feat: allow path params for verse navigation

scripts/sb_extract.py

@@ -555,8 +555,38 @@ def extract_verses(soup: BeautifulSoup, max_canto=99, max_chapter=999, max_verse
         if state == "SUMMARY":
             continue
 
-        # --- TEXT N header (p21) ---
+        # --- TEXT N header (p21) — authoritative verse number ---
+        # The p20 reference sometimes has wrong verse numbers (duplicates).
+        # The p21 TEXT header always has the correct verse number.
         if cls == TEXT_HEADER_CLASS and current_verse:
+            m = re.match(r"TEXTS?\s+(\d+)(?:\s*[-–]\s*(\d+))?", text, re.IGNORECASE)
+            if m:
+                correct_verse = int(m.group(1))
+                correct_end = int(m.group(2)) if m.group(2) else None
+
+                # Fix the verse number if p20 was wrong
+                if current_verse["verse"] != correct_verse:
+                    current_verse["verse"] = correct_verse
+
+                # Fix verse_end
+                if correct_end is not None:
+                    current_verse["verse_end"] = correct_end
+                elif (
+                    "verse_end" in current_verse
+                    and current_verse.get("verse_end") != correct_end
+                ):
+                    if correct_end is None and "verse_end" in current_verse:
+                        del current_verse["verse_end"]
+
+                # Rebuild the clean reference with corrected numbers
+                canto = current_verse["canto"]
+                chapter = current_verse["chapter"]
+                if correct_end is not None:
+                    current_verse["reference"] = (
+                        f"SB {canto}.{chapter}.{correct_verse}-{correct_end}"
+                    )
+                else:
+                    current_verse["reference"] = f"SB {canto}.{chapter}.{correct_verse}"
             continue
 
         # --- TEXT label (p22) ---
@@ -691,7 +721,7 @@ def write_json(data, filepath):
 def main():
     base_dir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
     html_path = os.path.join(base_dir, "public", "html", "sb.html")
-    output_dir = os.path.join(base_dir, "app", "lib", "books", "sb")
+    output_dir = os.path.join(base_dir, "lib", "jsons", "books", "sb")
 
     print(f"Reading {html_path}...")
     with open(html_path, "r", encoding="utf-8") as f: