Breaking in StdLib while not stepping into

Abierto

#854 abierto el 4/2/2022 6 comentarios 0 reacciones 0 asignados Ver en GitHub

Nadie ha tomado este issue todavía.

Evaluación

Dificultad: 5/5
Tiempo estimado: Más de una semana
Aptitud para principiantes: 30/100

Tipo de issue: Error
Claridad: Necesita aclaración
Estado de actividad: Estancado
Stack tecnológico: jupyter, jupyter-notebook, python
Área: backend, devtools

Línea de trabajo

Empieza leyendo los PRs #812 y #839 de ipykernel, el issue #832 y el PR #11566 enlazado de JupyterLab para entender el comportamiento modificado del stepping. El issue no menciona archivos fuente ni tests; se consideraría terminado definir y verificar una forma de detenerse en archivos de userland y de la biblioteca estándar sin avanzar inesperadamente por esos archivos.

Escrito por el modelo de indexación a partir del texto del issue.

Descripción

JupyterLab 3.3.0a3 has a kernel sources panel which displays the loaded py files loaded (so the core python libraries, and userland libraries). This aims to offer the user the ability to add breakpoints and debug to see what happens in those userland libraries.

https://github.com/ipython/ipykernel/pull/812 has added the needed option to be able break into those stdlib files, while https://github.com/ipython/ipykernel/pull/839 has removed it. The removal is due to https://github.com/ipython/ipykernel/issues/832 where the step was going to the line in the call stack. Interesting to note, today the side effect describe in https://github.com/jupyterlab/jupyterlab/pull/11566#issuecomment-983668657 (breaking in selector.py) has disappeared.

Thinking a bit more, the behavior is expected. When we say to debugpy to break into all files, we also say to step into all files, meaning that the taken step goes to the next in the call stack, including the StdLib files.

@JohanMabille You say on https://github.com/ipython/ipykernel/issues/832#issuecomment-1010414146 that you would think to a solution ...changing the step over behavior, or filter stop events. Is there anything we can do in the short therm?

Lenguaje dominante: Python
Estrellas: 734
Forks: 411
Merge medio: 1 d 2 h
PR fusionados (30 d): 9