kodit 0.1.14__py3-none-any.whl → 0.1.16__py3-none-any.whl

kodit/source/source_service.py

@@ -109,6 +109,8 @@ class SourceService:
                 uri_or_path_like = uri_or_path_like + ".git"
                 try:
                     return await self._create_git_source(uri_or_path_like)
+                except git.GitCommandError:
+                    raise
                 except ValueError:
                     pass
 
@@ -197,11 +199,14 @@ class SourceService:
             clone_path.mkdir(parents=True, exist_ok=True)
 
             try:
-                # Clone the repository
+                self.log.info("Cloning repository", uri=uri, clone_path=str(clone_path))
                 git.Repo.clone_from(uri, clone_path)
             except git.GitCommandError as e:
-                msg = f"Failed to clone repository: {e}"
-                raise ValueError(msg) from e
+                if "already exists and is not an empty directory" in str(e):
+                    self.log.info("Repository already exists, reusing...", uri=uri)
+                else:
+                    msg = f"Failed to clone repository: {e}"
+                    raise ValueError(msg) from e
 
             source = await self.repository.create_source(
                 Source(uri=uri, cloned_path=str(clone_path)),
@@ -212,6 +217,7 @@ class SourceService:
             file_count = sum(1 for _ in clone_path.rglob("*") if _.is_file())
 
             # Process each file in the source directory
+            self.log.info("Inspecting files", source_id=source.id)
             for path in tqdm(clone_path.rglob("*"), total=file_count, leave=False):
                 await self._process_file(source.id, path.absolute())
 

kodit/util/__init__.py

@@ -0,0 +1 @@
+"""Utility functions and classes."""

kodit/util/spinner.py

@@ -0,0 +1,59 @@
+"""Spinner for long-running tasks."""
+
+import itertools
+import sys
+import threading
+import time
+
+
+class Spinner:
+    """Spinner for long-running tasks."""
+
+    def __init__(self, delay: float = 0.1) -> None:
+        """Initialize the spinner."""
+        self.spinner = itertools.cycle(["-", "/", "|", "\\"])
+        self.delay = delay
+        self.busy = False
+        self.spinner_visible = False
+
+    def write_next(self) -> None:
+        """Write the next character of the spinner."""
+        with self._screen_lock:
+            if not self.spinner_visible:
+                sys.stdout.write(next(self.spinner))
+                self.spinner_visible = True
+                sys.stdout.flush()
+
+    def remove_spinner(self, cleanup: bool = False) -> None:  # noqa: FBT001, FBT002
+        """Remove the spinner."""
+        with self._screen_lock:
+            if self.spinner_visible:
+                sys.stdout.write("\b")
+                self.spinner_visible = False
+                if cleanup:
+                    sys.stdout.write(" ")  # overwrite spinner with blank
+                    sys.stdout.write("\r")  # move to next line
+                sys.stdout.flush()
+
+    def spinner_task(self) -> None:
+        """Task that runs the spinner."""
+        while self.busy:
+            self.write_next()
+            time.sleep(self.delay)
+            self.remove_spinner()
+
+    def __enter__(self) -> None:
+        """Enter the context manager."""
+        if sys.stdout.isatty():
+            self._screen_lock = threading.Lock()
+            self.busy = True
+            self.thread = threading.Thread(target=self.spinner_task)
+            self.thread.start()
+
+    def __exit__(self, exception: object, value: object, tb: object) -> None:
+        """Exit the context manager."""
+        if sys.stdout.isatty():
+            self.busy = False
+            self.remove_spinner(cleanup=True)
+        else:
+            sys.stdout.write("\r")

{kodit-0.1.14.dist-info → kodit-0.1.16.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: kodit
-Version: 0.1.14
+Version: 0.1.16
 Summary: Code indexing for better AI code generation
 Project-URL: Homepage, https://docs.helixml.tech/kodit/
 Project-URL: Documentation, https://docs.helixml.tech/kodit/
@@ -15,12 +15,14 @@ Keywords: ai,indexing,mcp,rag
 Classifier: Development Status :: 2 - Pre-Alpha
 Classifier: Intended Audience :: Developers
 Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
 Classifier: Topic :: Software Development :: Code Generators
 Requires-Python: >=3.12
 Requires-Dist: aiofiles>=24.1.0
 Requires-Dist: aiosqlite>=0.20.0
 Requires-Dist: alembic>=1.15.2
 Requires-Dist: asgi-correlation-id>=4.3.4
+Requires-Dist: asyncpg>=0.30.0
 Requires-Dist: better-exceptions>=0.3.3
 Requires-Dist: bm25s[core]>=0.2.12
 Requires-Dist: click>=8.1.8
@@ -41,6 +43,7 @@ Requires-Dist: sqlalchemy[asyncio]>=2.0.40
 Requires-Dist: structlog>=25.3.0
 Requires-Dist: tdqm>=0.0.1
 Requires-Dist: tiktoken>=0.9.0
+Requires-Dist: transformers>=4.51.3
 Requires-Dist: tree-sitter-language-pack>=0.7.3
 Requires-Dist: tree-sitter>=0.24.0
 Requires-Dist: uritools>=5.0.0

kodit-0.1.16.dist-info/RECORD

@@ -0,0 +1,64 @@
+kodit/.gitignore,sha256=ztkjgRwL9Uud1OEi36hGQeDGk3OLK1NfDEO8YqGYy8o,11
+kodit/__init__.py,sha256=aEKHYninUq1yh6jaNfvJBYg-6fenpN132nJt1UU6Jxs,59
+kodit/_version.py,sha256=VYJNWHISWEW-KD_clKUYcTY_Z30r993Sjws4URJIL0g,513
+kodit/app.py,sha256=Mr5BFHOHx5zppwjC4XPWVvHjwgl1yrKbUjTWXKubJQM,891
+kodit/cli.py,sha256=i7eEt0FdIQGEfXKFte-8fBcZZGE8BPXBp40aGwJDQGI,11323
+kodit/config.py,sha256=2W2u5J8j-Mbt-C4xzOuK-PeuDCx0S_rnCXPhBwvfLT4,4353
+kodit/database.py,sha256=WB1KpVxUYPgiJGU0gJa2hqytYB8wJEJ5z3WayhWzNMU,2403
+kodit/log.py,sha256=HU1OmuxO4FcVw61k4WW7Y4WM7BrDaeplw1PcBHhuIZY,5434
+kodit/mcp.py,sha256=QruyPskWB0_x59pkfj5BBeXuR13GMny5TAZEa2j4U9s,5752
+kodit/middleware.py,sha256=I6FOkqG9-8RH5kR1-0ZoQWfE4qLCB8lZYv8H_OCH29o,2714
+kodit/bm25/__init__.py,sha256=j8zyriNWhbwE5Lbybzg1hQAhANlU9mKHWw4beeUR6og,19
+kodit/bm25/keyword_search_factory.py,sha256=rp-wx3DJsc2KlELK1V337EyeYvmwnMQwUqOo1WVPSmg,631
+kodit/bm25/keyword_search_service.py,sha256=aBbWQKgQmi2re3EIHdXFS00n7Wj3b2D0pZsLZ4qmHfE,754
+kodit/bm25/local_bm25.py,sha256=AAbFhbQDqyL3d7jsPL7W4HsLxdoYctaDsREUXOLy6jM,3260
+kodit/bm25/vectorchord_bm25.py,sha256=_nGrkUReYLLV-L8RIuIVLwjuhSYZl9T532n5OVf0kWs,6393
+kodit/embedding/__init__.py,sha256=h9NXzDA1r-K23nvBajBV-RJzHJN0p3UJ7UQsmdnOoRw,24
+kodit/embedding/embedding_factory.py,sha256=UGnFRyyQXazSUOwyW4Hg7Vq2-kfAoDj9lD4CTLu8x04,1630
+kodit/embedding/embedding_models.py,sha256=rN90vSs86dYiqoawcp8E9jtwY31JoJXYfaDlsJK7uqc,656
+kodit/embedding/embedding_repository.py,sha256=-ux3scpBzel8c0pMH9fNOEsSXFIzl-IfgaWrkTb1szo,6907
+kodit/embedding/local_vector_search_service.py,sha256=hkF0qlfzjyGt400qIX9Mr6B7b7i8WvYIYWN2Z2C_pcs,1907
+kodit/embedding/vector_search_service.py,sha256=pQJ129QjGrAWOXzqkywmgtDRpy8_gtzYgkivyqF9Vrs,1009
+kodit/embedding/vectorchord_vector_search_service.py,sha256=KSs0IMFHHIllwq2d3A0LGqGGZDqO1Ht6K-BCfBBWW0Y,5051
+kodit/embedding/embedding_provider/__init__.py,sha256=h9NXzDA1r-K23nvBajBV-RJzHJN0p3UJ7UQsmdnOoRw,24
+kodit/embedding/embedding_provider/embedding_provider.py,sha256=Tf3bwUsUMzAgoyLFM5qBtOLqPp1qr03TzrwGczkDvy0,1835
+kodit/embedding/embedding_provider/hash_embedding_provider.py,sha256=nAhlhh8j8PqqCCbhVl26Y8ntFBm2vJBCtB4X04g5Wwg,2638
+kodit/embedding/embedding_provider/local_embedding_provider.py,sha256=4ER-UPq506Y0TWU6qcs0nUqw6bSKQkSrdog-DhNQWM8,1906
+kodit/embedding/embedding_provider/openai_embedding_provider.py,sha256=V_jdUXiaGdslplwxMlfgFc4_hAVS2eaJXMTs2C7RiLI,2666
+kodit/enrichment/__init__.py,sha256=vBEolHpKaHUhfINX0dSGyAPlvgpLNAer9YzFtdvCB24,18
+kodit/enrichment/enrichment_factory.py,sha256=vKjkUTdhj74IW2S4GENDWdWMJx6BwUSZjJGDC0i7DSk,787
+kodit/enrichment/enrichment_service.py,sha256=87Sd3gGbEMJYb_wVrHG8L1yGIZmQNR7foUS4_y94azI,977
+kodit/enrichment/enrichment_provider/__init__.py,sha256=klf8iuLVWX4iRz-DZQauFFNAoJC5CByczh48TBZPW-o,27
+kodit/enrichment/enrichment_provider/enrichment_provider.py,sha256=E0H5rq3OENM0yYbA8K_3nSnj5lUHCpoIOqpWLo-2MVU,413
+kodit/enrichment/enrichment_provider/local_enrichment_provider.py,sha256=bR6HR1gH7wtZdMLOwaKdASjvllRo1FlNW9GyZC11zAM,2164
+kodit/enrichment/enrichment_provider/openai_enrichment_provider.py,sha256=gYuFTAeIVdQNlCUvNSPgRoiRwCvRD0C8419h8ubyABA,2725
+kodit/indexing/__init__.py,sha256=cPyi2Iej3G1JFWlWr7X80_UrsMaTu5W5rBwgif1B3xo,75
+kodit/indexing/fusion.py,sha256=TZb4fPAedXdEUXzwzOofW98QIOymdbclBOP1KOijuEk,1674
+kodit/indexing/indexing_models.py,sha256=6NX9HVcj6Pu9ePwHC7n-PWSyAgukpJq0nCNmUIigtbo,1282
+kodit/indexing/indexing_repository.py,sha256=GYHoACUWYKQdVTwP7tfik_TMUD1WUK76nywH88eCSwg,7006
+kodit/indexing/indexing_service.py,sha256=tKcZpi0pzsmF6OpqnqF0Q5HfSXxi5iLTysrVSou4JiQ,10579
+kodit/migrations/README,sha256=ISVtAOvqvKk_5ThM5ioJE-lMkvf9IbknFUFVU_vPma4,58
+kodit/migrations/__init__.py,sha256=lP5MuwlyWRMO6UcDWnQcQ3G-GYHcFb6rl9gYPHJ1sjo,40
+kodit/migrations/env.py,sha256=w1M7OZh-ZeR2dPHS0ByXAUxQjfZQ8xIzMseWuzLDTWw,2469
+kodit/migrations/script.py.mako,sha256=zWziKtiwYKEWuwPV_HBNHwa9LCT45_bi01-uSNFaOOE,703
+kodit/migrations/versions/7c3bbc2ab32b_add_embeddings_table.py,sha256=-61qol9PfQKILCDQRA5jEaats9aGZs9Wdtp-j-38SF4,1644
+kodit/migrations/versions/85155663351e_initial.py,sha256=Cg7zlF871o9ShV5rQMQ1v7hRV7fI59veDY9cjtTrs-8,3306
+kodit/migrations/versions/__init__.py,sha256=9-lHzptItTzq_fomdIRBegQNm4Znx6pVjwD4MiqRIdo,36
+kodit/snippets/__init__.py,sha256=-2coNoCRjTixU9KcP6alpmt7zqf37tCRWH3D7FPJ8dg,48
+kodit/snippets/method_snippets.py,sha256=EVHhSNWahAC5nSXv9fWVFJY2yq25goHdCSCuENC07F8,4145
+kodit/snippets/snippets.py,sha256=mwN0bM1Msu8ZeEsUHyQ7tx3Hj3vZsm8G7Wu4eWSkLY8,1539
+kodit/snippets/languages/__init__.py,sha256=Bj5KKZSls2MQ8ZY1S_nHg447MgGZW-2WZM-oq6vjwwA,1187
+kodit/snippets/languages/csharp.scm,sha256=gbBN4RiV1FBuTJF6orSnDFi8H9JwTw-d4piLJYsWUsc,222
+kodit/snippets/languages/go.scm,sha256=SEX9mTOrhP2KiQW7oflDKkd21u5dK56QbJ4LvTDxY8A,533
+kodit/snippets/languages/python.scm,sha256=ee85R9PBzwye3IMTE7-iVoKWd_ViU3EJISTyrFGrVeo,429
+kodit/source/__init__.py,sha256=1NTZyPdjThVQpZO1Mp1ColVsS7sqYanOVLqnoqV9Ipo,83
+kodit/source/source_models.py,sha256=xb42CaNDO1CUB8SIW-xXMrB6Ji8cFw-yeJ550xBEg9Q,2398
+kodit/source/source_repository.py,sha256=0EksMpoLzdkfe8S4eeCm4Sf7TuxsOzOzaF4BBsMYo-4,3163
+kodit/source/source_service.py,sha256=u_GaH07ewakThQJRfT8O_yZ54A52qLtJuM1bF3xUT2A,9633
+kodit/util/__init__.py,sha256=bPu6CtqDWCRGU7VgW2_aiQrCBi8G89FS6k1PjvDajJ0,37
+kodit/util/spinner.py,sha256=R9bzrHtBiIH6IfLbmsIVHL53s8vg-tqW4lwGGALu4dw,1932
+kodit-0.1.16.dist-info/METADATA,sha256=1lR4ZSTiRBzUv9Gj8FPspv4GU2vWGQU6HSiffWgU2Do,2467
+kodit-0.1.16.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+kodit-0.1.16.dist-info/entry_points.txt,sha256=hoTn-1aKyTItjnY91fnO-rV5uaWQLQ-Vi7V5et2IbHY,40
+kodit-0.1.16.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+kodit-0.1.16.dist-info/RECORD,,

kodit/embedding/embedding.py

@@ -1,203 +0,0 @@
-"""Embedding service."""
-
-import asyncio
-import os
-from abc import ABC, abstractmethod
-from collections.abc import AsyncGenerator
-from typing import NamedTuple
-
-import structlog
-import tiktoken
-from openai import AsyncOpenAI
-from sentence_transformers import SentenceTransformer
-
-TINY = "tiny"
-CODE = "code"
-TEST = "test"
-
-COMMON_EMBEDDING_MODELS = {
-    TINY: "ibm-granite/granite-embedding-30m-english",
-    CODE: "flax-sentence-embeddings/st-codesearch-distilroberta-base",
-    TEST: "minishlab/potion-base-4M",
-}
-
-
-class EmbeddingInput(NamedTuple):
-    """Input for embedding."""
-
-    id: int
-    text: str
-
-
-class EmbeddingOutput(NamedTuple):
-    """Output for embedding."""
-
-    id: int
-    embedding: list[float]
-
-
-class Embedder(ABC):
-    """Embedder interface."""
-
-    @abstractmethod
-    def embed(
-        self, data: list[EmbeddingInput]
-    ) -> AsyncGenerator[EmbeddingOutput, None]:
-        """Embed a list of documents.
-
-        The embedding service accepts a massive list of id,strings to embed. Behind the
-        scenes it batches up requests and parallelizes them for performance according to
-        the specifics of the embedding service.
-
-        The id reference is required because the parallelization may return results out
-        of order.
-        """
-
-    @abstractmethod
-    def query(self, data: list[str]) -> AsyncGenerator[list[float], None]:
-        """Query the embedding model."""
-
-
-def embedding_factory(openai_client: AsyncOpenAI | None = None) -> Embedder:
-    """Create an embedding service."""
-    if openai_client is not None:
-        return OpenAIEmbedder(openai_client)
-    return LocalEmbedder(model_name=TINY)
-
-
-class LocalEmbedder(Embedder):
-    """Local embedder."""
-
-    def __init__(self, model_name: str) -> None:
-        """Initialize the local embedder."""
-        self.log = structlog.get_logger(__name__)
-        self.log.info("Creating local embedder", model_name=model_name)
-        self.model_name = COMMON_EMBEDDING_MODELS.get(model_name, model_name)
-        self.embedding_model = None
-        self.encoding = tiktoken.encoding_for_model("text-embedding-3-small")
-
-    def _model(self) -> SentenceTransformer:
-        """Get the embedding model."""
-        if self.embedding_model is None:
-            os.environ["TOKENIZERS_PARALLELISM"] = "false"  # Avoid warnings
-            self.embedding_model = SentenceTransformer(
-                self.model_name,
-                trust_remote_code=True,
-                device="cpu",  # Force CPU so we don't have to install accelerate, etc.
-            )
-        return self.embedding_model
-
-    async def embed(
-        self, data: list[EmbeddingInput]
-    ) -> AsyncGenerator[EmbeddingOutput, None]:
-        """Embed a list of documents."""
-        model = self._model()
-
-        batched_data = _split_sub_batches(self.encoding, data)
-
-        for batch in batched_data:
-            embeddings = model.encode(
-                [i.text for i in batch], show_progress_bar=False, batch_size=4
-            )
-            for i, x in zip(batch, embeddings, strict=False):
-                yield EmbeddingOutput(i.id, [float(y) for y in x])
-
-    async def query(self, data: list[str]) -> AsyncGenerator[list[float], None]:
-        """Query the embedding model."""
-        model = self._model()
-        embeddings = model.encode(data, show_progress_bar=False, batch_size=4)
-        for embedding in embeddings:
-            yield [float(x) for x in embedding]
-
-
-OPENAI_MAX_EMBEDDING_SIZE = 8192
-OPENAI_NUM_PARALLEL_TASKS = 10
-
-
-def _split_sub_batches(
-    encoding: tiktoken.Encoding, data: list[EmbeddingInput]
-) -> list[list[EmbeddingInput]]:
-    """Split a list of strings into smaller sub-batches."""
-    log = structlog.get_logger(__name__)
-    result = []
-    data_to_process = [s for s in data if s.text.strip()]  # Filter out empty strings
-
-    while data_to_process:
-        next_batch = []
-        current_tokens = 0
-
-        while data_to_process:
-            next_item = data_to_process[0]
-            item_tokens = len(encoding.encode(next_item.text))
-
-            if item_tokens > OPENAI_MAX_EMBEDDING_SIZE:
-                log.warning("Skipping too long snippet", snippet=data_to_process.pop(0))
-                continue
-
-            if current_tokens + item_tokens > OPENAI_MAX_EMBEDDING_SIZE:
-                break
-
-            next_batch.append(data_to_process.pop(0))
-            current_tokens += item_tokens
-
-        if next_batch:
-            result.append(next_batch)
-
-    return result
-
-
-class OpenAIEmbedder(Embedder):
-    """OpenAI embedder."""
-
-    def __init__(
-        self, openai_client: AsyncOpenAI, model_name: str = "text-embedding-3-small"
-    ) -> None:
-        """Initialize the OpenAI embedder."""
-        self.log = structlog.get_logger(__name__)
-        self.log.info("Creating OpenAI embedder", model_name=model_name)
-        self.openai_client = openai_client
-        self.encoding = tiktoken.encoding_for_model(model_name)
-        self.log = structlog.get_logger(__name__)
-
-    async def embed(
-        self,
-        data: list[EmbeddingInput],
-    ) -> AsyncGenerator[EmbeddingOutput, None]:
-        """Embed a list of documents."""
-        # First split the list into a list of list where each sublist has fewer than
-        # max tokens.
-        batched_data = _split_sub_batches(self.encoding, data)
-
-        # Process batches in parallel with a semaphore to limit concurrent requests
-        sem = asyncio.Semaphore(OPENAI_NUM_PARALLEL_TASKS)
-
-        async def process_batch(batch: list[EmbeddingInput]) -> list[EmbeddingOutput]:
-            async with sem:
-                try:
-                    response = await self.openai_client.embeddings.create(
-                        model="text-embedding-3-small",
-                        input=[i.text for i in batch],
-                    )
-                    return [
-                        EmbeddingOutput(i.id, x.embedding)
-                        for i, x in zip(batch, response.data, strict=False)
-                    ]
-                except Exception as e:
-                    self.log.exception("Error embedding batch", error=str(e))
-                    return []
-
-        # Create tasks for all batches
-        tasks = [process_batch(batch) for batch in batched_data]
-
-        # Process all batches and yield results as they complete
-        for task in asyncio.as_completed(tasks):
-            embeddings = await task
-            for e in embeddings:
-                yield e
-
-    async def query(self, data: list[str]) -> AsyncGenerator[list[float], None]:
-        """Query the embedding model."""
-        async for e in self.embed(
-            [EmbeddingInput(i, text) for i, text in enumerate(data)]
-        ):
-            yield e.embedding

kodit/search/__init__.py

@@ -1 +0,0 @@
-"""Search for relevant snippets."""

kodit/search/search_service.py

@@ -1,147 +0,0 @@
-"""Search service."""
-
-from pathlib import Path
-
-import pydantic
-import structlog
-
-from kodit.bm25.bm25 import BM25Service
-from kodit.embedding.embedding import Embedder
-from kodit.embedding.embedding_models import EmbeddingType
-from kodit.search.search_repository import SearchRepository
-
-
-class SearchRequest(pydantic.BaseModel):
-    """Request for a search."""
-
-    code_query: str | None = None
-    keywords: list[str] | None = None
-    top_k: int = 10
-
-
-class SearchResult(pydantic.BaseModel):
-    """Data transfer object for search results.
-
-    This model represents a single search result, containing both the file path
-    and the matching snippet content.
-    """
-
-    id: int
-    uri: str
-    content: str
-
-
-class Snippet(pydantic.BaseModel):
-    """Snippet model."""
-
-    content: str
-    file_path: str
-
-
-class SearchService:
-    """Service for searching for relevant data."""
-
-    def __init__(
-        self,
-        repository: SearchRepository,
-        data_dir: Path,
-        embedding_service: Embedder,
-    ) -> None:
-        """Initialize the search service."""
-        self.repository = repository
-        self.log = structlog.get_logger(__name__)
-        self.bm25 = BM25Service(data_dir)
-        self.code_embedding_service = embedding_service
-
-    async def search(self, request: SearchRequest) -> list[SearchResult]:
-        """Search for relevant data."""
-        fusion_list = []
-        if request.keywords:
-            snippet_ids = await self.repository.list_snippet_ids()
-
-            # Gather results for each keyword
-            result_ids: list[tuple[int, float]] = []
-            for keyword in request.keywords:
-                results = self.bm25.retrieve(snippet_ids, keyword, request.top_k)
-                result_ids.extend(results)
-
-            # Sort results by score
-            result_ids.sort(key=lambda x: x[1], reverse=True)
-
-            self.log.debug("Search results (BM25)", results=result_ids)
-
-            bm25_results = [x[0] for x in result_ids]
-            fusion_list.append(bm25_results)
-
-        # Compute embedding for semantic query
-        semantic_results = []
-        if request.code_query:
-            query_embedding = await anext(
-                self.code_embedding_service.query([request.code_query])
-            )
-
-            query_results = await self.repository.list_semantic_results(
-                EmbeddingType.CODE, query_embedding, top_k=request.top_k
-            )
-
-            # Sort results by score
-            query_results.sort(key=lambda x: x[1], reverse=True)
-
-            # Extract the snippet ids from the query results
-            semantic_results = [x[0] for x in query_results]
-            fusion_list.append(semantic_results)
-
-        if len(fusion_list) == 0:
-            return []
-
-        # Combine all results together with RFF if required
-        final_results = reciprocal_rank_fusion(fusion_list, k=60)
-
-        # Extract ids from final results
-        final_ids = [x[0] for x in final_results]
-
-        # Get snippets from database (up to top_k)
-        search_results = await self.repository.list_snippets_by_ids(
-            final_ids[: request.top_k]
-        )
-
-        return [
-            SearchResult(
-                id=snippet.id,
-                uri=file.uri,
-                content=snippet.content,
-            )
-            for file, snippet in search_results
-        ]
-
-
-def reciprocal_rank_fusion(
-    rankings: list[list[int]], k: float = 60
-) -> list[tuple[int, float]]:
-    """RRF prioritises results that are present in all results.
-
-    Args:
-        rankings: List of rankers, each containing a list of document ids. Top of the
-        list is considered to be the best result.
-        k: Parameter for RRF.
-
-    Returns:
-        Dictionary of ids and their scores.
-
-    """
-    scores = {}
-    for ranker in rankings:
-        for rank in ranker:
-            scores[rank] = float(0)
-
-    for ranker in rankings:
-        for i, rank in enumerate(ranker):
-            scores[rank] += 1.0 / (k + i)
-
-    # Create a list of tuples of ids and their scores
-    results = [(rank, scores[rank]) for rank in scores]
-
-    # Sort results by score
-    results.sort(key=lambda x: x[1], reverse=True)
-
-    return results

kodit-0.1.14.dist-info/RECORD

@@ -1,44 +0,0 @@
-kodit/.gitignore,sha256=ztkjgRwL9Uud1OEi36hGQeDGk3OLK1NfDEO8YqGYy8o,11
-kodit/__init__.py,sha256=aEKHYninUq1yh6jaNfvJBYg-6fenpN132nJt1UU6Jxs,59
-kodit/_version.py,sha256=O_r2EWoixTKREu-RyeL8e93UHfqprj1LCIlwiWXfHcg,513
-kodit/app.py,sha256=Mr5BFHOHx5zppwjC4XPWVvHjwgl1yrKbUjTWXKubJQM,891
-kodit/cli.py,sha256=VLoXFS1xJnQ0TLy3_cO8-B9tCb4NJHiYPfzZtHxpgRY,7784
-kodit/config.py,sha256=TDcLt6fiJn9cI1PoO5AqBqsL_Bxmm9JV5GqRxhj1tLw,4202
-kodit/database.py,sha256=kekSdyEATdb47jxzQemkSOXMNOwnUwmVVTpn9hYaDK8,2356
-kodit/log.py,sha256=HU1OmuxO4FcVw61k4WW7Y4WM7BrDaeplw1PcBHhuIZY,5434
-kodit/mcp.py,sha256=ot5CIH240mSXK3sJcxTf4lBfthq0tcMS8XBGTaHY-n8,5088
-kodit/middleware.py,sha256=I6FOkqG9-8RH5kR1-0ZoQWfE4qLCB8lZYv8H_OCH29o,2714
-kodit/bm25/__init__.py,sha256=j8zyriNWhbwE5Lbybzg1hQAhANlU9mKHWw4beeUR6og,19
-kodit/bm25/bm25.py,sha256=JtgJfsHz-2SHx96zxWjkPFSH7fXkahFMp01cDwl4YBg,2298
-kodit/embedding/__init__.py,sha256=h9NXzDA1r-K23nvBajBV-RJzHJN0p3UJ7UQsmdnOoRw,24
-kodit/embedding/embedding.py,sha256=EMJpHK8ICZk_FjiO9Aqr2IO20qkGOmj_PfA1hyfI7Vk,6745
-kodit/embedding/embedding_models.py,sha256=rN90vSs86dYiqoawcp8E9jtwY31JoJXYfaDlsJK7uqc,656
-kodit/indexing/__init__.py,sha256=cPyi2Iej3G1JFWlWr7X80_UrsMaTu5W5rBwgif1B3xo,75
-kodit/indexing/indexing_models.py,sha256=6NX9HVcj6Pu9ePwHC7n-PWSyAgukpJq0nCNmUIigtbo,1282
-kodit/indexing/indexing_repository.py,sha256=7bkAiBwtr3qlkdhNIalwMwbxezVz_RQGOhLVWPKHwNk,5506
-kodit/indexing/indexing_service.py,sha256=VGfKgbkYEAYP_gIubvhMxo3yThT20ndS5xdg2LxwRgA,6685
-kodit/migrations/README,sha256=ISVtAOvqvKk_5ThM5ioJE-lMkvf9IbknFUFVU_vPma4,58
-kodit/migrations/__init__.py,sha256=lP5MuwlyWRMO6UcDWnQcQ3G-GYHcFb6rl9gYPHJ1sjo,40
-kodit/migrations/env.py,sha256=w1M7OZh-ZeR2dPHS0ByXAUxQjfZQ8xIzMseWuzLDTWw,2469
-kodit/migrations/script.py.mako,sha256=zWziKtiwYKEWuwPV_HBNHwa9LCT45_bi01-uSNFaOOE,703
-kodit/migrations/versions/7c3bbc2ab32b_add_embeddings_table.py,sha256=-61qol9PfQKILCDQRA5jEaats9aGZs9Wdtp-j-38SF4,1644
-kodit/migrations/versions/85155663351e_initial.py,sha256=Cg7zlF871o9ShV5rQMQ1v7hRV7fI59veDY9cjtTrs-8,3306
-kodit/migrations/versions/__init__.py,sha256=9-lHzptItTzq_fomdIRBegQNm4Znx6pVjwD4MiqRIdo,36
-kodit/search/__init__.py,sha256=4QbdjbrlhNKMovmuKHxJnUeZT7KNjTTFU0GdnuwUHdQ,36
-kodit/search/search_repository.py,sha256=r1fkV6-cy9BKsy5J4WTHaY_FcjMaT1PV5qqqq0gvjZw,5833
-kodit/search/search_service.py,sha256=KePkqCAc3CUcrpNsbDc5DqbF6W2m0TG6TDa9-VSJZS0,4227
-kodit/snippets/__init__.py,sha256=-2coNoCRjTixU9KcP6alpmt7zqf37tCRWH3D7FPJ8dg,48
-kodit/snippets/method_snippets.py,sha256=EVHhSNWahAC5nSXv9fWVFJY2yq25goHdCSCuENC07F8,4145
-kodit/snippets/snippets.py,sha256=mwN0bM1Msu8ZeEsUHyQ7tx3Hj3vZsm8G7Wu4eWSkLY8,1539
-kodit/snippets/languages/__init__.py,sha256=Bj5KKZSls2MQ8ZY1S_nHg447MgGZW-2WZM-oq6vjwwA,1187
-kodit/snippets/languages/csharp.scm,sha256=gbBN4RiV1FBuTJF6orSnDFi8H9JwTw-d4piLJYsWUsc,222
-kodit/snippets/languages/python.scm,sha256=ee85R9PBzwye3IMTE7-iVoKWd_ViU3EJISTyrFGrVeo,429
-kodit/source/__init__.py,sha256=1NTZyPdjThVQpZO1Mp1ColVsS7sqYanOVLqnoqV9Ipo,83
-kodit/source/source_models.py,sha256=xb42CaNDO1CUB8SIW-xXMrB6Ji8cFw-yeJ550xBEg9Q,2398
-kodit/source/source_repository.py,sha256=0EksMpoLzdkfe8S4eeCm4Sf7TuxsOzOzaF4BBsMYo-4,3163
-kodit/source/source_service.py,sha256=qBV9FCFQbJppeFrVo4uMgvC_mzWRIKldymp5yqLx9pw,9255
-kodit-0.1.14.dist-info/METADATA,sha256=acFpcf0ODyUSnA1hg4BPlLexpOEh-0yuaqsaWUNopOs,2349
-kodit-0.1.14.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-kodit-0.1.14.dist-info/entry_points.txt,sha256=hoTn-1aKyTItjnY91fnO-rV5uaWQLQ-Vi7V5et2IbHY,40
-kodit-0.1.14.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-kodit-0.1.14.dist-info/RECORD,,