Enlightened
-
Given that all experiences we've had in Samsung with EFL (not just me - literally everyone who worked with EFL) show that it's neither stable, powerful (actually it's absurdly not powerful when compared to others) nor lightweight, you must have had some sad experiences in your life.
-
Maybe he's coming directly from Windows ME.
-
Windows ME.
I seem to be the only person that liked Windows ME. I thought it was a nice change from 98 SE and very stable. Must have had the magic combination of components it worked well on or something.
-
I seem to be the only person that liked Windows ME.
Yes.
very stable.
No.
-
I seem to be the only person that liked Windows ME.
And I though I was weird for liking Vista.
-
I came from Xfce 4.
-
I seem to be the only person that liked Windows ME. I thought it was a nice change from 98 SE and very stable.
/me falls down in a dead faint
-
You should try Cinnamon then, modern but classic.
-
I find it [ E20 ] to be a a desktop that's stable, powerful (in terms of features / configuration options) as well as lightweight.
@Samuel_Hodgkins said:I came from Xfce 4.
I would also label Xfce 4.12 as such, though I do tend to stick to a Windows XP layout with a vertical task bar on the left...
-
I think that using JavaScript with a native UI toolkit is questionable, but eh.
Depends on what you use it for and how good the JS interpreter is, I guess.
<fanboy>
You can use JS for scripting in Qt as well. Works well enough and as long as you don't push it too far it's not that much of a performance impact. Of course, if you want to do heavy processing you should defer it C++ (you can expose C++ methods to JS). It's damned convenient for some stuff, like pulling data from RESTful services - instead of messing with creating connections and whatnot, you can just write an AJAX request in JS and be done with it.I am a bit dubious about the addition of canvas you're supposed to draw to from JS in 5.5 though. Being able to just port code that uses something like three.js to Qt is nice and all, but polishing Qt3D would be a better course of action, IMHO.
</fanboy>
-
I didn't mean E20 (it's Alpha software after all.) - I am currently using E19.10.
-
Oops, I misunderstood something in the rather long reply chain.
Again, I'm sticking to Xfce as it does what I want it to do. The last Enlightement I used was the decrepit E16, after that I never bothered looking back as they were stating the newer versions were WIP.
-
Xfce is pretty neat. I hope you enjoy it :)
-
Enlightenment and Samsung professionalism at its best.
-
part name of the swallow
So, we pass stuff like "beak" or "wing" or what?
-
Pretty sure you are supposed to input the air-speed velocity of an unladen swallow
Filed Under: /Python
-
African or European?
-
Korean
Filed Under: Obviously
-
Well, just as long as Julius Squeezer doesn't try to eat it.
Filed Under: what do you mean, wrong python?
-
Enlightenment and Samsung professionalism at its best.
Hardware manufacturers were always terrible at doing software
-
-
Official Samsung code for the Tizen.
-
This is C exception handling, or is there some other you are pointing to?
-
It's a macro called goto_if that will expand to that yellow code. trying to extend the language with macros like that is frowned upon, and has great potential for some hard to detect and/or debug errors.
-
It hides the ugly fact that it uses
goto
for exception handling, as long as it is documented and is used consistently I prefer it to different incantations of summoning thegoto
demon by copypasta.
-
It hides the ugly fact that it uses goto for exception handling, as long as it is documented and is used consistently I prefer it to different incantations of summoning the goto demon by copypasta.
the C programmers I know are used to goto and don't care much about this, because goto is great for error treatment.
-
C programmers should die.
-
There is no better way than
goto
. But packaging that to agoto_if
macro is not a bad practice, it limits the scope of its use to the only legitimate use-case still valid. As long as it is enforced throughout the project and is documented.Otherwise people may change innocent:
if (error) goto ERROR;
to
if (error) { if (crap) goto NO_ERROR; goto ERROR; }
The latter is abuse, it is better to hide
goto
branching and think it does not exist.
-
C programmers should die.
Help, help, I'm being repressed
-
C programmers should die.
We all should die. While we do live however there is no choice but to usegoto
-
It's releasing universal resources in case of a failure in the most inept method.
A. It's not reusable. These resources have to be released when the application closes gracefully.
So the code is duplicated for the app_exit event.B. Possibly unmaintanable as ad is a global variable. Other things that need be released could be added later.
C. Should be done automatically in destructor, but remember they hate C++.
D. goto in a macro ? For a mobile phone ? What year is this ?
-
It's releasing universal resources in case of a failure in the most inept method.A. It's not reusable. These resources have to be released when the application closes gracefully.So the code is duplicated for the app_exit event
If you're talking about the "goto", I don't know what is the better alternative. Several nested ifs is worse IMHO.
I'm not going in the other points you mention, I'm already convinced that Tizen is
-
if (!ad->something)
return releaseGlobalResources();
-
Bloody peasant!
-
Explain!
-
-
If you're talking about the "goto", I don't know what is the better alternative.
The better alternative is to either use a higher-level language that doesn't need to use a basic control flow primitive, or to restructure the code so that you can do it without needing to do that sort of thing. Often, refactoring into more functions can help, but only if the things you refactor into have good coherence as well; if you can't describe what the contract of the function is, it's unlikely to serve you well.
-
-
The better alternative is to either use a higher-level language that doesn't need to use a basic control flow primitive,
This is not a reasonable answer. The scope is C. A C vs C++ or whatever would be a long long discussion. (Or a int64_t discussion)
-
This is not a reasonable answer.
Only bothering to read half what I wrote isn't especially reasonable either. It's not like you even bothered to reach the end of the sentence.
-
what I wrote isn't especially reasonable either.
The rest was more reasonable but kind of subjective. Even in a simple function goto is better than nested ifs.
The anti-goto cult started for much worse abuses of goto, that I dont see happening that much these days.
It was a time when I used msx-basic, there wasn't functions or even blocks. The interpreter removed any attempt at indentention and gosub and return with line numbers were a thing.
-
-
I'll use it as an avatar
-
Even in a simple function goto is better than nested ifs.
You seem to be in a mood to throw out babies with bathwater there. I'm not saying either use or don't use
goto
. I'm saying that it's pretty low-level and isn't necessarily the right answer; a solution with nestedif
s might be better.I was also saying that other languages that have higher-level control-flow structures need
goto
less, to the point where they don't have to have it at all. Sometimes, the right solution is to stop using C and to use one of these other languages instead.Ask your doctor if
goto
is right for you. Side effects of repeated exposure togoto
include faster code, more complicated debugging, and death due to velociraptors.
-
-
I find in many cases more, smaller functions also help avoid goto's, providing you don't need to pass too many parameters.
-
This thread is like the gold standard of ery.
Is there a way to jeff unrelated goto bashing to keep this for the future generations who come here to only laugh at Samsung's incompetence?
-
PHP now has
goto
. PHP also has velociraptors on its manual page.
-
I find in many cases more, smaller functions also help avoid goto's, providing you don't need to pass too many parameters.
That works, provided you can also get good coherence in what the function actually does. Approximately, that means giving it a good descriptive name. Otherwise, you too easily get a maze of tiny functions calling each other and you've just moved the spaghetti from the control flow into the calling pattern, which is not an improvement. (I call this particular anti-pattern ravioli code; it's still dratted codepasta.)
-
That code complete book claims that the ideal maximum size for functions is around 200 lines. It mentions some scientific studies.