Comparison with py-spy?

[nalzok]

I wonder how does this project compare to py-spy?

[pablogsal]

The main difference is that py-spy is a profiler while pystack is a debugger. This means that while both share functionality, pystack focus is getting as much information as possible even in the most complicated situations at the expense of time. This means that pystack works in some situations where py-spy does not. Currently, pystack is the only one to support core files as well (although IIRC they have plans to support some core file formats in the future). Here is a short table comparing what I know are the difference between both tools:

Feature	py-spy	pystack
Works on Linux	✔️	✔️
Works on MacOS	✔️	❌
Works on Windows	✔️	❌
Performance profiler (focus on speed)	✔️	❌
Debugger (focus on information and correctness)	❌	✔️
Can profile a process to tell why is slow	✔️	❌
Has an htop like interface	✔️	❌
Works with running processes	✔️	✔️
Works with core dump files	❌	✔️
Shows Python GIL status	✔️	✔️
Shows garbage collection cycles	✔️	✔️
Shows native function calls	✔️	✔️
Demangles symbols in native stack	✔️	✔️
Includes inlined functions in native stack	❌	✔️
Shows values of local variables and function arguments	✔️	✔️
Safe to use on running processes	✔️	✔️
Perform Python stack analysis without pausing process	✔️	✔️
Works with optimized Python interpreter binaries	❓	✔️
Works without symbols or debug information	❓	✔️
Tolerates memory corruption	❌	✔️
Uses PEP 657 information in tracebacks	❌	✔️