On Tue, 8 May 2007, Johannes Schindelin wrote: > > It might be a much better idea to write something a la git-fetch--tool, > which is a helper in C (thus very fast and memory efficient), outputting > easily parseable data. Well, we actually do have that. "git log" (or "git-rev-list") really does all the heavy lifting. The reason you can do things like "gitk --merge" is not because gitk itself has _any_ idea about anything, but because it just passes the arguments down to git-rev-list (and hopefully soon git log), which really does all the complex stuff. But gitk still ends up having a big memory footpring, simply because if you get the data for a few hundred thousand commits (with commit messages etc), and have to keep track of the relationships between them, you are going to easily use hundreds of megs of memory. Linus - To unsubscribe from this list: send the line "unsubscribe git" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
