Hacker News new | comments | show | ask | jobs | submit login
[dupe] Tom's Obvious, Minimal Language (like INI, only better) (github.com)
33 points by telemachos 1608 days ago | hide | past | web | 26 comments | favorite

Possible improvements occur to me:

1. If indentation is significant and required, dotted notation should be optional:

The parser should be able to understand that this results in a foo.bar.car structure without explicitly using dotted notation in the keys.

2. Support include files. With #1 implemented, includes would be extremely powerful and flexible. Since indentation is required, no special syntax is needed to relatively position the included files. And without requiring dotted notation in keys, include files can be reused in a configuration (such as common settings in the [servers] section of the example).

Whoops. I tend to rely on the site itself to block (recent) dupes. I guess the different title threw it, though I would have expected the identical URL to get caught.

I don't know if this is the answer, but I'm a fan of someone trying.

I personally never jumped on YAML or JSON. I liked that they weren't XML, but they each felt like XML++ too much for me. There was still too much, I don't know, weirdness/typing/oops! Don't get me wrong, I think they were better than XML, but only marginally. It would be nice to see more of a phase change type of approach. I'd LOVE to see something like this (or something else) that involves much less typing/special characters/etc than YAML and JSON.

Personal opinion.

How many less characters could there be while still maintaining meaning? Could you show an example of what you'd prefer?

The format could be very simple, but... allowing comments after values make it unnecessarily complex to parse. For example:

     value = "example # no that's not a comment" # but here's one
If inline comments where not allowed we could just check the first and last character for a quote and we know we have a string, same for arrays (check for [ and ]). But with inline comments, we need to parse the whole string.

As far as I can see both the C# and Python parsers don't handle this kind of comment correctly at the moment, and probably many implementations will have troubles with that. They should just drop this option in my opinion.

That's not hard to parse. Inline comments are nice. You'll be doing more reading of files that writing of parsers.

WAT?. Python can parse the above very nicely, thank you.

    >>> value = "example # no that's not a comment" # but here's one
    >>> value
    "example # no that's not a comment"

At the time I wrote that, the only Python parser was removing the comments by splitting the line on "#", which mean it wouldn't work with the example I mentioned.

It seems there are now several other Python parsers, so some of them are probably getting it right.

So the objection to JSON comes down to a couple of things: no comments and fragility in face of a trailing comma. I wonder if the right thing though would be to fix JSON (maybe even call it something else) rather than throw the baby out with the bathwater.

Personally, of course, I think we should just use a slightly modified version of s-exprs which would be easier to parse.

(:keyword value)

where value can be an atom, list of atoms or a list of keyword value pairs. Comments could follow the Lisp convention and start with semi-colon and end with newline. Only 5 characters to escape, namely, '(', ')', ':', ' ' and ';'.

INI, JSON and YAML all have problems. None of them are serious enough for me to consider this for a small project with one or two config files required. Good to be able to point to a config file syntax done right though.

But the combination of Tom's reputation and the nightmare security situation currently unfolding with YAML means this project has a very high likelihood of becoming defacto standard in the Ruby community. The thought of being able to get rid of YAML from my code base entirely is an appealing end unto itself. Security aside, YAML is just ridiculously complex and error-prone.

Can you tell at a glance what is wrong with this YAML snippet:

    SE: Sweden
    NO: Norway
    FI: Finland

I'm not familiar enough with YAML to see the problem at a glance but parsing the snippet with an online parser [1] suggests that unless you put quotation marks around "NO" it gets parsed into a boolean False. Looks precarious, especially since not requiring strings to be quoted is kind of a selling point of YAML.

[1] https://yaml-online-parser.appspot.com/

That's it, but you cheated :)

I had to spend half an hour debugging a bizarre side effect of that parsing to pinpoint the problem.

That is utterly stupid isn't it, how could the designers of YAML count a capital "NO" as a key as False. Wow.

Edit, holy fuck all these are booleans:

y|Y|yes|Yes|YES|n|N|no|No|NO |true|True|TRUE|false|False|FALSE |on|On|ON|off|Off|OFF

Do tell about the problems with JSON for use as a configuration object? Almost every language has a library that can parse JSON.

It’s hard to author by hand. Ever added a comma after the last element in a list? It’s not very forgiving.

This is actually one of the worst things about it to me, other than the lack of comments. A language that has comma-separated lists and doesn't allow trailing commas on the list as a whole is frustrating to use in a version controlled context because adding an item requires modifying the item before, which leads to more (and less obviously solvable) merge conflicts.

I've seen a style of list initialization that sort of addresses this issue. I like it, but I don't see it used very often. Darn you standards!

    kittyBreeds = ["Himalayan"
                  ,"Cornish Rex" //rawr
    //This is not a real cat! >:(
    //              ,"lolcat"
In this way the issue with the commas is relegated to the first element, rather than the last. Since appending is more common, this seems a better style.

Interesting approach. This strikes me as kind of like the classic "put the constant on the left" advice in C/C++ that, no matter how many people agree it's a good idea, no one ever does because the cognitive effort of understanding it is high enough to diminish the correctness benefit.

In other words, you know you're just gonna end up with this in your code after a few months:

kittyBreeds = ["Himalayan", ,"Bengal", "Siamese", "Cornish Rex", ,"Manx"]

and not only is it now even worse for changing lines (correcting it and adding another will be -1+3 instead of just -1+2, the optimal +1 a distant memory), but it's just plain a mess and probably people don't know why they need to do it.

That's JavaScript not just JSON, if you install JSHint or JSLint in your editor you'll spot the error right away.

I like to put comments in my configuration files. How do you comment out blocks of a JSON configuration file?

Pipe it through JSmin or something else that strips comments.

You seem picky. :) But check out libconfig. Maybe it'll work for you.

I see your mixed use of square brackets and raise you a syntax error.

Yeah, that's gonna be a tough one.

But I think the "square bracket list" things only ever occur in value position, and the headers only ever occur in root position, so it's not that hard.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact