end
it "removes the file from s3 on another multisite db" do
test_multisite_connection('second') do
upload.update!(url: "//s3-upload-bucket.s3.dualstack.us-west-1.amazonaws.com/#{upload_path}/original/1X/#{upload.sha1}.png")
tombstone_key = "uploads/tombstone/second/original/1X/#{upload.sha1}.png"
bucket = prepare_fake_s3
expect(bucket.find_object(upload_key)).to be_present
expect(bucket.find_object(tombstone_key)).to be_nil
store.remove_upload(upload)
expect(bucket.find_object(upload_key)).to be_nil
expect(bucket.find_object(tombstone_key)).to be_present
end
end
describe "when s3_upload_bucket includes folders path" do
let(:upload_key) { "discourse-uploads/#{upload_path}/original/1X/#{upload.sha1}.png" }
before do
SiteSetting.s3_upload_bucket = "s3-upload-bucket/discourse-uploads"
end
it "removes the file from s3 on multisite" do
test_multisite_connection('default') do
upload.update!(url: "//s3-upload-bucket.s3.dualstack.us-west-1.amazonaws.com/discourse-uploads/#{upload_path}/original/1X/#{upload.sha1}.png")
tombstone_key = "discourse-uploads/uploads/tombstone/default/original/1X/#{upload.sha1}.png"
bucket = prepare_fake_s3
expect(bucket.find_object(upload_key)).to be_present
expect(bucket.find_object(tombstone_key)).to be_nil
store.remove_upload(upload)
expect(bucket.find_object(upload_key)).to be_nil
expect(bucket.find_object(tombstone_key)).to be_present
end
end
end
end
end
context 'secure uploads' do
let(:store) { FileStore::S3Store.new }
let(:client) { Aws::S3::Client.new(stub_responses: true) }
let(:resource) { Aws::S3::Resource.new(client: client) }
let(:s3_bucket) { resource.bucket("some-really-cool-bucket") }
let(:s3_helper) { store.s3_helper }
let(:s3_object) { stub }
before(:each) do
setup_s3
SiteSetting.s3_upload_bucket = "some-really-cool-bucket"
SiteSetting.authorized_extensions = "pdf|png|jpg|gif"
end
before do
s3_object.stubs(:put).returns(Aws::S3::Types::PutObjectOutput.new(etag: "etag"))
end
describe "when secure attachments are enabled" do
it "returns signed URL with correct path" do
test_multisite_connection('default') do
upload = Fabricate(:upload, original_filename: "small.pdf", extension: "pdf", secure: true)
path = Discourse.store.get_path_for_upload(upload)
s3_helper.expects(:s3_bucket).returns(s3_bucket).at_least_once
s3_bucket.expects(:object).with("#{upload_path}/#{path}").returns(s3_object).at_least_once
s3_object.expects(:presigned_url).with(:get, expires_in: SiteSetting.s3_presigned_get_url_expires_after_seconds)
upload.url = store.store_upload(uploaded_file, upload)
expect(upload.url).to eq(
"//some-really-cool-bucket.s3.dualstack.us-west-1.amazonaws.com/#{upload_path}/#{path}"
)
expect(store.url_for(upload)).not_to eq(upload.url)
end
end
end
describe "when secure media are enabled" do
before do
SiteSetting.login_required = true
SiteSetting.secure_media = true
s3_helper.stubs(:s3_client).returns(client)
Discourse.stubs(:store).returns(store)
end
it "returns signed URL with correct path" do
test_multisite_connection('default') do
upload = Fabricate.build(:upload_s3, sha1: upload_sha1, id: 1)
signed_url = Discourse.store.signed_url_for_path(upload.url)
expect(signed_url).to match(/Amz-Expires/)
expect(signed_url).to match("#{upload_path}")
end
test_multisite_connection('second') do
upload_path = Discourse.store.upload_path
upload = Fabricate.build(:upload_s3, sha1: upload_sha1, id: 1)
signed_url = Discourse.store.signed_url_for_path(upload.url)
expect(signed_url).to match(/Amz-Expires/)
expect(signed_url).to match("#{upload_path}")
end
end
end
describe "#update_upload_ACL" do
it "updates correct file for default and second multisite db" do
test_multisite_connection('default') do
upload = build_upload
upload.update!(original_filename: "small.pdf", extension: "pdf", secure: true)
s3_helper.expects(:s3_bucket).returns(s3_bucket).at_least_once
s3_bucket.expects(:object).with("#{upload_path}/original/1X/#{upload.sha1}.pdf").returns(s3_object)
s3_object.expects(:acl).returns(s3_object)
s3_object.expects(:put).with(acl: "private").returns(s3_object)
expect(store.update_upload_ACL(upload)).to be_truthy
end
test_multisite_connection('second') do
upload_path = Discourse.store.upload_path
upload = build_upload
upload.update!(original_filename: "small.pdf", extension: "pdf", secure: true)
s3_helper.expects(:s3_bucket).returns(s3_bucket).at_least_once
s3_bucket.expects(:object).with("#{upload_path}/original/1X/#{upload.sha1}.pdf").returns(s3_object)
s3_object.expects(:acl).returns(s3_object)
s3_object.expects(:put).with(acl: "private").returns(s3_object)
expect(store.update_upload_ACL(upload)).to be_truthy
end
end
end
end
describe "#has_been_uploaded?" do
before do
setup_s3
SiteSetting.s3_upload_bucket = "s3-upload-bucket/test"
end
let(:store) { FileStore::S3Store.new }
it "returns false for blank urls and bad urls" do
expect(store.has_been_uploaded?("")).to eq(false)
expect(store.has_been_uploaded?("http://test@test.com:test/test.git")).to eq(false)
expect(store.has_been_uploaded?("http:///+test@test.com/test.git")).to eq(false)
end
it "returns true if the base hostname is the same for both urls" do
url = "https://s3-upload-bucket.s3.dualstack.us-west-1.amazonaws.com/test/original/2X/d/dd7964f5fd13e1103c5244ca30abe1936c0a4b88.png"
expect(store.has_been_uploaded?(url)).to eq(true)
end
it "returns false if the base hostname is the same for both urls BUT the bucket name is different in the path" do
bucket = "someotherbucket"
url = "https://s3-upload-bucket.s3.dualstack.us-west-1.amazonaws.com/#{bucket}/original/2X/d/dd7964f5fd13e1103c5244ca30abe1936c0a4b88.png"
expect(store.has_been_uploaded?(url)).to eq(false)
end
it "returns false if the hostnames do not match and the s3_cdn_url is blank" do
url = "https://www.someotherhostname.com/test/original/2X/d/dd7964f5fd13e1103c5244ca30abe1936c0a4b88.png"
expect(store.has_been_uploaded?(url)).to eq(false)
end
it "returns true if the s3_cdn_url is present and matches the url hostname" do
SiteSetting.s3_cdn_url = "https://www.someotherhostname.com"
url = "https://www.someotherhostname.com/test/original/2X/d/dd7964f5fd13e1103c5244ca30abe1936c0a4b88.png"
expect(store.has_been_uploaded?(url)).to eq(true)
end
it "returns false if the URI is an invalid mailto link" do
link = 'mailto: roman;@test.com'
expect(store.has_been_uploaded?(link)).to eq(false)
end
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
describe "#signed_url_for_temporary_upload" do
before do
setup_s3
end
let(:store) { FileStore::S3Store.new }
context "for a bucket with no folder path" do
before { SiteSetting.s3_upload_bucket = "s3-upload-bucket" }
it "returns a presigned url with the correct params and the key for the temporary file" do
url = store.signed_url_for_temporary_upload("test.png")
FEATURE: Direct S3 multipart uploads for backups (#14736) This PR introduces a new `enable_experimental_backup_uploads` site setting (default false and hidden), which when enabled alongside `enable_direct_s3_uploads` will allow for direct S3 multipart uploads of backup .tar.gz files. To make multipart external uploads work with both the S3BackupStore and the S3Store, I've had to move several methods out of S3Store and into S3Helper, including: * presigned_url * create_multipart * abort_multipart * complete_multipart * presign_multipart_part * list_multipart_parts Then, S3Store and S3BackupStore either delegate directly to S3Helper or have their own special methods to call S3Helper for these methods. FileStore.temporary_upload_path has also removed its dependence on upload_path, and can now be used interchangeably between the stores. A similar change was made in the frontend as well, moving the multipart related JS code out of ComposerUppyUpload and into a mixin of its own, so it can also be used by UppyUploadMixin. Some changes to ExternalUploadManager had to be made here as well. The backup direct uploads do not need an Upload record made for them in the database, so they can be moved to their final S3 resting place when completing the multipart upload. This changeset is not perfect; it introduces some special cases in UploadController to handle backups that was previously in BackupController, because UploadController is where the multipart routes are located. A subsequent pull request will pull these routes into a module or some other sharing pattern, along with hooks, so the backup controller and the upload controller (and any future controllers that may need them) can include these routes in a nicer way.
2021-11-11 06:25:31 +08:00
key = store.s3_helper.path_from_url(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
expect(url).to match(/Amz-Expires/)
expect(key).to match(/temp\/uploads\/default\/test_[0-9]\/[a-zA-z0-9]{0,32}\/[a-zA-z0-9]{0,32}.png/)
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
end
it "presigned url contans the metadata when provided" do
url = store.signed_url_for_temporary_upload("test.png", metadata: { "test-meta": "testing" })
expect(url).to include("&x-amz-meta-test-meta=testing")
end
end
context "for a bucket with a folder path" do
before { SiteSetting.s3_upload_bucket = "s3-upload-bucket/site" }
it "returns a presigned url with the correct params and the key for the temporary file" do
url = store.signed_url_for_temporary_upload("test.png")
FEATURE: Direct S3 multipart uploads for backups (#14736) This PR introduces a new `enable_experimental_backup_uploads` site setting (default false and hidden), which when enabled alongside `enable_direct_s3_uploads` will allow for direct S3 multipart uploads of backup .tar.gz files. To make multipart external uploads work with both the S3BackupStore and the S3Store, I've had to move several methods out of S3Store and into S3Helper, including: * presigned_url * create_multipart * abort_multipart * complete_multipart * presign_multipart_part * list_multipart_parts Then, S3Store and S3BackupStore either delegate directly to S3Helper or have their own special methods to call S3Helper for these methods. FileStore.temporary_upload_path has also removed its dependence on upload_path, and can now be used interchangeably between the stores. A similar change was made in the frontend as well, moving the multipart related JS code out of ComposerUppyUpload and into a mixin of its own, so it can also be used by UppyUploadMixin. Some changes to ExternalUploadManager had to be made here as well. The backup direct uploads do not need an Upload record made for them in the database, so they can be moved to their final S3 resting place when completing the multipart upload. This changeset is not perfect; it introduces some special cases in UploadController to handle backups that was previously in BackupController, because UploadController is where the multipart routes are located. A subsequent pull request will pull these routes into a module or some other sharing pattern, along with hooks, so the backup controller and the upload controller (and any future controllers that may need them) can include these routes in a nicer way.
2021-11-11 06:25:31 +08:00
key = store.s3_helper.path_from_url(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
expect(url).to match(/Amz-Expires/)
expect(key).to match(/temp\/site\/uploads\/default\/test_[0-9]\/[a-zA-z0-9]{0,32}\/[a-zA-z0-9]{0,32}.png/)
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
end
end
context "for a multisite site" do
before { SiteSetting.s3_upload_bucket = "s3-upload-bucket/standard99" }
it "returns a presigned url with the correct params and the key for the temporary file" do
test_multisite_connection('second') do
url = store.signed_url_for_temporary_upload("test.png")
FEATURE: Direct S3 multipart uploads for backups (#14736) This PR introduces a new `enable_experimental_backup_uploads` site setting (default false and hidden), which when enabled alongside `enable_direct_s3_uploads` will allow for direct S3 multipart uploads of backup .tar.gz files. To make multipart external uploads work with both the S3BackupStore and the S3Store, I've had to move several methods out of S3Store and into S3Helper, including: * presigned_url * create_multipart * abort_multipart * complete_multipart * presign_multipart_part * list_multipart_parts Then, S3Store and S3BackupStore either delegate directly to S3Helper or have their own special methods to call S3Helper for these methods. FileStore.temporary_upload_path has also removed its dependence on upload_path, and can now be used interchangeably between the stores. A similar change was made in the frontend as well, moving the multipart related JS code out of ComposerUppyUpload and into a mixin of its own, so it can also be used by UppyUploadMixin. Some changes to ExternalUploadManager had to be made here as well. The backup direct uploads do not need an Upload record made for them in the database, so they can be moved to their final S3 resting place when completing the multipart upload. This changeset is not perfect; it introduces some special cases in UploadController to handle backups that was previously in BackupController, because UploadController is where the multipart routes are located. A subsequent pull request will pull these routes into a module or some other sharing pattern, along with hooks, so the backup controller and the upload controller (and any future controllers that may need them) can include these routes in a nicer way.
2021-11-11 06:25:31 +08:00
key = store.s3_helper.path_from_url(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
expect(url).to match(/Amz-Expires/)
expect(key).to match(/temp\/standard99\/uploads\/second\/test_[0-9]\/[a-zA-z0-9]{0,32}\/[a-zA-z0-9]{0,32}.png/)
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
end
end
end
end
end