On Thursday, February 01, 2018 02:42:52 Adam D. Ruppe via Digitalmars-d wrote:
> On Thursday, 1 February 2018 at 02:24:44 UTC, H. S. Teoh wrote:
> > I thought that *is* the color support that was added?  If you're expecting IDE-style syntax highlighting, I think you're setting your expectations a little high for something that ostensibly was banged out in a couple of hours.
>
> Nope, the dmd compiler will highlight code between `` in its outputted error messages. dmd already knows how to lex D code (obviously) and even has a highlighter for ddoc, so it just pipes it through that existing function.
>
> Not all error messages had the `` around its code though. That's what Mike did in most his PRs.

I haven't see any syntax highlighting in any error messages. The closest I've seen is a type name or symbol name being in white instead of grey when it's mentioned explicitly by name outside of code snippets, but no code snippets have been anything but grey (which is the default text color for the console). So, either none of the errors I've gotten are marked up with `` correctly, or there's something with my setup which makes it so that the coloring doesn't work.

- Jonathan M Davis

On 1/31/2018 5:58 PM, H. S. Teoh wrote:
> cosmetic features.

I tough lesson I've learned is that cosmetics matter, a lot. Sometimes much more than substance. There's no getting away from it.

On 1/31/2018 6:14 PM, Jonathan M Davis wrote:
> I have to wonder if my settings are right. I've never noticed any color in
> error messages. Messing around with some errors right now, the only color I
> see is that "Error:" is in red, and some of the text is bolded, so it's
> white instead of the grey that text is normally on my console. Maybe my
> console's settings aren't interacting with the color stuff very well.

Nah, it's just the code in dmd:

   if (username == "Jonathan")
       color = off;    // muwa-ha-ha-hah!

On 1/31/2018 3:38 PM, H. S. Teoh wrote:
> A "small fry" like myself wouldn't dare
> to push the merge button on changes of this kind of magnitude, since it
> could have drastic consequences that I can't foresee due to not having a
> full grasp of the full scale of what is being changed.
https://youtu.be/amclN9RG49c?t=146

On Wednesday, January 31, 2018 19:01:46 Walter Bright via Digitalmars-d wrote:
> On 1/31/2018 6:14 PM, Jonathan M Davis wrote:
> > I have to wonder if my settings are right. I've never noticed any color in error messages. Messing around with some errors right now, the only color I see is that "Error:" is in red, and some of the text is bolded, so it's white instead of the grey that text is normally on my console. Maybe my console's settings aren't interacting with the color stuff very well.
> Nah, it's just the code in dmd:
>
>     if (username == "Jonathan")
>         color = off;    // muwa-ha-ha-hah!

Clearly, there are downsides to actually knowing the people who work on the compiler. ;)

- Jonathan M Davis

On Wednesday, 31 January 2018 at 21:19:50 UTC, H. S. Teoh wrote:
> On Wed, Jan 31, 2018 at 08:05:37PM +0000, Adam D. Ruppe via Digitalmars-d wrote:
>> On Wednesday, 31 January 2018 at 19:59:23 UTC, Seb wrote:
>> > ... and Mike did put _a lot_ of effort in pushing colorful error messages:
>> 
>> Yes, that was a direct result of that forum post I alluded to. It isn't something he (or most anyone else, in my experience) really cares about, but he felt if he did it then many Walter wouldn't waste any more time on it.
>
> In fact, I find the colors distracting.  Pretty? Certainly.  But helpful?  Not really, at least not for me.  I'd rather stick with just B&W. But all things considered, it's a pretty minor issue, and one that's not worth the time to complain about.  (Haha, and look what I just did. :-P)
>
>
> T

BTW run.dlang.io supports DMD's colorized error messages since a while too, e.g.

https://run.dlang.io/is/neLaOB

If anyone minds, send a PR to https://github.com/dlang-tour/core/blob/master/public/static/css/ansi.css

On Wed, 2018-01-31 at 16:13 +0000, John Gabriele via Digitalmars-d wrote:
> […]
> this older language from times past, before C++11, and using ddoc
> for docs instead of markdown contributes to this perception. Let
> me know if you'd like help in translating D website and doc pages
> from ddoc to markdown.

I am sure Markdown is find for single page HTML pages, but for bigger documents that need to render to HTML or PDF (or other e-publishing formats) surely AsciiDoctor and XeLaTeX are the only choices.

-- 
Russel.
===========================================
Dr Russel Winder      t: +44 20 7585 2200
41 Buckmaster Road    m: +44 7770 465 077
London SW11 1EN, UK   w: www.russel.org.uk

On Wednesday, 31 January 2018 at 18:35:50 UTC, Seb wrote:
[...]
>> Like: https://github.com/adamdruppe/arsd/blob/master/simpledisplay.d
>> And the examples from D-Lang Tour.
>>
>> So you only push a button [try D], and get a running environment to play around.
>
> Like this?
>
> https://tour.dlang.org/tour/en/dub/mir
>
> It's a small series since today. Any help with filling the blank content or new pages is welcome.
>
> See https://forum.dlang.org/post/acovehcwaxjykmhekuwh@forum.dlang.org for adding new libraries to run.dlang.io

This looks very promising!
Have not been on the dang tour page for several month, it shows an amazing progress!!

Idea: There should be some kind of news ticker for all enhancements and important decisions, maybe at first just via twitter  with a special #tag  beside #dlang where all updates are announced. And a place on the homepage, where this feed is displayed separately.

On the other side a voting mechanism in the forum would be very useful, so readers can mark a post as valuable maybe to be be displayed in a special feed. I know this is difficult because the forum has more than only the web frontend.
But why not translate the [I recommend to read this]-Button into a mail or post, with the content ... **recommend** this. And probably back? from short mail only with **KEYWORD**.
It would fill up the mailboxes of the readers, but it would be easier to count than this "me too ++" posts.

On Wed, 2018-01-31 at 13:14 +0000, Michael via Digitalmars-d wrote:
> 
[…]
> I agree that marketing is a pretty serious problem for D. Many people just aren't aware of it. Even for people who do not […]

Some suggestions:

More people putting more good projects using D on GitLab, BitBucket, Launchpad, and GitHub.

People writing more articles using D in publications such as CVu and Overload, etc.

People putting more sessions into conferences such as ACCU using D – there hasn't been a single D-related session since Andrei's keynote at ACCU 2016.

My own small contribution is to answer all the ACCU C++ Code Critiques with answers that say do not use C++, but use Python, D, Go, Rust, so as to get more non-C++ into C++ programmers minds. Clearly hardened C++ folk look only to C++ and never consider any language other than assembly language, but a lot of currently C++ folk are not hardened to the C++-only mindset. And ACCU is about programming with an emphasis on C++. This is clearly a great place for D content.

D-only conferences, as (C++|Go|Rust)-only ones, have their place, but their main role is to reinforce the community. Marketing requires people to take the message outside the community. ACCU conferences and journals provide an excellent place to market D to people outside the D community.

This year ACCU has some good Rust content. I wanted some Go content but it seems they are as reluctant to come to a polyglto conference as D people seem to be. :-(

-- 
Russel.
===========================================
Dr Russel Winder      t: +44 20 7585 2200
41 Buckmaster Road    m: +44 7770 465 077
London SW11 1EN, UK   w: www.russel.org.uk

On Wed, 2018-01-31 at 13:54 +0000, Jack Stouffer via Digitalmars-d wrote:
> On Wednesday, 31 January 2018 at 07:56:37 UTC, Andrew Benton wrote:
> > E.g. three compilers
> 
> Every other compiled language (and a lot of scripting ones) uses the fact of multiple compilers for the language as a sign of adoption and ecosystem growth.

Definitely.

> I've only ever seen people complain about D in this area. Never once have I seen someone argue that the existence of PyPy hurts Python or gogcc hurts Go.

D should be proud of having a reference compiler, a GCC-based one, and an LLVM-based one. This is a Good Thing™, let no-one undermine this.

-- 
Russel.
===========================================
Dr Russel Winder      t: +44 20 7585 2200
41 Buckmaster Road    m: +44 7770 465 077
London SW11 1EN, UK   w: www.russel.org.uk