cppcheck analysis takes around 1.5-2 hours for my project. Second run
without any changes takes about 30 minutes. It would be really useful
if results cache would work faster.
About my use case:
- using cppcheck cli
- solution with about 8k files (4.4k after excluding not interesting files)
- using .sln argument, --enable=all, -j 8 and --cppcheck-build
I briefly checked hashing approach, it looks like hash function is
called for parsed tokens.
Token extraction takes a significant part of time in this case.
That can be improved by calling hash function for whole file content
at the very beginning. Downside in that case would be that any minor
comment/formatting/styling change would invalidate cached data, but
from my experience that is not a common change.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
cppcheck analysis takes around 1.5-2 hours for my project. Second run
without any changes takes about 30 minutes. It would be really useful
if results cache would work faster.
About my use case:
- using cppcheck cli
- solution with about 8k files (4.4k after excluding not interesting files)
- using .sln argument, --enable=all, -j 8 and --cppcheck-build
I briefly checked hashing approach, it looks like hash function is
called for parsed tokens.
Token extraction takes a significant part of time in this case.
That can be improved by calling hash function for whole file content
at the very beginning. Downside in that case would be that any minor
comment/formatting/styling change would invalidate cached data, but
from my experience that is not a common change.