root 8cfcdb8e5f G0 D2: storaged S3 GET/PUT/LIST/DELETE · 3-lineage scrum · 4 fixes applied
Phase G0 Day 2 ships storaged: aws-sdk-go-v2 wrapper + chi routes
binding 127.0.0.1:3211 with 256 MiB MaxBytesReader, Content-Length
up-front 413, and a 4-slot non-blocking semaphore returning 503 +
Retry-After:5 when full. Acceptance smoke (6/6 probes) PASSES against
the dedicated MinIO bucket lakehouse-go-primary, isolated from the
Rust system's lakehouse bucket during coexistence.

Cross-lineage scrum on the shipped code:
  - Opus 4.7 (opencode): 1 BLOCK + 3 WARN + 3 INFO
  - Qwen3-coder (openrouter): 2 BLOCK + 1 WARN + 1 INFO (3 false positives)
  - Kimi K2-0905 (openrouter, after route-shopping past opencode's 4k
    cap and the direct adapter's empty-content reasoning bug):
    1 BLOCK + 2 WARN + 1 INFO

Fixed:
  C1 buildRegistry ctx cancel footgun → context.Background()
     (Opus + Kimi convergent; future credential refresh chains)
  C2 MaxBytesReader unwrap through manager.Uploader multipart
     goroutines → Content-Length up-front 413 + string-suffix fallback
     (Opus + Kimi convergent; latent 500-instead-of-413 in 5-256 MiB range)
  C3 Bucket.List unbounded accumulation → MaxListResults=10_000 cap
     (Opus + Kimi convergent; OOM guard)
  S1 PUT response Content-Type: application/json (Opus single-reviewer)

Strict validateKey policy (J approved): rejects empty, >1024B, NUL,
leading "/", ".." path components, CR/LF/tab control characters.
DELETE exposed at HTTP layer (J approved option A) for symmetry +
smoke ergonomics.

Build clean, vet clean, all unit tests pass, smoke 6/6 PASS after
every fix round. go.mod 1.23 → 1.24 (required by aws-sdk-go-v2).

Process finding worth recording: opencode caps non-streaming Kimi at
max_tokens=4096; the direct kimi.com adapter consumed 8192 tokens of
reasoning but surfaced empty content; openrouter/moonshotai/kimi-k2-0905
delivered structured output in ~33s. Future Kimi scrums should default
to that route.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-28 23:23:03 -05:00

65 lines
1.9 KiB
Go

// BucketRegistry resolves logical bucket names → *Bucket. G0 holds
// exactly one bucket ("primary"); G2 (multi-bucket federation, per
// Rust ADR-017) extends this to many. Keeping the registry as a
// distinct type now means cmd/storaged never has to be refactored
// when G2 lands — every handler already routes through Resolve.
package storaged
import (
"errors"
"fmt"
"sync"
)
var ErrBucketUnknown = errors.New("storaged: unknown bucket")
// BucketRegistry is a thread-safe map of logical name → *Bucket.
// In G0 only "primary" is registered; the registry exists so that
// when G2 multi-bucket lands we add buckets in registry.go and the
// HTTP layer is unchanged.
type BucketRegistry struct {
mu sync.RWMutex
buckets map[string]*Bucket
}
func NewRegistry() *BucketRegistry {
return &BucketRegistry{buckets: make(map[string]*Bucket)}
}
// Register adds a bucket under its logical name. Re-registering the
// same name is rejected — G0 services build the registry once at
// startup and a duplicate is almost always a config bug.
func (r *BucketRegistry) Register(b *Bucket) error {
r.mu.Lock()
defer r.mu.Unlock()
if _, exists := r.buckets[b.name]; exists {
return fmt.Errorf("bucket %q already registered", b.name)
}
r.buckets[b.name] = b
return nil
}
// Resolve returns the bucket for a logical name. G0 callers pass
// "primary" since that's the only one registered.
func (r *BucketRegistry) Resolve(name string) (*Bucket, error) {
r.mu.RLock()
defer r.mu.RUnlock()
b, ok := r.buckets[name]
if !ok {
return nil, fmt.Errorf("%w: %q", ErrBucketUnknown, name)
}
return b, nil
}
// Names returns the registered logical names (sorted-order not
// guaranteed). Useful for /health surface or debug dumps.
func (r *BucketRegistry) Names() []string {
r.mu.RLock()
defer r.mu.RUnlock()
out := make([]string, 0, len(r.buckets))
for name := range r.buckets {
out = append(out, name)
}
return out
}