Update utils.py

utils.py

@@ -335,8 +335,8 @@ def document_loading_splitting():
 
     # Dateien im Hugging Face Space auflisten
     files_in_repo = list_repo_files(repo_id=STORAGE_REPO_ID, repo_type="space", token=hf_token)
-    pdf_files = [f for f in files_in_repo if f.endswith('.pdf') and f.startswith("chroma/kkg/pdf/")]
-    word_files = [f for f in files_in_repo if f.endswith('.docx') and f.startswith("chroma/kkg/word/")]
+    pdf_files = [f for f in files_in_repo if f.endswith('.pdf') and f.startswith("chroma/demo/pdf/")]
+    word_files = [f for f in files_in_repo if f.endswith('.docx') and f.startswith("chroma/demo/word/")]
 
     
     # Erstellen von DirectoryLoader für jeden Dateityp
@@ -403,7 +403,7 @@ def document_storage_chroma(splits):
 ########################################################
 #Splits für den Vektorstore speichern - bzw. laden
 ########################################################
-def save_splits(preprocessed_splits, original_splits, directory="chroma/kkg", preprocessed_filename="preprocessed_splits.pkl", original_filename="original_splits.pkl"):
+def save_splits(preprocessed_splits, original_splits, directory="chroma/demo", preprocessed_filename="preprocessed_splits.pkl", original_filename="original_splits.pkl"):
     # Erstellen des Verzeichnisses, falls es nicht existiert
     if not os.path.exists(directory):
         os.makedirs(directory)
@@ -422,7 +422,7 @@ def save_splits(preprocessed_splits, original_splits, directory="chroma/kkg", pr
     upload_file_to_huggingface(preprocessed_filepath, f"{directory}/{preprocessed_filename}")
     upload_file_to_huggingface(original_filepath, f"{directory}/{original_filename}")
 
-def load_splits(directory="chroma/kkg", preprocessed_filename="preprocessed_splits.pkl", original_filename="original_splits.pkl"):
+def load_splits(directory="chroma/demo", preprocessed_filename="preprocessed_splits.pkl", original_filename="original_splits.pkl"):
     preprocessed_splits = None
     original_splits = None
 
@@ -457,7 +457,7 @@ def load_splits(directory="chroma/kkg", preprocessed_filename="preprocessed_spli
 ########################################
 #das Mapping der orginal-Splits und der preprocessed Splits speichern - und laden
 ########################################
-def save_split_to_original_mapping(mapping, directory="chroma/kkg", filename="mapping.pkl"):
+def save_split_to_original_mapping(mapping, directory="chroma/demo", filename="mapping.pkl"):
     # Erstellen des Verzeichnisses, falls es nicht existiert
     if not os.path.exists(directory):
         os.makedirs(directory)
@@ -471,7 +471,7 @@ def save_split_to_original_mapping(mapping, directory="chroma/kkg", filename="ma
     upload_file_to_huggingface(filepath, f"{directory}/{filename}")
 
 
-def load_split_to_original_mapping(directory="chroma/kkg", filename="mapping.pkl"):
+def load_split_to_original_mapping(directory="chroma/demo", filename="mapping.pkl"):
     try:
         # Laden des Mappings aus dem Hugging Face Repository
         file_path = hf_hub_download(
@@ -739,9 +739,9 @@ def download_link(doc):
 
     # Bestimmen des Dokumenttyps und Anpassen des Pfads
     if doc_path.lower().endswith('.pdf'):
-        file_url = f"{base_url}/chroma/kkg/pdf/{quote(title)}?token={hf_token}"
+        file_url = f"{base_url}/chroma/demo/pdf/{quote(title)}?token={hf_token}"
     elif doc_path.lower().endswith('.docx'):
-        file_url = f"{base_url}/chroma/kkg/word/{quote(title)}?token={hf_token}"
+        file_url = f"{base_url}/chroma/demo/word/{quote(title)}?token={hf_token}"
     else:
         # Fallback für andere Dateitypen
         file_url = f"{base_url}/{quote(doc_path)}?token={hf_token}"
@@ -760,7 +760,7 @@ def display_files():
     
     # PDF-Dateien
     files_table += "<tr style='background-color: #930BBA; color: white; font-weight: bold; font-size: larger;'><th>Dateiname - PDF-Ordner</th></tr>"
-    pdf_files = [f for f in list_repo_files(repo_id=STORAGE_REPO_ID, repo_type="space", token=hf_token) if f.endswith('.pdf') and f.startswith("chroma/kkg/pdf/")]
+    pdf_files = [f for f in list_repo_files(repo_id=STORAGE_REPO_ID, repo_type="space", token=hf_token) if f.endswith('.pdf') and f.startswith("chroma/demo/pdf/")]
     for i, file in enumerate(pdf_files):
         row_color = "#4f4f4f" if i % 2 == 0 else "#3a3a3a"
         files_table += f"<tr style='background-color: {row_color}; border-bottom: 1px solid #ddd;'>"
@@ -768,7 +768,7 @@ def display_files():
     
     # Word-Dateien
     files_table += "<tr style='background-color: #930BBA; color: white; font-weight: bold; font-size: larger;'><th>Dateiname - Word-Ordner</th></tr>"
-    word_files = [f for f in list_repo_files(repo_id=STORAGE_REPO_ID, repo_type="space", token=hf_token) if f.endswith('.docx') and f.startswith("chroma/kkg/word/")]
+    word_files = [f for f in list_repo_files(repo_id=STORAGE_REPO_ID, repo_type="space", token=hf_token) if f.endswith('.docx') and f.startswith("chroma/demo/word/")]
     for i, file in enumerate(word_files):
         row_color = "#4f4f4f" if i % 2 == 0 else "#3a3a3a"
         files_table += f"<tr style='background-color: {row_color}; border-bottom: 1px solid #ddd;'>"