Wednesday, March 1, 2017

Digest for comp.lang.c++@googlegroups.com - 11 updates in 4 topics

Ian Collins <ian-news@hotmail.com>: Mar 02 08:04AM +1300

On 03/ 1/17 11:11 PM, Hergen Lehmann wrote:
> in any list. The last one i remember off the cuff were several missing
> conversion specifiers in strftime, which are supposed to be available
> since C++11.
 
The lack of C99 support is the compiler's main weakness. This goes back
to Microsoft's unfortunate disregard for C and is probably due to the
lack of C99 support in the system libraries.
 
A C++ beginner probably won't encounter too many of these and the
quality of the IDE (modulo it's awful build system) is still sufficient
to justify a recommendation.
 
VS has one huge advantage over the likes of Eclipse or NetBeans: it
isn't written in (spit) Java :)
 
>> (our code is built with clang on Linux and VS on Windows).
 
> I'm doing this, too. And still need several "#ifdef _MSC_VER" here and
> there... :-(
 
Here to, mainly for silencing warnings...
 
--
Ian
David Brown <david.brown@hesbynett.no>: Mar 01 09:09PM +0100

On 01/03/17 18:56, Manfred wrote:
> Specifically about IDEs, I mostly use Eclipse for regular projects.
> For tiny projects I would mention Komodo has a reasonably usable editor
> too.
 
The OP posted from Linux, so it's a fair bet he/she is going to be using
Linux for programming.
 
Yes, gcc is the obvious choice of compiler - clang and icc might be fun
to play with, but gcc is the norm and is probably the best in terms of
solid code generation, top-class standards conformance, static warnings,
and extra extensions for those that like that sort of thing (they can
improve your code, at the cost of reducing portability).
 
As for IDEs, I agree that Eclipse is the "big" one - it has a lot of
functionality, though it is quite demanding on resources. Netbeans is
another choice that comes recommended.
Manfred <mx2927@gmail.com>: Mar 01 10:31PM +0100

On 03/01/2017 09:09 PM, David Brown wrote:
>> Specifically about IDEs, I mostly use Eclipse for regular projects.
>> For tiny projects I would mention Komodo has a reasonably usable editor
>> too.
 
<snip>
> As for IDEs, I agree that Eclipse is the "big" one - it has a lot of
> functionality, though it is quite demanding on resources. Netbeans is
> another choice that comes recommended.
Tried Netbeans as well, but IME, although both are coded in Java, it was
significantly more demanding than Eclipse.
(the other I mentioned, komodo, even if not comparable in functionality,
is coded in C++ and does not suffer from these issues)
Vir Campestris <vir.campestris@invalid.invalid>: Mar 01 09:33PM

On 01/03/2017 07:48, Marcel Mueller wrote:
> But have a look at Eclipse CDT. It is free and very powerful. One of the
> benefits is that it supports several languages, so it is easier to
> switch if required.
 
I use Eclipse - but for me it isn't an IDE, just a good editor. I'm
building Android kernel stuff (mostly C :( ) and the build system isn't
compatible with Eclipse - or AFAIK any other IDE.
 
I miss Visual Studio.
 
Andy
red floyd <dont.bother@its.invalid>: Mar 01 01:36PM -0800

On 3/1/2017 1:31 PM, Manfred wrote:
> significantly more demanding than Eclipse.
> (the other I mentioned, komodo, even if not comparable in functionality,
> is coded in C++ and does not suffer from these issues)
 
Anyone use Code::Blocks (or whatever its successor is)?
Paavo Helde <myfirstname@osa.pri.ee>: Mar 01 11:55PM +0200

On 1.03.2017 23:36, red floyd wrote:
>> (the other I mentioned, komodo, even if not comparable in functionality,
>> is coded in C++ and does not suffer from these issues)
 
> Anyone use Code::Blocks (or whatever its successor is)?
 
Unfortunately yes. Most of our C++ development happens on Windows so it
is a bit unclear why our Linux builds are using C::B. Anyway, it crashes
about half of times when doing automated builds, and typically freezes
when I try to debug with it. So whenever there are problems in our Linux
build I use Emacs and gdb instead. C::B's only useful feature was to
find the first g++ error in the pile of warnings, but lately we have
succeeded in cutting down the amount of warnings so it's not so
important any more.
 
We are planning to switch back to plain make, but lacking manpower to
make changes.
Bo Persson <bop@gmb.dk>: Mar 01 11:34PM +0100

On 2017-03-01 14:42, Scott Lurndal wrote:
 
> Unlike a graphics-heavy IDE, one can use the above tools to develop
> software remotely (e.g. over secure shell and low-speed network
> connection) if required.
 
Some of us have our own PC and don't need remote shell access.
Especially not newbies.
 
 
 
Bo Persson
ruben safir <ruben@mrbrklyn.com>: Mar 01 05:32PM -0500


> Brian
 
> "But the path of the righteous is like the light of dawn,
> That shines brighter and brighter until the full day." Proverbs 4:18
 
fuck you and fuck your christian bullshit
"Chris M. Thomasson" <invalid@invalid.invalid>: Mar 01 02:04PM -0800

On 2/25/2017 7:46 PM, Chris M. Thomasson wrote:
 
> You can use the original CAS based version by Dmitry Vyukov.
 
> Or, my fast-pathed wait-free alteration:
 
> <pseudo code, membars aside for a moment>
 
 
Here are the membars required wrt the pseudo code of the
multi-producer/consumer atomic bounded queue:
> cell cells[N]; // N must be a power of 2
 
> void init() {
> for (uint32_t i = 0; i < N; ++i) cells[i].ver = i;
 
// RELEASE (depending on how this was created! dynamic, and/or before
threads?)
 
> uint32_t ver = XADD(&head, 1);
> cell& c = cells[ver & (N - 1)];
> while (LOAD(&c.ver) != ver) backoff();
// ACQUIRE
 
> c.state = state;
// RELEASE
 
> uint32_t ver = XADD(&tail, 1);
> cell& c = cells[ver & (N - 1)];
> while (LOAD(&c.ver) != ver + 1) backoff();
// ACQUIRE
 
> double state = c.state;
// RELEASE
 
> return state;
> }
> ______________________________________________
 
 
This is the basic acquire/release model.
"Chris M. Thomasson" <invalid@invalid.invalid>: Mar 01 02:07PM -0800

On 3/1/2017 2:04 PM, Chris M. Thomasson wrote:
 
> // RELEASE (depending on how this was created! dynamic, and/or before
> threads?)
 
>> }
 
Ummm, that before should read as:
 
before/after
Vir Campestris <vir.campestris@invalid.invalid>: Mar 01 09:44PM

On 28/02/2017 10:58, Alf P. Steinbach wrote:
> In the Expressive C++ announce posting I wrote that "It implements a
> less unsafe, more convenient and readable-for-non-C++-expert-folks
> dialect of C++".
Alf, I like your posts, and I respect your advice. But I found your
Expressive code impossible to read. It requires knowledge of all sorts
of things that I don't know, and I don't have time to read.
 
It's not C++.
 
I come here to keep my hand in, when my current job is 90% C. I want to
read about C++.
 
<snip>
> So, I've expended some effort on making the code shorter, that is,
> terseness, which therefore seems to be a benefit to me, one that comes
> in addition to the safety, convenience and readability benefits
 
Be careful. Remember APL...
 
Andy
You received this digest because you're subscribed to updates for this group. You can change your settings on the group membership page.
To unsubscribe from this group and stop receiving emails from it send an email to comp.lang.c+++unsubscribe@googlegroups.com.
Posted by at

No comments:

Post a Comment

Subscribe to: Post Comments (Atom)