added storing of usage data to hf dataset with commitscheduler

app.py

@@ -1,10 +1,30 @@
 from gradio_client import Client
 import gradio as gr
-import os
-import random
+import os, uuid, json, random, time
 import datetime
-from huggingface_hub import hf_api
-import time
+from huggingface_hub import hf_api, CommitScheduler, HfApi
+from pathlib import Path
+
+feedback_file = Path("output_data/") / f"output_{uuid.uuid4()}.json"
+feedback_folder = feedback_file.parent
+
+scheduler = CommitScheduler(
+    repo_id="eth-zurich-cle/deckify-dataset",
+    repo_type="dataset",
+    folder_path=feedback_folder,
+    path_in_repo="output_data",
+    every=10,
+)
+
+# scheduler = CommitScheduler(
+#     repo_id="eth-zurich-cle/deckify-dataset",
+#     repo_type="dataset",
+#     folder_path=feedback_folder,
+#     path_in_repo="input_data",
+#     every=10,
+# )
+
+api = HfApi()
 
 def check_password(username, password):
     if password == os.environ["ACCESS"]:
@@ -12,10 +32,26 @@ def check_password(username, password):
     else:
         return False
 
-
 def func(file, number_of_pages, secret):
+    
     if secret != os.environ["ACCESS"]:
         return "Wrong password, please try again"
+
+    date_string = datetime.datetime.now().strftime("%Y-%m-%d-%H-%M-%S")
+    
+    print(file)
+    unique_filename = f"{file.split('.')[0]}_{date_string}.{file.split('.')[-1]}"
+    
+    print(unique_filename)
+    
+    api.upload_file(
+        # path_or_fileobj="/path/to/local/folder/README.md",
+        path_or_fileobj=file,
+        path_in_repo=f"input_files/{unique_filename}",
+        repo_id="eth-zurich-cle/deckify-dataset",
+        repo_type="dataset",
+    )
+    
     
     space_runtime = hf_api.get_space_runtime("ByMatthew/deckify_private", token=read_key)
     print(f"Space runtime: {space_runtime}")
@@ -44,6 +80,7 @@ def func(file, number_of_pages, secret):
     
     print(f"Client: {client}")
 
+
     output = client.predict(file, number_of_pages)
     if "Error" in output:
         return output   
@@ -52,14 +89,26 @@ def func(file, number_of_pages, secret):
     # with open(f"{s}.tex", "w", encoding="utf-8") as f:
     #     f.write(text)
 
-    date_string = datetime.datetime.now().strftime("%Y-%m-%d-%H-%M-%S")
-    temp_string = "% The following slide is generated with [[SCIDECK]](https://huggingface.co/spaces/Nauryzbay/Scideck)"
+    temp_string = "% The following slides are generated with [[SCIDECK]](https://huggingface.co/spaces/ByMatthew/Scideck)"
     temp_string += "\n% Generated on " + date_string
     temp_string += "\n%" + "-"*100 + "\n"
     output = temp_string + output 
+    
+    save_output(unique_filename, output, number_of_pages, date_string)
+    
     return output
 
+def save_output(unique_filename: str, output: str, num_pages:int, date_string: str) -> None:
+
+    # Append outputs and using a thread lock to avoid concurrent writes from different users.
+    with scheduler.lock:
+        with feedback_file.open("a") as f:
+            f.write(json.dumps({"input_name": unique_filename, "output": output, "num_pages": num_pages, "timestamp": date_string}))
+            f.write("\n")
+
+
 def upload_file(file):
+    print(file)
     return file.name
 
 # 📝 If you get an error message, you can send me email with the PDF file attached to this email address: <b>nkoisheke [at] ethz [dot] ch</b>, and I will generate the slides for you. If there are any other issues or questions, please do not hesitate to contact me 🤗 <br>