fix: subdirectory dated files now decay by filename date

Author: sachinsharma9780

README.md

@@ -152,7 +152,8 @@ The SQLite index is a **derived cache** — always rebuildable from the Markdown
 | `memory/MEMORY.md` | **Evergreen** — never decays, write-protected during `flush()` |
 | `memory/2026-03-21.md` | **Dated** — subject to temporal decay (older memories rank lower) |
 | `memory/researcher_agent/` | **Agent-scoped** — isolated namespace per agent |
-| `memory/episodes/event.md` | **Episodic** — named events, timestamped |
+| `memory/episodes/known-facts.md` | **Evergreen** — non-dated file in a subdirectory, always full score |
+| `memory/sessions/2026-04-01.md` | **Dated** — subdirectory dated file, decays by filename date |
 
 Evergreen files hold foundational facts that should always surface at full score. Dated files accumulate daily learning and fade naturally — recent memories rank higher.
 
@@ -248,9 +249,10 @@ With the default `half_life_days=30`:
 
 | File | Age source | Decays? |
 |------|------------|---------|
-| `memory/MEMORY.md`, `memory/architecture.md` (any non-dated file under `memory/`) | — | **No** — evergreen, always full score |
+| `memory/MEMORY.md`, `memory/architecture.md` (any non-dated file directly under `memory/`) | — | **No** — evergreen, always full score |
+| `memory/agents/notes.md` (non-dated file in any `memory/` subdirectory) | — | **No** — evergreen, same rule as root non-dated files |
 | `memory/2026-03-21.md` (dated daily log) | Date parsed from filename | **Yes** |
-| `sessions/foo.md`, `memory/agents/notes.md` (undated, non-evergreen) | File `mtime` on disk | **Yes** |
+| `memory/sessions/2026-03-21.md` (dated file in any `memory/` subdirectory) | Date parsed from filename | **Yes** — same rule as root dated files |
 
 Evergreen files hold foundational facts — stack choices, hard constraints, permanent preferences — that should always surface at full score regardless of when they were written. Daily logs capture evolving context and fade naturally as new sessions add fresher knowledge.
 

memweave/search/temporal_decay.py

@@ -33,8 +33,9 @@
 from memweave.search.strategy import RawSearchRow
 
 # Regex that matches dated memory paths: memory/YYYY-MM-DD.md
-# Works with both ``memory/2026-03-28.md`` and ``./memory/2026-03-28.md``
-_DATED_PATH_RE = re.compile(r"(?:^|\/)memory\/(\d{4})-(\d{2})-(\d{2})\.md$")
+# Also matches one level of subdirectory: memory/sessions/YYYY-MM-DD.md
+# Works with both ``memory/2026-03-28.md`` and ``./memory/sessions/2026-03-28.md``
+_DATED_PATH_RE = re.compile(r"(?:^|\/)memory\/(?:[^/]+\/)?(\d{4})-(\d{2})-(\d{2})\.md$")
 
 _DAY_SECONDS = 86_400.0
 
@@ -162,7 +163,7 @@ def is_evergreen_path(file_path: str) -> bool:
         return True
     if not normalized.startswith("memory/"):
         return False
-    # Under memory/ but not dated → evergreen
+    # Under memory/ but not dated → evergreen (applies at any depth)
     return _DATED_PATH_RE.search(file_path.replace("\\", "/")) is None
 
 
@@ -190,12 +191,18 @@ async def _extract_date(
     """Extract the effective date for a file — from path first, then mtime.
 
     Steps:
-    1. Parse date from filename (``memory/YYYY-MM-DD.md``). Return if found.
+    1. Parse date from filename (``memory/YYYY-MM-DD.md`` or
+       ``memory/<subdir>/YYYY-MM-DD.md``). Return if found.
     2. If the file is evergreen → return ``None`` (no decay applies).
+       All ``memory/`` files with non-dated filenames are evergreen at any depth.
     3. If ``workspace_dir`` is ``None`` → return ``None`` (cannot resolve path).
     4. Resolve the absolute path and ``stat()`` the file.
     5. Return the mtime as a :class:`datetime.date`, or ``None`` on error.
 
+    Steps 3–5 (mtime fallback) are only reached for files outside ``memory/``
+    (e.g. ``extra_paths`` external files with no date in their filename).
+    All ``memory/``-managed files are fully resolved by steps 1–2.
+
     Args:
         file_path:     Relative or absolute file path.
         workspace_dir: Root workspace directory used to resolve relative paths.

tests/integration/test_temporal_decay.py

@@ -86,3 +86,55 @@ async def test_temporal_decay(workspace: Path, embedding_model: str) -> None:
         old_long = long_scores.get(old_path, scores[old_path])
 
         assert old_long > old_decay, "Longer half-life should preserve more of the old file's score"
+
+
+_SESSION_OLD_DATE = "2026-01-01"  # ~100 days before 2026-04-11
+_SESSION_NEW_DATE = "2026-04-01"  # ~10 days before 2026-04-11
+_SESSION_CONTENT = (
+    "We confirmed the rollback plan: revert the migration and restart the service.\n"
+    "The on-call engineer will monitor for 24 hours after deployment.\n"
+)
+
+
+@pytest.mark.asyncio
+async def test_session_subdir_dated_decay(workspace: Path, embedding_model: str) -> None:
+    """Dated files under memory/sessions/ decay by filename date — same rule as memory/ root."""
+    sessions_dir = workspace / "memory" / "sessions"
+    sessions_dir.mkdir()
+
+    (sessions_dir / f"{_SESSION_OLD_DATE}.md").write_text(_SESSION_CONTENT)
+    (sessions_dir / f"{_SESSION_NEW_DATE}.md").write_text(_SESSION_CONTENT)
+
+    config = MemoryConfig(
+        workspace_dir=workspace,
+        embedding=EmbeddingConfig(model=embedding_model),
+        query=QueryConfig(min_score=0.0, max_results=10),
+    )
+
+    query = "rollback plan on-call deployment"
+    old_path = f"memory/sessions/{_SESSION_OLD_DATE}.md"
+    new_path = f"memory/sessions/{_SESSION_NEW_DATE}.md"
+
+    async with MemWeave(config) as mem:
+        await mem.index()
+
+        # Without decay: identical content → nearly identical scores
+        r_no_decay = await mem.search(query, min_score=0.0)
+        scores = {r.path: r.score for r in r_no_decay}
+        assert old_path in scores, f"{old_path} not found in results"
+        assert new_path in scores, f"{new_path} not found in results"
+        diff = abs(scores[old_path] - scores[new_path])
+        assert diff < 0.05, f"Identical content without decay should score similarly, diff={diff:.4f}"
+
+        # With aggressive decay: ~100-day-old session file should drop sharply
+        # half_life=7d → 2^(-100/7) ≈ 0.000054 multiplier
+        r_decay = await mem.search(query, min_score=0.0, decay_half_life_days=7.0)
+        decay_scores = {r.path: r.score for r in r_decay}
+
+        old_decay = decay_scores.get(old_path, 0.0)
+        new_decay = decay_scores.get(new_path, scores.get(new_path, 0.0))
+
+        assert new_decay > old_decay * 5, (
+            f"With half_life=7d, new session should score >5× old. "
+            f"new={new_decay:.4f} old={old_decay:.6f}"
+        )

tests/unit/test_search_temporal_decay.py

@@ -156,6 +156,12 @@ def test_empty_string(self):
     def test_session_file(self):
         assert parse_date_from_path("sessions/2026-03-28.md") is None
 
+    def test_subdir_dated_path(self):
+        assert parse_date_from_path("memory/sessions/2026-03-28.md") == date(2026, 3, 28)
+
+    def test_subdir_dated_path_researcher(self):
+        assert parse_date_from_path("memory/researcher/2026-04-01.md") == date(2026, 4, 1)
+
 
 # ── is_evergreen_path ─────────────────────────────────────────────────────────
 
@@ -185,6 +191,21 @@ def test_session_file_not_evergreen(self):
     def test_arbitrary_file_not_evergreen(self):
         assert is_evergreen_path("docs/readme.md") is False
 
+    def test_sessions_subdir_nondated_is_evergreen(self):
+        assert is_evergreen_path("memory/sessions/foo.md") is True
+
+    def test_sessions_subdir_dated_not_evergreen(self):
+        assert is_evergreen_path("memory/sessions/2026-03-29.md") is False
+
+    def test_researcher_subdir_nondated_is_evergreen(self):
+        assert is_evergreen_path("memory/researcher/analysis.md") is True
+
+    def test_researcher_subdir_dated_not_evergreen(self):
+        assert is_evergreen_path("memory/researcher/2026-04-01.md") is False
+
+    def test_custom_subdir_nondated_is_evergreen(self):
+        assert is_evergreen_path("memory/episodes/standup.md") is True
+
 
 # ── age_in_days ───────────────────────────────────────────────────────────────
 
@@ -312,6 +333,26 @@ async def test_session_file_fresh_score_nearly_unchanged(self, tmp_path: Path):
         result = await apply_temporal_decay(rows, half_life_days=30, workspace_dir=tmp_path)
         assert result[0].score > 0.9
 
+    async def test_memory_sessions_nondated_is_evergreen(self):
+        """Non-dated memory/sessions/ file is evergreen — score unchanged."""
+        rows = [_row("s_nd", "memory/sessions/known-facts.md", 0.8)]
+        result = await apply_temporal_decay(rows, half_life_days=30, now=date(2026, 3, 28))
+        assert result[0].score == pytest.approx(0.8)
+
+    async def test_memory_sessions_dated_file_decays_by_date(self):
+        """Dated memory/sessions/YYYY-MM-DD.md decays by filename date, no mtime needed."""
+        rows = [_row("s_d", "memory/sessions/2026-01-01.md", 1.0)]
+        result = await apply_temporal_decay(rows, half_life_days=30, now=date(2026, 3, 28))
+        expected = apply_decay_to_score(1.0, 86.0, 30.0)
+        assert result[0].score == pytest.approx(expected, rel=1e-6)
+
+    async def test_memory_researcher_dated_file_decays_by_date(self):
+        """Dated memory/researcher/YYYY-MM-DD.md decays by filename date."""
+        rows = [_row("r_d", "memory/researcher/2026-01-01.md", 1.0)]
+        result = await apply_temporal_decay(rows, half_life_days=30, now=date(2026, 3, 28))
+        expected = apply_decay_to_score(1.0, 86.0, 30.0)
+        assert result[0].score == pytest.approx(expected, rel=1e-6)
+
     async def test_mtime_cache_deduplicates_stat_calls(self, tmp_path: Path, monkeypatch):
         """Multiple chunks from the same file path are stat-ted only once."""
         import aiofiles.os as aio_os