kodit 0.3.1__py3-none-any.whl → 0.3.2__py3-none-any.whl

kodit/_version.py

@@ -17,5 +17,5 @@ __version__: str
 __version_tuple__: VERSION_TUPLE
 version_tuple: VERSION_TUPLE
 
-__version__ = version = '0.3.1'
-__version_tuple__ = version_tuple = (0, 3, 1)
+__version__ = version = '0.3.2'
+__version_tuple__ = version_tuple = (0, 3, 2)

kodit/application/services/code_indexing_application_service.py

@@ -157,7 +157,7 @@ class CodeIndexingApplicationService:
             snippet_results = await self.snippet_domain_service.search_snippets(
                 prefilter_request
             )
-            filtered_snippet_ids = [snippet.id for snippet in snippet_results]
+            filtered_snippet_ids = [snippet.snippet.id for snippet in snippet_results]
 
         # Gather results from different search modes
         fusion_list: list[list[FusionRequest]] = []
@@ -225,9 +225,20 @@ class CodeIndexingApplicationService:
         return [
             MultiSearchResult(
                 id=result.snippet.id,
-                uri=result.file.uri,
                 content=result.snippet.content,
                 original_scores=fr.original_scores,
+                # Enhanced fields
+                source_uri=result.source.uri,
+                relative_path=MultiSearchResult.calculate_relative_path(
+                    result.file.cloned_path, result.source.cloned_path
+                ),
+                language=MultiSearchResult.detect_language_from_extension(
+                    result.file.extension
+                ),
+                authors=[author.name for author in result.authors],
+                created_at=result.snippet.created_at,
+                # Summary from snippet entity
+                summary=result.snippet.summary,
             )
             for result, fr in zip(search_results, final_results, strict=True)
         ]
@@ -300,16 +311,8 @@ class CodeIndexingApplicationService:
         async for result in self.enrichment_service.enrich_documents(
             enrichment_request
         ):
-            # Update snippet content through domain service
-            enriched_content = (
-                result.text
-                + "\n\n```\n"
-                + next(s.content for s in snippets if s.id == result.snippet_id)
-                + "\n```"
-            )
-
-            await self.snippet_domain_service.update_snippet_content(
-                result.snippet_id, enriched_content
+            await self.snippet_domain_service.update_snippet_summary(
+                result.snippet_id, result.text
             )
 
             processed += 1

kodit/cli.py

@@ -20,7 +20,11 @@ from kodit.config import (
 )
 from kodit.domain.errors import EmptySourceError
 from kodit.domain.services.source_service import SourceService
-from kodit.domain.value_objects import MultiSearchRequest, SnippetSearchFilters
+from kodit.domain.value_objects import (
+    MultiSearchRequest,
+    MultiSearchResult,
+    SnippetSearchFilters,
+)
 from kodit.infrastructure.ui.progress import (
     create_lazy_progress_callback,
     create_multi_stage_progress_callback,
@@ -219,6 +223,7 @@ def _parse_filters(
 @click.option(
     "--source-repo", help="Filter by source repository (e.g., github.com/example/repo)"
 )
+@click.option("--output-format", default="text", help="Format to display snippets in")
 @with_app_context
 @with_session
 async def code(  # noqa: PLR0913
@@ -231,6 +236,7 @@ async def code(  # noqa: PLR0913
     created_after: str | None,
     created_before: str | None,
     source_repo: str | None,
+    output_format: str,
 ) -> None:
     """Search for snippets using semantic code search.
 
@@ -259,8 +265,10 @@ async def code(  # noqa: PLR0913
         click.echo("No snippets found")
         return
 
-    for snippet in snippets:
-        click.echo(str(snippet))
+    if output_format == "text":
+        click.echo(MultiSearchResult.to_string(snippets))
+    elif output_format == "json":
+        click.echo(MultiSearchResult.to_jsonlines(snippets))
 
 
 @search.command()
@@ -279,6 +287,7 @@ async def code(  # noqa: PLR0913
 @click.option(
     "--source-repo", help="Filter by source repository (e.g., github.com/example/repo)"
 )
+@click.option("--output-format", default="text", help="Format to display snippets in")
 @with_app_context
 @with_session
 async def keyword(  # noqa: PLR0913
@@ -291,6 +300,7 @@ async def keyword(  # noqa: PLR0913
     created_after: str | None,
     created_before: str | None,
     source_repo: str | None,
+    output_format: str,
 ) -> None:
     """Search for snippets using keyword search."""
     log_event("kodit.cli.search.keyword")
@@ -316,8 +326,10 @@ async def keyword(  # noqa: PLR0913
         click.echo("No snippets found")
         return
 
-    for snippet in snippets:
-        click.echo(str(snippet))
+    if output_format == "text":
+        click.echo(MultiSearchResult.to_string(snippets))
+    elif output_format == "json":
+        click.echo(MultiSearchResult.to_jsonlines(snippets))
 
 
 @search.command()
@@ -336,6 +348,7 @@ async def keyword(  # noqa: PLR0913
 @click.option(
     "--source-repo", help="Filter by source repository (e.g., github.com/example/repo)"
 )
+@click.option("--output-format", default="text", help="Format to display snippets in")
 @with_app_context
 @with_session
 async def text(  # noqa: PLR0913
@@ -348,6 +361,7 @@ async def text(  # noqa: PLR0913
     created_after: str | None,
     created_before: str | None,
     source_repo: str | None,
+    output_format: str,
 ) -> None:
     """Search for snippets using semantic text search.
 
@@ -376,8 +390,10 @@ async def text(  # noqa: PLR0913
         click.echo("No snippets found")
         return
 
-    for snippet in snippets:
-        click.echo(str(snippet))
+    if output_format == "text":
+        click.echo(MultiSearchResult.to_string(snippets))
+    elif output_format == "json":
+        click.echo(MultiSearchResult.to_jsonlines(snippets))
 
 
 @search.command()
@@ -398,6 +414,7 @@ async def text(  # noqa: PLR0913
 @click.option(
     "--source-repo", help="Filter by source repository (e.g., github.com/example/repo)"
 )
+@click.option("--output-format", default="text", help="Format to display snippets in")
 @with_app_context
 @with_session
 async def hybrid(  # noqa: PLR0913
@@ -412,6 +429,7 @@ async def hybrid(  # noqa: PLR0913
     created_after: str | None,
     created_before: str | None,
     source_repo: str | None,
+    output_format: str,
 ) -> None:
     """Search for snippets using hybrid search."""
     log_event("kodit.cli.search.hybrid")
@@ -446,8 +464,10 @@ async def hybrid(  # noqa: PLR0913
         click.echo("No snippets found")
         return
 
-    for snippet in snippets:
-        click.echo(str(snippet))
+    if output_format == "text":
+        click.echo(MultiSearchResult.to_string(snippets))
+    elif output_format == "json":
+        click.echo(MultiSearchResult.to_jsonlines(snippets))
 
 
 @cli.group()
@@ -458,6 +478,7 @@ def show() -> None:
 @show.command()
 @click.option("--by-path", help="File or directory path to search for snippets")
 @click.option("--by-source", help="Source URI to filter snippets by")
+@click.option("--output-format", default="text", help="Format to display snippets in")
 @with_app_context
 @with_session
 async def snippets(
@@ -465,6 +486,7 @@ async def snippets(
     app_context: AppContext,
     by_path: str | None,
     by_source: str | None,
+    output_format: str,
 ) -> None:
     """Show snippets with optional filtering by path or source."""
     log_event("kodit.cli.show.snippets")
@@ -478,8 +500,10 @@ async def snippets(
         source_service=source_service,
     )
     snippets = await service.list_snippets(file_path=by_path, source_uri=by_source)
-    for snippet in snippets:
-        click.echo(str(snippet))
+    if output_format == "text":
+        click.echo(MultiSearchResult.to_string(snippets))
+    elif output_format == "json":
+        click.echo(MultiSearchResult.to_jsonlines(snippets))
 
 
 @cli.command()

kodit/domain/entities.py

@@ -183,10 +183,18 @@ class Snippet(Base, CommonMixin):
     file_id: Mapped[int] = mapped_column(ForeignKey("files.id"), index=True)
     index_id: Mapped[int] = mapped_column(ForeignKey("indexes.id"), index=True)
     content: Mapped[str] = mapped_column(UnicodeText, default="")
+    summary: Mapped[str] = mapped_column(UnicodeText, default="")
 
-    def __init__(self, file_id: int, index_id: int, content: str) -> None:
+    def __init__(
+        self,
+        file_id: int,
+        index_id: int,
+        content: str,
+        summary: str = "",
+    ) -> None:
         """Initialize the snippet."""
         super().__init__()
         self.file_id = file_id
         self.index_id = index_id
         self.content = content
+        self.summary = summary

kodit/domain/repositories.py

@@ -13,7 +13,7 @@ from kodit.domain.entities import (
 )
 from kodit.domain.value_objects import (
     MultiSearchRequest,
-    SnippetListItem,
+    SnippetWithContext,
 )
 
 T = TypeVar("T")
@@ -92,7 +92,7 @@ class SnippetRepository(GenericRepository[Snippet]):
 
     async def list_snippets(
         self, file_path: str | None = None, source_uri: str | None = None
-    ) -> Sequence[SnippetListItem]:
+    ) -> Sequence[SnippetWithContext]:
         """List snippets with optional filtering by file path and source URI.
 
         Args:
@@ -102,19 +102,19 @@ class SnippetRepository(GenericRepository[Snippet]):
             all sources.
 
         Returns:
-            A sequence of SnippetListItem instances matching the criteria
+            A sequence of SnippetWithContext instances matching the criteria
 
         """
         raise NotImplementedError
 
-    async def search(self, request: MultiSearchRequest) -> Sequence[SnippetListItem]:
+    async def search(self, request: MultiSearchRequest) -> Sequence[SnippetWithContext]:
         """Search snippets with filters.
 
         Args:
             request: The search request containing queries and optional filters.
 
         Returns:
-            A sequence of SnippetListItem instances matching the search criteria.
+            A sequence of SnippetWithContext instances matching the search criteria.
 
         """
         raise NotImplementedError

kodit/domain/services/indexing_service.py

@@ -8,7 +8,7 @@ from kodit.domain.value_objects import (
     FusionResult,
     IndexCreateRequest,
     IndexView,
-    SnippetWithFile,
+    SnippetWithContext,
 )
 
 
@@ -52,7 +52,7 @@ class IndexRepository(ABC):
         """Update the content of an existing snippet."""
 
     @abstractmethod
-    async def list_snippets_by_ids(self, ids: list[int]) -> list[SnippetWithFile]:
+    async def list_snippets_by_ids(self, ids: list[int]) -> list[SnippetWithContext]:
         """List snippets by IDs."""
 
 
@@ -191,7 +191,7 @@ class IndexingDomainService:
         """
         return self.fusion_service.reciprocal_rank_fusion(rankings, k)
 
-    async def get_snippets_by_ids(self, ids: list[int]) -> list[SnippetWithFile]:
+    async def get_snippets_by_ids(self, ids: list[int]) -> list[SnippetWithContext]:
         """Get snippets by IDs.
 
         Args:

kodit/domain/services/snippet_service.py

@@ -16,7 +16,7 @@ from kodit.domain.value_objects import (
     MultiSearchRequest,
     MultiSearchResult,
     SnippetExtractionRequest,
-    SnippetListItem,
+    SnippetWithContext,
 )
 from kodit.reporting import Reporter
 
@@ -93,6 +93,7 @@ class SnippetDomainService:
                         file_id=file.id,
                         index_id=index_id,
                         content=snippet_content,
+                        summary="",  # Initially empty, will be populated by enrichment
                     )
                     saved_snippet = await self.snippet_repository.save(snippet)
                     created_snippets.append(saved_snippet)
@@ -128,22 +129,16 @@ class SnippetDomainService:
         # This delegates to the repository but provides a domain-level interface
         return list(await self.snippet_repository.get_by_index(index_id))
 
-    async def update_snippet_content(self, snippet_id: int, content: str) -> None:
-        """Update the content of an existing snippet.
-
-        Args:
-            snippet_id: The ID of the snippet to update
-            content: The new content for the snippet
-
-        """
+    async def update_snippet_summary(self, snippet_id: int, summary: str) -> None:
+        """Update the summary of an existing snippet."""
         # Get the snippet first to ensure it exists
         snippet = await self.snippet_repository.get(snippet_id)
         if not snippet:
             msg = f"Snippet not found: {snippet_id}"
             raise ValueError(msg)
 
-        # Update the content
-        snippet.content = content
+        # Update the summary
+        snippet.summary = summary
         await self.snippet_repository.save(snippet)
 
     async def delete_snippets_for_index(self, index_id: int) -> None:
@@ -157,14 +152,14 @@ class SnippetDomainService:
 
     async def search_snippets(
         self, request: MultiSearchRequest
-    ) -> list[SnippetListItem]:
+    ) -> list[SnippetWithContext]:
         """Search snippets with filters.
 
         Args:
             request: The search request containing filters
 
         Returns:
-            List of matching snippet items
+            List of matching snippet items with context
 
         """
         return list(await self.snippet_repository.search(request))
@@ -185,13 +180,22 @@ class SnippetDomainService:
         snippet_items = await self.snippet_repository.list_snippets(
             file_path, source_uri
         )
-        # Convert SnippetListItem to MultiSearchResult for unified display format
+        # Convert SnippetWithContext to MultiSearchResult for unified display format
         return [
             MultiSearchResult(
-                id=item.id,
-                uri=item.source_uri,
-                content=item.content,
-                original_scores=[],
+                id=item.snippet.id,
+                content=item.snippet.content,
+                original_scores=[],  # No scores for list operation
+                source_uri=item.source.uri,
+                relative_path=MultiSearchResult.calculate_relative_path(
+                    item.file.cloned_path, item.source.cloned_path
+                ),
+                language=MultiSearchResult.detect_language_from_extension(
+                    item.file.extension
+                ),
+                authors=[author.name for author in item.authors],
+                created_at=item.snippet.created_at,
+                summary=item.snippet.summary,
             )
             for item in snippet_items
         ]

kodit/domain/value_objects.py

@@ -1,5 +1,6 @@
 """Domain value objects and DTOs."""
 
+import json
 from dataclasses import dataclass
 from datetime import datetime
 from enum import Enum
@@ -9,7 +10,7 @@ from typing import Any, ClassVar
 from sqlalchemy import JSON, DateTime, Integer, Text
 from sqlalchemy.orm import Mapped, mapped_column
 
-from kodit.domain.entities import Base
+from kodit.domain.entities import Author, Base, File, Snippet, Source
 from kodit.domain.enums import SnippetExtractionStrategy
 
 
@@ -175,25 +176,90 @@ class MultiSearchRequest:
 
 @dataclass
 class MultiSearchResult:
-    """Domain model for multi-modal search result."""
+    """Enhanced search result with comprehensive snippet metadata."""
 
     id: int
-    uri: str
     content: str
     original_scores: list[float]
+    source_uri: str
+    relative_path: str
+    language: str
+    authors: list[str]
+    created_at: datetime
+    summary: str
 
     def __str__(self) -> str:
-        """Return formatted string representation for all snippet display."""
+        """Return enhanced formatted string representation."""
         lines = [
-            "-" * 80,
-            f"ID: {self.id} | {self.uri}",
-            f"Original scores: {self.original_scores}",
-            self.content,
-            "-" * 80,
-            "",
+            "---",
+            f"id: {self.id}",
+            f"source: {self.source_uri}",
+            f"path: {self.relative_path}",
+            f"lang: {self.language}",
+            f"created: {self.created_at.isoformat()}",
+            f"authors: {', '.join(self.authors)}",
+            f"scores: {self.original_scores}",
+            "---",
+            f"{self.summary}\n",
+            f"```{self.language}",
+            f"{self.content}",
+            "```\n",
         ]
         return "\n".join(lines)
 
+    def to_json(self) -> str:
+        """Return LLM-optimized JSON representation following the compact schema."""
+        json_obj = {
+            "id": self.id,
+            "source": self.source_uri,
+            "path": self.relative_path,
+            "lang": self.language.lower(),
+            "created": self.created_at.isoformat() if self.created_at else "",
+            "author": ", ".join(self.authors),
+            "score": self.original_scores,
+            "code": self.content,
+            "summary": self.summary,
+        }
+
+        return json.dumps(json_obj, separators=(",", ":"))
+
+    @classmethod
+    def to_jsonlines(cls, results: list["MultiSearchResult"]) -> str:
+        """Convert multiple MultiSearchResult objects to JSON Lines format.
+
+        Args:
+            results: List of MultiSearchResult objects
+            include_summary: Whether to include summary fields
+
+        Returns:
+            JSON Lines string (one JSON object per line)
+
+        """
+        return "\n".join(result.to_json() for result in results)
+
+    @classmethod
+    def to_string(cls, results: list["MultiSearchResult"]) -> str:
+        """Convert multiple MultiSearchResult objects to a string."""
+        return "\n\n".join(str(result) for result in results)
+
+    @staticmethod
+    def calculate_relative_path(file_path: str, source_path: str) -> str:
+        """Calculate relative path from source root."""
+        try:
+            return str(Path(file_path).relative_to(Path(source_path)))
+        except ValueError:
+            # If file_path is not relative to source_path, return the file name
+            return Path(file_path).name
+
+    @staticmethod
+    def detect_language_from_extension(extension: str) -> str:
+        """Detect programming language from file extension."""
+        try:
+            return LanguageMapping.get_language_for_extension(extension).title()
+        except ValueError:
+            # Unknown extension, return a default
+            return "Unknown"
+
 
 @dataclass
 class FusionRequest:
@@ -292,36 +358,13 @@ class IndexView:
 
 
 @dataclass
-class SnippetListItem:
-    """Domain model for snippet list item with file information."""
-
-    id: int
-    file_path: str
-    content: str
-    source_uri: str
-
-
-@dataclass
-class FileInfo:
-    """Domain model for file information."""
-
-    uri: str
-
-
-@dataclass
-class SnippetInfo:
-    """Domain model for snippet information."""
-
-    id: int
-    content: str
-
-
-@dataclass
-class SnippetWithFile:
-    """Domain model for snippet with associated file information."""
+class SnippetWithContext:
+    """Domain model for snippet with associated context information."""
 
-    file: FileInfo
-    snippet: SnippetInfo
+    source: Source
+    file: File
+    authors: list[Author]
+    snippet: Snippet
 
 
 class LanguageMapping:

kodit/infrastructure/indexing/index_repository.py

@@ -6,9 +6,20 @@ from typing import TypeVar
 from sqlalchemy import delete, func, select
 from sqlalchemy.ext.asyncio import AsyncSession
 
-from kodit.domain.entities import Embedding, File, Index, Snippet, Source
+from kodit.domain.entities import (
+    Author,
+    AuthorFileMapping,
+    Embedding,
+    File,
+    Index,
+    Snippet,
+    Source,
+)
 from kodit.domain.services.indexing_service import IndexRepository
-from kodit.domain.value_objects import FileInfo, IndexView, SnippetInfo, SnippetWithFile
+from kodit.domain.value_objects import (
+    IndexView,
+    SnippetWithContext,
+)
 
 T = TypeVar("T")
 
@@ -202,6 +213,7 @@ class SQLAlchemyIndexRepository(IndexRepository):
             file_id=snippet["file_id"],
             index_id=snippet["index_id"],
             content=snippet["content"],
+            summary=snippet.get("summary", ""),
         )
         self.session.add(db_snippet)
 
@@ -221,30 +233,31 @@ class SQLAlchemyIndexRepository(IndexRepository):
             snippet.content = content
             # SQLAlchemy will automatically track this change
 
-    async def list_snippets_by_ids(self, ids: list[int]) -> list[SnippetWithFile]:
-        """List snippets by IDs.
-
-        Args:
-            ids: List of snippet IDs to retrieve.
-
-        Returns:
-            List of SnippetWithFile objects containing file and snippet information.
-
-        """
+    async def list_snippets_by_ids(self, ids: list[int]) -> list[SnippetWithContext]:
+        """List snippets by IDs."""
         query = (
-            select(Snippet, File)
+            select(Snippet, File, Source, Author)
             .where(Snippet.id.in_(ids))
             .join(File, Snippet.file_id == File.id)
+            .join(Source, File.source_id == Source.id)
+            .outerjoin(AuthorFileMapping, AuthorFileMapping.file_id == File.id)
+            .outerjoin(Author, AuthorFileMapping.author_id == Author.id)
         )
         rows = await self.session.execute(query)
 
-        # Create a dictionary for O(1) lookup of results by ID
-        id_to_result = {}
-        for snippet, file in rows.all():
-            id_to_result[snippet.id] = SnippetWithFile(
-                file=FileInfo(uri=file.uri),
-                snippet=SnippetInfo(id=snippet.id, content=snippet.content)
-            )
+        # Group results by snippet ID and collect authors
+        id_to_result: dict[int, SnippetWithContext] = {}
+        for snippet, file, source, author in rows.all():
+            if snippet.id not in id_to_result:
+                id_to_result[snippet.id] = SnippetWithContext(
+                    snippet=snippet,
+                    file=file,
+                    source=source,
+                    authors=[],
+                )
+            # Add author if it exists (outer join might return None)
+            if author is not None:
+                id_to_result[snippet.id].authors.append(author)
 
         # Check that all IDs are present
         if len(id_to_result) != len(ids):

kodit/infrastructure/sqlalchemy/snippet_repository.py

@@ -1,7 +1,9 @@
 """SQLAlchemy implementation of snippet repository."""
 
+import builtins
 from collections.abc import Sequence
 from pathlib import Path
+from typing import Any
 
 from sqlalchemy import delete, or_, select
 from sqlalchemy.ext.asyncio import AsyncSession
@@ -18,7 +20,7 @@ from kodit.domain.repositories import SnippetRepository
 from kodit.domain.value_objects import (
     LanguageMapping,
     MultiSearchRequest,
-    SnippetListItem,
+    SnippetWithContext,
 )
 
 
@@ -102,7 +104,7 @@ class SqlAlchemySnippetRepository(SnippetRepository):
 
     async def list_snippets(
         self, file_path: str | None = None, source_uri: str | None = None
-    ) -> Sequence[SnippetListItem]:
+    ) -> Sequence[SnippetWithContext]:
         """List snippets with optional filtering by file path and source URI.
 
         Args:
@@ -112,20 +114,11 @@ class SqlAlchemySnippetRepository(SnippetRepository):
             all sources.
 
         Returns:
-            A sequence of SnippetListItem instances matching the criteria
+            A sequence of SnippetWithContext instances matching the criteria
 
         """
-        # Build the base query
-        query = (
-            select(
-                Snippet,
-                File.cloned_path,
-                Source.cloned_path.label("source_cloned_path"),
-                Source.uri.label("source_uri"),
-            )
-            .join(File, Snippet.file_id == File.id)
-            .join(Source, File.source_id == Source.id)
-        )
+        # Build the base query with joins for all required entities
+        query = self._build_base_query()
 
         # Apply filters
         if file_path is not None:
@@ -140,20 +133,7 @@ class SqlAlchemySnippetRepository(SnippetRepository):
             query = query.where(Source.uri == source_uri)
 
         result = await self.session.execute(query)
-        return [
-            SnippetListItem(
-                id=snippet.id,
-                file_path=self._get_relative_path(file_cloned_path, source_cloned_path),
-                content=snippet.content,
-                source_uri=source_uri_val,
-            )
-            for (
-                snippet,
-                file_cloned_path,
-                source_cloned_path,
-                source_uri_val,
-            ) in result.all()
-        ]
+        return self._process_results(result)
 
     def _get_relative_path(self, file_path: str, source_path: str) -> str:
         """Calculate the relative path of a file from the source root.
@@ -174,57 +154,98 @@ class SqlAlchemySnippetRepository(SnippetRepository):
             # If the file is not relative to the source, return the filename
             return Path(file_path).name
 
-    async def search(self, request: MultiSearchRequest) -> Sequence[SnippetListItem]:
-        """Search snippets with filters.
+    def _apply_filters(self, query: Any, filters: Any) -> Any:
+        """Apply filters to the query.
 
         Args:
-            request: The search request containing queries and optional filters.
+            query: The base query to apply filters to
+            filters: The filters to apply
 
         Returns:
-            A sequence of SnippetListItem instances matching the search criteria.
+            The modified query with filters applied
 
         """
-        # Build the base query with joins
-        query = (
-            select(
-                Snippet,
-                File.cloned_path,
-                Source.cloned_path.label("source_cloned_path"),
-                Source.uri.label("source_uri"),
-            )
+        if not filters:
+            return query
+
+        # Language filter (using file extension)
+        if filters.language:
+            extensions = LanguageMapping.get_extensions_with_fallback(filters.language)
+            query = query.where(File.extension.in_(extensions))
+
+        # Author filter
+        if filters.author:
+            query = query.where(Author.name.ilike(f"%{filters.author}%"))
+
+        # Date filters
+        if filters.created_after:
+            query = query.where(Snippet.created_at >= filters.created_after)
+
+        if filters.created_before:
+            query = query.where(Snippet.created_at <= filters.created_before)
+
+        # Source repository filter
+        if filters.source_repo:
+            query = query.where(Source.uri.like(f"%{filters.source_repo}%"))
+
+        return query
+
+    def _build_base_query(self) -> Any:
+        """Build the base query with joins for all required entities.
+
+        Returns:
+            The base query with joins
+
+        """
+        return (
+            select(Snippet, File, Source, Author)
             .join(File, Snippet.file_id == File.id)
             .join(Source, File.source_id == Source.id)
+            .outerjoin(AuthorFileMapping, AuthorFileMapping.file_id == File.id)
+            .outerjoin(Author, AuthorFileMapping.author_id == Author.id)
         )
 
-        # Apply filters if provided
-        if request.filters:
-            filters = request.filters
+    def _process_results(self, result: Any) -> builtins.list[SnippetWithContext]:
+        """Process query results into SnippetWithContext objects.
 
-            # Language filter (using file extension)
-            if filters.language:
-                extensions = LanguageMapping.get_extensions_with_fallback(
-                    filters.language
-                )
-                query = query.where(File.extension.in_(extensions))
-
-            # Author filter
-            if filters.author:
-                query = (
-                    query.join(AuthorFileMapping, File.id == AuthorFileMapping.file_id)
-                    .join(Author, AuthorFileMapping.author_id == Author.id)
-                    .where(Author.name.ilike(f"%{filters.author}%"))
+        Args:
+            result: The query result
+
+        Returns:
+            List of SnippetWithContext objects
+
+        """
+        # Group results by snippet ID and collect authors
+        id_to_result: dict[int, SnippetWithContext] = {}
+        for snippet, file, source, author in result.all():
+            if snippet.id not in id_to_result:
+                id_to_result[snippet.id] = SnippetWithContext(
+                    snippet=snippet,
+                    file=file,
+                    source=source,
+                    authors=[],
                 )
+            # Add author if it exists (outer join might return None)
+            if author is not None:
+                id_to_result[snippet.id].authors.append(author)
+
+        return list(id_to_result.values())
 
-            # Date filters
-            if filters.created_after:
-                query = query.where(Snippet.created_at >= filters.created_after)
+    async def search(self, request: MultiSearchRequest) -> Sequence[SnippetWithContext]:
+        """Search snippets with filters.
 
-            if filters.created_before:
-                query = query.where(Snippet.created_at <= filters.created_before)
+        Args:
+            request: The search request containing queries and optional filters.
 
-            # Source repository filter
-            if filters.source_repo:
-                query = query.where(Source.uri.like(f"%{filters.source_repo}%"))
+        Returns:
+            A sequence of SnippetWithContext instances matching the search criteria.
+
+        """
+        # Build the base query with joins for all required entities
+        query = self._build_base_query()
+
+        # Apply filters if provided
+        query = self._apply_filters(query, request.filters)
 
         # Only apply top_k limit if there are no search queries
         # This ensures that when used for pre-filtering (with search queries),
@@ -235,17 +256,4 @@ class SqlAlchemySnippetRepository(SnippetRepository):
             query = query.limit(request.top_k)
 
         result = await self.session.execute(query)
-        return [
-            SnippetListItem(
-                id=snippet.id,
-                file_path=self._get_relative_path(file_cloned_path, source_cloned_path),
-                content=snippet.content,
-                source_uri=source_uri_val,
-            )
-            for (
-                snippet,
-                file_cloned_path,
-                source_cloned_path,
-                source_uri_val,
-            ) in result.all()
-        ]
+        return self._process_results(result)

kodit/log.py

@@ -190,11 +190,14 @@ def _from_sysfs() -> list[int]:
     macs: list[int] = []
     for iface in base.iterdir():
         try:
+            # Skip if iface is not a directory (e.g., bonding_masters is a file)
+            if not iface.is_dir():
+                continue
             with (base / iface / "address").open() as f:
                 content = f.read().strip()
             if _MAC_RE.fullmatch(content):
                 macs.append(_mac_int(content))
-        except (FileNotFoundError, PermissionError):
+        except (FileNotFoundError, PermissionError, NotADirectoryError):
             pass
     return macs
 

kodit/mcp.py

@@ -195,17 +195,12 @@ async def search(  # noqa: PLR0913
     snippets = await service.search(request=search_request)
 
     log.debug("Fusing output")
-    output = output_fusion(snippets=snippets)
+    output = MultiSearchResult.to_jsonlines(results=snippets)
 
     log.debug("Output", output=output)
     return output
 
 
-def output_fusion(snippets: list[MultiSearchResult]) -> str:
-    """Fuse the snippets into a single output."""
-    return "\n\n".join(str(snippet) for snippet in snippets)
-
-
 @mcp.tool()
 async def get_version() -> str:
     """Get the version of the kodit project."""

kodit/migrations/versions/4552eb3f23ce_add_summary.py

@@ -0,0 +1,34 @@
+# ruff: noqa
+"""add summary
+
+Revision ID: 4552eb3f23ce
+Revises: 9e53ea8bb3b0
+Create Date: 2025-06-30 16:32:49.293087
+
+"""
+
+from typing import Sequence, Union
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision: str = '4552eb3f23ce'
+down_revision: Union[str, None] = '9e53ea8bb3b0'
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    """Upgrade schema."""
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.add_column('snippets', sa.Column('summary', sa.UnicodeText(), nullable=False))
+    # ### end Alembic commands ###
+
+
+def downgrade() -> None:
+    """Downgrade schema."""
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.drop_column('snippets', 'summary')
+    # ### end Alembic commands ###

{kodit-0.3.1.dist-info → kodit-0.3.2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: kodit
-Version: 0.3.1
+Version: 0.3.2
 Summary: Code indexing for better AI code generation
 Project-URL: Homepage, https://docs.helixml.tech/kodit/
 Project-URL: Documentation, https://docs.helixml.tech/kodit/

{kodit-0.3.1.dist-info → kodit-0.3.2.dist-info}/RECORD

@@ -1,34 +1,34 @@
 kodit/.gitignore,sha256=ztkjgRwL9Uud1OEi36hGQeDGk3OLK1NfDEO8YqGYy8o,11
 kodit/__init__.py,sha256=aEKHYninUq1yh6jaNfvJBYg-6fenpN132nJt1UU6Jxs,59
-kodit/_version.py,sha256=lOWWIGJeBi0KkFopWU_n3GH71C1PsaZ-ZYDfxFkne6c,511
+kodit/_version.py,sha256=5NopxuphNnyGZECYEkvIUFi0KZxwtDHmTpW5R266eSo,511
 kodit/app.py,sha256=uv67TE83fZE7wrA7cz-sKosFrAXlKRr1B7fT-X_gMZQ,2103
-kodit/cli.py,sha256=fGk2VMJDrgaj0T9w-97Xh6LVqus6vVehWJUTkjgWtyk,16013
+kodit/cli.py,sha256=a-bJQ_Jyz201TEbgJPNvPDa0Qyt3kfSKqBuypeVqG_k,17219
 kodit/config.py,sha256=VUoUi2t2yGhqOtm5MSZuaasNSklH50hfWn6GOrz3jnU,7518
 kodit/database.py,sha256=kI9yBm4uunsgV4-QeVoCBL0wLzU4kYmYv5qZilGnbPE,1740
-kodit/log.py,sha256=sHPHYetlMcKTor2VaFLMyao1_fZ_xhuzqXCAt5F5UMU,8575
-kodit/mcp.py,sha256=4rvj76SUbO-FfoJgOY31CR5r0y0VxMAKkVKNuKLgwB0,6431
+kodit/log.py,sha256=WOsLRitpCBtJa5IcsyZpKr146kXXHK2nU5VA90gcJdQ,8736
+kodit/mcp.py,sha256=6gCJvjTqWGWUicuidbpMPtM1Vtqvlc0fKUua3l-EVPQ,6273
 kodit/middleware.py,sha256=I6FOkqG9-8RH5kR1-0ZoQWfE4qLCB8lZYv8H_OCH29o,2714
 kodit/reporting.py,sha256=icce1ZyiADsA_Qz-mSjgn2H4SSqKuGfLKnw-yrl9nsg,2722
 kodit/application/__init__.py,sha256=mH50wTpgP9dhbKztFsL8Dda9Hi18TSnMVxXtpp4aGOA,35
 kodit/application/factories/__init__.py,sha256=bU5CvEnaBePZ7JbkCOp1MGTNP752bnU2uEqmfy5FdRk,37
 kodit/application/factories/code_indexing_factory.py,sha256=pyGcTmqhBRjw0tDvp5UpG0roBf3ROqYvBcHyvaLZ-qQ,4927
 kodit/application/services/__init__.py,sha256=p5UQNw-H5sxQvs5Etfte93B3cJ1kKW6DNxK34uFvU1E,38
-kodit/application/services/code_indexing_application_service.py,sha256=HGaZDZG0GPLxxQ1-GUgeEx_AsZhG0FAtdUugoD68uI0,12521
+kodit/application/services/code_indexing_application_service.py,sha256=PXnBbDnaYqU6xnKGTcOjmYZbjcQZed-_ehf6Uzhx5v4,12809
 kodit/domain/__init__.py,sha256=TCpg4Xx-oF4mKV91lo4iXqMEfBT1OoRSYnbG-zVWolA,66
-kodit/domain/entities.py,sha256=fErA9ZTAqlofkqhBte8FOnV0PHf1MUORb37bW0-Dgc4,5624
+kodit/domain/entities.py,sha256=6UBPi7zH9bCIgeXg0Poq6LQu01O5JvoHaWqNusNJ3iA,5787
 kodit/domain/enums.py,sha256=Ik_h3D3eZ0FsSlPsU0ikm-Yv3Rmvzicffi9yBn19UIE,191
 kodit/domain/errors.py,sha256=yIsgCjM_yOFIg8l7l-t7jM8pgeAX4cfPq0owf7iz3DA,106
 kodit/domain/interfaces.py,sha256=Jkd0Ob4qSvhZHI9jRPFQ1n5Cv0SvU-y3Z-HCw2ikc4I,742
-kodit/domain/repositories.py,sha256=KAIx_-qZD68pAByc1JNVxSCRLjseayHKn5ykqsE6uWw,3781
-kodit/domain/value_objects.py,sha256=7OtjtgdVB0rhWwezJB_-A6cCqk_ijmkefMRv2ql5AQ0,13479
+kodit/domain/repositories.py,sha256=VgNV4NXywh6LtxN1GU2fg8bn8mNZ2wgVXZEugqOOb1M,3796
+kodit/domain/value_objects.py,sha256=h9KMAB0neX3gQT2mTC8JxyxxuDXuJ2lyG0czUhkZc0E,15575
 kodit/domain/services/__init__.py,sha256=Q1GhCK_PqKHYwYE4tkwDz5BIyXkJngLBBOHhzvX8nzo,42
 kodit/domain/services/bm25_service.py,sha256=nsfTan3XtDwXuuAu1LUv-6Jukm6qFKVqqCVymjyepZQ,3625
 kodit/domain/services/embedding_service.py,sha256=Wh6Y2NR_GRnud8dq1Q7S6F40aNe-S2UyD5Nqz9LChTM,4507
 kodit/domain/services/enrichment_service.py,sha256=XsXg3nV-KN4rqtC7Zro_ZiZ6RSq-1eA1MG6IDzFGyBA,1316
 kodit/domain/services/ignore_service.py,sha256=boEN-IRLmUtwO9ZnuACaVFZbIKrtUG8YwnsXKEDIG28,1136
-kodit/domain/services/indexing_service.py,sha256=vBjg9G75XoNfwH7m43l16zEmKdemHkzrgwunguiWix8,5911
+kodit/domain/services/indexing_service.py,sha256=7Yb6lyyd_VpZldK_CVMeOXpzXq-08Et-WRhulCWDQdM,5920
 kodit/domain/services/snippet_extraction_service.py,sha256=QW_99bXWpr8g6ZI-hp4Aj57VCSrUf71dLwQca5T6pyg,3065
-kodit/domain/services/snippet_service.py,sha256=OyiDPJx5O2I1EyYnH_W-rvggti0DUefp4JaKIblHhR8,6867
+kodit/domain/services/snippet_service.py,sha256=EyJQoT9UkJdMM2yfC1cFlj0yZVxK5a7NzleeM8lqWR0,7355
 kodit/domain/services/source_service.py,sha256=9XGS3imJn65v855cztsJSaaFod6LhkF2xfUVMaytx-A,3068
 kodit/infrastructure/__init__.py,sha256=HzEYIjoXnkz_i_MHO2e0sIVYweUcRnl2RpyBiTbMObU,28
 kodit/infrastructure/bm25/__init__.py,sha256=DmGbrEO34FOJy4e685BbyxLA7gPW1eqs2gAxsp6JOuM,34
@@ -64,7 +64,7 @@ kodit/infrastructure/ignore/ignore_pattern_provider.py,sha256=9m2XCsgW87UBTfzHr6
 kodit/infrastructure/indexing/__init__.py,sha256=7UPRa2jwCAsa0Orsp6PqXSF8iIXJVzXHMFmrKkI9yH8,38
 kodit/infrastructure/indexing/auto_indexing_service.py,sha256=uXggladN3PTU5Jzhz0Kq-0aObvq3Dq9YbjYKCSkaQA8,3131
 kodit/infrastructure/indexing/fusion_service.py,sha256=mXUUcx3-8e75mWkxXMfl30HIoFXrTNHzB1w90MmEbak,1806
-kodit/infrastructure/indexing/index_repository.py,sha256=qs1RiFuf29kbKX4unP98_4-f9vQCmpCo-q-zvCCFPCE,8287
+kodit/infrastructure/indexing/index_repository.py,sha256=4m_kFHQ3OSQdf2pgR1RM72g-k4UZHyHbtYKUwJ8huRs,8719
 kodit/infrastructure/indexing/indexing_factory.py,sha256=LPjPCps_wJ9M_fZGRP02bfc2pvYc50ZSTYI99XwRRPg,918
 kodit/infrastructure/indexing/snippet_domain_service_factory.py,sha256=OMp9qRJSAT3oWqsMyF1fgI2Mb_G-SA22crbbaCb7c-Q,1253
 kodit/infrastructure/snippet_extraction/__init__.py,sha256=v6KqrRDjSj0nt87m7UwRGx2GN_fz_14VWq9Q0uABR_s,54
@@ -82,7 +82,7 @@ kodit/infrastructure/sqlalchemy/__init__.py,sha256=UXPMSF_hgWaqr86cawRVqM8XdVNum
 kodit/infrastructure/sqlalchemy/embedding_repository.py,sha256=u29RVt4W0WqHj6TkrydMHw2iF5_jERHtlidDjWRQvqc,7886
 kodit/infrastructure/sqlalchemy/file_repository.py,sha256=9_kXHJ1YiWA1ingpvBNq8cuxkMu59PHwl_m9_Ttnq2o,2353
 kodit/infrastructure/sqlalchemy/repository.py,sha256=EpZnOjR3wfPEqIauWw_KczpkSqBQPTq5sIyCpJCuW2w,4565
-kodit/infrastructure/sqlalchemy/snippet_repository.py,sha256=4Hb_K7rPevzCgqZDIJPBTM3lzF36QXjmUES9WwW9E2k,8252
+kodit/infrastructure/sqlalchemy/snippet_repository.py,sha256=aBsr2U6RUQftWnkOHka809WH9YxS4Tpg34knZ--WNms,8473
 kodit/infrastructure/ui/__init__.py,sha256=CzbLOBwIZ6B6iAHEd1L8cIBydCj-n_kobxJAhz2I9_Y,32
 kodit/infrastructure/ui/progress.py,sha256=BaAeMEgXlSSb0c_t_NPxnThIktkzzCS9kegb5ExULJs,4791
 kodit/infrastructure/ui/spinner.py,sha256=GcP115qtR0VEnGfMEtsGoAUpRzVGUSfiUXfoJJERngA,2357
@@ -90,13 +90,14 @@ kodit/migrations/README,sha256=ISVtAOvqvKk_5ThM5ioJE-lMkvf9IbknFUFVU_vPma4,58
 kodit/migrations/__init__.py,sha256=lP5MuwlyWRMO6UcDWnQcQ3G-GYHcFb6rl9gYPHJ1sjo,40
 kodit/migrations/env.py,sha256=j89vEWdSgfnreTAz5ZvFAPlsMGI8SfKti0MlWhm7Jbc,2364
 kodit/migrations/script.py.mako,sha256=zWziKtiwYKEWuwPV_HBNHwa9LCT45_bi01-uSNFaOOE,703
+kodit/migrations/versions/4552eb3f23ce_add_summary.py,sha256=_saoHs5HGzc_z2OzBkFKrifTLQfoNox3BpSBeiKg_f8,870
 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/9e53ea8bb3b0_add_authors.py,sha256=a32Zm8KUQyiiLkjKNPYdaJDgjW6VsV-GhaLnPnK_fpI,3884
 kodit/migrations/versions/__init__.py,sha256=9-lHzptItTzq_fomdIRBegQNm4Znx6pVjwD4MiqRIdo,36
 kodit/migrations/versions/c3f5137d30f5_index_all_the_things.py,sha256=rI8LmjF-I2OMxZ2nOIF_NRmqOLXe45hL_iz_nx97DTQ,1680
-kodit-0.3.1.dist-info/METADATA,sha256=iI5UCxq-ih4cF-GVJLyNQ8wdRxwC907Jj6pddwiIsJo,6358
-kodit-0.3.1.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-kodit-0.3.1.dist-info/entry_points.txt,sha256=hoTn-1aKyTItjnY91fnO-rV5uaWQLQ-Vi7V5et2IbHY,40
-kodit-0.3.1.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-kodit-0.3.1.dist-info/RECORD,,
+kodit-0.3.2.dist-info/METADATA,sha256=JDWIO27pGDjCMUm5gRWUWjdQhRgEGC8J0O3gMFki6p8,6358
+kodit-0.3.2.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+kodit-0.3.2.dist-info/entry_points.txt,sha256=hoTn-1aKyTItjnY91fnO-rV5uaWQLQ-Vi7V5et2IbHY,40
+kodit-0.3.2.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+kodit-0.3.2.dist-info/RECORD,,