chunksilo 2.1.0__tar.gz → 2.1.2__tar.gz

{chunksilo-2.1.0/src/chunksilo.egg-info → chunksilo-2.1.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: chunksilo
-Version: 2.1.0
+Version: 2.1.2
 Summary: Local RAG-based semantic document search with MCP server interface
 Author: Fredrik Reveny
 License-Expression: Apache-2.0
@@ -26,7 +26,6 @@ Requires-Dist: pillow<11,>=10.3.0
 Requires-Dist: pypdf<7,>=5.1.0
 Requires-Dist: python-docx<2,>=1.1.0
 Requires-Dist: mcp<2,>=1.0.0
-Requires-Dist: python-dotenv<2,>=1.0.0
 Requires-Dist: huggingface-hub<2,>=0.22.0
 Requires-Dist: flashrank<1,>=0.1.0
 Requires-Dist: fastembed<1,>=0.5.0
@@ -41,7 +40,7 @@ Requires-Dist: requests<3,>=2.31.0; extra == "test"
 Dynamic: license-file
 
 <p align="center">
-  <img src="chunksilo.png" alt="ChunkSilo Logo" width="500">
+  <img src="https://raw.githubusercontent.com/Chetic/chunksilo/main/chunksilo.png" alt="ChunkSilo Logo" width="500">
 </p>
 
 # ChunkSilo MCP Server

{chunksilo-2.1.0 → chunksilo-2.1.2}/README.md

@@ -1,5 +1,5 @@
 <p align="center">
-  <img src="chunksilo.png" alt="ChunkSilo Logo" width="500">
+  <img src="https://raw.githubusercontent.com/Chetic/chunksilo/main/chunksilo.png" alt="ChunkSilo Logo" width="500">
 </p>
 
 # ChunkSilo MCP Server

{chunksilo-2.1.0 → chunksilo-2.1.2}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "chunksilo"
-version = "2.1.0"
+version = "2.1.2"
 description = "Local RAG-based semantic document search with MCP server interface"
 license = "Apache-2.0"
 requires-python = ">=3.11"

{chunksilo-2.1.0 → chunksilo-2.1.2}/requirements.txt

@@ -6,7 +6,6 @@ pillow>=10.3.0,<11
 pypdf>=5.1.0,<7
 python-docx>=1.1.0,<2
 mcp>=1.0.0,<2
-python-dotenv>=1.0.0,<2
 huggingface-hub>=0.22.0,<2
 flashrank>=0.1.0,<1
 fastembed>=0.5.0,<1

{chunksilo-2.1.0 → chunksilo-2.1.2}/src/chunksilo/__init__.py

@@ -1,4 +1,4 @@
 # SPDX-License-Identifier: Apache-2.0
 """ChunkSilo - Local RAG-based semantic document search."""
 
-__version__ = "2.1.0"
+__version__ = "2.1.2"

{chunksilo-2.1.0 → chunksilo-2.1.2}/src/chunksilo/search.py

@@ -738,8 +738,8 @@ def _search_jira(query: str, config: dict[str, Any]) -> list[NodeWithScore]:
 
     Configuration Requirements:
         config["jira"]["url"]: Jira base URL (empty = disabled)
-        config["jira"]["username"]: Jira username or email
-        config["jira"]["api_token"]: Jira API token (not password)
+        config["jira"]["username"]: Jira username/email (required for Cloud, optional for Server PAT)
+        config["jira"]["api_token"]: API token (Cloud) or Personal Access Token (Server/Data Center)
         config["jira"]["max_results"]: Maximum issues to return
         config["jira"]["projects"]: List of project keys (empty = all)
         config["jira"]["include_comments"]: Include issue comments
@@ -763,8 +763,12 @@ def _search_jira(query: str, config: dict[str, Any]) -> list[NodeWithScore]:
         - Automatically configured through jira_options["verify"]
 
     Authentication:
-        - Uses basic auth (username + API token)
-        - Works for both Jira Cloud and Data Center/Server
+        - Jira Cloud: Set both username (email) and api_token (API token)
+          Uses basic auth internally
+        - Jira Server/Data Center with PAT: Set only api_token (Personal Access Token)
+          Leave username empty; uses bearer token auth internally
+        - Jira Server/Data Center with password: Set username and api_token (password)
+          Uses basic auth internally (if basic auth enabled on server)
 
     References:
         - Jira REST API: https://developer.atlassian.com/cloud/jira/platform/rest/v3/
@@ -790,13 +794,10 @@ def _search_jira(query: str, config: dict[str, Any]) -> list[NodeWithScore]:
     ca_bundle_path = config["ssl"]["ca_bundle_path"] or None
 
     # Validate required credentials are present
-    if not (base_url and username and api_token):
-        missing = []
-        if not username:
-            missing.append("jira.username")
-        if not api_token:
-            missing.append("jira.api_token")
-        logger.warning(f"Jira search skipped: missing {', '.join(missing)} in config")
+    # For Jira Cloud: both username and api_token required (basic auth)
+    # For Jira Server/Data Center with PAT: only api_token required (token auth)
+    if not api_token:
+        logger.warning("Jira search skipped: missing jira.api_token in config")
         return []
 
     try:
@@ -806,12 +807,23 @@ def _search_jira(query: str, config: dict[str, Any]) -> list[NodeWithScore]:
         if ca_bundle_path:
             jira_options["verify"] = ca_bundle_path
 
-        # Use basic auth (username + API token) for authentication
-        # Works for both Jira Cloud and Data Center/Server
-        jira_client = JIRA(
-            options=jira_options,
-            basic_auth=(username, api_token)
-        )
+        # Choose authentication method based on credentials provided:
+        # - Username + API token: Use basic auth (Jira Cloud, or Server with password)
+        # - API token only: Use token auth (Jira Server/Data Center with PAT)
+        if username:
+            # Basic auth for Jira Cloud (username + API token)
+            # Also works for Jira Server with username + password
+            jira_client = JIRA(
+                options=jira_options,
+                basic_auth=(username, api_token)
+            )
+        else:
+            # Token auth for Jira Server/Data Center Personal Access Tokens (PAT)
+            # PATs are used alone without a username
+            jira_client = JIRA(
+                options=jira_options,
+                token_auth=api_token
+            )
 
         # Construct JQL with text search and project filtering
         jql = _prepare_jira_jql_query(query, config)
@@ -1116,6 +1128,7 @@ def run_search(
                 rerank_request = RerankRequest(query=enhanced_query, passages=passages)
                 reranked_results = reranker.rerank(rerank_request)
 
+                # Build text-to-node mapping for fallback text matching
                 text_to_indices: dict[str, list[tuple[int, NodeWithScore]]] = {}
                 for idx, node in enumerate(nodes):
                     node_text = node.node.get_content() or ""
@@ -1126,20 +1139,34 @@ def run_search(
                 reranked_nodes = []
                 seen_indices: set[int] = set()
                 for result in reranked_results:
-                    doc_text = result.get("text", "")
                     score = result.get("score", 0.0)
-
-                    if doc_text in text_to_indices:
-                        for idx, node in text_to_indices[doc_text]:
-                            if idx not in seen_indices:
-                                reranked_nodes.append(node)
-                                rerank_scores[id(node)] = float(score)
-                                seen_indices.add(idx)
-                                break
-
+                    result_idx = result.get("id")
+
+                    # Primary: match by index (flashrank returns original passage index)
+                    if result_idx is not None and 0 <= result_idx < len(nodes):
+                        if result_idx not in seen_indices:
+                            node = nodes[result_idx]
+                            reranked_nodes.append(node)
+                            rerank_scores[id(node)] = float(score)
+                            seen_indices.add(result_idx)
+                    else:
+                        # Fallback: match by text content
+                        doc_text = result.get("text", "")
+                        if doc_text in text_to_indices:
+                            for idx, node in text_to_indices[doc_text]:
+                                if idx not in seen_indices:
+                                    reranked_nodes.append(node)
+                                    rerank_scores[id(node)] = float(score)
+                                    seen_indices.add(idx)
+                                    break
+
+                # Add remaining unmatched nodes with minimum matched score
+                # This ensures Jira/Confluence results aren't dropped due to text mismatch
+                min_score = min(rerank_scores.values()) if rerank_scores else 0.0
                 for idx, node in enumerate(nodes):
                     if idx not in seen_indices:
                         reranked_nodes.append(node)
+                        rerank_scores[id(node)] = min_score
 
                 nodes = reranked_nodes[:rerank_limit]
             except Exception as e:

{chunksilo-2.1.0 → chunksilo-2.1.2/src/chunksilo.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: chunksilo
-Version: 2.1.0
+Version: 2.1.2
 Summary: Local RAG-based semantic document search with MCP server interface
 Author: Fredrik Reveny
 License-Expression: Apache-2.0
@@ -26,7 +26,6 @@ Requires-Dist: pillow<11,>=10.3.0
 Requires-Dist: pypdf<7,>=5.1.0
 Requires-Dist: python-docx<2,>=1.1.0
 Requires-Dist: mcp<2,>=1.0.0
-Requires-Dist: python-dotenv<2,>=1.0.0
 Requires-Dist: huggingface-hub<2,>=0.22.0
 Requires-Dist: flashrank<1,>=0.1.0
 Requires-Dist: fastembed<1,>=0.5.0
@@ -41,7 +40,7 @@ Requires-Dist: requests<3,>=2.31.0; extra == "test"
 Dynamic: license-file
 
 <p align="center">
-  <img src="chunksilo.png" alt="ChunkSilo Logo" width="500">
+  <img src="https://raw.githubusercontent.com/Chetic/chunksilo/main/chunksilo.png" alt="ChunkSilo Logo" width="500">
 </p>
 
 # ChunkSilo MCP Server

{chunksilo-2.1.0 → chunksilo-2.1.2}/src/chunksilo.egg-info/requires.txt

@@ -6,7 +6,6 @@ pillow<11,>=10.3.0
 pypdf<7,>=5.1.0
 python-docx<2,>=1.1.0
 mcp<2,>=1.0.0
-python-dotenv<2,>=1.0.0
 huggingface-hub<2,>=0.22.0
 flashrank<1,>=0.1.0
 fastembed<1,>=0.5.0

{chunksilo-2.1.0 → chunksilo-2.1.2}/test/test_rag_metrics.py

@@ -21,9 +21,6 @@ from typing import Any, Dict, List, Optional, Set, Tuple
 from urllib.parse import urlparse
 
 import requests
-from dotenv import load_dotenv
-
-load_dotenv()
 
 # Set up logging
 logging.basicConfig(

{chunksilo-2.1.0 → chunksilo-2.1.2}/test/test_retrieval_only.py

@@ -2,7 +2,6 @@
 """Test the RAG system in retrieval-only mode (no LLM in the MCP server)."""
 import traceback
 from pathlib import Path
-from dotenv import load_dotenv
 import pytest
 
 from chunksilo.index import load_index_config, build_index
@@ -11,8 +10,6 @@ from chunksilo.cfgload import load_config
 
 STORAGE_DIR = Path(load_config()["storage"]["storage_dir"])
 
-load_dotenv()
-
 
 def test_ingestion():
     """Test the ingestion pipeline."""