‹ BackHN Continuity

Thread

Git 3.0's upcoming SHA-256 default will be a costly mistake

570 points · 536 comments · chmaynard

  1. pavon · · focus · HN ↗
    Ugh, I didn't know that SHA-1 submodules wouldn't be supported in SHA-256 repos. That changes the transition from painless to a major dumpster fire. Having to maintain converted forks, and use different hashes from upstream is going to be a mess.
    1. hnlmorg · · focus · HN ↗
      git submodules has always been a major dumpster fire. You’re honestly better avoiding regardless of SHA-256 incompatibilities
      1. xyzsparetimexyz · · focus · HN ↗
        The majority of C++ projects I work with use them. Telling people not to use a core git feature isn't viable
        1. cerved · · focus · HN ↗
          I wouldn't call it a core feature -- more like a convenience hack.
        2. hnlmorg · · focus · HN ↗
          The only reason C++ projects depend upon submodules is because C++ is just as archaic as COBOL these days.

          I used to be one of the more vocal defenders of C++ but honestly, if the reason you’re defending git submodules is because there’s no better option in C++; then you’ve basically already lost the argument.

          Nearly every other language has found a better alternative. Even Go has, and that’s largely mocked for its prehistoric approach to modern programming language design.

Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.