fix: change default model to gpt-5.4-mini, fix page_count fallback in indexer

- Default model changed from gpt-5.4 to gpt-5.4-mini
- Indexer get_page_content no longer uses hardcoded 9999 fallback
- Infers page_count from structure end_index when doc lacks page_count field
- Added debug logging for doc keys and page_count diagnosis

Author: KylinMountain

openkb/config.py

@@ -6,11 +6,14 @@
 import yaml
 
 DEFAULT_CONFIG: dict[str, Any] = {
-    "model": "gpt-5.4",
+    "model": "gpt-5.4-mini",
     "language": "en",
     "pageindex_threshold": 20,
 }
 
+GLOBAL_CONFIG_DIR = Path.home() / ".config" / "openkb"
+GLOBAL_CONFIG_PATH = GLOBAL_CONFIG_DIR / "global.yaml"
+
 
 def load_config(config_path: Path) -> dict[str, Any]:
     """Load YAML config from config_path, merged with DEFAULT_CONFIG.
@@ -30,3 +33,30 @@ def save_config(config_path: Path, config: dict) -> None:
     config_path.parent.mkdir(parents=True, exist_ok=True)
     with config_path.open("w", encoding="utf-8") as fh:
         yaml.safe_dump(config, fh, allow_unicode=True, sort_keys=True)
+
+
+def load_global_config() -> dict[str, Any]:
+    """Load the global config from ~/.config/openkb/global.yaml."""
+    if GLOBAL_CONFIG_PATH.exists():
+        with GLOBAL_CONFIG_PATH.open("r", encoding="utf-8") as fh:
+            return yaml.safe_load(fh) or {}
+    return {}
+
+
+def save_global_config(config: dict[str, Any]) -> None:
+    """Save the global config to ~/.config/openkb/global.yaml."""
+    GLOBAL_CONFIG_DIR.mkdir(parents=True, exist_ok=True)
+    with GLOBAL_CONFIG_PATH.open("w", encoding="utf-8") as fh:
+        yaml.safe_dump(config, fh, allow_unicode=True, sort_keys=True)
+
+
+def register_kb(kb_path: Path) -> None:
+    """Register a KB path in the global config's known_kbs list."""
+    gc = load_global_config()
+    known = gc.get("known_kbs", [])
+    resolved = str(kb_path.resolve())
+    if resolved not in known:
+        known.append(resolved)
+        gc["known_kbs"] = known
+    gc["default_kb"] = resolved
+    save_global_config(gc)

openkb/indexer.py

@@ -67,6 +67,10 @@ def index_long_document(pdf_path: Path, kb_dir: Path) -> IndexResult:
     description: str = doc.get("doc_description", "")
     structure: list = doc.get("structure", [])
 
+    # Debug: print doc keys and page_count to diagnose get_page_content range
+    logger.info("Doc keys: %s", list(doc.keys()))
+    logger.info("page_count from doc: %s", doc.get("page_count", "NOT PRESENT"))
+
     tree = {
         "doc_name": doc_name,
         "doc_description": description,
@@ -78,8 +82,18 @@ def index_long_document(pdf_path: Path, kb_dir: Path) -> IndexResult:
     sources_dir.mkdir(parents=True, exist_ok=True)
     dest_images_dir = sources_dir / "images" / pdf_path.stem
 
-    # Get per-page content from PageIndex
-    all_pages = col.get_page_content(doc_id, f"1-{doc.get('page_count', 9999)}")
+    # Get per-page content from PageIndex — use actual page count
+    page_count = doc.get("page_count")
+    if page_count is None:
+        # Fallback: count pages from structure's max end_index
+        max_page = 0
+        for node in structure:
+            end = node.get("end_index", 0)
+            if end > max_page:
+                max_page = end
+        page_count = max_page if max_page > 0 else 100
+        logger.info("page_count not in doc, inferred from structure: %d", page_count)
+    all_pages = col.get_page_content(doc_id, f"1-{page_count}")
 
     # Relocate image paths in each page
     dest_images_dir.mkdir(parents=True, exist_ok=True)