This post introduces Optique, a new library created to address the pervasive problem of repetitive and often messy validation code in CLI tools. The author was motivated by the observation that nearly every CLI tool reinvents the wheel with similar validation patterns for dependent options, mutually exclusive options, and environment-specific requirements. Optique leverages parser combinators and TypeScript's type inference to ensure that CLI arguments are parsed directly into valid configurations, eliminating the need for manual validation. By describing the desired CLI configuration with Optique, TypeScript automatically infers the types and constraints, catching potential bugs at compile time. The author shares their experience of deleting large chunks of validation code and simplifying refactoring tasks. Optique aims to provide a more robust and maintainable approach to CLI argument parsing, potentially saving developers from writing the same validation logic repeatedly.
  • azertyfun@sh.itjust.works
    251·
    21 days ago

    Counterpoint: Yes, parse don’t validate, but CLIs should not be dealing with dependency management.

    I love Python’s argparse because:

    • It’s “Parse, don’t validate” (even supports FileType as a target)
    • It enforces or strongly encourages good CLI design
      • Required arguments should in most situations be positional arguments, not flags. It’s curl <URL> not curl --url <URL>.
      • Flags should not depend on each other. That usually indicates spaghetti CLI design. Don’t do server --serve --port 8080 and server --reload with rules for mix-and-matching those, do server serve --port 8080 and server reload with two separate subparsers.
      • Mutually exclusive flags sometimes make sense but usually don’t. Don’t do --xml --json, do -f [xml|json].
      • This or( pattern of yours IMO should always be replaced by a subparser (which can use inheritance!). As a user the options’ data model should be immediately intuitive to me as I look at the --help and having mutually exclusive flags forces the user to do the extra work of dependency management. Don’t do server --env prod --auth abc --ssl, do server serve prod --auth abc --ssl where prod is its own subparser inheriting from AbstractServeParser or whatever.

    Thinking of CLI flags as a direct mapping to runtime variables is the fundamental mistake here I think. A CLI should be a mapping to the set(s) of behavior(s) of your application. A good CLI may have mandatory positional arguments but has 0 mandatory flags, 0 mutually exclusive flags, and if it implements multiple separate behaviors should be a tree of subparsers. Any mandatory or mutually exclusive flags should be an immediate warning that you’re not being very UNIX-y in your CLI design.

  • Asetru@feddit.org
    225·
    21 days ago

    The or() combinator means exactly one succeeds.

    Using “or” to define a function that does “xor”… Did that guy never hear about formal logic? That’s, like, first or second semester stuff…

    Here’s the thing: I don’t have a CS degree.

    sigh

    • Glitchvid@lemmy.worldEnglish
      21·
      20 days ago

      It’s an understandable interpretation for the lexical use of or which can imply exclusive disjunction.

      In Rust the result type has the method .or() which returns either Ok(A) or Ok(B) (but not both), and I don’t see clambering to change it to xor, because the exclusive nature is implicit both linguistically and in the type state.

      • Asetru@feddit.org
        1·
        19 days ago

        The result type in rust does not return a true/false but a type. More importantly though, it doesn’t return err if both values are set but simply returns the first value:

        So… It’s not only not mapping your input to truth values, it also behaves more like I’d expect an “or” to behave, which is not “xor” or, if there’s more than two inputs, “exactly one”, but succeeding if any input is set.

        • Glitchvid@lemmy.worldEnglish
          1·
          18 days ago

          …Which is basically how the OP’s or function also works, it takes several Option<T>s and returns the first valid one (and only that one), it doesn’t operate on boolean logic types — it’s a valid lexical use of or.

          • Asetru@feddit.org
            1·
            18 days ago

            Absolutely not.

            Mutually exclusive options

            Another classic. Pick one output format: JSON, YAML, or XML. But definitely not two.

            Emphasis mine.

            It takes the input and fails if there is more than one valid one, which decidedly isn’t what’s an “or” in comp sci.

    • kibiz0r@midwest.socialEnglish
      71·
      21 days ago

      Not exactly an answer, but I’ll take the opportunity to point out that Bun has a shell feature which makes it easy to mix and match JS and Bash in the same script, and it provides a compatibility layer for Windows users so that you don’t have to worry about platform differences in shell capabilities. https://bun.sh/guides/runtime/shell

  • Serinus@lemmy.world
    71·
    21 days ago

    I’m not sure the value added is worth the extra layer.

    I guess my command line options just aren’t all that complicated.

  • TehPers@beehaw.orgEnglish
    3·
    21 days ago

    I like the concept, and it’s great in TS. Unfortunately, not as doable in other languages.

    I’m a bit curious if it’s possible to extend clap to do this in Rust though (specifically mutually-exclusive arg groups).

      • TehPers@beehaw.orgEnglish
        9·
        21 days ago

        This doesn’t represent the mutual exclusivity through the type system (which is what the article is all about).

        I love clap and I use it a lot, but the only way to represent the exclusivity through the type system in Rust is through an enum.

        • ExFed@programming.dev
          5·
          21 days ago

          Agreed. As nice as clap is, it’s not a combinator. Parser combinators have a the really nice feature of sharing the same “shape” as the data they parse, which makes them trivial to generate from a schema … or to just use them to represent your schema in the first place ;) .

      • TehPers@beehaw.orgEnglish
        7·
        21 days ago

        Mentioned this to the other commenter, but this doesn’t use the type system to enforce the mutual exclusivity constraint. In Rust, the main way to do that via the type system is through enums.

        • Ephera@lemmy.mlEnglish
          3·
          21 days ago

          Ah, fair enough. Not sure how to do that then.

          I was gonna say, I feel like the current method does a good enough job documenting that validation has happened, but I guess you do want it reflected in the structure of the type, so that the code that takes the information from the struct can safely make the assumption that some of the options don’t exist. And then, yeah, it would be nice to not need a separate parsing step for that.