deepresearch-flow 0.6.1__py3-none-any.whl → 0.7.1__py3-none-any.whl

deepresearch_flow/__init__.py

@@ -2,4 +2,4 @@
 
 __all__ = ["__version__"]
 
-__version__ = "0.1.0"
+__version__ = "0.7.1"

deepresearch_flow/paper/db.py

@@ -424,6 +424,87 @@ def register_db_commands(db_group: click.Group) -> None:
         click.echo(f"Wrote snapshot DB: {opts.output_db}")
         click.echo(f"Wrote static export: {opts.static_export_dir}")
 
+    @snapshot_group.group("unpack")
+    def snapshot_unpack_group() -> None:
+        """Unpack snapshot artifacts."""
+
+    @snapshot_unpack_group.command("md")
+    @click.option("--snapshot-db", "snapshot_db", required=True, help="Path to snapshot database")
+    @click.option(
+        "--static-export-dir",
+        "static_export_dir",
+        required=True,
+        help="Path to static export directory",
+    )
+    @click.option(
+        "--pdf-root",
+        "pdf_roots",
+        multiple=True,
+        required=True,
+        help="PDF root directories for name alignment (repeatable)",
+    )
+    @click.option("--md-output-dir", "md_output_dir", required=True, help="Output directory for Markdown")
+    @click.option(
+        "--md-translated-output-dir",
+        "md_translated_output_dir",
+        required=True,
+        help="Output directory for translated Markdown",
+    )
+    def snapshot_unpack_md(
+        snapshot_db: str,
+        static_export_dir: str,
+        pdf_roots: tuple[str, ...],
+        md_output_dir: str,
+        md_translated_output_dir: str,
+    ) -> None:
+        """Unpack source/translated markdown and align filenames to PDFs."""
+        from deepresearch_flow.paper.snapshot.unpacker import SnapshotUnpackMdOptions, unpack_md
+
+        opts = SnapshotUnpackMdOptions(
+            snapshot_db=Path(snapshot_db),
+            static_export_dir=Path(static_export_dir),
+            pdf_roots=[Path(path) for path in pdf_roots],
+            md_output_dir=Path(md_output_dir),
+            md_translated_output_dir=Path(md_translated_output_dir),
+        )
+        unpack_md(opts)
+
+    @snapshot_unpack_group.command("info")
+    @click.option("--snapshot-db", "snapshot_db", required=True, help="Path to snapshot database")
+    @click.option(
+        "--static-export-dir",
+        "static_export_dir",
+        required=True,
+        help="Path to static export directory",
+    )
+    @click.option(
+        "--pdf-root",
+        "pdf_roots",
+        multiple=True,
+        required=True,
+        help="PDF root directories for name alignment (repeatable)",
+    )
+    @click.option("--template", "template", required=True, help="Summary template tag")
+    @click.option("--output-json", "output_json", required=True, help="Output JSON file path")
+    def snapshot_unpack_info(
+        snapshot_db: str,
+        static_export_dir: str,
+        pdf_roots: tuple[str, ...],
+        template: str,
+        output_json: str,
+    ) -> None:
+        """Unpack aggregated paper_infos.json from snapshot summaries."""
+        from deepresearch_flow.paper.snapshot.unpacker import SnapshotUnpackInfoOptions, unpack_info
+
+        opts = SnapshotUnpackInfoOptions(
+            snapshot_db=Path(snapshot_db),
+            static_export_dir=Path(static_export_dir),
+            pdf_roots=[Path(path) for path in pdf_roots],
+            template=template,
+            output_json=Path(output_json),
+        )
+        unpack_info(opts)
+
     @db_group.group("api")
     def api_group() -> None:
         """Read-only JSON API server backed by a snapshot DB."""

deepresearch_flow/paper/snapshot/api.py

@@ -11,8 +11,9 @@ from starlette.applications import Starlette
 from starlette.middleware.cors import CORSMiddleware
 from starlette.requests import Request
 from starlette.responses import JSONResponse, Response
-from starlette.routing import Route
+from starlette.routing import Mount, Route
 
+from deepresearch_flow.paper.snapshot.common import ApiLimits, _open_ro_conn
 from deepresearch_flow.paper.snapshot.text import merge_adjacent_markers, remove_cjk_spaces, rewrite_search_query
 
 _WHITESPACE_RE = re.compile(r"\s+")
@@ -87,13 +88,6 @@ _FACET_TYPE_TO_KEY = {
 }
 
 
-@dataclass(frozen=True)
-class ApiLimits:
-    max_query_length: int = 500
-    max_page_size: int = 100
-    max_pagination_offset: int = 10_000  # page * page_size
-
-
 @dataclass(frozen=True)
 class SnapshotApiConfig:
     snapshot_db: Path
@@ -110,12 +104,6 @@ def _json_error(status_code: int, *, error: str, detail: str) -> JSONResponse:
     return JSONResponse({"error": error, "detail": detail}, status_code=status_code)
 
 
-def _open_ro_conn(db_path: Path) -> sqlite3.Connection:
-    uri = f"file:{db_path.as_posix()}?mode=ro"
-    conn = sqlite3.connect(uri, uri=True)
-    conn.row_factory = sqlite3.Row
-    conn.execute("PRAGMA query_only=ON;")
-    return conn
 
 
 def _snapshot_build_id(conn: sqlite3.Connection) -> str:
@@ -917,6 +905,22 @@ def create_app(
         limits=limits or ApiLimits(),
     )
 
+    # Lazy import to avoid circular dependency
+    from deepresearch_flow.paper.snapshot.mcp_server import (
+        McpSnapshotConfig,
+        create_mcp_app,
+        resolve_static_export_dir,
+    )
+
+    mcp_config = McpSnapshotConfig(
+        snapshot_db=snapshot_db,
+        static_base_url=_normalize_base_url(static_base_url),
+        static_export_dir=resolve_static_export_dir(),
+        limits=limits or ApiLimits(),
+        origin_allowlist=cors_allowed_origins or ["*"],
+    )
+    mcp_app, mcp_lifespan = create_mcp_app(mcp_config)
+
     routes = [
         Route("/api/v1/config", _api_config, methods=["GET"]),
         Route("/api/v1/search", _api_search, methods=["GET"]),
@@ -927,9 +931,15 @@ def create_app(
         Route("/api/v1/facets/{facet:str}/{facet_id:str}/stats", _api_facet_stats, methods=["GET"]),
         Route("/api/v1/facets/{facet:str}/by-value/{value:str}/papers", _api_facet_by_value_papers, methods=["GET"]),
         Route("/api/v1/facets/{facet:str}/by-value/{value:str}/stats", _api_facet_by_value_stats, methods=["GET"]),
+        Mount("/mcp", app=mcp_app),
     ]
 
-    app = Starlette(routes=routes)
+    # Pass MCP lifespan to ensure session manager initializes properly
+    # https://gofastmcp.com/deployment/http#mounting-in-starlette
+    app = Starlette(
+        routes=routes,
+        lifespan=mcp_lifespan,
+    )
     if cfg.cors_allowed_origins:
         app.add_middleware(
             CORSMiddleware,

deepresearch_flow/paper/snapshot/common.py

@@ -0,0 +1,34 @@
+"""Shared utilities for snapshot API and MCP server.
+
+This module contains common types, configuration, and utilities used by both
+the snapshot REST API and the MCP server to avoid circular imports.
+"""
+
+from __future__ import annotations
+
+from dataclasses import dataclass
+from pathlib import Path
+import sqlite3
+
+
+@dataclass(frozen=True)
+class ApiLimits:
+    """API rate and size limits."""
+
+    max_query_length: int = 500
+    max_page_size: int = 100
+    max_pagination_offset: int = 10_000
+
+
+def _open_ro_conn(db_path: Path) -> sqlite3.Connection:
+    """Open a read-only SQLite connection with Row factory.
+
+    Args:
+        db_path: Path to the SQLite database file.
+
+    Returns:
+        sqlite3.Connection: A read-only connection with row_factory set to Row.
+    """
+    conn = sqlite3.connect(f"file:{db_path}?mode=ro", uri=True, check_same_thread=False)
+    conn.row_factory = sqlite3.Row
+    return conn