feat(arns): resolve ArNS names to IPFS CIDs (ArNS→IPFS, Phase 2)

CLAUDE.md

@@ -74,7 +74,10 @@ yarn service:start / stop / restart / status / logs
   bundle unbundling, verification, and webhooks. Controlled by `START_WRITERS`.
 - IPFS serving (`src/ipfs/`) is opt-in via `IPFS_ENABLED`. Uses a Kubo sidecar
   for content retrieval with its own cache, rate limiter, and blocklist. Routes
-  mount before ArNS in `app.ts`. See `docs/ipfs-integration.md`.
+  mount before ArNS in `app.ts`. ArNS names whose ANT record has
+  `targetProtocol: ipfs` resolve to a CID and are routed to the same IPFS
+  handler by the ArNS middleware (`src/middleware/arns.ts`); the on-demand
+  resolver reads `targetProtocol`. See `docs/ipfs-integration.md`.
 - Responses include trust headers indicating verification status.
 - HTTPSIG signs response headers (RFC 9421); `Content-Digest` is in
   `CO_SIGNABLE_HEADERS` so when present it binds the body to the signature.

docker-compose.yaml

@@ -183,9 +183,6 @@ services:
       - CHUNK_OFFSET_CHAIN_FALLBACK_TX_OFFSET_CACHE_TTL_MS=${CHUNK_OFFSET_CHAIN_FALLBACK_TX_OFFSET_CACHE_TTL_MS:-}
       - ARWEAVE_PEER_CHUNK_GET_MAX_PEER_ATTEMPT_COUNT=${ARWEAVE_PEER_CHUNK_GET_MAX_PEER_ATTEMPT_COUNT:-}
       - ARWEAVE_PEER_CHUNK_GET_PEER_SELECTION_COUNT=${ARWEAVE_PEER_CHUNK_GET_PEER_SELECTION_COUNT:-}
-      - ARWEAVE_PEER_CHUNK_POST_CONCURRENCY_LIMIT=${ARWEAVE_PEER_CHUNK_POST_CONCURRENCY_LIMIT:-}
-      - ARWEAVE_PEER_CHUNK_POST_MAX_PEER_ATTEMPT_COUNT=${ARWEAVE_PEER_CHUNK_POST_MAX_PEER_ATTEMPT_COUNT:-}
-      - ARWEAVE_PEER_CHUNK_POST_MIN_SUCCESS_COUNT=${ARWEAVE_PEER_CHUNK_POST_MIN_SUCCESS_COUNT:-}
       - ARWEAVE_CHUNK_GET_GEOMETRY_TIMEOUT_MS=${ARWEAVE_CHUNK_GET_GEOMETRY_TIMEOUT_MS:-}
       - ARWEAVE_CHUNK_GET_GEOMETRY_RETRY_COUNT=${ARWEAVE_CHUNK_GET_GEOMETRY_RETRY_COUNT:-}
       - WEBHOOK_TARGET_SERVERS=${WEBHOOK_TARGET_SERVERS:-}
@@ -241,6 +238,11 @@ services:
       - AWS_SECRET_ACCESS_KEY=${AWS_SECRET_ACCESS_KEY:-}
       - AWS_REGION=${AWS_REGION:-}
       - AWS_ENDPOINT=${AWS_ENDPOINT:-}
+      - TURBO_AWS_REGION=${TURBO_AWS_REGION:-}
+      - TURBO_AWS_ENDPOINT=${TURBO_AWS_ENDPOINT:-}
+      - TURBO_AWS_ACCESS_KEY_ID=${TURBO_AWS_ACCESS_KEY_ID:-}
+      - TURBO_AWS_SECRET_ACCESS_KEY=${TURBO_AWS_SECRET_ACCESS_KEY:-}
+      - TURBO_AWS_SESSION_TOKEN=${TURBO_AWS_SESSION_TOKEN:-}
       - AWS_S3_CONTIGUOUS_DATA_BUCKET=${AWS_S3_CONTIGUOUS_DATA_BUCKET:-}
       - AWS_S3_CONTIGUOUS_DATA_PREFIX=${AWS_S3_CONTIGUOUS_DATA_PREFIX:-}
       - AWS_S3_TURBO_CONTIGUOUS_DATA_BUCKET=${AWS_S3_TURBO_CONTIGUOUS_DATA_BUCKET:-}
@@ -383,6 +385,8 @@ services:
       - CLICKHOUSE_QUERY_TIMEOUT_SECONDS=${CLICKHOUSE_QUERY_TIMEOUT_SECONDS:-}
       - CLICKHOUSE_GQL_MAX_ROWS_TO_READ=${CLICKHOUSE_GQL_MAX_ROWS_TO_READ:-}
       - CLICKHOUSE_GQL_DEDUPE_HEADROOM=${CLICKHOUSE_GQL_DEDUPE_HEADROOM:-}
+      - CLICKHOUSE_GQL_OWNER_PROJECTION_ROUTING_ENABLED=${CLICKHOUSE_GQL_OWNER_PROJECTION_ROUTING_ENABLED:-}
+      - CLICKHOUSE_GQL_OWNER_PROJECTION_ENTITY_TYPES=${CLICKHOUSE_GQL_OWNER_PROJECTION_ENTITY_TYPES:-}
       - CLICKHOUSE_SQLITE_CIRCUIT_BREAKER_TIMEOUT_MS=${CLICKHOUSE_SQLITE_CIRCUIT_BREAKER_TIMEOUT_MS:-}
       - CLICKHOUSE_SQLITE_CIRCUIT_BREAKER_ERROR_THRESHOLD_PERCENTAGE=${CLICKHOUSE_SQLITE_CIRCUIT_BREAKER_ERROR_THRESHOLD_PERCENTAGE:-}
       - CLICKHOUSE_SQLITE_CIRCUIT_BREAKER_ROLLING_COUNT_TIMEOUT_MS=${CLICKHOUSE_SQLITE_CIRCUIT_BREAKER_ROLLING_COUNT_TIMEOUT_MS:-}

docs/arweave/transaction-and-chunk-offsets.md

@@ -338,6 +338,26 @@ for (const range of ranges) {
 1. **Direct Chunk Access**: Using absolute offsets allows O(1) chunk lookup
 2. **Minimal Data Transfer**: Only fetch chunks containing requested bytes
 3. **Streaming Support**: Process chunks as they arrive without buffering all
+4. **Local offset→block resolution**: When data is retrieved by absolute offset
+   (the cold-data chunk-by-offset path), the gateway must first find the block
+   containing the offset. Rather than walking the chain with ~log₂(height)
+   sequential `GET /block/height/{h}` requests, the Arweave client first
+   consults a local index over `stable_blocks.weave_size`
+   (`getBlockByWeaveOffset`, backed by `stable_blocks_weave_size_idx`): the
+   containing block is the lowest block whose cumulative `weave_size` reaches
+   the offset. The local result is only trusted when the immediately-preceding
+   block is also present and ends before the offset (a tight bracket, so no
+   missing block can hide the true container); otherwise — and for offsets in
+   the not-yet-stable chain tip — it falls back to the chain binary search.
+   This only changes how the block is *found*; the block returned is identical.
+
+   Note this resolves offset→**block** only. Resolving offset→**transaction**
+   within that block cannot be done from local block/transaction columns,
+   because per-transaction offsets follow the block's binary tx-ID sort order
+   (see [Transaction Order vs Offset Order](#transaction-order-vs-offset-order))
+   and a transaction's true weave contribution is not always captured by
+   `data_size` (e.g. v1 inline data). Per-transaction offsets are therefore
+   taken from the chain's authoritative `/tx/{id}/offset` endpoint.
 
 ### Edge Cases