sampler-cli 0.4.1__tar.gz → 0.4.2__tar.gz

{sampler_cli-0.4.1 → sampler_cli-0.4.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sampler-cli
-Version: 0.4.1
+Version: 0.4.2
 Summary: Token-efficient CLI for indexing and searching code symbols (Python-first, designed for minimal LLM/agent context size)
 Author: Samuel Ignacio Carmona Rodriguez
 License: MIT
@@ -52,7 +52,7 @@ Dynamic: license-file
 
 Token-efficient CLI for indexing and searching code symbols across multiple projects.
 
-Current version: 0.4.1
+Current version: 0.4.2
 
 Designed for humans and agents: compact default output, short paths, and low-noise symbol views.
 
@@ -92,7 +92,7 @@ sampler overview src/main.py
 ## Command Overview
 
 Core:
-- `sampler version`
+- `sampler version [--plain]`
 - `sampler init`
 - `sampler index <project>`
 - `sampler search <query> [--project <name>] [--type <t>] [--limit <n>] [--semantic] [--style plain|bars]`
@@ -168,6 +168,13 @@ Offline / air-gapped: `provider: hash` (or just don't install the embeddings ext
 
 - function is called from test files
 - function has zero non-test callers in project call graph
+- symbol is defined in production code (symbols defined in test files are excluded)
+
+Test file detection supports common multi-language patterns:
+
+- Python: `tests/`, `test_*.py`, `*_test.py`
+- Go: `*_test.go`
+- TypeScript/JavaScript: `__tests__/`, `test/`, `spec/`, `*.test.*`, `*.spec.*`
 
 This is heuristic signal, not guaranteed dead-code proof.
 

{sampler_cli-0.4.1 → sampler_cli-0.4.2}/README.md

@@ -2,7 +2,7 @@
 
 Token-efficient CLI for indexing and searching code symbols across multiple projects.
 
-Current version: 0.4.1
+Current version: 0.4.2
 
 Designed for humans and agents: compact default output, short paths, and low-noise symbol views.
 
@@ -42,7 +42,7 @@ sampler overview src/main.py
 ## Command Overview
 
 Core:
-- `sampler version`
+- `sampler version [--plain]`
 - `sampler init`
 - `sampler index <project>`
 - `sampler search <query> [--project <name>] [--type <t>] [--limit <n>] [--semantic] [--style plain|bars]`
@@ -118,6 +118,13 @@ Offline / air-gapped: `provider: hash` (or just don't install the embeddings ext
 
 - function is called from test files
 - function has zero non-test callers in project call graph
+- symbol is defined in production code (symbols defined in test files are excluded)
+
+Test file detection supports common multi-language patterns:
+
+- Python: `tests/`, `test_*.py`, `*_test.py`
+- Go: `*_test.go`
+- TypeScript/JavaScript: `__tests__/`, `test/`, `spec/`, `*.test.*`, `*.spec.*`
 
 This is heuristic signal, not guaranteed dead-code proof.
 

{sampler_cli-0.4.1 → sampler_cli-0.4.2}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "sampler-cli"
-version = "0.4.1"
+version = "0.4.2"
 description = "Token-efficient CLI for indexing and searching code symbols (Python-first, designed for minimal LLM/agent context size)"
 readme = "README.md"
 license = { text = "MIT" }

{sampler_cli-0.4.1 → sampler_cli-0.4.2}/src/sampler/__init__.py

@@ -1,3 +1,3 @@
 __all__ = ["__version__"]
 
-__version__ = "0.4.1"
+__version__ = "0.4.2"

{sampler_cli-0.4.1 → sampler_cli-0.4.2}/src/sampler/cli/main.py

@@ -1,3 +1,4 @@
+import sys
 from pathlib import Path
 
 import typer
@@ -6,7 +7,6 @@ from rich.console import Console
 from sampler import __version__
 from sampler.config import ConfigManager
 from sampler.db import Database
-from sampler.indexer.builder import IndexBuilder
 from sampler.query.engine import QueryEngine
 
 # Embeddings provider support (lazy, optional)
@@ -89,9 +89,17 @@ def _format_line_range(start_line: int | None, end_line: int | None) -> str:
 
 
 @app.command("version")
-def version() -> None:
+def version(
+    plain: bool = typer.Option(False, "--plain", help="Plain text output (version number only)"),
+) -> None:
     """Show installed sampler version."""
-    console.print(f"sampler {__version__}")
+    if plain or not sys.stdout.isatty():
+        console.print(f"sampler {__version__}")
+        return
+
+    from sampler.viz.headline import print_version_card
+
+    print_version_card(console, __version__)
 
 
 @app.command("init")
@@ -381,9 +389,25 @@ def symbols(
         console.print(line)
 
 
+def _build_embedder():
+    from sampler.indexer.embedder import Embedder
+
+    if get_embedding_provider is not None:
+        try:
+            return Embedder(provider=get_embedding_provider())
+        except Exception:
+            return Embedder()
+    return Embedder()
+
+
 @app.command("index")
-def index(project: str) -> None:
-    """Index selected project."""
+def index(
+    project: str,
+    plain: bool = typer.Option(False, "--plain", help="Compact output without Live visualization (for CI/scripts)"),
+    batch_size: int = typer.Option(32, "--batch-size", help="Batch size for embedding generation"),
+    force: bool = typer.Option(False, "--force", help="Re-index all files regardless of hash"),
+) -> None:
+    """Index project and generate embeddings (Live visualization when attached to a TTY)."""
     config = ConfigManager()
     project_cfg = config.get_project(project)
     if project_cfg is None:
@@ -394,17 +418,46 @@ def index(project: str) -> None:
             "Use 'sampler project list' to see registered projects."
         )
 
-    builder = IndexBuilder(db=_database())
-    stats = builder.index_project(
-        project_name=project_cfg.name,
-        project_path=project_cfg.path,
-        language=project_cfg.language,
-    )
-    console.print(
-        f"[green]✓[/green] Indexed [bold]{stats['project']}[/bold]: "
-        f"discovered={stats['discovered']} indexed={stats['indexed']} "
-        f"skipped={stats['skipped']} failed={stats['failed']}"
-    )
+    embedder = _build_embedder()
+    use_plain = plain or not sys.stdout.isatty()
+
+    try:
+        embedder.provider.embed("sampler health probe", for_query=True)
+    except RuntimeError as exc:
+        from rich.panel import Panel
+
+        console.print(Panel.fit(str(exc), title="Error", border_style="red"))
+        raise typer.Exit(code=1)
+
+    from sampler.viz.pipeline import run_index_pipeline
+
+    try:
+        stats = run_index_pipeline(
+            db=_database(),
+            project_cfg=project_cfg,
+            embedder=embedder,
+            force=force,
+            batch_size=batch_size,
+            plain=use_plain,
+            console=console,
+        )
+    except RuntimeError as exc:
+        from rich.panel import Panel
+
+        console.print(Panel.fit(str(exc), title="Error", border_style="red"))
+        raise typer.Exit(code=1)
+
+    if use_plain:
+        console.print(
+            f"[green]✓[/green] Indexed [bold]{stats['project']}[/bold]: "
+            f"discovered={stats['discovered']} indexed={stats['indexed']} "
+            f"skipped={stats['skipped']} failed={stats['failed']}"
+        )
+        prov_name = getattr(embedder.provider, "name", "hash")
+        console.print(
+            f"[green]✓[/green] Embedded [bold]{stats['embed_count']}[/bold] symbols "
+            f"using [bold]{prov_name}[/bold] ({stats['model']}) in [bold]{stats['elapsed']:.1f}s[/bold]"
+        )
 
 
 @app.command("embed")

{sampler_cli-0.4.1 → sampler_cli-0.4.2}/src/sampler/db.py

@@ -611,6 +611,57 @@ class Database:
         with self.connect() as conn:
             return conn.execute(sql, (project_name,)).fetchall()
 
+    def get_project_index_stats(self, project_name: str) -> dict[str, int]:
+        """Aggregate counts for index pipeline / status displays."""
+        sql = """
+                SELECT
+                    (SELECT COUNT(*) FROM files f
+                     JOIN projects p ON f.project_id = p.id WHERE p.name = ?) AS files,
+                    (SELECT COUNT(*) FROM symbols s
+                     JOIN files f ON s.file_id = f.id
+                     JOIN projects p ON f.project_id = p.id WHERE p.name = ?) AS symbols,
+                    (SELECT COUNT(*) FROM relationships r
+                     JOIN symbols src ON r.source_id = src.id
+                     JOIN files f ON src.file_id = f.id
+                     JOIN projects p ON f.project_id = p.id WHERE p.name = ?) AS relationships,
+                    (SELECT COUNT(*) FROM embeddings e
+                     JOIN symbols s ON e.symbol_id = s.id
+                     JOIN files f ON s.file_id = f.id
+                     JOIN projects p ON f.project_id = p.id WHERE p.name = ?) AS embeddings
+                """
+        with self.connect() as conn:
+            row = conn.execute(sql, (project_name, project_name, project_name, project_name)).fetchone()
+        if row is None:
+            return {"files": 0, "symbols": 0, "relationships": 0, "embeddings": 0}
+        return {
+            "files": int(row["files"]),
+            "symbols": int(row["symbols"]),
+            "relationships": int(row["relationships"]),
+            "embeddings": int(row["embeddings"]),
+        }
+
+    def get_top_symbols_by_degree(self, project_name: str, limit: int = 80) -> list[sqlite3.Row]:
+        """Top symbols by in+out relationship degree for graph preview."""
+        sql = """
+                SELECT
+                    s.id,
+                    s.name,
+                    s.qualified_name,
+                    s.type,
+                    (
+                        SELECT COUNT(*) FROM relationships r
+                        WHERE r.source_id = s.id OR r.target_id = s.id
+                    ) AS degree
+                FROM symbols s
+                JOIN files f ON s.file_id = f.id
+                JOIN projects p ON f.project_id = p.id
+                WHERE p.name = ?
+                ORDER BY degree DESC, s.qualified_name, s.name
+                LIMIT ?
+                """
+        with self.connect() as conn:
+            return conn.execute(sql, (project_name, limit)).fetchall()
+
     def get_project_call_edges(self, project_name: str) -> list[sqlite3.Row]:
         """Return CALLS edges within project, with caller/target symbol + file context.
 

{sampler_cli-0.4.1 → sampler_cli-0.4.2}/src/sampler/indexer/builder.py

@@ -4,6 +4,8 @@ import hashlib
 import re
 from pathlib import Path
 
+from typing import TYPE_CHECKING
+
 from sampler.db import Database
 from sampler.indexer.discover import discover_files, discover_files_multi
 from sampler.indexer.imports import extract_imports
@@ -11,6 +13,11 @@ from sampler.indexer.parsers.go import GoParser
 from sampler.indexer.parsers.python import PythonParser
 from sampler.indexer.parsers.typescript import TypeScriptParser
 from sampler.indexer.store import SymbolStore
+from sampler.viz.discover_emit import emit_discover
+from sampler.viz.events import FileParsing, LogLine, Stage, StageChanged
+
+if TYPE_CHECKING:
+    from sampler.viz.bus import EventBus, NullEventBus
 
 
 class IndexBuilder:
@@ -24,7 +31,14 @@ class IndexBuilder:
             "javascript": TypeScriptParser(),
         }
 
-    def index_project(self, project_name: str, project_path: str, language: str, force: bool = False) -> dict:
+    def index_project(
+        self,
+        project_name: str,
+        project_path: str,
+        language: str,
+        force: bool = False,
+        event_bus: EventBus | NullEventBus | None = None,
+    ) -> dict:
         is_auto = language.lower() == "auto"
         if not is_auto and language not in self.parsers:
             raise ValueError(f"Unsupported language: {language}")
@@ -38,10 +52,17 @@ class IndexBuilder:
         else:
             file_entries = [(f, language) for f in discover_files(project_path=project_abs_path, language=language)]
 
+        bus = event_bus
+        if bus is not None:
+            emit_discover(bus, project_abs_path, file_entries)
+            bus.emit(StageChanged(Stage.PARSING))
+
         indexed = 0
         skipped = 0
         failed = 0
         all_imports: set[str] = set()
+        total = len(file_entries)
+        parse_idx = 0
 
         for filepath, file_language in file_entries:
             parser = self.parsers.get(file_language)
@@ -63,7 +84,13 @@ class IndexBuilder:
                 skipped += 1
                 continue
 
+            if bus is not None:
+                bus.emit(FileParsing(path=filepath, index=parse_idx, total=total))
+                bus.emit(LogLine(message=f"+ parser {Path(filepath).name}"))
+
             symbols, relationships = parser.parse(content=content, filepath=filepath)
+            if bus is not None:
+                bus.emit(StageChanged(Stage.RELATIONSHIPS))
             self.store.save_symbols(
                 project_id=project_id,
                 filepath=filepath,
@@ -71,8 +98,10 @@ class IndexBuilder:
                 file_hash=file_hash,
                 symbols=symbols,
                 relationships=relationships,
+                event_bus=bus,
             )
             indexed += 1
+            parse_idx += 1
 
         self.db.update_project_file_count(project_id)
         self._resolve_project_dependencies(project_id, project_name, all_imports)

{sampler_cli-0.4.1 → sampler_cli-0.4.2}/src/sampler/indexer/embedder.py

@@ -2,9 +2,13 @@ from __future__ import annotations
 
 import hashlib
 import re
-from typing import Callable
+from typing import TYPE_CHECKING, Callable
 
 from sampler.db import Database
+from sampler.viz.events import EmbeddingGenerated
+
+if TYPE_CHECKING:
+    from sampler.viz.bus import EventBus, NullEventBus
 
 # --- Public constants kept for backward compat ---
 DEFAULT_BATCH_SIZE = 32
@@ -200,6 +204,7 @@ class Embedder:
         project_name: str,
         batch_size: int = DEFAULT_BATCH_SIZE,
         on_progress: Callable[[int, int], None] | None = None,
+        event_bus: EventBus | NullEventBus | None = None,
     ) -> int:
         """Generate and store embeddings for every symbol in a project using the active provider.
 
@@ -220,6 +225,7 @@ class Embedder:
             # Use provider batch (supports for_query=False for document storage)
             vectors = provider.embed_batch(texts, for_query=False)
 
+            done_before = start
             for row, vec in zip(batch, vectors):
                 try:
                     import numpy as np
@@ -234,6 +240,12 @@ class Embedder:
                     dim=dim,
                     vector=vec_bytes,
                 )
+                done_before += 1
+                if event_bus is not None:
+                    name = row["qualified_name"] or row["name"]
+                    event_bus.emit(
+                        EmbeddingGenerated(name=name, index=done_before, total=total)
+                    )
 
             if on_progress is not None:
                 on_progress(min(start + batch_size, total), total)

{sampler_cli-0.4.1 → sampler_cli-0.4.2}/src/sampler/indexer/store.py

@@ -1,4 +1,12 @@
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
 from sampler.db import Database
+from sampler.viz.events import RelationshipCreated, SymbolExtracted
+
+if TYPE_CHECKING:
+    from sampler.viz.bus import EventBus, NullEventBus
 
 
 class SymbolStore:
@@ -87,6 +95,7 @@ class SymbolStore:
         file_hash: str,
         symbols: list[dict],
         relationships: list[dict],
+        event_bus: EventBus | NullEventBus | None = None,
     ) -> None:
         file_id = self.db.upsert_file(project_id=project_id, path=filepath, language=language, file_hash=file_hash)
         self.db.clear_file_data(file_id)
@@ -103,6 +112,14 @@ class SymbolStore:
                 local_by_name.setdefault(name, []).append(inserted_id)
                 if name not in symbol_id_map:
                     symbol_id_map[name] = inserted_id
+            if event_bus is not None:
+                event_bus.emit(
+                    SymbolExtracted(
+                        name=qualified or name or "symbol",
+                        symbol_type=symbol.get("type") or "symbol",
+                        file_path=filepath,
+                    )
+                )
 
         for relation in relationships:
             source_key = relation.get("source")
@@ -131,3 +148,11 @@ class SymbolStore:
                 continue
 
             self.db.insert_relationship(source_id=source_id, target_id=target_id, relation=relation)
+            if event_bus is not None:
+                event_bus.emit(
+                    RelationshipCreated(
+                        source=source_key,
+                        target=target_key,
+                        relation_type=relation.get("type") or "CALLS",
+                    )
+                )

{sampler_cli-0.4.1 → sampler_cli-0.4.2}/src/sampler/query/engine.py

@@ -119,7 +119,31 @@ class QueryEngine:
     def _is_test_path(path: str) -> bool:
         p = path.replace("\\", "/").lower()
         name = Path(path).name.lower()
-        return "/tests/" in p or name.startswith("test_") or name.endswith("_test.py")
+        if any(seg in p for seg in ("/tests/", "/test/", "/__tests__/", "/spec/")):
+            return True
+
+        if name.startswith("test_"):
+            return True
+
+        if name.endswith(("_test.py", "_test.go")):
+            return True
+
+        return name.endswith(
+            (
+                ".test.ts",
+                ".test.tsx",
+                ".test.js",
+                ".test.jsx",
+                ".test.mjs",
+                ".test.cjs",
+                ".spec.ts",
+                ".spec.tsx",
+                ".spec.js",
+                ".spec.jsx",
+                ".spec.mjs",
+                ".spec.cjs",
+            )
+        )
 
     def stale_code_candidates(self, project_name: str) -> list[dict]:
         """Detect code likely stale: function/method called by tests but not by non-test code."""
@@ -156,6 +180,9 @@ class QueryEngine:
 
         stale: list[dict] = []
         for entry in by_target.values():
+            # Test-file symbols (helpers, fixtures) are expected to be test-only callers.
+            if self._is_test_path(entry["file_path"]):
+                continue
             if entry["test_callers"] and not entry["non_test_callers"]:
                 stale.append(
                     {

sampler_cli-0.4.2/src/sampler/viz/__init__.py

@@ -0,0 +1,6 @@
+from sampler.viz.bus import EventBus
+from sampler.viz.engine import AnimationEngine
+from sampler.viz.events import Stage
+from sampler.viz.live import IndexLiveSession
+
+__all__ = ["AnimationEngine", "EventBus", "IndexLiveSession", "Stage"]

sampler_cli-0.4.2/src/sampler/viz/bus.py

@@ -0,0 +1,29 @@
+from __future__ import annotations
+
+from collections.abc import Callable
+
+from sampler.viz.events import IndexEvent
+
+
+class EventBus:
+    """Synchronous pub/sub for index pipeline events."""
+
+    def __init__(self) -> None:
+        self._handlers: list[Callable[[IndexEvent], None]] = []
+
+    def subscribe(self, handler: Callable[[IndexEvent], None]) -> None:
+        self._handlers.append(handler)
+
+    def emit(self, event: IndexEvent) -> None:
+        for handler in self._handlers:
+            handler(event)
+
+
+class NullEventBus:
+    """No-op bus for plain/headless indexing."""
+
+    def subscribe(self, handler: Callable[[IndexEvent], None]) -> None:
+        return None
+
+    def emit(self, event: IndexEvent) -> None:
+        return None