The biggest bottleneck to pulling up every process' libraries and all their packages are all the dpkg -S queries.
- If it turns out that the same libraries are included in lots of processes, we can cache path <-> package relationships to speed this up.
- We may also just be able to store a db of these relationships in memory if it's not too big. We can research how
dpkg -S works and also apt-file.
- Since
apt-file stores an offline database for all of the packages in the repo, we could potentially do this check on the backend and not have libspector be responsible for it (although this would make symlinks and packages installed from outside the default repos problematic).
The biggest bottleneck to pulling up every process' libraries and all their packages are all the
dpkg -Squeries.dpkg -Sworks and alsoapt-file.apt-filestores an offline database for all of the packages in the repo, we could potentially do this check on the backend and not have libspector be responsible for it (although this would make symlinks and packages installed from outside the default repos problematic).