On 2015-11-28 18:29, BLM768 wrote:

> Hmm. Pretty, standardized, similar to JSON. I like it!
>
> ...but we don't have a parser yet, do we? That might be an obstacle.

Not as far as I know.

-- 
/Jacob Carlborg

On Friday, 27 November 2015 at 22:05:05 UTC, terchestor wrote:
> On Friday, 27 November 2015 at 20:16:20 UTC, Minas Mina wrote:
>> SDLang is fine. If someone wants to use D, it won't be SDLang that will stop him.
>>
>> Keep calm and use SDLang.
>
> No. STDLang is a terrible idea. Walter's arguments are right on.

He argued that {"comment":"blabla"} is a comment, and the sad thing is he's serious. That's a really shocking thing from someone like him!

And then he suggested we could just use a standard JSON parses, totally ignoring the reasons SDL was introduced in the first place, namely that comments can't be used in standard JSON, and it forces commas and braces all over the place while disallowing trailing commas.

The third argument I've seen from him boils down to that it's an uncommon format (although at first he claimed it was invented for DUB, which is clearly wrong). While this may be a legitimate concern, it is a very minor one IMO.

So, how can you conclude these arguments "are right on"? In fact, they're at best unimportant, but mostly non-sensical or based on wrong assumptions.

So we reached +200 replies...

I'm here wondering if you will keep going with all this drama about this config format, remembering that you can choose one if you don't like the other.

And all this drama while this group is small, I'm imagining what happens on a large group or language like the C++...

Bubba.

On Sunday, 29 November 2015 at 06:03:21 UTC, Poyeyo wrote:
> On Saturday, 28 November 2015 at 23:38:35 UTC, UserAbcabc123 wrote:
>> On Saturday, 28 November 2015 at 22:00:32 UTC, Poyeyo wrote:
>>> Dubconf seems to me a good replacement name for the format.
>>
>> Hilarous, I guess you didn't read the topic at all...
>
> I've read it all. But I'm not joining the group in arguing against the format. And yes, someone said .dubconf should be an analog for.

DUB people can't change the format name. They are not the SDL authors.
This is what I find hilarous because let's say 12 pages before there's been a misunderstanding caused by the fact that some people thought the format is propietary, thus DUB people have have been unfairly blamed...

The whole discussion is not about the file extention. The File extention is not a problem at all, the discussion was more about the format itself.

Again, it is not only about selecting config format. For me the main
problem is to have more than one format.
Dne 29. 11. 2015 16:51 napsal uživatel "bubbasaur via Digitalmars-d" <
digitalmars-d@puremagic.com>:

> So we reached +200 replies...
>
> I'm here wondering if you will keep going with all this drama about this config format, remembering that you can choose one if you don't like the other.
>
> And all this drama while this group is small, I'm imagining what happens on a large group or language like the C++...
>
> Bubba.
>

On Sunday, 29 November 2015 at 16:07:05 UTC, UserAbcabc123 wrote:
> On Sunday, 29 November 2015 at 06:03:21 UTC, Poyeyo wrote:
>> On Saturday, 28 November 2015 at 23:38:35 UTC, UserAbcabc123 wrote:
>>> On Saturday, 28 November 2015 at 22:00:32 UTC, Poyeyo wrote:
>>>> Dubconf seems to me a good replacement name for the format.
>>>
>>> Hilarous, I guess you didn't read the topic at all...
>>
>> I've read it all. But I'm not joining the group in arguing against the format. And yes, someone said .dubconf should be an analog for.
>
> DUB people can't change the format name. They are not the SDL authors.
> This is what I find hilarous because let's say 12 pages before there's been a misunderstanding caused by the fact that some people thought the format is propietary, thus DUB people have have been unfairly blamed...
>
> The whole discussion is not about the file extention. The File extention is not a problem at all, the discussion was more about the format itself.

By the way, I'm curious to see if A.D.Ruppe will summarize the whole topic in the "signifiant NG discussion" section in the next "this week in D" issue...

I you do so, good luck !

On Sunday, 29 November 2015 at 16:07:05 UTC, UserAbcabc123 wrote:
> On Sunday, 29 November 2015 at 06:03:21 UTC, Poyeyo wrote:
>> On Saturday, 28 November 2015 at 23:38:35 UTC, UserAbcabc123 wrote:
>>> On Saturday, 28 November 2015 at 22:00:32 UTC, Poyeyo wrote:
>>>> Dubconf seems to me a good replacement name for the format.
>>>
>>> Hilarous, I guess you didn't read the topic at all...
>>
>> I've read it all. But I'm not joining the group in arguing against the format. And yes, someone said .dubconf should be an analog for.
>
> DUB people can't change the format name. They are not the SDL authors.

Of course they can, it's called a fork. It's even logical to change the name if they choose to add features to it. Examples of forks: Iceweasel and Pale Moon are forks of Firefox.

> This is what I find hilarous because let's say 12 pages before there's been a misunderstanding caused by the fact that some people thought the format is propietary, thus DUB people have have been unfairly blamed...
>
> The whole discussion is not about the file extention. The File extention is not a problem at all, the discussion was more about the format itself.

Well, I'm not the only one to mention that the only really bad thing about sdlang is the name.

Let's go back to a general point of view.

As far as dependency managers go, the language used varies greatly:

C/C++ make/cmake/nmake -> here be dragons
perl CPANfile -> something perly
java maven -> xml
ruby gemfile -> ruby
python pip -> python egg
php composer -> json
node.js npm -> json
go godep -> json
rust cargo -> rust manifest
d dub -> json and sdlang

Looking at the two main D competitors, go uses json and rust uses a proprietary format.

Both languages seem to be used without issues in each project, may be the only issue for dub is supporting two formats instead of one.

Regards,

On Sunday, 29 November 2015 at 17:25:04 UTC, Poyeyo wrote:
> Looking at the two main D competitors, go uses json and rust uses a proprietary format.

Tangential to your point, but both Go and Rust suffers from "our policy is to be weird" in many ways that are "off-putting", and D also suffers from a rather broad stroke of not-invented-here.

Wouldn't it be refreshingly progressive if D just picked _one_ canonical format for _all metadata_ that isn't D code as a policy?

Something flexible like XML? That way you could wrap up all meta for an entire project and extract information from it using widespread standards like XPath and XQuery.

One problem with JSON is that it depends on meta information being relayed in HTTP headers. It is not a file format. Given a bunch of JSON files, you can't really tell what they are...

JSON is suitable as an alternative formulation, but not for a canonical format. Both JSON and YAML can be translated into corresponding XML too. Just provide tooling for JSON->XML and YAML->XML.

On Sun, 2015-11-29 at 17:25 +0000, Poyeyo via Digitalmars-d wrote:
> 
[…]
> C/C++ make/cmake/nmake -> here be dragons

Or SCons if you want to be cool. I guess Bazel (and maybe Tup) might
become trendy.

> perl CPANfile -> something perly
> java maven -> xml

Does anyone still use Maven – surely the world has moved to Gradle with it's Groovy scripts using the Gradle DSL.

> ruby gemfile -> ruby
> python pip -> python egg

Python has moved to wheels, eggs were always crap.

> php composer -> json
> node.js npm -> json
> go godep -> json

I am not sure this as as mainstream as this comment implies. Also a lot of people are using gb.

> rust cargo -> rust manifest

Which is TOML.

> d dub -> json and sdlang

Personally I abhor JSON for this kind of specification, it a transfer notation between computers, cf. XML. On the other hand I couldn't get SDL specs working. I will undoubtedly try again as SDL is just so much nicer than JSON for this. On the third hand lots of people seem addicted to JSON. On the fourth hand I cannot get worked up about this, it is just a build specification script which really ought to be written in D. cf. SBT for Scala uses Scala. Leiningen for Clojure uses Clojure. These languages have the right idea.

Oh, I just got worked up about this.

-- 
Russel. ============================================================================= Dr Russel Winder      t: +44 20 7585 2200   voip: sip:russel.winder@ekiga.net 41 Buckmaster Road    m: +44 7770 465 077   xmpp: russel@winder.org.uk London SW11 1EN, UK   w: www.russel.org.uk  skype: russel_winder

On Sunday, 29 November 2015 at 16:19:11 UTC, UserAbcabc123 wrote:
> By the way, I'm curious to see if A.D.Ruppe will summarize the whole topic in the "signifiant NG discussion" section in the next "this week in D" issue...

I'm catching up on the email for today then going to write it.

Actually, my summary of this will probably be just a couple sentences. I doubt much will actually change after this thread.