@xdarkicex/openclaw-memory-libravdb 1.3.17 → 1.3.18

package/README.md

@@ -185,6 +185,59 @@ The formal math lives in:
 - [docs/ast-v2.md](./docs/ast-v2.md)
 - [docs/elevated-guidance.md](./docs/elevated-guidance.md)
 
+## LongMemEval Harness
+
+For internal tuning, the repo includes a local LongMemEval harness that runs the
+dataset through the plugin layer and measures whether the assembled prompt still
+contains the evidence turns.
+
+The benchmark runner is committed, but the dataset and generated reports are not.
+Keep downloaded data and local outputs under `benchmarks/longmemeval/`, which is
+ignored by default.
+
+The harness writes JSONL incrementally, so partial results survive if a transient
+daemon failure interrupts a long run.
+
+The run summary now prints a compact table with total questions, processed rows,
+skipped abstentions, errors, session hit rate, turn hit rate, and average prompt
+size.
+
+Run it with:
+
+```bash
+LONGMEMEVAL_DATA_FILE=/path/to/longmemeval_oracle.json pnpm run benchmark:longmemeval
+```
+
+If you already have a daemon running and do not want the benchmark to spawn
+another one, set:
+
+```bash
+LONGMEMEVAL_USE_EXISTING_DAEMON=1 LONGMEMEVAL_SIDECAR_PATH=unix:/path/to/libravdb.sock
+```
+
+If the local test daemon drops mid-run, the benchmark will restart it and retry
+the current instance once before recording an error result.
+
+Optional outputs:
+
+- `LONGMEMEVAL_LIMIT` to cap the number of questions
+- `LONGMEMEVAL_TOPK` to change the search budget
+- `LONGMEMEVAL_OUT_FILE` to write JSONL records for analysis
+
+To score a hypothesis JSONL file with the official LongMemEval evaluator, point
+the repo at a local checkout of the benchmark and run:
+
+```bash
+LONGMEMEVAL_EVAL_REPO=/path/to/LongMemEval \
+LONGMEMEVAL_HYPOTHESIS_FILE=/path/to/hypotheses.jsonl \
+LONGMEMEVAL_DATA_FILE=/path/to/longmemeval_oracle.json \
+OPENAI_API_KEY=... \
+pnpm run benchmark:longmemeval:score
+```
+
+That scorer wrapper shells out to the official Python evaluation script and then
+prints the aggregate metrics from the generated log when available.
+
 ## Compaction Model
 
 This system does not treat long chats as append-only forever.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@xdarkicex/openclaw-memory-libravdb",
-  "version": "1.3.17",
+  "version": "1.3.18",
   "type": "module",
   "publishConfig": {
     "access": "public"
@@ -31,6 +31,9 @@
     "gate:assemble_optimization": "./.ts-toolchain/node_modules/.bin/tsc -p tsconfig.tests.json && OPENCLAW_PROFILE_ASSEMBLE=1 OPENCLAW_ENFORCE_ASSEMBLE_EVIDENCE_GATE=1 node --test --test-name-pattern=\"real sidecar mid-sized session search benchmark\" .ts-build/test/integration/host-flow.test.js",
     "probe:session_recall": "./.ts-toolchain/node_modules/.bin/tsc -p tsconfig.tests.json && OPENCLAW_PROFILE_ASSEMBLE=1 node --test --test-name-pattern=\"real sidecar mid-sized session search benchmark\" .ts-build/test/integration/host-flow.test.js",
     "probe:session_recall_threshold": "./.ts-toolchain/node_modules/.bin/tsc -p tsconfig.tests.json && OPENCLAW_PROFILE_ASSEMBLE=1 node --test --test-name-pattern=\"real sidecar session_recall index threshold probe\" .ts-build/test/integration/host-flow.test.js",
+    "benchmark:longmemeval": "./.ts-toolchain/node_modules/.bin/tsc -p tsconfig.tests.json && node --test .ts-build/test/integration/longmemeval-benchmark.test.js",
+    "benchmark:longmemeval:score": "node scripts/longmemeval-score.mjs",
+    "benchmark:longmemeval:diagnose": "node scripts/longmemeval-diagnose.mjs",
     "build:daemon": "bash scripts/build-daemon.sh"
   },
   "dependencies": {

package/packaging/homebrew/libravdbd.rb.tmpl

@@ -32,12 +32,12 @@ class Libravdbd < Formula
     if Hardware::CPU.arm?
       resource "onnxruntime" do
         url "https://github.com/microsoft/onnxruntime/releases/download/v1.23.0/onnxruntime-linux-aarch64-1.23.0.tgz"
-        sha256 :no_check
+        sha256 :no_check # TODO: pin real checksum when Linux ARM64 CI is available
       end
     else
       resource "onnxruntime" do
         url "https://github.com/microsoft/onnxruntime/releases/download/v1.23.0/onnxruntime-linux-x64-1.23.0.tgz"
-        sha256 :no_check
+        sha256 :no_check # TODO: pin real checksum when Linux AMD64 CI is available
       end
     end
   end
@@ -54,12 +54,12 @@ class Libravdbd < Formula
 
   resource "all-minilm-l6-v2-model" do
     url "https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2/resolve/main/onnx/model.onnx"
-    sha256 "759c3cd2b7fe7e93933ad23c4c9181b7396442a2ed746ec7c1d46192c469c46e"
+    sha256 "6fd5d72fe4589f189f8ebc006442dbb529bb7ce38f8082112682524616046452"
   end
 
   resource "all-minilm-l6-v2-tokenizer" do
     url "https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2/resolve/main/tokenizer.json"
-    sha256 "da0e79933b9ed51798a3ae27893d3c5fa4a201126cef75586296df9b4d2c62a0"
+    sha256 "be50c3628f2bf5bb5e3a7f17b1f74611b2561a3a27eeab05e5aa30f411572037"
   end
 
   resource "t5-small-encoder" do
@@ -88,7 +88,7 @@ class Libravdbd < Formula
   end
 
   resource "provision" do
-    url "https://github.com/xDarkicex/openclaw-memory-libravdb/releases/download/v#{version}/provision.sh"
+    url "https://github.com/xDarkicex/openclaw-memory-libravdb/releases/download/v__VERSION__/provision.sh"
     sha256 "__SHA256_PROVISION__"
   end
 
@@ -107,7 +107,13 @@ class Libravdbd < Formula
     t5_dir.mkpath
 
     resource("onnxruntime").stage do
-      cp_r Dir["onnxruntime-*"].first, runtime_dir
+      # Homebrew may auto-strip the top-level dir from the tgz
+      subdir = Dir["onnxruntime-*"].first
+      if subdir
+        cp_r "#{subdir}/.", runtime_dir
+      else
+        cp_r ".", runtime_dir
+      end
     end
 
     resource("nomic-embed-text-v1.5-model").stage do
@@ -143,12 +149,14 @@ class Libravdbd < Formula
     end
     write_summarizer_manifest(t5_dir, "t5-small")
 
-    libexec.install resource("provision")
+    resource("provision").stage do
+      libexec.install "provision.sh"
+    end
     chmod 0755, libexec/"provision.sh"
   end
 
   def post_install
-    (var/"clawdb/data").mkpath
+    (var/"clawdb").mkpath
     (var/"clawdb/run").mkpath
   end
 
@@ -160,7 +168,8 @@ class Libravdbd < Formula
 
         #{libexec}/provision.sh --target #{prefix}/models
 
-      Data directory:   #{var}/clawdb/data
+      Data directory:   #{var}/clawdb
+      Database file:    #{var}/clawdb/data.libravdb
       Socket directory: #{var}/clawdb/run
     EOS
   end
@@ -202,7 +211,8 @@ class Libravdbd < Formula
   service do
     run [opt_bin/"libravdbd", "serve"]
     environment_variables LIBRAVDB_RPC_ENDPOINT: "unix:#{var}/clawdb/run/libravdb.sock",
-                          LIBRAVDB_DB_PATH: "#{var}/clawdb/data",
+                          LIBRAVDB_DB_PATH: "#{var}/clawdb/data.libravdb",
+                          LIBRAVDB_ONNX_RUNTIME: "#{opt_prefix}/models/onnxruntime/lib/libonnxruntime.dylib",
                           LIBRAVDB_SUMMARIZER_BACKEND: "bundled"
     keep_alive true
     working_dir var/"clawdb"