Re: DragonEgg has been revived

[Lucretia <laguest9000@googlemail.com>]

On Friday, 25 May 2018 14:16:55 UTC+1, Simon Clubley  wrote:
> On 2018-05-24, Luke A  Guest <> wrote:
> > Chris M Moore <m> wrote:
> >
> >> 
> >> I mean who uses "overtly clearly"?
> >> 
> >
> > It?s like reading posts from a spambot, isn?t it? My eyes glaze and I just
> > tl;rcfbb
> 
> I think I prefer the Sky59 person. (And no, that's not an invitation
> for him to start posting again.)

Oh god no! ;P
 
> I don't care if the compiler itself is GPL.
> 
> I _do_ care if the RTS or anything else is GPL (or even LGPL) and as
> a result there are constraints imposed on my binaries or source code
> when I use the compiler. I want to be able to compile programs using
> the compiler without having to do anything else other than ship the
> binary for my program.

Major issue I agree with.
 
> This is why I talk about generated code and not the compiler.

Yes, because this is the issue with the GPL and GNAT/GCC as they stand. It is possible to have compilers which are not encumbered in this way. One of the things I want to do is to generate SPIR-V from Ada, that would be cool AF!
 
> The FSF runtime exception is fine for me.

For now.
 
> I would like to be able to use Ada in all the places I can currently
> use C and C++ code, including bare metal targets.

Yes, including mobile.
 
> If I start using Ada in those places, I would be like to sure that
> I can still build Ada code for those targets in 2-5 years using the
> current toolchains of the day, just like I can with C and C++ code.

A clear branching/tagging on the compiler sources where you can grab an earlier version if you need to, then fix up later versions if they fail for some reason.
 
> It would be nice if it was as easy to port a compiler toolchain to
> a new OS or architecture as it is to port a RTOS to a new target.
> This would be one answer to the lack of targets for an Ada compiler.

This has to be a major factor.

My uni thesis was supposed to be a Java compiler, but my "instructor" at the time wanted me to use GCC, I spent so much time just trying to work out GCC I didn't get it done. If I'd just done it with Flex/Bison, it would've been done. GCC is way too big to handle for such a short timeframe, IMO.
 
> For those of you who have not done RTOS based development, an RTOS is

Haven't, intend to.

> typically very cleanly divided internally into libraries of generic
> 
> It would be nice if an Ada compiler was also that clean internally
> and as well documented so that you could easily port it to a new OS
> or environment yourself if you needed.

The documentation needs to be external to the source, so you don't have to go around the sources looking for stuff. This documentation can be generated from the source though.

I agree with everything you said, apart from the sky invite &;P

Luke.