Merge pull request 'speed up CI dependency installation' (#29) from ci-speed-optimizations into master

Reviewed-on: #29

.gitea/workflows/deploy.yml

@@ -15,12 +15,12 @@ jobs:
       - name: Install dependencies
         shell: bash
         run: |
-          apt update && apt upgrade -y
-          apt install rustc cargo python-is-python3 pip python3-venv python3-virtualenv libmagic-dev git -y
-          virtualenv .venv
+          apt update -qq
+          apt install -y -qq python-is-python3 pip python3-venv libmagic-dev git
+          python -m venv .venv
           source .venv/bin/activate
-          pip install --upgrade pip
-          pip install -r requirements.txt
+          pip install --upgrade pip -q
+          pip install -r requirements.txt --prefer-binary -q
           git config --global user.name "Blog Creator"
           git config --global user.email "ridgway.infrastructure@gmail.com"
           git config --global push.autoSetupRemote true

requirements.txt

@@ -1,4 +1,5 @@
 ollama
+httpx
 trilium-py
 gitpython
 PyGithub

src/ai_generators/blog_flow.py

@@ -17,14 +17,19 @@ journalist draft is chunked, embedded, and stored in a collection; the editor
 receives the top-N most relevant chunks as context.
 """
 
+from __future__ import annotations
+
 import json
 import os
 import random
 import re
 import string
 from datetime import datetime
+from typing import TYPE_CHECKING
+
+if TYPE_CHECKING:
+    import chromadb  # noqa: F811
 
-import chromadb
 from crewai.flow.flow import Flow, listen, start
 from ollama import Client
 from pydantic import BaseModel, ConfigDict
@@ -85,7 +90,21 @@ class BlogFlow(Flow[BlogFlowState]):
         )
 
     @staticmethod
-    def _get_chroma_client() -> chromadb.HttpClient:
+    def _get_chroma_client() -> "chromadb.HttpClient":
+        # Lazily import chromadb here rather than at module level.
+        # chromadb unconditionally loads hnswlib (a native C++ library
+        # compiled with AVX instructions) even when using HttpClient
+        # to talk to an external ChromaDB server.  Deferring the import
+        # avoids "Illegal instruction" (SIGILL) crashes in environments
+        # that lack AVX support (e.g. act, older CI runners).
+        try:
+            import chromadb
+        except ImportError as exc:
+            raise RuntimeError(
+                "chromadb is required for the editor phase but could not be "
+                f"imported: {exc}"
+            ) from exc
+
         chroma_port = int(os.environ["CHROMA_PORT"])
         return chromadb.HttpClient(host=os.environ["CHROMA_HOST"], port=chroma_port)
 
@@ -127,7 +146,7 @@ class BlogFlow(Flow[BlogFlowState]):
             print(f"Error generating embeddings: {exc}")
             return []
 
-    def _load_drafts_to_vector_db(self, drafts: list[str]) -> chromadb.Collection:
+    def _load_drafts_to_vector_db(self, drafts: list[str]) -> "chromadb.Collection":
         """Load journalist drafts into a new ChromaDB collection and return it."""
         chroma = self._get_chroma_client()
         collection_name = (
@@ -165,7 +184,7 @@ class BlogFlow(Flow[BlogFlowState]):
         return collection
 
     @staticmethod
-    def _query_vector_db(collection: chromadb.Collection, query_text: str) -> str:
+    def _query_vector_db(collection: "chromadb.Collection", query_text: str) -> str:
         """Query the ChromaDB collection and return the most relevant
         document chunks joined as a single string."""
         ollama_client = BlogFlow._get_ollama_client()

src/ai_generators/ollama_md_generator.py

@@ -140,8 +140,8 @@ class OllamaGenerator:
             )
             return response["message"]["content"]
 
-        # Retry mechanism with 30-minute timeout (same as the original)
-        timeout_seconds = 30 * 60
+        # Retry mechanism with 60-minute timeout (bumped for large context models)
+        timeout_seconds = 60 * 60
         max_retries = 3
 
         for attempt in range(max_retries):

src/ai_generators/tools/ollama_web_search_tool.py

@@ -4,17 +4,25 @@ Custom CrewAI tool that wraps Ollama's native web search API.
 This tool allows CrewAI agents to perform web searches using an Ollama
 subscription instead of third-party services like Serper or EXA.
 
+Uses direct HTTP requests via httpx with explicit Authorization: Bearer
+header to ensure the OLLAMA_API_KEY is properly passed to the Ollama cloud
+API endpoint (https://ollama.com/api/web_search).
+
 Requires:
-- Ollama Python library: pip install ollama
+- httpx library: pip install httpx (already a transitive dependency of crewai)
 - OLLAMA_API_KEY environment variable set with your Ollama API key
+
+Reference: https://docs.ollama.com/capabilities/web-search
 """
 
 import os
 
-import ollama
+import httpx
 from crewai.tools import BaseTool
 from pydantic import BaseModel, Field
 
+OLLAMA_WEB_SEARCH_URL = "https://ollama.com/api/web_search"
+
 
 class OllamaWebSearchInput(BaseModel):
     """Input schema for OllamaWebSearchTool."""
@@ -42,6 +50,10 @@ class OllamaWebSearchTool(BaseTool):
     The tool requires an Ollama subscription and the OLLAMA_API_KEY environment
     variable to be set.
 
+    Authentication is handled by sending the OLLAMA_API_KEY as a Bearer token
+    in the Authorization header, as documented at:
+    https://docs.ollama.com/capabilities/web-search
+
     Example usage:
         from ai_generators.tools.ollama_web_search_tool import OllamaWebSearchTool
 
@@ -65,6 +77,9 @@ class OllamaWebSearchTool(BaseTool):
         """
         Execute a web search and return formatted results.
 
+        Makes a POST request to https://ollama.com/api/web_search with the
+        OLLAMA_API_KEY as a Bearer token in the Authorization header.
+
         Args:
             query: The search query string
             max_results: Maximum number of results to return (1-10)
@@ -72,16 +87,40 @@ class OllamaWebSearchTool(BaseTool):
         Returns:
             Formatted string with search results, each containing title, URL, and content
         """
+        api_key = os.environ.get("OLLAMA_API_KEY")
+        if not api_key:
+            return (
+                "Error: OLLAMA_API_KEY environment variable is not set. "
+                "Please set your Ollama API key."
+            )
+
         try:
-            # Ensure API key is set
-            if not os.environ.get("OLLAMA_API_KEY"):
-                return "Error: OLLAMA_API_KEY environment variable is not set. Please set your Ollama API key."
+            response = httpx.post(
+                OLLAMA_WEB_SEARCH_URL,
+                json={"query": query, "max_results": max_results},
+                headers={
+                    "Authorization": f"Bearer {api_key}",
+                    "Content-Type": "application/json",
+                },
+                timeout=60.0,
+            )
 
-            # Perform the web search
-            response = ollama.web_search(query=query, max_results=max_results)
+            # Raise for HTTP errors so we can catch them with specific messages
+            if response.status_code == 401:
+                return (
+                    "Authentication error: OLLAMA_API_KEY was rejected. "
+                    "Your key may be invalid or expired. Please verify it at "
+                    "https://ollama.com/settings/keys"
+                )
+            if response.status_code == 429:
+                return (
+                    "Rate limit exceeded: Too many search requests. "
+                    "Please wait a moment and try again."
+                )
+            response.raise_for_status()
 
-            # Extract and format results
-            results = response.get("results", [])
+            data = response.json()
+            results = data.get("results", [])
 
             if not results:
                 return f"No search results found for query: '{query}'"
@@ -98,27 +137,20 @@ class OllamaWebSearchTool(BaseTool):
 
             return "\n".join(formatted_results)
 
-        except Exception as exc:
-            return f"Error performing web search: {exc}"
-
-    def _handle_exception(self, exc: Exception) -> str:
-        """Handle exceptions gracefully and return a user-friendly error message."""
-        error_message = str(exc)
-
-        # Check for common error types
-        if "authentication" in error_message.lower() or "401" in error_message:
+        except httpx.TimeoutException:
             return (
-                "Authentication error: Your OLLAMA_API_KEY may be invalid or expired. "
-                "Please check your API key and ensure it's set correctly in the environment."
+                "Timeout error: The web search request timed out. "
+                "Please try again with a simpler query."
             )
-        elif "rate limit" in error_message.lower() or "429" in error_message:
-            return "Rate limit exceeded: Too many search requests. Please wait a moment and try again."
-        elif (
-            "network" in error_message.lower() or "connection" in error_message.lower()
-        ):
+        except httpx.ConnectError:
             return (
                 "Network error: Unable to connect to Ollama's web search service. "
                 "Please check your internet connection and try again."
             )
-        else:
-            return f"Search failed: {error_message}"
+        except httpx.HTTPStatusError as exc:
+            return (
+                f"HTTP error {exc.response.status_code} from Ollama web search API: "
+                f"{exc.response.text}"
+            )
+        except Exception as exc:
+            return f"Error performing web search: {exc}"

src/repo_management/repo_manager.py

@@ -42,7 +42,7 @@ class GitRepository:
             print(f"Cloning failed: {e}")
             return False
 
-    def fetch(self, remote_name="origin", ref_name="main"):
+    def fetch(self, remote_name="origin", ref_name="master"):
         """Fetch updates from a remote repository with authentication"""
         try:
             self.repo.remotes[remote_name].fetch(ref_name=ref_name)
@@ -51,7 +51,7 @@ class GitRepository:
             print(f"Fetching failed: {e}")
             return False
 
-    def pull(self, remote_name="origin", ref_name="main"):
+    def pull(self, remote_name="origin", ref_name="master"):
         """Pull updates from a remote repository with authentication"""
         print("Pulling Latest Updates (if any)")
         try:
@@ -93,8 +93,8 @@ class GitRepository:
             self.repo.git.checkout(title)
             self.pull(ref_name=title)
         else:
-            # New branch, create from main
-            self.repo.git.checkout("-b", title, "origin/main")
+            # New branch, create from master
+            self.repo.git.checkout("-b", title, "origin/master")
 
         # Ensure destination directory exists
         dest_dir = f"{self.repo_path}src/content/"