On Tue, 2015-12-01 at 17:24 +0000, Chris Wright via Digitalmars-d wrote:
> […]
> Bob's your uncle.

Is Uncle Bob your uncle?


PS Sorry for that, I am in a weird state just now.

-- 
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 11/25/2015 06:53 AM, Suliman wrote:
>> I find the SDLang format much cleaner to use than JSON
> But it's dead format! Nobody do not use it. JSON easy to read, there is
> a lot of it's checkers and formating tools.
>
> Yes, it's not perfect, but now it's _standard_. Personally I'd prefer
> yaml, because it's much easier to read for humans.
>
> But what we will do with SDL? Who know how to parse, validate it with D,
> and with another language? Even ini is better, because everybody know it.
>

This whole debate is completely moronic.

1. With DUB, which format is "default" means next to nothing.

2. I don't know where in the world you've gotten the idea you can no longer just copy-paste deps. That's patently BS.

3. SDLang is fucking trivial. Any programmer worth at least half their salt (ie anyone here, including you) could've learned it in same time you've already spent bellyaching about it.

4. Fuck "standard/popular/common". Seriously, fuck it. That sort of bullshit nonsense attitude is EXACTLY why half our industry is as completely fucked as it is with complete and total shit like PHP, JS, Node, JVM, Angular, JS Toolkit #five-billion-and-one, gaudy metro colors, meaningless hieroglyphs, walled gardens, web pages with near-zero content just empty space and screen-sized images with a quick slogan or two, text entry boxes that are *literally* slower than a goddamn Apple II, etc, etc etc... Seriously, enough of this goddamn "fashion before engineering" bullshit. That load of crap is why I'm right on the verge of completely jumping ship from what's left of this goddamn industry. If we start pulling that shit as a matter of course here too, I'm fucking gone, good riddance.

On Wednesday, 2 December 2015 at 16:15:04 UTC, Nick Sabalausky wrote:
> On 11/25/2015 06:53 AM, Suliman wrote:
>>> I find the SDLang format much cleaner to use than JSON
>> But it's dead format! Nobody do not use it. JSON easy to read, there is
>> a lot of it's checkers and formating tools.
>>
>> Yes, it's not perfect, but now it's _standard_. Personally I'd prefer
>> yaml, because it's much easier to read for humans.
>>
>> But what we will do with SDL? Who know how to parse, validate it with D,
>> and with another language? Even ini is better, because everybody know it.
>>
>
> This whole debate is completely moronic.
>
> 1. With DUB, which format is "default" means next to nothing.
>
> 2. I don't know where in the world you've gotten the idea you can no longer just copy-paste deps. That's patently BS.
>
> 3. SDLang is fucking trivial. Any programmer worth at least half their salt (ie anyone here, including you) could've learned it in same time you've already spent bellyaching about it.
>
> 4. Fuck "standard/popular/common". Seriously, fuck it. That sort of bullshit nonsense attitude is EXACTLY why half our industry is as completely fucked as it is with complete and total shit like PHP, JS, Node, JVM, Angular, JS Toolkit #five-billion-and-one, gaudy metro colors, meaningless hieroglyphs, walled gardens, web pages with near-zero content just empty space and screen-sized images with a quick slogan or two, text entry boxes that are *literally* slower than a goddamn Apple II, etc, etc etc... Seriously, enough of this goddamn "fashion before engineering" bullshit. That load of crap is why I'm right on the verge of completely jumping ship from what's left of this goddamn industry. If we start pulling that shit as a matter of course here too, I'm fucking gone, good riddance.

The issue is not with humans reading and writing SDLang files - like you said, the syntax is not hard, and besides - the default should be enough for most of the basic learning projects one can make, so by the time you actually need to edit dub.sdl you should know enough D to not be learning two things at once.

No - the problem is with making tools(IDEs/editor plugins/scripts) parse and emit it. Yes, computers can read&write SDLang files - but you need to find a library(or write one yourself) that fits your language of choice.


I, for example, have a Ruby script that creates small DUB projects for me when I want to test stuff. Since usually I want to interact with small parts of the huge application, I need to do some specific configuration on the DUB project file so I have my Ruby script load it with it's JSON module, perform the necessary changes, and save over the original file.

Back then, when DUB moved to SDLang, obviously my script crashed because `dub init` now creates SDLang config files by default. This can be changed, but I thought I would go with the flow and modify my script to edit the SDLang file. I'm always eager to test new technologies(not adapt - test. I want to experience them before I decide whether to adapt them or not).

So, the first step was to google for a Ruby Gem that deals with SDLang. Should be easy, right? Wrong! Neither "Ruby SDL" nor "Ruby SDLang" yielded any relevant results(at least not in the first pages of google), and only recently, when this thread started and I decided to attempt again, I figured out you need to search for "Ruby Simple Declarative Language". Or to use rubygems.org's search engine. But that doesn't matter - back then I couldn't find it so instead just set my script to run `dub init` with `--format=json`, and now I no longer feel motivated to convert my script to use SDLang...


If you try to push a new technology, and when people complain about problems of that technology your reply is that they can simply use the better alternative instead - well, if you think this is good for the ecosystem let me tell you about a project called Tango...

On Wednesday, 2 December 2015 at 20:45:33 UTC, Idan Arye wrote:
> On Wednesday, 2 December 2015 at 16:15:04 UTC, Nick Sabalausky wrote:
>> On 11/25/2015 06:53 AM, Suliman wrote:
>>>> I find the SDLang format much cleaner to use than JSON
>>> But it's dead format! Nobody do not use it. JSON easy to read, there is
>>> a lot of it's checkers and formating tools.
>>>
>>> Yes, it's not perfect, but now it's _standard_. Personally I'd prefer
>>> yaml, because it's much easier to read for humans.
>>>
>>> But what we will do with SDL? Who know how to parse, validate it with D,
>>> and with another language? Even ini is better, because everybody know it.
>>>
>>
>> This whole debate is completely moronic.
>>
>> 1. With DUB, which format is "default" means next to nothing.
>>
>> 2. I don't know where in the world you've gotten the idea you can no longer just copy-paste deps. That's patently BS.
>>
>> 3. SDLang is fucking trivial. Any programmer worth at least half their salt (ie anyone here, including you) could've learned it in same time you've already spent bellyaching about it.
>>
>> 4. Fuck "standard/popular/common". Seriously, fuck it. That sort of bullshit nonsense attitude is EXACTLY why half our industry is as completely fucked as it is with complete and total shit like PHP, JS, Node, JVM, Angular, JS Toolkit #five-billion-and-one, gaudy metro colors, meaningless hieroglyphs, walled gardens, web pages with near-zero content just empty space and screen-sized images with a quick slogan or two, text entry boxes that are *literally* slower than a goddamn Apple II, etc, etc etc... Seriously, enough of this goddamn "fashion before engineering" bullshit. That load of crap is why I'm right on the verge of completely jumping ship from what's left of this goddamn industry. If we start pulling that shit as a matter of course here too, I'm fucking gone, good riddance.
>
> The issue is not with humans reading and writing SDLang files - like you said, the syntax is not hard, and besides - the default should be enough for most of the basic learning projects one can make, so by the time you actually need to edit dub.sdl you should know enough D to not be learning two things at once.
>
> No - the problem is with making tools(IDEs/editor plugins/scripts) parse and emit it. Yes, computers can read&write SDLang files - but you need to find a library(or write one yourself) that fits your language of choice.
>
>
> I, for example, have a Ruby script that creates small DUB projects for me when I want to test stuff. Since usually I want to interact with small parts of the huge application, I need to do some specific configuration on the DUB project file so I have my Ruby script load it with it's JSON module, perform the necessary changes, and save over the original file.
>
> Back then, when DUB moved to SDLang, obviously my script crashed because `dub init` now creates SDLang config files by default. This can be changed, but I thought I would go with the flow and modify my script to edit the SDLang file. I'm always eager to test new technologies(not adapt - test. I want to experience them before I decide whether to adapt them or not).
>
> So, the first step was to google for a Ruby Gem that deals with SDLang. Should be easy, right? Wrong! Neither "Ruby SDL" nor "Ruby SDLang" yielded any relevant results(at least not in the first pages of google), and only recently, when this thread started and I decided to attempt again, I figured out you need to search for "Ruby Simple Declarative Language". Or to use rubygems.org's search engine. But that doesn't matter - back then I couldn't find it so instead just set my script to run `dub init` with `--format=json`, and now I no longer feel motivated to convert my script to use SDLang...
>
>
> If you try to push a new technology, and when people complain about problems of that technology your reply is that they can simply use the better alternative instead - well, if you think this is good for the ecosystem let me tell you about a project called Tango...

I am most worried that Sonke refuses to respond to very reasonable arguments like this. He seems to dismiss the entire discussion as a troll. Clearly that is in part true but we have a right to hear his response on the valid points. And no, sorry but just because he has already done the work we won't suck it. At least please revert the default back to json.

Sonke, please reply.

On Wednesday, 2 December 2015 at 20:45:33 UTC, Idan Arye wrote:
> On Wednesday, 2 December 2015 at 16:15:04 UTC, Nick Sabalausky wrote:
>> [...]
>
> The issue is not with humans reading and writing SDLang files - like you said, the syntax is not hard, and besides - the default should be enough for most of the basic learning projects one can make, so by the time you actually need to edit dub.sdl you should know enough D to not be learning two things at once.
>
> [...]

Where you looking for this:

https://github.com/ikayzo/SDL.rb

On Wednesday, 2 December 2015 at 22:57:31 UTC, CraigDillabaugh wrote:
> On Wednesday, 2 December 2015 at 20:45:33 UTC, Idan Arye wrote:
>> On Wednesday, 2 December 2015 at 16:15:04 UTC, Nick Sabalausky wrote:
>>> [...]
>>
>> The issue is not with humans reading and writing SDLang files - like you said, the syntax is not hard, and besides - the default should be enough for most of the basic learning projects one can make, so by the time you actually need to edit dub.sdl you should know enough D to not be learning two things at once.
>>
>> [...]
>
> Where you looking for this:
>
> https://github.com/ikayzo/SDL.rb

Yes, and eventually I found it - when I searched with "Ruby Simple Declarative Language". My point was not that it doesn't exist, but that it much harder than it should have been to search for it.


BTW - I tried, just to see what I get, to search for a Python implementation:

"Python SDL", as expected, yields only results related to Simple DirectMedia Layer.

"Python Simple Declarative Language" the only related thing I find ikayzo's github page(which contains SDLang implementations for Java, .NET and Ruby - but not for Python).


And here comes the fun part:

"Python SDLang" does not find the SDLang implementation for Python - at least not on the first page.

But - the first 3 results are about an SDL implementation...

... it's SDLang-D!

Yes, you got that right - I searched for something related to Python(!!!) and got a result for D. So yea, maybe SDLang wasn't created specifically for DUB, but it might as well have been. Either that, or D suddenly became more popular than Python. I'll let you judge which of these two alternatives is more probable.

On 12/02/2015 07:29 PM, Idan Arye wrote:
> On Wednesday, 2 December 2015 at 22:57:31 UTC, CraigDillabaugh wrote:
>> On Wednesday, 2 December 2015 at 20:45:33 UTC, Idan Arye wrote:
>>> On Wednesday, 2 December 2015 at 16:15:04 UTC, Nick Sabalausky wrote:
>>>> [...]
>>>
>>> The issue is not with humans reading and writing SDLang files - like you said, the syntax is not hard, and besides - the default should be enough for most of the basic learning projects one can make, so by the time you actually need to edit dub.sdl you should know enough D to not be learning two things at once.
>>>
>>> [...]
>>
>> Where you looking for this:
>>
>> https://github.com/ikayzo/SDL.rb
> 
> Yes, and eventually I found it - when I searched with "Ruby Simple Declarative Language". My point was not that it doesn't exist, but that it much harder than it should have been to search for it.
> 
> 
> BTW - I tried, just to see what I get, to search for a Python implementation:
> 
> "Python SDL", as expected, yields only results related to Simple DirectMedia Layer.
> 
> "Python Simple Declarative Language" the only related thing I find ikayzo's github page(which contains SDLang implementations for Java, .NET and Ruby - but not for Python).
> 
> 
> And here comes the fun part:
> 
> "Python SDLang" does not find the SDLang implementation for Python - at least not on the first page.
> 
> But - the first 3 results are about an SDL implementation...
> 
> ... it's SDLang-D!
> 
> Yes, you got that right - I searched for something related to Python(!!!) and got a result for D. So yea, maybe SDLang wasn't created specifically for DUB, but it might as well have been. Either that, or D suddenly became more popular than Python. I'll let you judge which of these two alternatives is more probable.

The fact that sdl shares its name with the Simple DirectMedia Language doesn't really help matters...

-- 
Matt Soucy
http://msoucy.me/

On 12/01/2015 07:22 AM, Ola Fosheim Grøstad wrote:
> On Tuesday, 1 December 2015 at 12:07:51 UTC, Russel Winder wrote:
>> I think you are meaning Bazel here. http://bazel.io/
>>
>> I haven't had chance to play with it as yet, and it changes massively every day – though I suspect it is the internal of the satisfaction engine that change not the specification notation (which looks a bit like a Python/SCons/Waf type thing). I will be playing with it over the next few weeks, so more news later.
> 
> 
> That's interesting, I didn't know they were going to make it publicly available. The FAQ says it supports Java, Objective-C and C++ out of the box. So I guess that means there is infrastructure for adding other languages too.
> 
> 
> 

Interestingly, bazel has D mentioned in its docs:

http://bazel.io/docs/be/d.html

-- 
Matt Soucy
http://msoucy.me/

On 11/30/2015 01:07 PM, Luis wrote:
> name "dedcpu"
> authors "Luis Panadero Guardeño"
> targetType "none"
> license "BSD 3-clause"
> description "DCPU-16 tools"
> 
> subPackage {
>    name "lem1802"
>    description "Visual LEM1802 font editor"
>    targetType "executable"
>    targetName "lem1802"
>    excludedSourceFiles "src/bconv.d"
>    excludedSourceFiles "src/ddis.d"
>    libs "gtkd" platform="windows"
> 
>    configuration "nogtk" {
>       platforms "windows"
>    }
> 
>    configuration "gtk" {
>       platforms "posix"
>       dependency "gtk-d:gtkd" version="~>3.2.0"
>    }
> }

name: "dedcpu"
authors: "Luis Panadero Guardeño"
targetType: "none"
license: "BSD 3-clause"
description: "DCPU-16 tools"

subPackages:
  - name: "lem1802"
    description: "Visual LEM1802 font editor"
    targetType: "executable"
    targetName: "lem1802"
    excludedSourceFiles:
      - "src/bconv.d"
      - "src/ddis.d"
    libs:
      "gtkd": {"platform": "windows"}
    configurations:
      - name: "nogtk"
        platforms: ["windows"]
      - name: "gtk"
        platforms: ["posix"]
        dependencies:
          "gtk-d:gtkd": {"version": "~>3.2.0"}

One important thing to note is that any valid JSON object is also valid YAML. This would have allowed a much cleaner upgrade path.

I have to argue this point nearly every semester - the courseware for certain courses here (disclaimer: I'm one of the maintainers) is open source and relies on students submitting assignment links into a YAML file.

-- 
Matt Soucy
http://msoucy.me/

On Thursday, 3 December 2015 at 01:40:12 UTC, Matt Soucy wrote:
> Interestingly, bazel has D mentioned in its docs:
>
> http://bazel.io/docs/be/d.html

Hah, yes! And Dub too, but they link to the json page...