Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

>Iirc google and facebook decided against using git because it didn't scale to stupidly enormous code bases.

Compared to mercurial, that Facebook uses?



And Facebook added a lot to Mercurial so that it would scale to their stupidly enormous code base.

https://code.facebook.com/posts/218678814984400/scaling-merc...


Meanwhile OpenJDK (which is another project composed of separate but interconnected components) wrote a mercurial extension to manage multiple repositories easily (http://openjdk.java.net/projects/code-tools/trees/). I think this is a much better solution than a monorepo, you can still see the "whole codebase" like it were a single giant repository but you don't have to deal with the scaling headaches in quite the same way.

I was pretty surprised when I made my own build of OpenJDK 9 how easy it was to work with, `hg tclone blah`, `hg tup jdk9.0.1+11`, `./configure blah; make images` and done. Even if git submodules were closer in functionality (checking out the same tag across multiple modules at once with ease) the song-and-dance with actually downloading the modules after cloning is annoying.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: