On: The Real Difference Between Git and Mercurial

[skrebbel]

Wait, Mercurial users start coding when they run into a problem with the tool?

I always start googling and hoping there's a stackoverflow question about it that has an answer that I don't get.

[gcp → skrebbel]

That certainly reflects my experience: I have a problem...

Mercurial: there's an extension for that!

Git: you can fix that just by adding --foo bar:buzz --derp herp to the commandline lol RTFM

[mariusmg → gcp]

Maybe somebody can add a extension to Mercurial that let's you use empty folders.....

[xentac → mariusmg]

Git and Mercurial both suffer from this "problem". Because manifests/trees must have contents (file hashes) to exist, you can't track an empty folder.

I suppose git's tree objects could point to the empty tree to record an empty folder, but most of the git code is comparing file blobs not trees.

[pavel_lishin → xentac]

In git, we solve this problem by sticking a .gitignore file in any directories we want to be part of the repo, but without the contents (cache folders, user-generated data destinations, etc.)

How do you get around this in mercurial?

[getsat → pavel_lishin]

It's usually .gitkeep from what I've seen. .gitignore usually lives in the repo root.

[tonfa → pavel_lishin]

> How do you get around this in mercurial?

Same (with a .empty or .keep file).

[lilyball → xentac]

Git's tree structure can certainly encode an empty directory, the problem is the index file can't.

[jonknee → skrebbel]

I think the author was referencing hg's numerous extensions. In reality I think both git and hg users turn to Google when they get stuck (same for bzr, svn, cvs, and every other scm).