beenencoded / filesystem-snapshot-creator Goto Github PK

As of right now, the program has to scan through the list of paths that represent the heart of a snapshot to get to the next snapshot's information. This is a problem, because a snapshot consists of a list of strings (a list of paths on the filesystem), and to go through them all is inefficient. To give you an idea of why this is so inefficient, I will tell you I have over 291,000 paths on my computer, which translate to a snapshot containing 291,000 strings (million+ characters at least).

I have an Idea: include the file postition on the next snapshot along with the snapshot information that preceds the list of paths. Then, when we load, we can just ifstream().seekg() to the next snapshot.

If anyone else has any other good ideas, I would be willing to hear them.

Also, about memory: Whenever a snapshot is taken, the entire thing is loaded into memory. I know it would be better to just save it to the file, but it also has to be completely loaded for snapshot comparisons, so that a report can be created listing the created and deleted paths. So far, upon the class' destruction there is 2,900 kb of memory still allocated to the program (ascertained through Task Manager... It may or may not be right...), and I want to make it more efficient that that.