coderay 1.0.2__tar.gz → 1.0.3__tar.gz

{coderay-1.0.2/src/coderay.egg-info → coderay-1.0.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: coderay
-Version: 1.0.2
+Version: 1.0.3
 Summary: X-ray your codebase — semantic search, code graphs, file skeletons, and MCP server
 Author-email: Bogdan Copocean <bogdancopocean@gmail.com>
 License-Expression: MIT
@@ -20,31 +20,31 @@ Classifier: Topic :: Text Processing :: Indexing
 Requires-Python: >=3.10
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: python-dotenv>=1.0.0
-Requires-Dist: networkx>=3.0
-Requires-Dist: tree-sitter>=0.24.0
+Requires-Dist: python-dotenv>=1.2.0
+Requires-Dist: networkx>=3.4
+Requires-Dist: tree-sitter>=0.25.0
 Requires-Dist: tree-sitter-python>=0.25.0
-Requires-Dist: lancedb>=0.5.0
-Requires-Dist: pyyaml>=6.0
-Requires-Dist: click>=8.0
-Requires-Dist: filelock>=3.0
-Requires-Dist: fastembed>=0.4.0
-Requires-Dist: mcp>=1.0.0
-Requires-Dist: watchdog>=4.0.0
-Requires-Dist: pathspec>=0.12.0
+Requires-Dist: lancedb>=0.29.0
+Requires-Dist: pyyaml>=6.0.3
+Requires-Dist: click>=8.3.0
+Requires-Dist: filelock>=3.25.0
+Requires-Dist: fastembed>=0.7.0
+Requires-Dist: fastmcp==3.1.0
+Requires-Dist: watchdog>=6.0.0
+Requires-Dist: pathspec>=1.0.0
 Provides-Extra: openai
-Requires-Dist: openai>=1.0.0; extra == "openai"
+Requires-Dist: openai>=2.0.0; extra == "openai"
 Provides-Extra: languages
-Requires-Dist: tree-sitter-javascript>=0.23.0; extra == "languages"
-Requires-Dist: tree-sitter-typescript>=0.23.0; extra == "languages"
-Requires-Dist: tree-sitter-go>=0.23.0; extra == "languages"
+Requires-Dist: tree-sitter-javascript>=0.25.0; extra == "languages"
+Requires-Dist: tree-sitter-typescript>=0.23.2; extra == "languages"
+Requires-Dist: tree-sitter-go>=0.25.0; extra == "languages"
 Provides-Extra: dev
 Requires-Dist: pytest>=7.0; extra == "dev"
 Requires-Dist: pytest-cov>=4.0; extra == "dev"
 Requires-Dist: ruff>=0.8.0; extra == "dev"
 Requires-Dist: mypy>=1.0.0; extra == "dev"
-Requires-Dist: openai>=1.0.0; extra == "dev"
-Requires-Dist: httpx>=0.27.0; extra == "dev"
+Requires-Dist: openai>=2.0.0; extra == "dev"
+Requires-Dist: httpx>=0.28.0; extra == "dev"
 Provides-Extra: maintain
 Requires-Dist: pylance>=0.15.0; extra == "maintain"
 Provides-Extra: all

{coderay-1.0.2 → coderay-1.0.3}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "coderay"
-version = "1.0.2"
+version = "1.0.3"
 description = "X-ray your codebase — semantic search, code graphs, file skeletons, and MCP server"
 readme = "README.md"
 license = "MIT"
@@ -25,18 +25,18 @@ classifiers = [
     "Topic :: Text Processing :: Indexing",
 ]
 dependencies = [
-    "python-dotenv>=1.0.0",
-    "networkx>=3.0",
-    "tree-sitter>=0.24.0",
+    "python-dotenv>=1.2.0",
+    "networkx>=3.4",
+    "tree-sitter>=0.25.0",
     "tree-sitter-python>=0.25.0",
-    "lancedb>=0.5.0",
-    "pyyaml>=6.0",
-    "click>=8.0",
-    "filelock>=3.0",
-    "fastembed>=0.4.0",
-    "mcp>=1.0.0",
-    "watchdog>=4.0.0",
-    "pathspec>=0.12.0",
+    "lancedb>=0.29.0",
+    "pyyaml>=6.0.3",
+    "click>=8.3.0",
+    "filelock>=3.25.0",
+    "fastembed>=0.7.0",
+    "fastmcp==3.1.0",
+    "watchdog>=6.0.0",
+    "pathspec>=1.0.0",
 ]
 
 [project.urls]
@@ -45,19 +45,19 @@ Repository = "https://github.com/bogdan-copocean/coderay"
 Issues = "https://github.com/bogdan-copocean/coderay/issues"
 
 [project.optional-dependencies]
-openai = ["openai>=1.0.0"]
+openai = ["openai>=2.0.0"]
 languages = [
-    "tree-sitter-javascript>=0.23.0",
-    "tree-sitter-typescript>=0.23.0",
-    "tree-sitter-go>=0.23.0",
+    "tree-sitter-javascript>=0.25.0",
+    "tree-sitter-typescript>=0.23.2",
+    "tree-sitter-go>=0.25.0",
 ]
 dev = [
     "pytest>=7.0",
     "pytest-cov>=4.0",
     "ruff>=0.8.0",
     "mypy>=1.0.0",
-    "openai>=1.0.0",
-    "httpx>=0.27.0",
+    "openai>=2.0.0",
+    "httpx>=0.28.0",
 ]
 maintain = [
     "pylance>=0.15.0",

{coderay-1.0.2 → coderay-1.0.3}/src/coderay/chunking/chunker.py

@@ -91,7 +91,6 @@ def _chunk_file_with_config(
                     start_line=start_line,
                     end_line=end_line,
                     symbol=symbol,
-                    language=lang_cfg.name,
                     content=text,
                 )
             )
@@ -107,7 +106,6 @@ def _chunk_file_with_config(
                 start_line=1,
                 end_line=root.end_point[0] + 1,
                 symbol="<module>",
-                language=lang_cfg.name,
                 content="\n".join(preamble_lines),
             ),
         )
@@ -118,7 +116,7 @@ def _chunk_file_with_config(
     return chunks
 
 
-def chunk_file(path: str | Path, content: str, language: str = "python") -> list[Chunk]:
+def chunk_file(path: str | Path, content: str) -> list[Chunk]:
     """Chunk a source file into semantic units (functions, classes, preamble)."""
     path_str = str(path) if isinstance(path, Path) else path
     if not (lang_cfg := get_language_for_file(path_str)):

{coderay-1.0.2 → coderay-1.0.3}/src/coderay/cli/commands.py

@@ -132,14 +132,12 @@ def update(ctx: click.Context, repo: Path) -> None:
 @click.argument("query_text", required=True)
 @click.option("--top-k", "top_k", default=10, help="Number of results")
 @click.option("--path-prefix", help="Filter by path prefix")
-@click.option("--language", help="Filter by language (e.g. python)")
 @click.pass_context
 def search_cmd(
     ctx: click.Context,
     query_text: str,
     top_k: int,
     path_prefix: str | None,
-    language: str | None,
 ) -> None:
     """Semantic search the index."""
     index_dir = ctx.obj["index_dir"]
@@ -162,7 +160,6 @@ def search_cmd(
         current_state=current_state,
         top_k=top_k,
         path_prefix=path_prefix,
-        language=language,
     )
     elapsed = time.perf_counter() - t0
     click.echo(_color(f"Query took {elapsed:.2f}s", BOLD))
@@ -171,31 +168,16 @@ def search_cmd(
         click.echo(_color("No results.", YELLOW))
         return
 
-    score_type = results[0].get("score_type", "cosine")
-    if score_type == "rrf":
-        click.echo(
-            _color("Scoring: hybrid (RRF) — relative ranking, not a percentage", CYAN)
-        )
-    else:
-        click.echo(_color("Scoring: cosine similarity (0-1)", CYAN))
-
     for i, r in enumerate(results, 1):
         path = r.get("path", "?")
         start = r.get("start_line", 0)
         end = r.get("end_line", 0)
         symbol = r.get("symbol", "?")
-        score = r.get("score", 0)
-        if score_type == "cosine":
-            score_str = f"score={score:.4f} ({score:.0%})"
-        else:
-            score_str = f"score={score:.4f} (rrf)"
         preview = (r.get("content") or "")[:200].replace("\n", " ")
         if len(r.get("content") or "") > 200:
             preview += "..."
         click.echo("")
-        click.echo(
-            _color(f"  {i}. {path}:{start}-{end} ({symbol})  {score_str}", GREEN)
-        )
+        click.echo(_color(f"  {i}. {path}:{start}-{end} ({symbol})", GREEN))
         click.echo(f"     {preview}")
 
 

{coderay-1.0.2 → coderay-1.0.3}/src/coderay/core/models.py

@@ -12,7 +12,6 @@ class Chunk:
     start_line: int
     end_line: int
     symbol: str
-    language: str
     content: str
 
     def line_range(self) -> tuple[int, int]:

{coderay-1.0.2 → coderay-1.0.3}/src/coderay/graph/code_graph.py

@@ -228,12 +228,15 @@ class CodeGraph:
                 very large sets.
         """
         resolved = self.resolve_symbol(symbol) or symbol
+        if resolved not in self._g:
+            return []
         visited: set[str] = set()
         frontier = {resolved}
         for _ in range(depth):
             next_frontier: set[str] = set()
             for nid in frontier:
-                # predecessors = nodes that have an edge pointing TO nid
+                if nid not in self._g:
+                    continue
                 for pred in self._g.predecessors(nid):
                     if pred not in visited:
                         visited.add(pred)

coderay-1.0.3/src/coderay/mcp_server/errors.py

@@ -0,0 +1,11 @@
+_DEFAULT_MSG = (
+    "Index not built. Ask the user to run 'coderay build' in their terminal, "
+    "then retry."
+)
+
+
+class IndexNotBuiltError(Exception):
+    """Raised when a tool requires a coderay index that has not been built."""
+
+    def __init__(self, message: str = _DEFAULT_MSG) -> None:
+        super().__init__(message)

coderay-1.0.3/src/coderay/mcp_server/server.py

@@ -0,0 +1,225 @@
+from __future__ import annotations
+
+import logging
+from pathlib import Path
+from typing import Annotated, Any
+
+from fastmcp import FastMCP
+from mcp.types import ToolAnnotations
+from pydantic import Field
+
+from coderay.mcp_server.errors import IndexNotBuiltError
+from coderay.retrieval.models import SearchResult
+
+logger = logging.getLogger(__name__)
+
+mcp = FastMCP(
+    name="coderay",
+    instructions=(
+        "CodeRay provides semantic code search, file skeletons, and "
+        "dependency impact analysis over a pre-built index.\n"
+        "\n"
+        "- semantic_search: search code by meaning. Best for "
+        "'how/where' questions. Use grep for exact symbol lookup.\n"
+        "- get_file_skeleton: signatures and docstrings only, no bodies. "
+        "Useful to check a file's API before reading full source. "
+        "Works without the index.\n"
+        "- get_impact_radius: reverse dependency traversal from the code "
+        "graph. Shows callers/dependents of a function or class.\n"
+        "\n"
+        "All tools except get_file_skeleton require a built index. "
+        "On index errors, ask the user to run 'coderay build'."
+    ),
+)
+
+DEFAULT_INDEX_DIR = ".index"
+
+_retrieval_cache: dict[Path, Any] = {}
+_state_machine_cache: dict[Path, Any] = {}
+
+
+def _resolve_index_dir() -> Path:
+    """Resolve the index directory to an absolute path."""
+    return Path(DEFAULT_INDEX_DIR).resolve()
+
+
+def _get_retrieval():
+    """Return a cached Retrieval instance for the given index directory."""
+    idx = _resolve_index_dir()
+    if idx not in _retrieval_cache:
+        from coderay.retrieval.search import Retrieval
+
+        _retrieval_cache[idx] = Retrieval(idx)
+    return _retrieval_cache[idx]
+
+
+def _load_graph():
+    """Load the code graph from disk, or return None if absent."""
+    from coderay.graph.builder import load_graph
+
+    return load_graph(_resolve_index_dir())
+
+
+def _get_state_machine():
+    """Return a cached StateMachine instance."""
+    idx = _resolve_index_dir()
+    if idx not in _state_machine_cache:
+        from coderay.state.machine import StateMachine
+
+        _state_machine_cache[idx] = StateMachine(idx)
+    return _state_machine_cache[idx]
+
+
+def _load_state():
+    """Load the current IndexMeta state, or None if no run has completed."""
+    return _get_state_machine().current_state
+
+
+READ_ONLY_ANNOTATIONS = ToolAnnotations(readOnlyHint=True, destructiveHint=False)
+
+
+@mcp.tool(
+    description=(
+        "Search code by meaning. Returns chunks ranked by relevance, "
+        "each with path, line range, symbol, and content. "
+        "Best for 'how/where' questions; use grep for exact symbols."
+    ),
+    annotations=READ_ONLY_ANNOTATIONS,
+    tags={"search"},
+)
+def semantic_search(
+    query: Annotated[
+        str,
+        Field(description="Natural language question about the code"),
+    ],
+    top_k: int = 5,
+    path_prefix: Annotated[
+        str | None,
+        Field(
+            description=(
+                "Filter to files under this directory, e.g. 'src/coderay/graph/'"
+            ),
+        ),
+    ] = None,
+) -> dict:
+    """Search the semantic index."""
+    retrieval = _get_retrieval()
+    state = _load_state()
+    if state is None:
+        raise IndexNotBuiltError()
+
+    raw_results = retrieval.search(
+        query=query,
+        current_state=state,
+        top_k=top_k,
+        path_prefix=path_prefix,
+    )
+    results = [SearchResult.from_raw(r) for r in raw_results]
+    return {"results": [r.to_dict() for r in results]}
+
+
+@mcp.tool(
+    description=(
+        "Extracts class/function signatures and docstrings from a "
+        "file — no bodies. Output is significantly shorter than "
+        "full source. Does not require the index."
+    ),
+    annotations=READ_ONLY_ANNOTATIONS,
+    tags={"analysis"},
+)
+def get_file_skeleton(
+    file_path: Annotated[
+        str,
+        Field(description="Absolute or relative path to the file"),
+    ],
+) -> str:
+    """Get the API surface of a file (signatures, no bodies)."""
+    from coderay.skeleton.extractor import extract_skeleton
+
+    p = Path(file_path)
+    if not p.is_file():
+        raise FileNotFoundError(f"File not found: {file_path}")
+    content = p.read_text(encoding="utf-8", errors="replace")
+    return extract_skeleton(p, content)
+
+
+@mcp.tool(
+    description=(
+        "Reverse dependency traversal: lists callers and dependents "
+        "of a function or class from the code graph. Returns empty "
+        "results when node_id has no dependents. "
+        "Static analysis only; dynamic dispatch may be missed."
+    ),
+    annotations=READ_ONLY_ANNOTATIONS,
+    tags={"analysis"},
+)
+def get_impact_radius(
+    node_id: Annotated[
+        str,
+        Field(
+            description=(
+                "Fully qualified node ID, e.g. "
+                "'src/utils.py::parse_config' or "
+                "'src/models.py::User.save'"
+            ),
+        ),
+    ],
+    max_depth: Annotated[
+        int,
+        Field(description="How many caller/dependent levels to traverse"),
+    ] = 2,
+) -> dict:
+    """Analyze the blast radius of changing a function or module."""
+    graph = _load_graph()
+    if graph is None:
+        raise IndexNotBuiltError(
+            "No graph found. Ask the user to run 'coderay build' "
+            "in their terminal, then retry."
+        )
+    impact = graph.get_impact_radius(node_id, depth=max_depth)
+    return {
+        "results": [n.to_dict() for n in impact],
+    }
+
+
+@mcp.resource(
+    "coderay://index/status",
+    description=("Index status: build state, branch, commit, and chunk count."),
+    tags={"status"},
+)
+def index_status() -> dict:
+    """Check health and status of the semantic index."""
+    state = _load_state()
+    if state is None:
+        raise IndexNotBuiltError()
+
+    from coderay.core.config import get_embedding_dimensions, load_config
+    from coderay.state.version import read_index_version
+    from coderay.storage.lancedb import index_exists as idx_exists
+
+    idx_dir = _resolve_index_dir()
+    has_store = idx_exists(idx_dir)
+    chunk_count = 0
+    if has_store:
+        from coderay.storage.lancedb import Store
+
+        config = load_config(idx_dir)
+        store = Store(idx_dir, dimensions=get_embedding_dimensions(config))
+        chunk_count = store.chunk_count()
+
+    return {
+        "status": state.state.value,
+        "branch": state.branch,
+        "last_commit": state.last_commit,
+        "chunk_count": chunk_count,
+        "schema_version": read_index_version(idx_dir),
+        "has_store": has_store,
+    }
+
+
+def main():
+    mcp.run(transport="stdio")
+
+
+if __name__ == "__main__":
+    main()

coderay-1.0.3/src/coderay/retrieval/models.py

@@ -0,0 +1,63 @@
+from __future__ import annotations
+
+from dataclasses import dataclass
+
+DEFAULT_MAX_CONTENT_LINES: int = 60
+
+
+@dataclass(frozen=True)
+class SearchResult:
+    """A single search hit with optional content truncation."""
+
+    path: str
+    start_line: int
+    end_line: int
+    symbol: str
+    content: str
+    truncated: bool = False
+
+    @classmethod
+    def from_raw(
+        cls,
+        row: dict,
+        *,
+        max_lines: int | None = DEFAULT_MAX_CONTENT_LINES,
+    ) -> SearchResult:
+        """Build from a raw dict returned by the storage layer.
+
+        Args:
+            row: Dict with keys path, start_line, end_line, symbol,
+                and content.
+            max_lines: Truncate content beyond this many lines.
+                None disables truncation.
+        """
+        content: str = row.get("content", "")
+        truncated = False
+
+        if max_lines is not None:
+            lines = content.split("\n")
+            if len(lines) > max_lines:
+                content = "\n".join(lines[:max_lines])
+                truncated = True
+
+        return cls(
+            path=row["path"],
+            start_line=row["start_line"],
+            end_line=row["end_line"],
+            symbol=row["symbol"],
+            content=content,
+            truncated=truncated,
+        )
+
+    def to_dict(self) -> dict:
+        """Serialize to a JSON-compatible dict for the MCP response."""
+        d: dict = {
+            "path": self.path,
+            "start_line": self.start_line,
+            "end_line": self.end_line,
+            "symbol": self.symbol,
+            "content": self.content,
+        }
+        if self.truncated:
+            d["truncated"] = True
+        return d

{coderay-1.0.2 → coderay-1.0.3}/src/coderay/retrieval/search.py

@@ -59,7 +59,6 @@ class Retrieval:
         *,
         top_k: int = 10,
         path_prefix: str | None = None,
-        language: str | None = None,
     ) -> list[dict[str, Any]]:
         """Semantic search over the index."""
         if not index_exists(self.index_dir):
@@ -82,7 +81,6 @@ class Retrieval:
             query_embedding=query_vectors[0],
             top_k=top_k,
             path_prefix=path_prefix,
-            language=language,
             query_text=query,
         )
         logger.info("Vector search took %.3fs", time.perf_counter() - t1)

{coderay-1.0.2 → coderay-1.0.3}/src/coderay/skeleton/extractor.py

@@ -11,17 +11,10 @@ logger = logging.getLogger(__name__)
 def extract_skeleton(
     path: str | Path,
     content: str,
-    language: str | None = None,
 ) -> str:
     """Extract the skeleton of a source file (signatures, no bodies)."""
     path_str = str(path)
-    lang_cfg = get_language_for_file(path_str) if language is None else None
-    if lang_cfg is None and language:
-        from coderay.chunking.registry import LANGUAGE_REGISTRY
-
-        lang_cfg = LANGUAGE_REGISTRY.get(language)
-    if lang_cfg is None:
-        lang_cfg = get_language_for_file(path_str)
+    lang_cfg = get_language_for_file(path_str)
     if lang_cfg is None:
         return content
 
@@ -33,7 +26,7 @@ def extract_skeleton(
     source_bytes = content.encode("utf-8")
     tree = parser.parse(source_bytes)
     lines: list[str] = []
-    _visit_skeleton(tree.root_node, source_bytes, lang_cfg.name, lines, depth=0)
+    _visit_skeleton(tree.root_node, source_bytes, lines, depth=0)
     return "\n".join(lines)
 
 
@@ -77,7 +70,6 @@ def _get_signature_line(node, source_bytes: bytes) -> str:
 def _visit_skeleton(
     node,
     source_bytes: bytes,
-    language: str,
     lines: list[str],
     depth: int,
 ) -> None:
@@ -118,7 +110,7 @@ def _visit_skeleton(
         for child in node.children:
             if child.type in ("block", "class_body", "statement_block"):
                 for member in child.children:
-                    _visit_skeleton(member, source_bytes, language, lines, depth + 1)
+                    _visit_skeleton(member, source_bytes, lines, depth + 1)
         lines.append("")
         return
 
@@ -137,4 +129,4 @@ def _visit_skeleton(
         return
 
     for child in node.children:
-        _visit_skeleton(child, source_bytes, language, lines, depth)
+        _visit_skeleton(child, source_bytes, lines, depth)

{coderay-1.0.2 → coderay-1.0.3}/src/coderay/storage/lancedb.py

@@ -64,7 +64,6 @@ class Store:
                     "start_line": chunk.start_line,
                     "end_line": chunk.end_line,
                     "symbol": chunk.symbol,
-                    "language": chunk.language,
                     "content": chunk.content,
                     "vector": emb,
                 }
@@ -115,7 +114,6 @@ class Store:
         query_embedding: list[float],
         top_k: int = 10,
         path_prefix: str | None = None,
-        language: str | None = None,
         query_text: str | None = None,
     ) -> list[dict[str, Any]]:
         """Nearest-neighbor search with optional hybrid scoring."""
@@ -146,9 +144,6 @@ class Store:
         if path_prefix:
             prefix = (path_prefix.rstrip("/") + "/").replace("'", "''")
             query = query.where(f"path LIKE '{prefix}%'")
-        if language:
-            lang = (language or "").replace("'", "''")
-            query = query.where(f"language = '{lang}'")
 
         query = query.limit(top_k)
         rows = query.to_list()
@@ -164,7 +159,6 @@ class Store:
             else:
                 score = row.pop("distance", 0.0)
             row["score"] = round(float(score), 4)
-            row["score_type"] = "rrf" if use_hybrid else "cosine"
             row.pop("vector", None)
             results.append(row)
 
@@ -189,7 +183,7 @@ class Store:
         if n == 0:
             return []
 
-        col_names = ["path", "start_line", "end_line", "symbol", "language"]
+        col_names = ["path", "start_line", "end_line", "symbol"]
 
         if path_prefix:
             prefix = (path_prefix.rstrip("/") + "/").replace("'", "''")

{coderay-1.0.2 → coderay-1.0.3/src/coderay.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: coderay
-Version: 1.0.2
+Version: 1.0.3
 Summary: X-ray your codebase — semantic search, code graphs, file skeletons, and MCP server
 Author-email: Bogdan Copocean <bogdancopocean@gmail.com>
 License-Expression: MIT
@@ -20,31 +20,31 @@ Classifier: Topic :: Text Processing :: Indexing
 Requires-Python: >=3.10
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: python-dotenv>=1.0.0
-Requires-Dist: networkx>=3.0
-Requires-Dist: tree-sitter>=0.24.0
+Requires-Dist: python-dotenv>=1.2.0
+Requires-Dist: networkx>=3.4
+Requires-Dist: tree-sitter>=0.25.0
 Requires-Dist: tree-sitter-python>=0.25.0
-Requires-Dist: lancedb>=0.5.0
-Requires-Dist: pyyaml>=6.0
-Requires-Dist: click>=8.0
-Requires-Dist: filelock>=3.0
-Requires-Dist: fastembed>=0.4.0
-Requires-Dist: mcp>=1.0.0
-Requires-Dist: watchdog>=4.0.0
-Requires-Dist: pathspec>=0.12.0
+Requires-Dist: lancedb>=0.29.0
+Requires-Dist: pyyaml>=6.0.3
+Requires-Dist: click>=8.3.0
+Requires-Dist: filelock>=3.25.0
+Requires-Dist: fastembed>=0.7.0
+Requires-Dist: fastmcp==3.1.0
+Requires-Dist: watchdog>=6.0.0
+Requires-Dist: pathspec>=1.0.0
 Provides-Extra: openai
-Requires-Dist: openai>=1.0.0; extra == "openai"
+Requires-Dist: openai>=2.0.0; extra == "openai"
 Provides-Extra: languages
-Requires-Dist: tree-sitter-javascript>=0.23.0; extra == "languages"
-Requires-Dist: tree-sitter-typescript>=0.23.0; extra == "languages"
-Requires-Dist: tree-sitter-go>=0.23.0; extra == "languages"
+Requires-Dist: tree-sitter-javascript>=0.25.0; extra == "languages"
+Requires-Dist: tree-sitter-typescript>=0.23.2; extra == "languages"
+Requires-Dist: tree-sitter-go>=0.25.0; extra == "languages"
 Provides-Extra: dev
 Requires-Dist: pytest>=7.0; extra == "dev"
 Requires-Dist: pytest-cov>=4.0; extra == "dev"
 Requires-Dist: ruff>=0.8.0; extra == "dev"
 Requires-Dist: mypy>=1.0.0; extra == "dev"
-Requires-Dist: openai>=1.0.0; extra == "dev"
-Requires-Dist: httpx>=0.27.0; extra == "dev"
+Requires-Dist: openai>=2.0.0; extra == "dev"
+Requires-Dist: httpx>=0.28.0; extra == "dev"
 Provides-Extra: maintain
 Requires-Dist: pylance>=0.15.0; extra == "maintain"
 Provides-Extra: all

{coderay-1.0.2 → coderay-1.0.3}/src/coderay.egg-info/SOURCES.txt

@@ -28,12 +28,14 @@ src/coderay/graph/builder.py
 src/coderay/graph/code_graph.py
 src/coderay/graph/extractor.py
 src/coderay/mcp_server/__init__.py
+src/coderay/mcp_server/errors.py
 src/coderay/mcp_server/server.py
 src/coderay/pipeline/__init__.py
 src/coderay/pipeline/indexer.py
 src/coderay/pipeline/watcher.py
 src/coderay/retrieval/__init__.py
 src/coderay/retrieval/boosting.py
+src/coderay/retrieval/models.py
 src/coderay/retrieval/search.py
 src/coderay/skeleton/__init__.py
 src/coderay/skeleton/extractor.py

coderay-1.0.3/src/coderay.egg-info/requires.txt

@@ -0,0 +1,34 @@
+python-dotenv>=1.2.0
+networkx>=3.4
+tree-sitter>=0.25.0
+tree-sitter-python>=0.25.0
+lancedb>=0.29.0
+pyyaml>=6.0.3
+click>=8.3.0
+filelock>=3.25.0
+fastembed>=0.7.0
+fastmcp==3.1.0
+watchdog>=6.0.0
+pathspec>=1.0.0
+
+[all]
+coderay[dev,languages,maintain,openai]
+
+[dev]
+pytest>=7.0
+pytest-cov>=4.0
+ruff>=0.8.0
+mypy>=1.0.0
+openai>=2.0.0
+httpx>=0.28.0
+
+[languages]
+tree-sitter-javascript>=0.25.0
+tree-sitter-typescript>=0.23.2
+tree-sitter-go>=0.25.0
+
+[maintain]
+pylance>=0.15.0
+
+[openai]
+openai>=2.0.0

coderay-1.0.2/src/coderay/mcp_server/server.py

@@ -1,178 +0,0 @@
-from __future__ import annotations
-
-import json
-import logging
-from pathlib import Path
-from typing import Any
-
-from mcp.server.fastmcp import FastMCP
-
-logger = logging.getLogger(__name__)
-
-mcp = FastMCP("coderay")
-
-DEFAULT_INDEX_DIR = ".index"
-
-_retrieval_cache: dict[Path, Any] = {}
-_state_machine_cache: dict[Path, Any] = {}
-
-
-def _resolve_index_dir(index_dir: str | None = None) -> Path:
-    """Resolve the index directory to an absolute path."""
-    return Path(index_dir or DEFAULT_INDEX_DIR).resolve()
-
-
-def _get_retrieval(index_dir: str | None = None):
-    """Return a cached Retrieval instance for the given index directory."""
-    idx = _resolve_index_dir(index_dir)
-    if idx not in _retrieval_cache:
-        from coderay.retrieval.search import Retrieval
-
-        _retrieval_cache[idx] = Retrieval(idx)
-    return _retrieval_cache[idx]
-
-
-def _load_graph(index_dir: str | None = None):
-    """Load the code graph from disk, or return None if absent."""
-    from coderay.graph.builder import load_graph
-
-    return load_graph(_resolve_index_dir(index_dir))
-
-
-def _get_state_machine(index_dir: str | None = None):
-    """Return a cached StateMachine instance for the given index directory."""
-    idx = _resolve_index_dir(index_dir)
-    if idx not in _state_machine_cache:
-        from coderay.state.machine import StateMachine
-
-        _state_machine_cache[idx] = StateMachine(idx)
-    return _state_machine_cache[idx]
-
-
-def _load_state(index_dir: str | None = None):
-    """Load the current IndexMeta state, or None if no run has completed."""
-    return _get_state_machine(index_dir).current_state
-
-
-@mcp.tool()
-def semantic_search(
-    query: str,
-    top_k: int = 10,
-    path_prefix: str | None = None,
-    language: str | None = None,
-    index_dir: str | None = None,
-) -> str:
-    """Search code by meaning."""
-    retrieval = _get_retrieval(index_dir)
-    state = _load_state(index_dir)
-    if state is None:
-        return json.dumps({"error": "No index state found. Run 'coderay build' first."})
-    try:
-        results = retrieval.search(
-            query,
-            state,
-            top_k=top_k,
-            path_prefix=path_prefix,
-            language=language,
-        )
-    except RuntimeError as e:
-        return json.dumps({"error": str(e)})
-    score_type = results[0].get("score_type", "cosine") if results else "cosine"
-    return json.dumps(
-        {
-            "score_type": score_type,
-            "score_description": (
-                "cosine similarity (0-1, higher = more similar)"
-                if score_type == "cosine"
-                else "RRF rank fusion (higher = more relevant, scale differs from cosine)"
-            ),
-            "results": results,
-        },
-        default=str,
-    )
-
-
-@mcp.tool()
-def get_file_skeleton(file_path: str) -> str:
-    """Get the API surface of a file (signatures, no bodies)."""
-    from coderay.skeleton.extractor import extract_skeleton
-
-    p = Path(file_path)
-    if not p.is_file():
-        return json.dumps({"error": f"File not found: {file_path}"})
-    content = p.read_text(encoding="utf-8", errors="replace")
-    return extract_skeleton(p, content)
-
-
-_STATIC_ANALYSIS_NOTE = (
-    "Based on static analysis of source code. Calls through dependency "
-    "injection, interfaces, dynamic dispatch (getattr), decorators, or "
-    "framework routing may not be detected."
-)
-
-
-@mcp.tool()
-def get_impact_radius(
-    node_id: str,
-    max_depth: int = 3,
-    index_dir: str | None = None,
-) -> str:
-    """Analyze the blast radius of changing a function or module."""
-    graph = _load_graph(index_dir)
-    if graph is None:
-        return json.dumps({"error": "No graph found. Run 'coderay build' first."})
-    impact = graph.get_impact_radius(node_id, depth=max_depth)
-    return json.dumps(
-        {
-            "results": [n.to_dict() for n in impact],
-            "note": _STATIC_ANALYSIS_NOTE,
-        }
-    )
-
-
-@mcp.tool()
-def index_status(index_dir: str | None = None) -> str:
-    """Check health and status of the semantic index."""
-    state = _load_state(index_dir)
-    if state is None:
-        return json.dumps({"status": "no_index", "message": "No index found."})
-
-    from coderay.core.config import get_embedding_dimensions, load_config
-    from coderay.state.version import read_index_version
-    from coderay.storage.lancedb import index_exists as idx_exists
-
-    idx_dir = _resolve_index_dir(index_dir)
-    has_store = idx_exists(idx_dir)
-    chunk_count = 0
-    if has_store:
-        from coderay.storage.lancedb import Store
-
-        config = load_config(idx_dir)
-        store = Store(idx_dir, dimensions=get_embedding_dimensions(config))
-        chunk_count = store.chunk_count()
-
-    return json.dumps(
-        {
-            "status": state.state.value,
-            "branch": state.branch,
-            "last_commit": state.last_commit,
-            "chunk_count": chunk_count,
-            "schema_version": read_index_version(idx_dir),
-            "has_store": has_store,
-        },
-        default=str,
-    )
-
-
-def main():
-    """Entry point for the coderay-mcp command."""
-    import sys
-
-    transport = "stdio"
-    if "--sse" in sys.argv:
-        transport = "sse"
-    mcp.run(transport=transport)
-
-
-if __name__ == "__main__":
-    main()

coderay-1.0.2/src/coderay.egg-info/requires.txt

@@ -1,34 +0,0 @@
-python-dotenv>=1.0.0
-networkx>=3.0
-tree-sitter>=0.24.0
-tree-sitter-python>=0.25.0
-lancedb>=0.5.0
-pyyaml>=6.0
-click>=8.0
-filelock>=3.0
-fastembed>=0.4.0
-mcp>=1.0.0
-watchdog>=4.0.0
-pathspec>=0.12.0
-
-[all]
-coderay[dev,languages,maintain,openai]
-
-[dev]
-pytest>=7.0
-pytest-cov>=4.0
-ruff>=0.8.0
-mypy>=1.0.0
-openai>=1.0.0
-httpx>=0.27.0
-
-[languages]
-tree-sitter-javascript>=0.23.0
-tree-sitter-typescript>=0.23.0
-tree-sitter-go>=0.23.0
-
-[maintain]
-pylance>=0.15.0
-
-[openai]
-openai>=1.0.0