aboutsummaryrefslogtreecommitdiffstats
path: root/lib/active_storage/service
Commit message (Collapse)AuthorAgeFilesLines
* Verify direct upload checksumsGeorge Claghorn2017-07-242-5/+5
| | | | Closes #74.
* Everything under app/ is eager loaded, don't want that for serviceDavid Heinemeier Hansson2017-07-245-0/+322
| | | | Since it references all the specific cloud services that are intended only to be loaded on demand.
* Move models and jobs to the app setupDavid Heinemeier Hansson2017-07-225-317/+0
| | | | Follow engine conventions more closely
* Accept S3 upload options (e.g. server_side_encryption)James Baer2017-07-201-3/+5
|
* Add direct upload support to GCS serviceMichael Herold2017-07-171-0/+11
|
* Splat optionsGeorge Claghorn2017-07-161-16/+2
|
* Fix a typo in S3Servicecolorfulfool2017-07-161-3/+1
|
* Swap branchesGeorge Claghorn2017-07-151-3/+3
|
* Merge pull request #48 from seanhandley/hosted_s3George Claghorn2017-07-151-2/+18
|\ | | | | Allow custom endpoints for S3.
| * Allow custom endpoints for S3.Sean Handley2017-07-151-2/+18
| |
* | Revert back to the original implementaion with varaiblesdixpac2017-07-141-2/+8
|/ | | | | | Revert `exist? and url` to the original implementation. Since the new one doesn't provide any benefits and makes implementation harder to follow.
* Added rubocop / codeclimate config and fixed current offenses (#45)Cristian Bica2017-07-142-5/+5
|
* Remove few ivars from gcs_service implementationdixpac2017-07-131-8/+2
|
* Merge pull request #42 from quartzmo/gcs-site-urlGeorge Claghorn2017-07-121-2/+2
|\ | | | | Update GCSService#url
| * Update GCSService#urlChris Smith2017-07-121-2/+2
| | | | | | | | | | Update google-cloud-storage dependency to 1.3 Refactor arguments to Google::Cloud::Storage::File#signed_url
* | Revert "S3: slim down service implementation (#40)" (#41)David Heinemeier Hansson2017-07-121-24/+36
|/ | | This reverts commit 6d3962461fb8d35fc9538d685fee96267663acf2.
* S3: slim down service implementation (#40)Jeremy Daer2017-07-111-36/+24
| | | | | | | * Use simple core API for duck-type compat with other clients * initialize: accept an existing client * initialize: accept arbitrary client args instead of a fixed, required set * download: use native get_object streaming, no need to implement range requests * exists?: use head_object (which returns immediately) rather than waiting for existence
* Expose chunked downloadsGeorge Claghorn2017-07-101-4/+4
|
* Direct uploads for S3David Heinemeier Hansson2017-07-092-1/+12
|
* Instrument and log the servicesDavid Heinemeier Hansson2017-07-093-32/+93
|
* Merge pull request #27 from dixpac/fix_configuration_keys_symbolizationDavid Heinemeier Hansson2017-07-091-1/+1
|\ | | | | Symbolize all keys inside configuration nested hash
| * Symbolize all keys inside configuration nested hashdixpac2017-07-091-1/+1
| | | | | | | | | | Since configuration is a nested hash we need to symbolize all keys of the hash. Othervise fetcing will fail on start
* | Ensure binary encoding for downloadingDavid Heinemeier Hansson2017-07-091-1/+1
|/
* Clarify how a service can build other composed servicesJeremy Daer2017-07-092-29/+22
| | | | | | | | | * Service.build takes the literal YAML config hash for the service and a reference to the Configurator that's doing the building. * Services that compose additional services can use the Configurator to look them up and build them by name. See MirrorService for an example. References #23
* Service: clarify Service.build argumentsJeremy Daer2017-07-091-3/+3
| | | | | | | First arg is config for the service we're instantiating. Second arg is service configurations so we can look up and configure other services by name.
* Configure services that reference other servicesJeremy Daer2017-07-082-0/+42
| | | | | | | | * Move service configuration from the Engine to Service * Delegate configuration mechanics to internal Service::Configurator * Delegate service building to the concrete Service classes, allowing them to configure composed services. * Implement for the Mirror service.
* Mirror: explicit primary service and list of mirrorsJeremy Daer2017-07-081-10/+9
| | | | | | | | | | | | Pass separate primary service and list of mirrors rather than treating the first of the services list as the primary. Nice fit for keyword args, and something we've long wanted in the equivalent Basecamp file repository. Upload returns the results of the underlying service uploads rather than the io.rewind result. Rewind before uploading rather than afterward, and demonstrate that behavior with a test. Test that more than one mirror works.
* Merge pull request #21 from jeremy/disk-binary-ioDavid Heinemeier Hansson2017-07-091-8/+4
|\ | | | | Disk service: use binary IO throughout, not UTF-8
| * Disk service: use binary IO throughout, not UTF-8Jeremy Daer2017-07-081-8/+4
| |
* | Disk storage: ensure URLs end with the blob filename since some user agents ↵Jeremy Daer2017-07-081-2/+2
|/ | | | don't respect Content-Disposition filename
* Use safe navigationGeorge Claghorn2017-07-061-1/+1
|
* Remove unnecessary method delegationsGeorge Claghorn2017-07-061-1/+1
|
* Fix testGeorge Claghorn2017-07-061-0/+2
| | | | | | | * S3 fails fast if the Content-MD5 header on an upload request is an invalid checksum. Send a valid but incorrect checksum. * Rescue the service-specific exception and raise the generic one.
* Merge branch 'master' into s3_integrityJohn Williams2017-07-061-2/+3
|\
| * StyleGeorge Claghorn2017-07-061-5/+3
| |
| * Adding server side integrity check for GCS ServiceStanislav Gospodinov2017-07-061-2/+5
| |
* | Send checksum to S3 to verify file integrityJohn Williams2017-07-061-2/+1
|/
* Fix hash usage consistency.Ra'Shaun Stovall2017-07-061-1/+1
| | | | | | | Unless this was intentional, being consistent with: https://github.com/rails/activestorage/blob/master/lib/active_storage/service/s3_service.rb#L8 Just showin' a lil' <3 while perusing the repo @dhh
* Check integrity after uploadsDavid Heinemeier Hansson2017-07-064-5/+15
|
* Need the byte helpersDavid Heinemeier Hansson2017-07-062-0/+2
|
* Compute checksum and byte_size client sideDavid Heinemeier Hansson2017-07-063-34/+0
| | | | | Then we can add integrity checks on uploads to prevent errors in transport.
* Remember to add streamingDavid Heinemeier Hansson2017-07-061-0/+1
|
* This is even more explaining and upload/download don't have some inherent ↵David Heinemeier Hansson2017-07-061-4/+2
| | | | need to synchronize chunk sizes anyway
* We have the technology!David Heinemeier Hansson2017-07-061-1/+1
|
* Convert magic number to constantDavid Heinemeier Hansson2017-07-061-2/+4
|
* Mirror: only hit all sites for upload and deleteGeorge Claghorn2017-07-061-21/+4
| | | | The mirror service exists for the purpose of migration, where all blobs exist in the primary subservice and a subset of blobs exist in the secondary subservice. Since the primary subservice is the source of truth until a migration is completed, operations like existence checks need not be performed against the secondary subservices.
* Rename from Site to Service now that we're called Active StorageDavid Heinemeier Hansson2017-07-064-0/+238