Conversation
S3's ListObjectsV2 has no default delimiter — omitting it returns a flat list of all matching objects. Multistore was defaulting to "/" which caused grouped responses (with CommonPrefixes) even when clients didn't request grouping. - Default delimiter to empty string instead of "/" - Pass None to PaginatedListOptions when delimiter is empty - Skip serializing <Delimiter> element in XML when empty Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
UnsignedUrlSigner::from_config was S3-centric, reading only "endpoint" and "bucket_name" options. For Azure backends, this produced incorrect URLs like https://s3.amazonaws.com/<key> instead of https://<account>.blob.core.windows.net/<container>/<key>. Dispatch on parsed_backend_type to construct the correct endpoint and bucket/container for each provider. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add `ResponseBody` associated type and `forward` method to the `ProxyBackend` trait, removing the standalone `Forwarder` trait. `ForwardResponse` and `ForwardRequest` types remain accessible via the `forwarder` module. Downstream code will be updated in follow-up tasks. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
ProxyBackend now owns forwarding (via forward() + ResponseBody), so ProxyGateway no longer needs a separate F type parameter. This removes the forwarder field, updates handle_request to use B::ResponseBody, replaces self.forwarder.forward() with self.backend.forward(), and merges MockForwarder into MockBackend in tests. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Move forward() and ResponseBody from separate Forwarder structs into each runtime's ProxyBackend impl. Remove ServerForwarder, LambdaForwarder, and WorkerForwarder. Update ProxyGateway construction to use 3 type params. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
ForwardResponse now lives in backend::mod alongside ProxyBackend that uses it. The standalone forwarder.rs (which only held ForwardResponse and a re-export after the Forwarder trait was merged into ProxyBackend) is deleted and all imports are updated. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Extract generic Cloudflare Workers runtime adapters from the cf-workers example into a proper library crate at crates/cf-workers/. This allows other CF Workers deployments (like data.source.coop) to depend on shared implementations without copy-pasting. Modules: - fetch_connector: HttpConnector for object_store using Workers Fetch API - body: JsBody zero-copy ReadableStream wrapper + collect helper - backend: WorkerBackend implementing ProxyBackend via Fetch API - tracing_layer: WorkerSubscriber routing tracing to console.log - response: web_sys::Response builders and header conversion helpers - noop_creds: NoopCredentialRegistry for anonymous-only deployments Also renames the example package to multistore-cf-workers-example to avoid the workspace name conflict. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Allow bucket registries to specify an optional display name that is used in the <Name> element of ListBucketResult XML responses instead of the backend bucket name. This enables virtual bucket naming where the client-facing name differs from the underlying storage bucket. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Replace inline WorkerBackend, FetchConnector, JsBody, WorkerSubscriber, and response helper implementations in the example with imports from the multistore-cf-workers crate. Remove client.rs, fetch_connector.rs, and tracing_layer.rs which are now provided by the crate. Keep application- specific code (OIDC wiring, rate limiting, bandwidth metering). Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Extracts the URL-path-to-bucket-name mapping logic into a reusable crate.
PathMapping translates hierarchical paths like /{account}/{product}/{key}
into flat bucket names like account--product, and MappedRegistry wraps any
BucketRegistry to apply display_name and ListRewrite automatically.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Forward azure and gcp features to multistore core so the WorkerBackend's create_paginated_store handles all provider builders. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
…ash in rewrite_key The top-level <Prefix> element in ListBucketResult XML was not including the add_prefix from ListRewrite, causing incorrect responses for prefix-routed list requests. Also fixed rewrite_key to avoid double-slashes when add_prefix already ends with '/'. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
|
🚀 Latest commit deployed to https://multistore-proxy-pr-18.development-seed.workers.dev
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Push common Cloudflare Workers tooling into a
multistore-cf-workerscrate