A streaming log viewer and profiler visualizer built for Python programs.
- 🔄 Streaming JSONL file processing — processes log files while loading
- 🌳 Virtualized expandable Call Tree — smooth scroll & expand
- 📊 Timeline view (Plotly powered) — visual execution timeline
- 🔥 FlameGraph View — visualizing time spent in functions.
- 📈 Summary table — aggregated total time per function
- ⚡ Fully browser-based — no backend server required
Profiler expects newline-delimited JSON arrays (.jsonl), where each line represents one event:
["0.4930", "0.0290", "dominates", 19, 5, 1]Flame Graph Tab:
See the program Flame Graph.
Timeline Tab:
See the visual execution timeline.
Summary Tab:
Overall summary for all functions (average runtimes, number of calls, ...).
Details Tab:
Click on any function in the call tree to see details.
Compare Tab:
Load two log files and see what has improved or worsened.
