1. 5
  1.  

  2. 2
    • streamlets (a way to enrich the stream with new attributes)
    • parallel processing (utilizes POSIX MQ)
    • SLEB128 (for encoding variable-length integers)
    • other improvements
    1. 2

      I added this to the end of my wiki page which lists similar projects that handle CSV, JSON, text, and binary:

      https://github.com/oilshell/oil/wiki/Structured-Data-in-Oil

      I don’t think binary is great for most use cases. Here is mine which I probably won’t get to any time soon:

      https://github.com/oilshell/oil/wiki/TSV2-Proposal

      1. 1

        This looks interesting, but why make it so cumbersome to get started? Instead of having to clone and checkout lots of files, why not pack every release in a tarball so that it’s easy to get started?