Merge pull request #1050 from MervinPraison/claude/pr-1049-20250723-0902

fix: Remove display_generating when stream=false to prevent streaming-like behavior

Author: MervinPraison

display_generating_fix_summary.md

@@ -0,0 +1,37 @@
+# Fix for display_generating Issue
+
+## Problem
+When `stream=false` but `verbose=true`, the system was still showing streaming-like visual behavior ("Generating... X.Xs") because the code was using `display_generating` function even when the user explicitly set `stream=false`.
+
+## Root Cause
+Two locations in `agent.py` were passing `display_generating` as `display_fn` when `stream=False` and `verbose=True`:
+
+- **Line 1073**: `display_fn=display_generating if (not stream and self.verbose) else None`
+- **Line 1172**: `display_fn=display_generating if (not stream and self.verbose) else None`
+
+This conflated two different concepts:
+- **Verbose**: Show detailed information  
+- **Visual Progress**: Show animated progress indicators
+
+## Solution
+Changed both locations to:
+```python
+display_fn=None,  # Don't use display_generating when stream=False to avoid streaming-like behavior
+```
+
+## Expected Behavior After Fix
+
+| Stream | Verbose | Visual Behavior |
+|--------|---------|----------------|
+| False  | False   | No display |
+| False  | True    | **No streaming-like behavior (FIXED)** |
+| True   | False   | Native streaming display |  
+| True   | True    | Native streaming display |
+
+## Files Modified
+- `src/praisonai-agents/praisonaiagents/agent/agent.py` - Lines 1073 and 1172
+
+## Verification
+- Test script: `test_display_generating_fix.py` 
+- All old problematic patterns removed
+- New safe patterns implemented at both locations

src/praisonai-agents/praisonaiagents/agent/agent.py

@@ -1070,7 +1070,7 @@ def _process_stream_response(self, messages, temperature, start_time, formatted_
             tools=formatted_tools,
             start_time=start_time,
             console=self.console,
-            display_fn=display_generating if (not stream and self.verbose) else None,  # stream is True in this context
+            display_fn=None,  # Don't use display_generating when stream=False to avoid streaming-like behavior
             reasoning_steps=reasoning_steps
         )
 
@@ -1109,9 +1109,9 @@ def _chat_completion(self, messages, temperature=0.2, tools=None, stream=True, r
                         reasoning_steps=reasoning_steps
                     )
                 else:
-                    # Non-streaming with custom LLM - add display functionality for verbose mode
-                    if (not stream and self.verbose) and self.console:
-                        # Show "Generating..." display for verbose mode like OpenAI path
+                    # Non-streaming with custom LLM - don't show streaming-like behavior
+                    if False:  # Don't use display_generating when stream=False to avoid streaming-like behavior
+                        # This block is disabled to maintain consistency with the OpenAI path fix
                         with Live(
                             display_generating("", start_time),
                             console=self.console,
@@ -1169,7 +1169,7 @@ def _chat_completion(self, messages, temperature=0.2, tools=None, stream=True, r
                     execute_tool_fn=self.execute_tool,
                     stream=stream,
                     console=self.console if (self.verbose or stream) else None,
-                    display_fn=display_generating if (not stream and self.verbose) else None,
+                    display_fn=None,  # Don't use display_generating when stream=False to avoid streaming-like behavior
                     reasoning_steps=reasoning_steps,
                     verbose=self.verbose,
                     max_iterations=10

test_display_generating_fix.py

@@ -0,0 +1,74 @@
+#!/usr/bin/env python3
+"""
+Test script to verify that display_generating fix is working correctly.
+This script checks that the problematic patterns have been removed from agent.py.
+"""
+
+import re
+import sys
+from pathlib import Path
+
+def test_display_generating_fix():
+    """Test that the display_generating fix has been applied correctly."""
+    
+    agent_file = Path("src/praisonai-agents/praisonaiagents/agent/agent.py")
+    
+    if not agent_file.exists():
+        print(f"❌ ERROR: {agent_file} not found")
+        return False
+    
+    content = agent_file.read_text()
+    
+    # Check that the old problematic patterns are gone
+    old_pattern = r"display_fn=display_generating if \(not stream and self\.verbose\) else None"
+    old_matches = re.findall(old_pattern, content)
+    
+    # Also check for the custom LLM path problematic pattern
+    old_custom_pattern = r"if \(not stream and self\.verbose\) and self\.console:\s*.*with Live\(\s*display_generating"
+    old_custom_matches = re.findall(old_custom_pattern, content, re.MULTILINE | re.DOTALL)
+    
+    if old_matches:
+        print(f"❌ FAILED: Found {len(old_matches)} instances of old problematic pattern:")
+        print(f"   'display_fn=display_generating if (not stream and self.verbose) else None'")
+        return False
+    
+    if old_custom_matches:
+        print(f"❌ FAILED: Found {len(old_custom_matches)} instances of old custom LLM problematic pattern:")
+        print(f"   'if (not stream and self.verbose) and self.console: ... display_generating'")
+        return False
+    
+    # Check that the new safe patterns are present
+    new_pattern = r"display_fn=None,\s*# Don't use display_generating when stream=False to avoid streaming-like behavior"
+    new_matches = re.findall(new_pattern, content)
+    
+    # Check for the new custom LLM path fix
+    new_custom_pattern = r"if False:\s*# Don't use display_generating when stream=False to avoid streaming-like behavior"
+    new_custom_matches = re.findall(new_custom_pattern, content)
+    
+    expected_total = 3  # 2 OpenAI path + 1 custom LLM path
+    actual_total = len(new_matches) + len(new_custom_matches)
+    
+    if actual_total < expected_total:
+        print(f"❌ FAILED: Expected at least {expected_total} instances of new patterns, found {actual_total}")
+        print("   Expected patterns:")
+        print("   - 'display_fn=None,  # Don't use display_generating when stream=False to avoid streaming-like behavior'")
+        print("   - 'if False:  # Don't use display_generating when stream=False to avoid streaming-like behavior'")
+        return False
+    
+    print("✅ SUCCESS: display_generating fix has been applied correctly!")
+    print(f"   - Removed old problematic patterns: 0 found (expected 0)")
+    print(f"   - Added new safe patterns: {actual_total} found (expected >= {expected_total})")
+    print(f"     * OpenAI path fixes: {len(new_matches)}")
+    print(f"     * Custom LLM path fixes: {len(new_custom_matches)}")
+    
+    # Show the context of the changes
+    lines = content.split('\n')
+    for i, line in enumerate(lines, 1):
+        if "Don't use display_generating when stream=False" in line:
+            print(f"   - Line {i}: {line.strip()}")
+    
+    return True
+
+if __name__ == "__main__":
+    success = test_display_generating_fix()
+    sys.exit(0 if success else 1)