sourcecode 1.32.4__tar.gz → 1.32.5__tar.gz

{sourcecode-1.32.4 → sourcecode-1.32.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sourcecode
-Version: 1.32.4
+Version: 1.32.5
 Summary: Deterministic codebase context for AI coding agents
 License-File: LICENSE
 Keywords: agents,ai,codebase,context,developer-tools,llm

{sourcecode-1.32.4 → sourcecode-1.32.5}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "sourcecode"
-version = "1.32.4"
+version = "1.32.5"
 description = "Deterministic codebase context for AI coding agents"
 readme = "README.md"
 requires-python = ">=3.9"

{sourcecode-1.32.4 → sourcecode-1.32.5}/src/sourcecode/__init__.py

@@ -1,3 +1,3 @@
 """sourcecode — Deterministic codebase context maps for AI coding agents."""
 
-__version__ = "1.32.4"
+__version__ = "1.32.5"

{sourcecode-1.32.4 → sourcecode-1.32.5}/src/sourcecode/cache.py

@@ -72,6 +72,11 @@ SCHEMA_VERSION: str = "2"
 #: Bump to invalidate all L1 core caches (independent of snapshot version).
 CORE_SCHEMA_VERSION: str = "1"
 
+#: Bump when analysis logic or output schema changes — NOT on every package release.
+#: This is the stable part of the L1 core cache key.  Package version bumps (patch,
+#: minor) must NOT bump this value unless the cached data format actually changed.
+ANALYZER_CACHE_VERSION: str = "1"
+
 #: Fields eligible for CAS deduplication (applied to top-level JSON dict keys).
 _CAS_FIELDS: frozenset[str] = frozenset([
     "file_paths",
@@ -511,7 +516,7 @@ def _cas_store_blob(cache_d: Path, serialised: str) -> str:
     path = _cas_path(cache_d, blob_hash)
     if not path.exists():
         path.parent.mkdir(parents=True, exist_ok=True)
-        path.write_bytes(gzip.compress(raw, compresslevel=6))
+        _atomic_write(path, gzip.compress(raw, compresslevel=6))
     return blob_hash
 
 
@@ -642,7 +647,11 @@ def _gc(cache_d: Path) -> None:
         # ── Pass 3: total size cap ──────────────────────────────────────────
         if max_size_bytes > 0:
             size_candidates = [p for p in surviving if p.exists()]
+            # Include CAS blobs in the size budget calculation
+            cas_d_sz = _cas_dir(cache_d)
+            cas_files = list(cas_d_sz.glob("*.gz")) if cas_d_sz.exists() else []
             total = sum(p.stat().st_size for p in size_candidates if not p.name.startswith("view-"))
+            total += sum(p.stat().st_size for p in cas_files if p.exists())
             if total > max_size_bytes:
                 # Sort oldest-first; evict core+snapshot files until under budget
                 size_candidates.sort(key=lambda p: p.stat().st_mtime)

{sourcecode-1.32.4 → sourcecode-1.32.5}/src/sourcecode/cli.py

@@ -208,6 +208,8 @@ _SUBCOMMANDS: frozenset[str] = frozenset(
         "onboard", "modernize", "fix-bug", "review-pr",
         # License
         "activate",
+        # Cache observability
+        "cache",
     }
 )
 
@@ -412,6 +414,9 @@ app.add_typer(telemetry_app, name="telemetry")
 mcp_app = typer.Typer(help="MCP integration: setup, status, serve, remove.", rich_markup_mode="rich")
 app.add_typer(mcp_app, name="mcp")
 
+cache_app = typer.Typer(help="Cache inspection and management.", rich_markup_mode="rich")
+app.add_typer(cache_app, name="cache")
+
 
 def _maybe_ask_consent() -> None:
     """Show first-run consent prompt once, on interactive TTYs only."""
@@ -1064,7 +1069,6 @@ def main(
             # Only cache when target IS the git repo root (not a subdir of one),
             # to avoid polluting sub-project directories used in tests.
             if _git_sha and (target / ".git").exists():
-                from sourcecode import __version__ as _sc_version
                 _excl_key = (
                     ",".join(sorted(e.strip() for e in exclude.split(",") if e.strip()))
                     if exclude else ""
@@ -1072,8 +1076,10 @@ def main(
 
                 # ── Core (analysis) flags: affect which analyzers run + scan config ──
                 # Use effective_depth (not raw depth) so Java auto-adjustment is captured.
+                # acv = ANALYZER_CACHE_VERSION — bumped only when analysis logic/schema
+                # changes, NOT on every package release.  Prevents patch-bump cache wipes.
                 _core_flags_str = (
-                    f"v={_sc_version},"
+                    f"acv={_cache_mod.ANALYZER_CACHE_VERSION},"
                     f"dep={dependencies},gm={graph_modules},"
                     f"docs={docs},fm={full_metrics},sem={semantics},"
                     f"arch={architecture},gc={git_context},em={env_map},"
@@ -4124,6 +4130,70 @@ def mcp_remove(
     typer.echo("  Re-add:  sourcecode mcp init")
 
 
+# ── Cache subcommands ─────────────────────────────────────────────────────────
+
+
+@cache_app.command("status")
+def cache_status_cmd(
+    path: Path = typer.Argument(Path("."), help="Repository path (default: current directory)"),
+    json_output: bool = typer.Option(False, "--json", help="Output as JSON."),
+) -> None:
+    """Show cache statistics for a repository."""
+    from sourcecode import cache as _cm
+    target = Path(path).resolve()
+    stats = _cm.status(target)
+    if json_output:
+        import json as _j
+        typer.echo(_j.dumps(stats, indent=2, ensure_ascii=False))
+    else:
+        typer.echo(f"Cache dir:   {stats['cache_dir']}")
+        typer.echo(f"Cores:       {stats['cores']}")
+        typer.echo(f"Snapshots:   {stats['snapshots']}")
+        typer.echo(f"Views:       {stats['views']}")
+        typer.echo(f"CAS blobs:   {stats['cas_blobs']}")
+        typer.echo(f"Total size:  {stats['total_size_mb']} MB")
+
+
+@cache_app.command("clear")
+def cache_clear_cmd(
+    path: Path = typer.Argument(Path("."), help="Repository path (default: current directory)"),
+    yes: bool = typer.Option(False, "--yes", "-y", help="Skip confirmation prompt."),
+) -> None:
+    """Delete all cached snapshots for a repository."""
+    from sourcecode import cache as _cm
+    target = Path(path).resolve()
+    if not yes:
+        typer.confirm(f"Delete all cache files for {target}?", abort=True)
+    removed = _cm.clear(target)
+    typer.echo(f"Removed {removed} file(s).")
+
+
+@cache_app.command("warm")
+def cache_warm_cmd(
+    path: Path = typer.Argument(Path("."), help="Repository path to warm (default: current directory)"),
+    compact: bool = typer.Option(True, "--compact/--no-compact", help="Warm compact view (default: on)."),
+    agent: bool = typer.Option(False, "--agent", help="Also warm agent view."),
+) -> None:
+    """Pre-populate the cache by running a fresh analysis."""
+    import subprocess as _sub
+    import sys as _sys
+    target = Path(path).resolve()
+    typer.echo(f"Warming cache for {target} …", err=True)
+    cmd = [_sys.executable, "-m", "sourcecode", str(target)]
+    if compact:
+        cmd.append("--compact")
+    if agent:
+        cmd.append("--agent")
+    result = _sub.run(cmd, capture_output=True, text=True)
+    if result.returncode == 0:
+        typer.echo("Cache warmed.", err=True)
+    else:
+        typer.echo(f"Warm failed (exit {result.returncode}).", err=True)
+        if result.stderr:
+            typer.echo(result.stderr.strip(), err=True)
+        raise typer.Exit(code=result.returncode)
+
+
 # ── Entry point ───────────────────────────────────────────────────────────────
 
 def main_entry() -> None:

{sourcecode-1.32.4 → sourcecode-1.32.5}/src/sourcecode/ris.py

@@ -329,10 +329,15 @@ def update_ris_api_surface(repo_root: Path, endpoints_data: dict) -> None:
 # ---------------------------------------------------------------------------
 
 def _current_git_head(repo_root: Path) -> str:
-    """Return current HEAD SHA.  Returns '' on any error or non-git directory."""
+    """Return current HEAD short SHA.  Returns '' on any error or non-git directory.
+
+    Uses --short to match the format stored in the RIS and used by cli.py
+    cache key computation — both sides must use the same format or staleness
+    checks will always return True.
+    """
     try:
         result = subprocess.run(
-            ["git", "-C", str(repo_root), "rev-parse", "HEAD"],
+            ["git", "-C", str(repo_root), "rev-parse", "--short", "HEAD"],
             capture_output=True,
             text=True,
             timeout=2,