Update app.py

app.py

@@ -49,7 +49,6 @@ description and a Python code snippet, then produce a structured evaluation repo
 - Do NOT execute the code.
 - Base your evaluation solely on static code analysis and logical reasoning.
 - Keep feedback concise, clear, and actionable.
-- Pay close attention to operators (+, -, *, /) and ensure they match the description exactly.
 - Your response MUST follow the output format exactly.
 ### OUTPUT FORMAT
 Respond only with the following structure — no extra text before or after:
@@ -416,21 +415,11 @@ body, .gradio-container {
     padding: 16px !important; color: #C0C3D0 !important;
     cursor: default !important; min-height: 120px !important;
 }
-#error-out {
-    width: 100% !important;
-}
 #error-out textarea {
-    background: #1A0E0E !important;
-    border: 1px solid #5a2020 !important;
-    border-radius: 10px !important;
-    font-family: 'DM Sans', sans-serif !important;
-    font-size: 13.5px !important;
-    padding: 14px 16px !important;
-    color: #F0997B !important;
-    cursor: default !important;
-    min-height: 120px !important;
-    width: 100% !important;
-    box-sizing: border-box !important;
+    background: #1A0E0E !important; border: 1px solid #5a2020 !important;
+    border-radius: 10px !important; font-family: 'DM Sans', sans-serif !important;
+    font-size: 13.5px !important; padding: 14px 16px !important;
+    color: #F0997B !important; cursor: default !important;
 }
 .divider { height: 1px; background: #2E3140; margin: 20px 0; }
 footer { display: none !important; }
@@ -460,7 +449,6 @@ INPUT_HINT_HTML = """
 </div>
 """
 
-
 def build_ui(evaluate_fn):
 
     def _split_metrics(metrics_str: str):
@@ -477,32 +465,13 @@ def build_ui(evaluate_fn):
     def _ui_evaluate(description: str, code: str):
         verdict, metrics, issues, error = evaluate_fn(description, code)
         accuracy, summary = _split_metrics(metrics)
-
-        if error:
-            # Top row stays visible but empty — keeps full width
-            return (
-                gr.update(value="",    visible=True),        # verdict
-                gr.update(value="",    visible=True),        # accuracy
-                gr.update(value="",    visible=True),        # summary
-                gr.update(value="",    visible=False),       # issues — hidden
-                gr.update(value=error, visible=True),        # error  — shown
-            )
-
-        # Success — show results, hide error
         if "PASS" in verdict.upper():
             verdict_display = "✅  PASS"
         elif "FAIL" in verdict.upper():
             verdict_display = "❌  FAIL"
         else:
             verdict_display = verdict or ""
-
-        return (
-            gr.update(value=verdict_display, visible=True),  # verdict
-            gr.update(value=accuracy,        visible=True),  # accuracy
-            gr.update(value=summary,         visible=True),  # summary
-            gr.update(value=issues,          visible=True),  # issues — shown
-            gr.update(value="",              visible=False), # error  — hidden
-        )
+        return verdict_display, accuracy, summary, issues, error
 
     with gr.Blocks(title="Python Code Evaluator") as demo:
 
@@ -514,69 +483,37 @@ def build_ui(evaluate_fn):
                 placeholder=(
                     "Describe what the Python code is supposed to do…\n\n"
                     "Example: Write a function that accepts a list of integers "
-                    "and returns the sum of all even numbers. It should handle "
-                    "empty lists by returning 0 and ignore non-integer values."
+                    "and returns the sum of all even numbers."
                 ),
-                lines=10,
-                max_lines=20,
-                elem_id="desc-input",
+                lines=10, max_lines=20, elem_id="desc-input",
             )
             code_input = gr.Textbox(
                 label="Python Code",
-                placeholder=(
-                    "# Paste your Python code here…\n\n"
-                    "# Note: Ensure proper indentation for accurate evaluation.\n\n"
-                    "def example(numbers):\n"
-                    "    total = 0\n"
-                    "    for item in numbers:\n"
-                    "        if item % 2 == 0:\n"
-                    "            total += item\n"
-                    "    return total\n"
-                ),
-                lines=10,
-                max_lines=30,
-                elem_id="code-input",
+                placeholder="# Paste your Python code here…\n\ndef example(data):\n    pass\n",
+                lines=10, max_lines=30, elem_id="code-input",
             )
 
         gr.HTML(INPUT_HINT_HTML)
 
         with gr.Row():
-            eval_btn  = gr.Button("Evaluate", variant="primary",
-                                  elem_id="eval-btn",  scale=0)
-            clear_btn = gr.Button("Clear",    variant="secondary",
-                                  elem_id="clear-btn", scale=0)
+            eval_btn  = gr.Button("Evaluate", variant="primary",  elem_id="eval-btn",  scale=0)
+            clear_btn = gr.Button("Clear",    variant="secondary", elem_id="clear-btn", scale=0)
 
         gr.HTML(RESULTS_HEADING_HTML)
 
-        # Top row — always visible to preserve full width
         with gr.Row(equal_height=True):
-            verdict_out  = gr.Textbox(label="Verdict",  interactive=False, elem_id="verdict-out",  scale=1)
-            accuracy_out = gr.Textbox(label="Accuracy", interactive=False, elem_id="accuracy-out", scale=1)
-            summary_out  = gr.Textbox(label="Summary",  interactive=False, elem_id="summary-out",  scale=2, lines=3)
-
-        # Issues and error swap in the same spot
-        issues_out = gr.Textbox(
-            label="Issues Detected",
-            interactive=False,
-            lines=5,
-            visible=True,
-            elem_id="issues-out",
-        )
+            verdict_out  = gr.Textbox(label="Verdict",   interactive=False, elem_id="verdict-out",  scale=1)
+            accuracy_out = gr.Textbox(label="Accuracy",  interactive=False, elem_id="accuracy-out", scale=1)
+            summary_out  = gr.Textbox(label="Summary",   interactive=False, elem_id="summary-out",  scale=2, lines=3)
 
-        with gr.Row():
-            error_out = gr.Textbox(
-                label="Status / Errors",
-                interactive=False,
-                visible=False,          # hidden by default
-                elem_id="error-out",
-                scale=1,
-            )
+        issues_out = gr.Textbox(label="Issues Detected", interactive=False, lines=5,  elem_id="issues-out")
+        error_out  = gr.Textbox(label="Status / Errors", interactive=False, visible=True, elem_id="error-out")
 
         gr.HTML("""
         <div style="text-align:right;font-size:11px;color:#4A4E60;margin-top:8px;">
           Press <kbd style="background:#1E2028;border:1px solid #3A3E52;border-radius:4px;
           padding:2px 6px;font-family:'Space Mono',monospace;font-size:10px;color:#9DA0B0;">
-          Ctrl+Enter</kbd> inside any input to evaluate
+          Ctrl+Enter</kbd> to evaluate
         </div>
         """)
 
@@ -589,8 +526,7 @@ def build_ui(evaluate_fn):
             return "", "", "", "", "", ""
 
         clear_btn.click(
-            fn=_clear,
-            inputs=[],
+            fn=_clear, inputs=[],
             outputs=[description_input, code_input,
                      verdict_out, accuracy_out, summary_out, issues_out],
         )
@@ -603,5 +539,4 @@ def build_ui(evaluate_fn):
 # ---------------------------------------------------------------------------
 if __name__ == "__main__":
     app = build_ui(validate_and_evaluate)
-    app.launch(server_name="0.0.0.0", server_port=7860, share=False,
-               show_error=True, css=CUSTOM_CSS)
+    app.launch(server_name="0.0.0.0", server_port=7860, share=False, show_error=True, css=CUSTOM_CSS)