fix: fix the ext of the generated summary file

file-summarizer/file_reader.py

@@ -3,6 +3,7 @@
 from summarize import DocumentSummarizer, MODEL, TIKTOKEN_MODEL, MAX_CHUNK_TOKENS, MAX_WORKERS
 import os
 import tiktoken
+import asyncio
 
 logger = setup_logger(__name__)
 
@@ -28,12 +29,15 @@ async def main():
         try:
             final_summary = summarizer.summarize(file_content)
         except Exception as e:
+            logger.error(f"Summarization failed: {e}")
             raise Exception(f"ERROR: Summarization failed: {e}")
 
-        response_str = f"Uploaded file {input_file} contains {len(tokens)} tokens.\n\n"
-        response_str += f"Summary of the file content:\n\n{final_summary}"
+        response_str = f"The uploaded file {input_file} contains too many tokens ({len(tokens)}), here is the summary of the file content:\n\n{final_summary}"
         print(response_str)
         return response_str
     else: # if the file has less than TOKEN_THRESHOLD tokens, directly return the file content
         print(file_content)
-        return file_content
+        return file_content
+
+if __name__ == "__main__":
+    asyncio.run(main())

file-summarizer/load_text.py

@@ -3,8 +3,9 @@
 import fitz  # PyMuPDF
 import docx
 from pptx import Presentation
-from helper import load_from_gptscript_workspace, save_to_gptscript_workspace
+from helper import load_from_gptscript_workspace, setup_logger
 
+logger = setup_logger(__name__)
 
 def extract_text_from_pdf(pdf_bytes: bytes) -> str:
     """Extracts text from a PDF file given as bytes."""
@@ -53,6 +54,7 @@ async def load_text_from_file(file_path: str) -> str:
     try:
         file_content = await load_from_gptscript_workspace(file_path)
     except Exception as e:
+        logger.error(f"Failed to load file from GPTScript workspace file {file_path}, Error: {e}")
         raise ValueError(
             f"Failed to load file from GPTScript workspace file {file_path}, Error: {e}"
         )

file-summarizer/summarize.py

@@ -295,7 +295,7 @@ async def main():
         if output_file == "":
             directory, file_name = os.path.split(input_file)
             name, ext = os.path.splitext(file_name)
-            summary_file_name = f"{name}_summary{ext}"
+            summary_file_name = f"{name}_summary.md"
             output_file = os.path.join(directory, summary_file_name)
 
         try:

file-summarizer/tool.gpt

@@ -1,20 +1,11 @@
----
 Name: File Summarizer
 Description: This tool summarizes the input file in the workspace, returns a text summary of the file content, either write to a file in the workspace or print to the console.
 Credential: sys.model.provider.credential
-Params: input_file: (Required) Name of the file in the workspace to summarize. Supported formats: [.md", ".txt", ".markdown", ".text", ".mdx", ".mdtxt", ".mdtxtx", ".docx", ".pdf", ".pptx"]. For any other file types, simply say it's not supported yet.
+Params: input_file: (Required) Name of the file in the workspace to summarize. Supported formats: [.md, .txt, .markdown, .text, .mdx, .mdtxt, .mdtxtx, .docx, .pdf, .pptx]. For any other file types, simply say it's not supported yet.
 Params: output_file: (Optional) Name of the file to save the summary, default to empty string. If not provided, a summary file will be created in the same directory as the input file. To print to the console, set this to "NONE".
 
 #!/usr/bin/env python3 ${GPTSCRIPT_TOOL_DIR}/summarize.py
 
----
-Name: File Reader
-Description: This tool reads the input file in the workspace, returns the file content and print to the console.
-Credential: sys.model.provider.credential
-Params: input_file: (Required) Name of the file in the workspace to summarize. Supported formats: [.md", ".txt", ".markdown", ".text", ".mdx", ".mdtxt", ".mdtxtx", ".docx", ".pdf", ".pptx"]. For any other file types, simply say it's not supported yet.
-
-#!/usr/bin/env python3 ${GPTSCRIPT_TOOL_DIR}/file_reader.py
-
 ---
 !metadata:*:category
 Utilities