fix: resolve linting, type errors, and optional dependencies

Author: SpectreDeath

stylometry/ai_interface.py

@@ -1,17 +1,14 @@
 import json
-import urllib.request
 import urllib.error
-from typing import Dict, List, Optional
+import urllib.request
+from typing import Dict, List, Optional, Any
+
 
-def analyze_stats_with_ai(
-    summary: Dict, 
-    api_base: str = "http://localhost:1234/v1", 
-    model: str = "local-model"
-) -> str:
+def analyze_stats_with_ai(summary: Dict[str, Any], api_base: str = "http://localhost:1234/v1", model: str = "local-model") -> str:
     """
     Sends stylometric summary to a local LLM (like LM Studio) for interpretation.
     """
-    
+
     # Construct the prompt
     stats_json = json.dumps(summary, indent=2)
     prompt = f"""
@@ -35,19 +32,19 @@ def analyze_stats_with_ai(
         "model": model,
         "messages": [
             {"role": "system", "content": "You are a professional linguistic analyst specializing in stylometry."},
-            {"role": "user", "content": prompt}
+            {"role": "user", "content": prompt},
         ],
-        "temperature": 0.7
+        "temperature": 0.7,
     }
-    
+
     data = json.dumps(payload).encode("utf-8")
     req = urllib.request.Request(f"{api_base.rstrip('/')}/chat/completions", data=data)
     req.add_header("Content-Type", "application/json")
-    
+
     try:
         with urllib.request.urlopen(req, timeout=120) as response:
             res_data = json.loads(response.read().decode("utf-8"))
-            return res_data["choices"][0]["message"]["content"]
+            return str(res_data["choices"][0]["message"]["content"])
     except urllib.error.URLError as e:
         return f"AI Analysis failed: Could not connect to LM Studio at {api_base}. Ensure the server is running. (Error: {e})"
     except Exception as e:

stylometry/boilerplate.py

@@ -1,34 +1,37 @@
 import collections
 from typing import List, Set
+
 from .models import DocRecord
 
+
 def find_boilerplate(docs: List[DocRecord], threshold: float = 0.5, min_len: int = 50) -> Set[int]:
     """
     Identifies paragraphs that appear in more than 'threshold' fraction of documents.
     Returns a set of hashes of boilerplate paragraphs.
     """
     if len(docs) < 3:
         return set()
-        
-    counts = collections.Counter()
+
+    counts: collections.Counter[str] = collections.Counter()
     for d in docs:
         paragraphs = {p.strip() for p in d.text.split("\n") if len(p.strip()) >= min_len}
         for p in paragraphs:
             counts[p] += 1
-            
+
     boilerplate_hashes = set()
     n_docs = len(docs)
     for p, count in counts.items():
         if count / n_docs >= threshold:
             boilerplate_hashes.add(hash(p))
-            
+
     return boilerplate_hashes
 
+
 def strip_boilerplate(docs: List[DocRecord], boilerplate_hashes: Set[int]):
     """Removes identified boilerplate paragraphs from documents in-place."""
     if not boilerplate_hashes:
         return
-        
+
     for d in docs:
         lines = d.text.split("\n")
         new_lines = []
@@ -39,7 +42,7 @@ def strip_boilerplate(docs: List[DocRecord], boilerplate_hashes: Set[int]):
             if hash(line.strip()) in boilerplate_hashes:
                 continue
             new_lines.append(line)
-        
+
         # Re-join and re-tokenize if changed
         new_text = "\n".join(new_lines).strip()
         if new_text != d.text.strip():