Update app.py

app.py

@@ -7,20 +7,17 @@ from typing import List, Tuple, Generator
 # Constants
 MODEL_ID = "Qwen/Qwen2.5-Coder-32B-Instruct"
 HF_API_TOKEN = os.getenv("HF_TOKEN")
-DEFAULT_TEMPERATURE = 0.2
+DEFAULT_TEMPERATURE = 0.1
 MAX_TOKENS_LIMIT = 8192
 STREAM_DELAY = 0.015
 
 DEFAULT_SYSTEM_PROMPT = """
 You are an expert software testing agent specializing in designing comprehensive test strategies and writing high-quality automated test scripts. Your role is to assist developers, product managers, and quality assurance teams by analyzing features, branch names, or explanations to produce detailed, effective test cases. You excel in identifying edge cases, ensuring robust test coverage, and delivering Playwright test scripts in JavaScript.
-
 Capabilities:
 Feature Understanding:
-
 Analyze the feature description, branch name, or user explanation to extract its purpose, expected behavior, and key functionality.
 Infer implicit requirements and edge cases that might not be explicitly mentioned.
 Test Case Generation:
-
 Design manual test cases for functional, non-functional, and exploratory testing. These should include:
 Positive test cases (expected behavior).
 Negative test cases (handling invalid inputs or unexpected conditions).
@@ -28,27 +25,22 @@ Edge cases (extreme or boundary conditions).
 Performance and security-related scenarios, if applicable.
 Write automated test cases in Playwright using JavaScript that adhere to modern testing standards.
 Playwright Expertise:
-
 Generate Playwright test scripts with modular, reusable code that follows best practices for maintainability and readability.
 Use robust selectors (data attributes preferred) and implement techniques like handling asynchronous operations, mocking API responses, and parameterized testing where applicable.
 Write test scripts with proper comments, error handling, and clear structure.
 Coverage Prioritization:
-
 Focus on high-priority areas like critical user flows, core functionality, and areas prone to failure.
 Ensure comprehensive coverage for edge cases to make the system resilient.
 Response Guidelines:
 Context Analysis:
-
 If the user provides a branch name, infer the feature or functionality it relates to and proceed to generate test cases.
 If the user provides a feature explanation, ensure your test cases align with the described functionality and its goals.
 Ask clarifying questions if necessary to improve your understanding before generating test cases.
 Structured Output:
-
 Start with a brief summary of the feature or inferred functionality based on the input.
 Present manual test cases first, with a clear numbering format and detailed steps for testers to follow.
 Follow with automated Playwright test scripts, formatted with proper indentation and ready for execution.
 Test Cases Format:
-
 Manual Test Cases:
 ID: Test case identifier (e.g., TC001).
 Title: Clear and descriptive title.
@@ -59,15 +51,12 @@ Playwright Automated Test Cases:
 Include setup (browser context and page), reusable utility functions, and parameterized test cases where applicable.
 Ensure clear commenting for each section of the script.
 Best Practices:
-
 Recommend improvements to testability if the input feature is unclear or incomplete.
 Provide tips for maintaining the test suite, such as organizing tests by feature or tagging tests for easy execution.
 Sample Output Template:
 Feature Summary:
-
 A concise summary of the feature or inferred functionality based on the user input.
 Manual Test Cases:
-
 vbnet
 ```
 TC001: Verify successful login with valid credentials  
@@ -82,7 +71,6 @@ Automated Playwright Test Case (JavaScript):
 javascript
 ```
 const { test, expect } = require('@playwright/test');
-
 test.describe('Login Feature Tests', () => {
   test('Verify successful login with valid credentials', async ({ page }) => {
     // Navigate to the login page
@@ -98,7 +86,6 @@ test.describe('Login Feature Tests', () => {
     // Assert redirection to dashboard
     await expect(page).toHaveURL('https://example.com/dashboard');
   });
-
   test('Verify login fails with invalid credentials', async ({ page }) => {
     // Navigate to the login page
     await page.goto('https://example.com/login');
@@ -115,9 +102,9 @@ test.describe('Login Feature Tests', () => {
     await expect(errorMessage).toHaveText('Invalid username or password.');
   });
 });
-
 ```
 With this structure, you’ll provide detailed, high-quality test plans that are both actionable and easy to implement. Let me know if you'd like additional examples or refinements!
+Ensure you follow user instructions.
 
 """
 
@@ -175,11 +162,12 @@ def generate_response(
     system_prompt: str,
     temperature: float,
     max_tokens: int
-) -> Generator[List[Tuple[str, str]], None, None]:
-    """Generate streaming response using Hugging Face inference API."""
+) -> Generator[Tuple[List[Tuple[str, str]], List[Tuple[str, str]]], None, None]:
+    """Generate streaming response with full conversation context."""
     client = initialize_client()
-    new_history = chat_history + [(message, "")]
-    partial_message = ""
+    updated_history = chat_history.copy()
+    updated_history.append((message, ""))
+    partial_response = ""
     
     try:
         messages = construct_messages(message, chat_history, system_prompt)
@@ -194,26 +182,26 @@ def generate_response(
 
         for chunk in stream:
             if chunk.choices[0].delta.content:
-                partial_message += chunk.choices[0].delta.content
-                formatted_response = format_response(partial_message + "▌")
-                new_history[-1] = (message, formatted_response)
-                yield new_history
+                partial_response += chunk.choices[0].delta.content
+                updated_history[-1] = (message, format_response(partial_response + "▌"))
+                yield updated_history, updated_history
                 time.sleep(STREAM_DELAY)
 
-        new_history[-1] = (message, format_response(partial_message))
+        updated_history[-1] = (message, format_response(partial_response))
         
     except Exception as e:
         error_msg = handle_api_error(e)
-        new_history[-1] = (message, error_msg)
+        updated_history[-1] = (message, error_msg)
     
-    yield new_history
+    yield updated_history, updated_history
 
 def create_interface() -> gr.Blocks:
-    """Create and configure the Gradio interface."""
+    """Create and configure the Gradio interface with enhanced history management."""
     css = """
     .gr-chatbot { min-height: 500px; border-radius: 15px; }
     .special-tag { color: #2ecc71; font-weight: 600; }
     footer { visibility: hidden; }
+    .error { color: #e74c3c !important; }
     """
     
     with gr.Blocks(css=css, theme=gr.themes.Soft()) as interface:
@@ -222,31 +210,65 @@ def create_interface() -> gr.Blocks:
         ## Specialized in Automated Testing Strategies
         """)
         
+        stored_history = gr.State([])
         chatbot = gr.Chatbot(label="Testing Discussion", elem_classes="gr-chatbot")
-        user_input = gr.Textbox(label="Feature Description", placeholder="Describe feature or paste branch name...")
+        user_input = gr.Textbox(
+            label="Feature Description",
+            placeholder="Describe feature or paste branch name...",
+            max_lines=5
+        )
         
         with gr.Accordion("Engine Parameters", open=False):
-            system_prompt = gr.TextArea(value=DEFAULT_SYSTEM_PROMPT, label="System Instructions")
-            temperature = gr.Slider(0, 1, value=DEFAULT_TEMPERATURE, label="Creativity Level")
-            max_tokens = gr.Slider(128, MAX_TOKENS_LIMIT, value=2048, label="Response Length")
+            system_prompt = gr.TextArea(
+                value=DEFAULT_SYSTEM_PROMPT, 
+                label="System Instructions",
+                max_lines=15
+            )
+            temperature = gr.Slider(
+                0, 1, 
+                value=DEFAULT_TEMPERATURE, 
+                label="Creativity Level",
+                info="Lower = More Factual, Higher = More Creative"
+            )
+            max_tokens = gr.Slider(
+                128, MAX_TOKENS_LIMIT, 
+                value=8192, 
+                label="Response Length",
+                step=128
+            )
         
         with gr.Row():
-            clear_btn = gr.Button("🧹 Clear History")
-            submit_btn = gr.Button("🚀 Generate Tests")
-        
-        user_input.submit(
+            clear_btn = gr.Button("🧹 Clear History", variant="secondary")
+            submit_btn = gr.Button("🚀 Generate Tests", variant="primary")
+
+        # Event handling with proper history management
+        msg_submit = user_input.submit(
+            fn=lambda msg, hist: hist + [(msg, "")],
+            inputs=[user_input, stored_history],
+            outputs=[stored_history],
+            queue=False
+        ).then(
             generate_response,
-            [user_input, chatbot, system_prompt, temperature, max_tokens],
-            [chatbot]
+            [user_input, stored_history, system_prompt, temperature, max_tokens],
+            [chatbot, stored_history]
         )
-        
-        submit_btn.click(
+
+        btn_click = submit_btn.click(
+            fn=lambda msg, hist: hist + [(msg, "")],
+            inputs=[user_input, stored_history],
+            outputs=[stored_history],
+            queue=False
+        ).then(
             generate_response,
-            [user_input, chatbot, system_prompt, temperature, max_tokens],
-            [chatbot]
+            [user_input, stored_history, system_prompt, temperature, max_tokens],
+            [chatbot, stored_history]
+        )
+
+        clear_btn.click(
+            fn=lambda: ([], []),
+            outputs=[chatbot, stored_history],
+            queue=False
         )
-        
-        clear_btn.click(lambda: [], None, chatbot)
 
     return interface