first update

app.py

@@ -1,4 +1,64 @@
 import gradio as gr
+from datasets import load_dataset, Dataset
+from collections import defaultdict
+import random
+
+# Load the source dataset
+source_dataset = load_dataset("vietdata/eng_echo", split="train")
+source_texts = source_dataset["query"]
+
+# Initialize variables
+translations = defaultdict(list)
+processed_data = []
+
+# Helper function to get the next text for translation
+def get_next_text(user_id):
+    # Filter texts that already have 10 translations
+    eligible_texts = [text for text in source_texts if len(translations[text]) < 10]
+    if not eligible_texts:
+        return "All texts are fully translated."
+
+    # Select a random eligible text for translation
+    next_text = random.choice(eligible_texts)
+    return next_text
+
+# Function to handle translation submission
+def submit_translation(user_id, original_text, translation):
+    # Check if text already has 10 translations
+    if len(translations[original_text]) < 10:
+        translations[original_text].append((user_id, translation))
+
+        # Check if 100 texts have enough translations to save
+        if len([t for t in translations if len(translations[t]) == 10]) >= 100:
+            save_to_translated_echo()
+
+        return "Translation submitted successfully."
+    else:
+        return "This text already has 10 translations. Please request a new text."
+
+# Function to save completed translations to 'translated_echo'
+def save_to_translated_echo():
+    global translations, processed_data
+
+    # Gather translations with exactly 10 versions
+    completed_translations = [
+        {"query": text, "translations": [t[1] for t in translations[text]]}
+        for text in translations if len(translations[text]) == 10
+    ]
+
+    # Append to processed data
+    processed_data.extend(completed_translations)
+
+    # Reset translations
+    translations = {text: val for text, val in translations.items() if len(val) < 10}
+
+    # Convert to Hugging Face dataset format
+    translated_dataset = Dataset.from_pandas(pd.DataFrame(processed_data))
+
+    # Append to Hugging Face dataset (dummy function call)
+    translated_dataset.push_to_hub("vietdata/translated_echo", split="train")
+
+import gradio as gr
 
 # Simulated user data for demonstration
 user_data = {"hello": "hello"}
@@ -9,13 +69,14 @@ english_text = "Translate this text to Vietnamese."
 # User session dictionary to store logged-in status
 user_sessions = {}
 
-def login(username, password):
+def login(username, state):
+    state[0] = username
     # Authenticate user
-    if username in user_data and user_data[username] == password:
-        user_sessions[username] = True
-        return f"Welcome, {username}!", gr.update(visible=False), gr.update(visible=True)
+    if True:
+        #user_sessions[username] = True
+        return f"Welcome, {username}!", gr.update(visible=False), gr.update(visible=True), get_next_text(username)
     else:
-        return "Invalid username or password.", gr.update(visible=True), gr.update(visible=False)
+        return "Invalid username or password.", gr.update(visible=True), gr.update(visible=False), ""
 
 def logout(username):
     # Log out user and reset session
@@ -23,22 +84,28 @@ def logout(username):
         del user_sessions[username]
     return "Logged out. Please log in again.", gr.update(visible=True), gr.update(visible=False)
 
-def submit_translation(translation):
-    # Save the translation and provide feedback
-    return f"Translation submitted: {translation}"
+def submit_translation(translation, state, job_input):
+    try:
+        submit_translation(state[0], job_input, translation)
+        origin = job_input
+        # Save the translation and provide feedback
+        return f"""Translation of "{origin}" submitted: {translation}""", get_next_text(state[0])
+    except Exception as e:
+        print(e)
+        return "Error please try submit again!", job_input
 
 # Define the Gradio interface
 with gr.Blocks() as demo:
+    state = gr.State([None])
     # Login section
     with gr.Column(visible=True) as login_section:
-        username_input = gr.Textbox(placeholder="Enter your username", label="Username")
-        password_input = gr.Textbox(placeholder="Enter your password", label="Password", type="password")
+        username_input = gr.Textbox(placeholder="Enter your token", label="Token ID")
         login_button = gr.Button("Login")
         login_output = gr.Textbox(label="Login Status", interactive=False)
 
     # Translation section (initially hidden)
     with gr.Column(visible=False) as translation_section:
-        gr.Textbox(value=english_text, label="English Text", interactive=False)
+        job_input = gr.Textbox(value=english_text, label="English Text", interactive=False)
         translation_input = gr.Textbox(placeholder="Enter your translation here", label="Your Translation")
         submit_button = gr.Button("Submit Translation")
         translation_output = gr.Textbox(label="Submission Status", interactive=False)
@@ -46,13 +113,13 @@ with gr.Blocks() as demo:
 
     # Button functions
     login_button.click(
-        login, inputs=[username_input, password_input], outputs=[login_output, login_section, translation_section]
+        login, inputs=[username_input, state], outputs=[login_output, login_section, translation_section, job_input]
     )
     submit_button.click(
-        submit_translation, inputs=translation_input, outputs=translation_output
+        submit_translation, inputs=[translation_input, state, job_input], outputs=[translation_output, job_input]
     )
     logout_button.click(
         logout, inputs=[username_input], outputs=[login_output, login_section, translation_section]
     )
 
-demo.launch()
+demo.launch(debug=True)