Porting to C++

For code related discussions and questions

Re: A question.

Postby bornemix » 13 Mar 2010, 16:25

I agree that it needs to be readable. (that's not my point)

My comment was addressing seismo's leaving contributing to the community because other don't refactor the code.

In my opinion; a good programmer should be able to either read the code, or rewrite it from scratch... both without going: :gonk: :scream:

So, no offense. I agree with your statements but wanted to address something different, that's all.
User avatar
bornemix
Code contributor
Code contributor
 
Posts: 56
Joined: 09 Mar 2010, 08:39

Re: A question.

Postby dmkp » 13 Mar 2010, 19:44

Oh right okay nevermind then :P
dmkp
Trained
Trained
 
Posts: 182
Joined: 08 Jan 2010, 19:50
Location: UK

Re: A question.

Postby Buginator » 13 Mar 2010, 19:46

bornemix wrote:I agree that it needs to be readable. (that's not my point)

My comment was addressing seismo's leaving contributing to the community because other don't refactor the code.

In my opinion; a good programmer should be able to either read the code, or rewrite it from scratch... both without going: :gonk: :scream:

So, no offense. I agree with your statements but wanted to address something different, that's all.

If only it was as easy as you make it seem.
The codebase was written by professional programmers, who got paid, they made the game, and moved on.
We don't have any of the info / resources that they did, it is all trial & error.
Yes, the code needs to be rewritten in allot of places, but, this codebase is highly volatile.

We do accept patches, and if they make sense, they do get applied.
If the patches would cause tons of merge conflicts, then we expect patches to fix those issues. If the patches cause allot of noise in svn's history without having any real benefit, then we tend not to apply those patches, unless they fix something important.

Make sense?
and it ends here.
User avatar
Buginator
Professional
Professional
 
Posts: 3285
Joined: 04 Nov 2007, 01:20

Re: A question.

Postby bornemix » 13 Mar 2010, 22:23

Hi Bug,

I'd say that what you described is how professional programmers do as well.

They make implementations if it hurts not to do them, and they choose not to do implementations that themselves hurt.

I'll stop talking now I think, I might come off like a total know-it-all.
User avatar
bornemix
Code contributor
Code contributor
 
Posts: 56
Joined: 09 Mar 2010, 08:39

Re: A question.

Postby Seismo » 13 Mar 2010, 23:59

Zarel wrote:...I'm afraid I don't understand your English. Can you speak in your native language?

May you tell me which part(s) you did not understand. However, i should rewrite it with shorter and a whole lot simpler sentences.

Zarel wrote:I'll try to guess at what you're trying to say:
- What issues are you talking about? I asked Buginator, and he says we've committed several of your patches.

So not all was to complicated? There are still enough left and some new warnings find a way in current trunk as well.

Zarel wrote:- What sort of ignorance are you encountering?

You may find it hardly worth mentioning, so i do not try to recapitulate it all. There were several nonsens disscussion belonging to complete different requirements. For e. g. mingw<->msvc<->linux comparisons.

Zarel wrote:Things about Qt, Linux, and branching are important. We are not going to commit a patch if it happens to break compatibility with Linux or something...

Well, thats not the thing i was talking about in detail. May we drop that here due to the fact that i am seriously not interested to help you anymore. I am sick of this pointy-headed discussions. If you are so cute, well, it must be my mistake.
Seismo
Trained
Trained
 
Posts: 70
Joined: 06 Feb 2010, 16:32

Re: A question.

Postby Zarel » 14 Mar 2010, 11:41

Seismo wrote:May you tell me which part(s) you did not understand.

Okay, the problem is, you're spending long paragraphs saying that our developers have done something wrong, without telling us what exactly has been done wrong.

Now, you say that you are no longer willing to help us. We are sorry such a situation has occurred. However, we'd greatly appreciate if you would at least explain what went wrong, so we know how to improve in the future.

dmkp wrote:I take it you don't do any programming then from your obvious ignorance shown here xD

I should point out that bournemix has submitted useful patches to the Warzone project. Have you?
User avatar
Zarel
Elite
Elite
 
Posts: 5770
Joined: 03 Jan 2008, 22:35
Location: Minnesota, USA

Re: A question.

Postby bornemix » 14 Mar 2010, 11:45

It sounds like an "admins-are-not-that-humble" problem, which can be hard to describe and communicate...
User avatar
bornemix
Code contributor
Code contributor
 
Posts: 56
Joined: 09 Mar 2010, 08:39

Re: A question.

Postby dmkp » 14 Mar 2010, 15:42

Zarel wrote:
dmkp wrote:I take it you don't do any programming then from your obvious ignorance shown here xD

I should point out that bournemix has submitted useful patches to the Warzone project. Have you?


I should point out that I don't really want to. Since you guys seem to have it all under control.

Completely irrelevant statement Zarel.

bornemix wrote:It sounds like an "admins-are-not-that-humble" problem

Yep.. seen that problem a lot.
dmkp
Trained
Trained
 
Posts: 182
Joined: 08 Jan 2010, 19:50
Location: UK

Re: A question.

Postby Buginator » 14 Mar 2010, 19:34

bornemix wrote:It sounds like an "admins-are-not-that-humble" problem, which can be hard to describe and communicate...


I think that cuts both ways.

Most of the time, we tend to bend over backwards to help people out. Yeah, sometimes, we get a bit annoyed, but we are all only doing this as a hobby--time permitted. Nobody is getting paid for anything around here.

On the other end of the stick, we got people who don't seem to comprehend all the issues involved, and they seem to take things personally.

Bottom line is, if you have a patch, put it on trac, not the forums. We will eventually get around to it, as soon as whomever has free time to look at it.

If you have a question or concern that isn't being handle correctly here, then either get on IRC, freenode, #warzone2100-dev and let us know, or send e-mail to the ML.

If you still feel your right, we are wrong, then the option is to fork, and start fresh from your perspective.
(all source code is open source, so everyone is free to do whatever they want with it) :)
and it ends here.
User avatar
Buginator
Professional
Professional
 
Posts: 3285
Joined: 04 Nov 2007, 01:20

Re: A question.

Postby Zarel » 15 Mar 2010, 08:23

dmkp wrote:I take it you don't do any programming then from your obvious ignorance shown here xD
Zarel wrote:I should point out that bournemix has submitted useful patches to the Warzone project. Have you?
dmkp wrote:I should point out that I don't really want to. Since you guys seem to have it all under control.

Completely irrelevant statement Zarel.

You're missing the point, which is that you are being rude to another forum member (which I note is against forum rules) and accusing him of inexperience when in fact he has presented solid evidence of his experience, and you have not.

So how's about you either put your actions where your mouth is, or you control that mouth a bit more, hmm?
User avatar
Zarel
Elite
Elite
 
Posts: 5770
Joined: 03 Jan 2008, 22:35
Location: Minnesota, USA

Re: Porting to C++

Postby Safety0ff » 15 Mar 2010, 14:15

What a lengthy digression! Feel free to ignore the following comments for the sake of bringing this thread back on topic:

Seismo, these passages where incomprehensible to me:
Seismo wrote:and no one is kick them or even setting things correct.
Which things need to be set right?
What has been done (in your opinion,) to warrant the removal of a member?
Seismo wrote: but when i said a, so i should also say b.
This simply didn't make any sense to me.

@Zarel on dmkp:
I interpreted the "nevermind" post as a retraction, is it really necessary to bring it up again in this thread?
==========================================================
Back to the topic at hand:

Ceiling Cat wrote:are there any plans on gradually porting this piece by piece to the more user-friendly alternative, or it's gonna stay as it is forever?
There aren't any plans that I know of, though I would be interested in anyone's plans to do so.

Also:
I don't think there would be any point to convert to C++ "gradually, piece by piece" because once you start, you might as well change the majority of the pieces it interfaces with, and then the pieces those pieces interfaced with, and so on.
Safety0ff
Trained
Trained
 
Posts: 397
Joined: 18 Jul 2009, 22:23

Re: Porting to C++

Postby Kamaze » 15 Mar 2010, 14:37

I'm fine with the C code. Here are also two nice quotes why:

The idea that new code is better than old is patently absurd. Old code has been used. It has been tested. Lots of bugs have been found, and they've been fixed. There's nothing wrong with it. It doesn't acquire bugs just by sitting around on your hard drive.


And:

When you throw away code and start from scratch, you are throwing away all that knowledge. All those collected bug fixes. Years of programming work.


http://www.joelonsoftware.com/articles/ ... 00069.html
Currently retired webmaster.
We all have the same heaven, but not the same horizon.
User avatar
Kamaze
Regular
Regular
 
Posts: 1017
Joined: 30 Jul 2006, 14:23
Location: Germany

Re: Porting to C++

Postby Per » 15 Mar 2010, 17:57

New code modules may be written in C++, several are already, and one happy shining day we might just port larger portions to C++, if there is some benefit in that, but that day is not on the current horizon. The salesman attitude of 'just use X and it will fix everything' is however rather annoying.
Per
Warzone 2100 Team Member
Warzone 2100 Team Member
 
Posts: 3134
Joined: 03 Aug 2006, 18:39

Re: Porting to C++

Postby Seismo » 16 Mar 2010, 02:31

Kamaze wrote:I'm fine with the C code. Here are also two nice quotes why:

The idea that new code is better than old is patently absurd. Old code has been used. It has been tested. Lots of bugs have been found, and they've been fixed. There's nothing wrong with it. It doesn't acquire bugs just by sitting around on your hard drive.


And:

When you throw away code and start from scratch, you are throwing away all that knowledge. All those collected bug fixes. Years of programming work.


http://www.joelonsoftware.com/articles/ ... 00069.html


Well, you are right that rebuilding an existing system complete from the scratch might be a mistake, but today, ten years later than your article, near by all software companies do code refactoring: http://en.wikipedia.org/wiki/Code_refactoring

Safety0ff wrote:What a lengthy digression! Feel free to ignore the following comments for the sake of bringing this thread back on topic:

Seismo, these passages where incomprehensible to me:
Seismo wrote:and no one is kick them or even setting things correct.
Which things need to be set right?


If there is no one in your team seeing that they are against evolution, you might not see that they do things not for the future but for holding there small world. Innovation in software development is not a bad thing at all. But when the developers are start to reject new ways and do not want to learn something new, they are dead and should better leave instead of doing something that no others will understand anymore. If you are good in assembler, ok, do it in assembler. If you are good in c, well, than try to learn c++ and the world will say thank you.
Seismo
Trained
Trained
 
Posts: 70
Joined: 06 Feb 2010, 16:32

Re: Porting to C++

Postby DuKe2112 » 16 Mar 2010, 16:18

Put it simply: old code works, .. for the purpose it was intented, but if you need it to do something it was not quite intended for, it gets increasingly difficult to adjust.
Also refactoring is everything but starting from scratch! You look what you did last time and try to (probably) do it better.
DuKe2112
Trained
Trained
 
Posts: 44
Joined: 01 Mar 2010, 21:01

PreviousNext

Return to Coding