Date: 2025-10-07

🚀 Highlights

• ⚡ Async/Await tracing with concurrency timeline and async stats
• 📊 Comparison Mode with side-by-side HTML, diff highlighting, and summary metrics
• 💾 Memory Leak Detection with object tracking, snapshots, growth analysis, and HTML reports
• 🎨 UI/UX fixes and improvements (module filtering, layouts, JSON export, CPU profiling panel)

---

⚡ Async/Await Tracing

New APIs

• trace_async — Decorator to trace async functions
• trace_scope_async() — Async context manager to trace async code
• gather_traced() — Traced version of asyncio.gather with concurrency tracking
• get_async_stats(graph) — Compute async stats (max concurrency, efficiency)
• AsyncCallGraph — Extended call graph with async metadata

Example

import asyncio
from callflow_tracer.async_tracer import trace_async, trace_scope_async, gather_traced, get_async_stats

@trace_async
async def fetch(i):
    await asyncio.sleep(0.05)
    return i

async def main():
    async with trace_scope_async("async_trace.html") as graph:
        results = await gather_traced(*[fetch(i) for i in range(10)])
    stats = get_async_stats(graph)
    print(stats)

asyncio.run(main())

---

📊 Comparison Mode

New APIs

• compare_graphs(before, after) — Compute function-level differences
• export_comparison_html(before, after, output_file, label1, label2, title) — Side-by-side HTML report

Features

• Diff highlighting (green = improvements, red = regressions)
• Summary metrics (time saved, added/removed/modified functions)
• Detailed comparison table

Example

from callflow_tracer import trace_scope
from callflow_tracer.comparison import export_comparison_html

with trace_scope() as before:
    slow()
with trace_scope() as after:
    fast()

export_comparison_html(before, after, "comparison.html", label1="Before", label2="After")

---

💾 Memory Leak Detection

New APIs

• MemoryLeakDetector — Orchestrates leak analysis and reporting
• detect_leaks(output_file=None) — Context manager for leak detection
• @track_allocations — Decorator to track allocations in a function
• MemorySnapshot(label) — Capture memory/object snapshots
• find_reference_cycles() — Detect cyclic references
• get_memory_growth(interval, iterations) — Monitor growth over time
• get_top_memory_consumers(limit) — Show top allocation sites

Example

from callflow_tracer.memory_leak_detector import detect_leaks

with detect_leaks("leak_report.html") as detector:
    leaked = []
    for i in range(10):
        leaked.append([0] * 10000)
        detector.take_snapshot(f"Iter_{i}")

Reports Include

• Memory growth chart (Chart.js)
• Object type distribution
• Snapshot comparisons (memory/objects diff)
• Suspected leaks and severity
• Recommendations

---

🎨 UI/UX Fixes in HTML Visualizations

• Module filter dropdown now functional (filters nodes/edges, smooth fit animation)
• Circular and timeline layouts fixed (correct positioning; physics toggled appropriately)
• JSON export fixed (no network.getData(); export uses original data with metadata)
• CPU profiling block restored with modern collapsible UI

---

🔧 Developer Notes

• Package version bumped to 0.2.4 in pyproject.toml and callflow_tracer/__init__.py
• New exports added to callflow_tracer/__init__.py for Async, Comparison, and Memory Leak APIs
• Added tests and examples for all new features

---

📚 Where To Start

• Async: examples/async_example.py
• Comparison: examples/comparison_example.py
• Memory Leaks: examples/memory_leak_example.py
• API Reference: docs/API_DOCUMENTATION.md

---

✅ Summary

v0.2.4 focuses on modern Python workloads (async), comparative performance analysis, and robustness via memory leak detection — all with polished visualization and documentation.