Calibre MCP Server

import os from calibre_mcp_server.core.service import LibraryResearchService def main() -> None: # Nimm hier GENAU denselben Pfad wie in CALIBRE_LIBRARY_PATH library_path = r"X:\E-Books" if not os.path.exists(os.path.join(library_path, "metadata.db")): print("metadata.db not found at:", library_path) return service = LibraryResearchService(calibre_root_path=library_path) # 1) Volltextsuchen-Test # Nimm etwas sehr Häufiges oder lass den Query leer für einen Rauchtest query = "der" # oder "" für 'alles' hits = service.fulltext_search(query=query, limit=5) print("Fulltext hits for query '%s':" % query) if not hits: print(" (no hits found)") else: for hit in hits: print(" book_id =", hit.book_id) print(" title =", hit.title) print(" isbn =", hit.isbn) print(" snippet =", hit.snippet[:120].replace("\n", " ")) print() # 2) Excerpt-Test: NIMM HIER EINE ECHTE ISBN AUS CALIBRE test_isbn = "9783446429338" # gleich unten ersetzen excerpt = service.get_excerpt_by_isbn(isbn=test_isbn, max_chars=400) print() print("Excerpt for ISBN %s:" % test_isbn) if excerpt is None: print(" (no book found or no comments/title to build excerpt)") else: print(" book_id =", excerpt.book_id) print(" title =", excerpt.title) print(" isbn =", excerpt.isbn) print(" source_hint=", excerpt.source_hint) print(" text =", excerpt.text) if __name__ == "__main__": main()