Merge branch 'master' into ci-speed-optimizations

Reviewed-on: #28

.gitea/workflows/deploy.yml

@@ -15,12 +15,12 @@ jobs:
       - name: Install dependencies
         shell: bash
         run: |
-          apt update && apt upgrade -y
+          apt update -qq
-          apt install rustc cargo python-is-python3 pip python3-venv python3-virtualenv libmagic-dev git -y
+          apt install -y -qq python-is-python3 pip python3-venv libmagic-dev git
-          virtualenv .venv
+          python -m venv .venv
           source .venv/bin/activate
-          pip install --upgrade pip
+          pip install --upgrade pip -q
-          pip install -r requirements.txt
+          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

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)
+        # Retry mechanism with 60-minute timeout (bumped for large context models)
-        timeout_seconds = 30 * 60
+        timeout_seconds = 60 * 60
         max_retries = 3
 
         for attempt in range(max_retries):

src/ai_generators/tools/ollama_web_search_tool.py

@@ -102,7 +102,7 @@ class OllamaWebSearchTool(BaseTool):
                     "Authorization": f"Bearer {api_key}",
                     "Content-Type": "application/json",
                 },
-                timeout=30.0,
+                timeout=60.0,
             )
 
             # Raise for HTTP errors so we can catch them with specific messages

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
+            # New branch, create from master
-            self.repo.git.checkout("-b", title, "origin/main")
+            self.repo.git.checkout("-b", title, "origin/master")
 
         # Ensure destination directory exists
         dest_dir = f"{self.repo_path}src/content/"