local-deep-research 0.3.6__tar.gz → 0.3.8__tar.gz

{local_deep_research-0.3.6 → local_deep_research-0.3.8}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: local-deep-research
-Version: 0.3.6
+Version: 0.3.8
 Summary: AI-powered research assistant with deep, iterative analysis using LLMs and web searches
 Author-Email: LearningCircuit <185559241+LearningCircuit@users.noreply.github.com>, HashedViking <6432677+HashedViking@users.noreply.github.com>
 License: MIT License
@@ -80,9 +80,14 @@ Description-Content-Type: text/markdown
 
 *AI-powered research assistant that performs deep, iterative analysis using multiple LLMs and web searches*
 
-<a href="https://www.youtube.com/watch?v=0ISreg9q0p0">
-  <img src="https://img.youtube.com/vi/0ISreg9q0p0/0.jpg" alt="Local Deep Research Demo" width="500">
-</a>
+<div align="center">
+  <a href="https://www.youtube.com/watch?v=0ISreg9q0p0">
+    <img src="https://img.youtube.com/vi/0ISreg9q0p0/0.jpg" alt="Local Deep Research">
+    <br>
+    <span>▶️ Watch Video</span>
+  </a>
+</div>
+
 
 </div>
 
@@ -167,6 +172,7 @@ print(results["summary"])
 
 **Windows**: Docker is the easiest option for Windows users. If preferred, a [Windows Installer](https://github.com/LearningCircuit/local-deep-research/releases/download/v0.1.0/LocalDeepResearch_Setup.exe) is also available.
 
+For more information on installation options, see [the wiki](https://github.com/LearningCircuit/local-deep-research/wiki/Installation).
 
 ## 🔍 Research Capabilities
 
@@ -273,9 +279,9 @@ For enhanced web search capabilities, you can configure these additional engines
 
 ```bash
 # Search API keys (if not using the web UI)
-SERP_API_KEY=your-key-here               # Google results via SerpAPI
-GOOGLE_PSE_API_KEY=your-key-here         # Google Programmable Search
-BRAVE_API_KEY=your-key-here              # Brave Search
+LDR_SEARCH_ENGINE_WEB_SERPAPI_API_KEY=your-key-here               # Google results via SerpAPI
+LDR_SEARCH_ENGINE_WEB_GOOGLE_PSE_API_KEY=your-key-here         # Google Programmable Search
+LDR_SEARCH_ENGINE_WEB_BRAVE_API_KEY=your-key-here              # Brave Search
 ```
 
 ### Search Engine Comparison

{local_deep_research-0.3.6 → local_deep_research-0.3.8}/README.md

@@ -9,9 +9,14 @@
 
 *AI-powered research assistant that performs deep, iterative analysis using multiple LLMs and web searches*
 
-<a href="https://www.youtube.com/watch?v=0ISreg9q0p0">
-  <img src="https://img.youtube.com/vi/0ISreg9q0p0/0.jpg" alt="Local Deep Research Demo" width="500">
-</a>
+<div align="center">
+  <a href="https://www.youtube.com/watch?v=0ISreg9q0p0">
+    <img src="https://img.youtube.com/vi/0ISreg9q0p0/0.jpg" alt="Local Deep Research">
+    <br>
+    <span>▶️ Watch Video</span>
+  </a>
+</div>
+
 
 </div>
 
@@ -96,6 +101,7 @@ print(results["summary"])
 
 **Windows**: Docker is the easiest option for Windows users. If preferred, a [Windows Installer](https://github.com/LearningCircuit/local-deep-research/releases/download/v0.1.0/LocalDeepResearch_Setup.exe) is also available.
 
+For more information on installation options, see [the wiki](https://github.com/LearningCircuit/local-deep-research/wiki/Installation).
 
 ## 🔍 Research Capabilities
 
@@ -202,9 +208,9 @@ For enhanced web search capabilities, you can configure these additional engines
 
 ```bash
 # Search API keys (if not using the web UI)
-SERP_API_KEY=your-key-here               # Google results via SerpAPI
-GOOGLE_PSE_API_KEY=your-key-here         # Google Programmable Search
-BRAVE_API_KEY=your-key-here              # Brave Search
+LDR_SEARCH_ENGINE_WEB_SERPAPI_API_KEY=your-key-here               # Google results via SerpAPI
+LDR_SEARCH_ENGINE_WEB_GOOGLE_PSE_API_KEY=your-key-here         # Google Programmable Search
+LDR_SEARCH_ENGINE_WEB_BRAVE_API_KEY=your-key-here              # Brave Search
 ```
 
 ### Search Engine Comparison

{local_deep_research-0.3.6 → local_deep_research-0.3.8}/pyproject.toml

@@ -57,7 +57,7 @@ dependencies = [
     "setuptools>=78.1.0",
     "flask-wtf>=1.2.2",
 ]
-version = "0.3.6"
+version = "0.3.8"
 
 [project.license]
 file = "LICENSE"

local_deep_research-0.3.8/src/local_deep_research/__version__.py

@@ -0,0 +1 @@
+__version__ = "0.3.8"

{local_deep_research-0.3.6 → local_deep_research-0.3.8}/src/local_deep_research/defaults/default_settings.json

@@ -2927,32 +2927,32 @@
         "value": "A locally-hosted meta-search engine.",
         "visible": false
     },
-    "search.engine.web.searxng.api_key": {
+    "search.engine.web.searxng.class_name": {
         "category": "searxng",
-        "description": "The SearXNG API key to use.",
+        "description": "Setting for searxng.class_name",
         "editable": true,
         "max_value": null,
         "min_value": null,
-        "name": "Api Key",
+        "name": "Class Name",
         "options": null,
         "step": null,
         "type": "SEARCH",
-        "ui_element": "password",
-        "value": "SEARXNG_API_KEY",
+        "ui_element": "text",
+        "value": "SearXNGSearchEngine",
         "visible": true
     },
-    "search.engine.web.searxng.class_name": {
+    "search.engine.web.searxng.default_params.instance_url": {
         "category": "searxng",
-        "description": "Setting for searxng.class_name",
+        "description": "The SearXNG API endpoint URL.",
         "editable": true,
         "max_value": null,
         "min_value": null,
-        "name": "Class Name",
+        "name": "Endpoint URL",
         "options": null,
         "step": null,
         "type": "SEARCH",
         "ui_element": "text",
-        "value": "SearXNGSearchEngine",
+        "value": "http://localhost:8080",
         "visible": true
     },
     "search.engine.web.searxng.default_params.categories": {

{local_deep_research-0.3.6 → local_deep_research-0.3.8}/src/local_deep_research/search_system.py

@@ -180,8 +180,11 @@ class AdvancedSearchSystem:
         self.questions_by_iteration = self.strategy.questions_by_iteration.copy()
         # Send progress message with search info
 
-        # if hasattr(self.strategy, "all_links_of_system"):
-        self.all_links_of_system.extend(self.strategy.all_links_of_system)
+        # Only extend if they're different objects in memory to avoid duplication
+        # This check prevents doubling the list when they reference the same object
+        # Fix for issue #301: "too many links in detailed report mode"
+        if id(self.all_links_of_system) != id(self.strategy.all_links_of_system):
+            self.all_links_of_system.extend(self.strategy.all_links_of_system)
 
         # Include the search system instance for access to citations
         result["search_system"] = self

{local_deep_research-0.3.6 → local_deep_research-0.3.8}/src/local_deep_research/utilities/db_utils.py

@@ -6,7 +6,7 @@ from typing import Any, Dict
 from sqlalchemy import create_engine
 from sqlalchemy.orm import Session, sessionmaker
 
-from ..web.services.settings_manager import SettingsManager, check_env_setting
+from ..web.services.settings_manager import SettingsManager
 
 logger = logging.getLogger(__name__)
 
@@ -38,7 +38,7 @@ def get_settings_manager() -> SettingsManager:
 
 
 def get_db_setting(
-    key: str, default_value: Any | None = None, check_env: bool = True
+    key: str, default_value: Any | None = None
 ) -> str | Dict[str, Any] | None:
     """
     Get a setting from the database with fallback to default value
@@ -46,15 +46,11 @@ def get_db_setting(
     Args:
         key: The setting key.
         default_value: If the setting is not found, it will return this instead.
-        check_env: If true, it will check the corresponding environment
-            variable before checking the DB and return that if it is set.
 
-    """
-    if check_env:
-        env_value = check_env_setting(key)
-        if env_value is not None:
-            return env_value
+    Returns:
+        The setting value.
 
+    """
     try:
         # Get settings manager which handles database access
         value = get_settings_manager().get_setting(key)

{local_deep_research-0.3.6 → local_deep_research-0.3.8}/src/local_deep_research/web/services/settings_manager.py

@@ -94,11 +94,18 @@ class SettingsManager:
                     return value
                 elif len(settings) > 1:
                     # This is a higher-level key.
-                    settings_map = {
-                        s.key.removeprefix(f"{key}."): s.value for s in settings
-                    }
-                    # We deliberately don't update the cache here to avoid
-                    # conflicts between low-level keys and their parent keys.
+                    settings_map = {}
+                    for setting in settings:
+                        output_key = setting.key.removeprefix(f"{key}.")
+                        value = setting.value
+
+                        if check_env:
+                            # Handle possible replacements from environment variables.
+                            env_value = check_env_setting(setting.key)
+                            if env_value is not None:
+                                value = env_value
+
+                        settings_map[output_key] = value
                     return settings_map
             except SQLAlchemyError as e:
                 logger.error(f"Error retrieving setting {key} from database: {e}")

{local_deep_research-0.3.6 → local_deep_research-0.3.8}/src/local_deep_research/web_search_engines/engines/search_engine_searxng.py

@@ -25,7 +25,7 @@ class SearXNGSearchEngine(BaseSearchEngine):
     def __init__(
         self,
         max_results: int = 15,
-        instance_url: Optional[str] = None,  # Can be None if using env var
+        instance_url: str = "http://localhost:8080",
         categories: Optional[List[str]] = None,
         engines: Optional[List[str]] = None,
         language: str = "en",
@@ -35,7 +35,6 @@ class SearXNGSearchEngine(BaseSearchEngine):
         llm: Optional[BaseLLM] = None,
         max_filtered_results: Optional[int] = None,
         include_full_content: bool = True,
-        api_key: Optional[str] = None,
     ):  # API key is actually the instance URL
         """
         Initialize the SearXNG search engine with ethical usage patterns.
@@ -52,7 +51,6 @@ class SearXNGSearchEngine(BaseSearchEngine):
             llm: Language model for relevance filtering
             max_filtered_results: Maximum number of results to keep after filtering
             include_full_content: Whether to include full webpage content in results
-            api_key: Alternative way to provide instance URL (takes precedence over instance_url)
         """
 
         # Initialize the BaseSearchEngine with LLM, max_filtered_results, and max_results
@@ -60,28 +58,25 @@ class SearXNGSearchEngine(BaseSearchEngine):
             llm=llm, max_filtered_results=max_filtered_results, max_results=max_results
         )
 
-        # Get instance URL from various sources in priority order:
-        # 1. api_key parameter (which is actually the instance URL)
-        # 2. SEARXNG_INSTANCE environment variable
-        # 3. instance_url parameter
-        # 4. Default to None, which will disable the engine
-        self.instance_url = api_key or os.getenv("SEARXNG_INSTANCE") or instance_url or "http://localhost:8080"
-
-        # Add debug logging for instance URL
-        logger.info(
-            f"SearXNG init - Instance URL sources: api_key={api_key}, env={os.getenv('SEARXNG_INSTANCE')}, param={instance_url}"
-        )
-
+        self.instance_url = instance_url
         # Validate and normalize the instance URL if provided
-        if self.instance_url:
-            self.instance_url = self.instance_url.rstrip("/")
-            self.is_available = True
-            logger.info(f"SearXNG initialized with instance URL: {self.instance_url}")
-        else:
+        self.instance_url = self.instance_url.rstrip("/")
+        logger.info(f"SearXNG initialized with instance URL: {self.instance_url}")
+        try:
+            # Make sure it's accessible.
+            response = requests.get(self.instance_url, timeout=5)
+            if response.status_code == 200:
+                logger.info("SearXNG instance is accessible.")
+                self.is_available = True
+            else:
+                self.is_available = False
+                logger.error(
+                    f"Failed to access SearXNG instance at {self.instance_url}. Status code: {response.status_code}"
+                )
+        except requests.RequestException as e:
             self.is_available = False
             logger.error(
-                "No SearXNG instance URL provided. The engine is disabled. "
-                "Set SEARXNG_INSTANCE environment variable or provide instance_url parameter."
+                f"Error while trying to access SearXNG instance at {self.instance_url}: {str(e)}"
             )
 
         # Add debug logging for all parameters

{local_deep_research-0.3.6 → local_deep_research-0.3.8}/src/local_deep_research/web_search_engines/search_engines_config.py

@@ -5,7 +5,6 @@ Loads search engine definitions from the user's configuration.
 
 import json
 import logging
-from functools import cache
 from typing import Any, Dict, List
 
 from ..utilities.db_utils import get_db_setting
@@ -25,16 +24,26 @@ def _extract_per_engine_config(raw_config: Dict[str, Any]) -> Dict[str, Dict[str
         Configuration dictionaries indexed by engine name.
 
     """
-    engine_config = {}
+    nested_config = {}
     for key, value in raw_config.items():
-        engine_name = key.split(".")[0]
-        setting_name = ".".join(key.split(".")[1:])
-        engine_config.setdefault(engine_name, {})[setting_name] = value
+        if "." in key:
+            # This is a higher-level key.
+            top_level_key = key.split(".")[0]
+            lower_keys = ".".join(key.split(".")[1:])
+            nested_config.setdefault(top_level_key, {})[lower_keys] = value
+        else:
+            # This is a low-level key.
+            nested_config[key] = value
 
-    return engine_config
+    # Expand all the lower-level keys.
+    for key, value in nested_config.items():
+        if isinstance(value, dict):
+            # Expand the child keys.
+            nested_config[key] = _extract_per_engine_config(value)
+
+    return nested_config
 
 
-@cache
 def search_config() -> Dict[str, Any]:
     """
     Returns:
@@ -105,7 +114,6 @@ def search_config() -> Dict[str, Any]:
     return search_engines
 
 
-@cache
 def default_search_engine() -> str:
     """
     Returns:
@@ -115,7 +123,6 @@ def default_search_engine() -> str:
     return get_db_setting("search.engine.DEFAULT_SEARCH_ENGINE", "wikipedia")
 
 
-@cache
 def local_search_engines() -> List[str]:
     """
     Returns:

local_deep_research-0.3.8/tests/fix_tests/README.md

@@ -0,0 +1,36 @@
+# Fix Tests
+
+This directory contains tests for specific bug fixes in the codebase.
+
+## Tests
+
+### test_duplicate_links_fix.py
+
+Demonstrates and tests the fix for GitHub issue [#301](https://github.com/LearningCircuit/local-deep-research/issues/301) - "too many links in detailed report mode".
+
+The test simulates the bug where links are duplicated in the `all_links_of_system` list when using detailed report mode. It compares the behavior before and after implementing the fix.
+
+#### Bug Description
+
+The bug occurred because `search_system.py` was unconditionally extending `self.all_links_of_system` with `self.strategy.all_links_of_system` in the `analyze_topic` method, even though they were the same list object (when initialized with the same reference).
+
+#### Fix
+
+The fix checks if the lists are the same object (have the same `id()`) before extending:
+
+```python
+# Only extend if they're different objects in memory to avoid duplication
+if id(self.all_links_of_system) != id(self.strategy.all_links_of_system):
+    self.all_links_of_system.extend(self.strategy.all_links_of_system)
+```
+
+This prevents duplicating the content when both lists are actually the same object.
+
+#### Running the Test
+
+```bash
+cd tests/fix_tests
+python test_duplicate_links_fix.py
+```
+
+The test demonstrates both the bug and the fixed behavior side-by-side.

local_deep_research-0.3.8/tests/fix_tests/test_duplicate_links_fix.py

@@ -0,0 +1,102 @@
+"""
+Test script to demonstrate the duplicate links issue and our fix for issue #301.
+
+This test demonstrates the problem where the search system duplicates links in detailed report mode
+by unconditionally extending the all_links_of_system list with itself.
+
+GitHub issue: https://github.com/LearningCircuit/local-deep-research/issues/301
+
+The fix is to check if the lists are the same object (have the same id()) before extending.
+"""
+
+class Strategy:
+    def __init__(self, all_links=None):
+        # Create a new list if None is provided
+        self.all_links_of_system = [] if all_links is None else all_links
+        print(f"Strategy initialized with list id: {id(self.all_links_of_system)}")
+        
+    def analyze_topic(self, query):
+        # Add some links 
+        self.all_links_of_system.extend([
+            {"title": "Link 1", "link": "http://example.com/1"},
+            {"title": "Link 2", "link": "http://example.com/2"},
+        ])
+        print(f"Strategy now has {len(self.all_links_of_system)} links")
+        return {"content": "Analysis results"}
+
+class AdvancedSearchSystem:
+    def __init__(self):
+        # Initialize with empty list
+        self.all_links_of_system = []
+        print(f"Search system initialized with list id: {id(self.all_links_of_system)}")
+        
+        # Create strategy with our list reference
+        self.strategy = Strategy(all_links=self.all_links_of_system)
+        
+        # Check if they're the same object
+        print(f"Are lists the same object? {id(self.all_links_of_system) == id(self.strategy.all_links_of_system)}")
+    
+    def analyze_topic_with_bug(self, query):
+        # Run the strategy
+        result = self.strategy.analyze_topic(query)
+        
+        # BUG: Unconditionally extend our list with the strategy's list
+        # This is problematic because they're the same list
+        print("\nBUG DEMO: Extending unconditionally")
+        before_count = len(self.all_links_of_system)
+        print(f"Before extending: {before_count} links")
+        
+        self.all_links_of_system.extend(self.strategy.all_links_of_system)
+        
+        after_count = len(self.all_links_of_system)
+        print(f"After extending: {after_count} links")
+        print(f"Added {after_count - before_count} links (duplicates)")
+        
+        # Return results
+        return {
+            "all_links_of_system": self.all_links_of_system,
+            "content": result["content"]
+        }
+    
+    def analyze_topic_with_fix(self, query):
+        # Run the strategy
+        result = self.strategy.analyze_topic(query)
+        
+        # FIX: Only extend if they're different objects
+        print("\nFIX DEMO: Only extending if lists are different")
+        before_count = len(self.all_links_of_system)
+        print(f"Before fix check: {before_count} links")
+        
+        if id(self.all_links_of_system) != id(self.strategy.all_links_of_system):
+            print("Lists are different objects - extending")
+            self.all_links_of_system.extend(self.strategy.all_links_of_system)
+        else:
+            print("Lists are the same object - not extending (avoiding duplicates)")
+        
+        after_count = len(self.all_links_of_system)
+        print(f"After fix check: {after_count} links")
+        print(f"Added {after_count - before_count} links")
+        
+        # Return results
+        return {
+            "all_links_of_system": self.all_links_of_system,
+            "content": result["content"]
+        }
+
+def test_bug_and_fix():
+    print("=== Testing Bug and Fix ===\n")
+    
+    print("1. Creating first search system to demonstrate the bug:")
+    search_system1 = AdvancedSearchSystem()
+    result1 = search_system1.analyze_topic_with_bug("What is quantum computing?")
+    print(f"\nFinal link count with bug: {len(result1['all_links_of_system'])}")
+    
+    print("\n2. Creating second search system to demonstrate the fix:")
+    search_system2 = AdvancedSearchSystem()
+    result2 = search_system2.analyze_topic_with_fix("What is quantum computing?")
+    print(f"\nFinal link count with fix: {len(result2['all_links_of_system'])}")
+    
+    print("\nTest complete!")
+
+if __name__ == "__main__":
+    test_bug_and_fix()

local_deep_research-0.3.6/src/local_deep_research/__version__.py

@@ -1 +0,0 @@
-__version__ = "0.3.6"