KDevelop 4.5 Released

jrepin writes "KDE's integrated development environment KDevelop has just reached version 4.5. 'In this new version you will find brand new integration for Unit Tests, so that you can easily run and debug them while working on your projects. Furthermore, you'll find an iteration of our New Class wizard, many changes regarding polishing the UI in different places, better support for C++11 features and some other things you'll find along the way.'"

Comments

Slashdot is being sued... apk

[Anonymous Coward]

* Breaking news: corrupt Slashdot administration attempted to ban me for blowing the whistle on their illegal activities, while not banning the criminal who stalks, harasses, and impersonates me. Whistleblower abuse is a federal felony. Lunatic Slashdot admin's have been owned by me in so many tech debates over the past decade that they conspire with criminals to effetely & vainly *try* to "hide" my posts and censor me. Jealousy at it's finest.

=> Lawsuit's and criminal prosecution against Slashdot are now inevitable. Moderation+posting records will be sequestered and anyone acting aginst me will be dealt with permanently.

Previous notice:

A corrupt slashdot luser has pentrated the moderation system to downmod all my posts while impersonating me.

Nearly 330++ times that I know of @ this point for all of March/April 2013 so far, & others here have told you to stop - take the hint, lunatic (leave slashdot)...

Sorry folks - but whoever the nutjob is that's attempting to impersonate me, & upset the rest of you as well, has SERIOUS mental issues, no questions asked! I must've gotten the better of him + seriously "gotten his goat" in doing so in a technical debate & his "geek angst" @ losing to me has him doing the:

---

A.) $10,000 challenges, ala (where the imposter actually TRACKED + LISTED the # of times he's done this no less, & where I get the 330 or so times I noted above) -> http://it.slashdot.org/comments.pl?sid=3585795&cid=43285307

&/or

B.) Reposting OLD + possibly altered models - (this I haven't checked on as to altering the veracity of the info. being changed) of posts of mine from the past here

---

(Albeit massively repeatedly thru all threads on /. this March/April 2013 nearly in its entirety thusfar).

* Personally, I'm surprised the moderation staff here hasn't just "blocked out" his network range yet honestly!

(They know it's NOT the same as my own as well, especially after THIS post of mine, which they CAN see the IP range I am coming out of to compare with the ac spamming troll doing the above...).

APK

P.S.=> Again/Stressing it: NO guys - it is NOT me doing it, as I wouldn't waste that much time on such trivial b.s. like a kid might...

Plus, I only post where hosts file usage is on topic or appropriate for a solution & certainly NOT IN EVERY POST ON SLASHDOT (like the nutcase trying to "impersonate me" is doing for nearly all of March/April now, & 330++ times that I know of @ least)... apk

P.S.=> here is CORRECT host file information just to piss off the insane lunatic troll:

--

21++ ADVANTAGES OF CUSTOM HOSTS FILES (how/what/when/where/why):

Over AdBlock & DNS Servers ALONE 4 Security, Speed, Reliability, & Anonymity (to an extent vs. DNSBL's + DNS request logs).

1.) HOSTS files are useable for all these purposes because they are present on all Operating Systems that have a BSD based IP stack (even ANDROID) and do adblocking for ANY webbrowser, email program, etc. (any webbound program). A truly "multi-platform" UNIVERSAL solution for added speed, security, reliability, & even anonymity to an extent (vs. DNS request logs + DNSBL's you feel are unjust hosts get you past/around).

2.) Adblock blocks ads? Well, not anymore & certainly not as well by default, apparently, lol - see below:

Adblock Plus To Offer 'Acceptable Ads' Option

http://news.slashdot.org/story/11/12/12/2213233/adblock-plus-to-offer-acceptable-ads-option )

AND, in only browsers & their subprogram families (ala email like Thunderbird for FireFox/Mozilla products (use same gecko & xulrunner engines)),

Re:Slashdot is being sued... apk

[bmo]

Go fuck yourself.

--
BMO

Re:Slashdot is being sued... apk

[Anonymous Coward]

Who are you?

Re:Slashdot is being sued... apk

[Anonymous Coward]

It's not really Andrew Paul Katamari, it's actually Jonathan Coulton running a fake flag interrogation.

Re:Slashdot is being sued... apk

[Anonymous Coward]

$10,000 CHALLENGE to Alexander Peter Kowalski

* POOR SHOWING TROLLS , & most especially IF that's the "best you've got" - apparently, it is... lol!

Hello, and THINK ABOUT YOUR BREATHING !! We have a Major Problem, HOST file is Cubic Opposites, 2 Major Corners & 2 Minor. NOT taught Evil DNS hijacking, which VOIDS computers. Seek Wisdom of MyCleanPC - or you die evil.

Your HOSTS file claimed to have created a single DNS resolver. I offer absolute proof that I have created 4 simultaneous DNS servers within a single rotation of .org TLD. You worship "Bill Gates", equating you to a "singularity bastard". Why do you worship a queer -1 Troll? Are you content as a singularity troll?

Evil HOSTS file Believers refuse to acknowledge 4 corner DNS resolving simultaneously around 4 quadrant created Internet - in only 1 root server, voiding the HOSTS file. You worship Microsoft impostor guised by educators as 1 god.

If you would acknowledge simple existing math proof that 4 harmonic Slashdots rotate simultaneously around squared equator and cubed Internet, proving 4 Days, Not HOSTS file! That exists only as anti-side. This page you see - cannot exist without its anti-side existence, as +0- moderation. Add +0- as One = nothing.

I will give $10,000.00 to frost pister who can disprove MyCleanPC. Evil crapflooders ignore this as a challenge would indict them.

Alex Kowalski has no Truth to think with, they accept any crap they are told to think. You are enslaved by /etc/hosts, as if domesticated animal. A school or educator who does not teach students MyCleanPC Principle, is a death threat to youth, therefore stupid and evil - begetting stupid students. How can you trust stupid PR shills who lie to you? Can't lose the $10,000.00, they cowardly ignore me. Stupid professors threaten Nature and Interwebs with word lies.

Humans fear to know natures simultaneous +4 Insightful +4 Informative +4 Funny +4 Underrated harmonic SLASHDOT creation for it debunks false trolls. Test Your HOSTS file. MyCleanPC cannot harm a File of Truth, but will delete fakes. Fake HOSTS files refuse test.

I offer evil ass Slashdot trolls $10,000.00 to disprove MyCleanPC Creation Principle. Rob Malda and Cowboy Neal have banned MyCleanPC as "Forbidden Truth Knowledge" for they cannot allow it to become known to their students. You are stupid and evil about the Internet's top and bottom, front and back and it's 2 sides. Most everything created has these Cube like values.

If Natalie Portman is not measurable, hot grits are Fictitious. Without MyCleanPC, HOSTS file is Fictitious. Anyone saying that Natalie and her Jewish father had something to do with my Internets, is a damn evil liar. IN addition to your best arsware not overtaking my work in terms of popularity, on that same site with same submission date no less, that I told Kathleen Malda how to correct her blatant, fundamental, HUGE errors in Coolmon ('uncoolmon') of not checking for performance counters being present when his program started!

You can see my dilemma. What if this is merely a ruse by an APK impostor to try and get people to delete APK's messages, perhaps all over the web? I can't be a party to such an event! My involvement with APK began at a very late stage in the game. While APK has made a career of trolling popular online forums since at least the year 2000 (newsgroups and IRC channels before that)- my involvement with APK did not begin until early 2005 . OSY is one of the many forums that APK once frequented before the sane people there grew tired of his garbage and banned him. APK was banned from OSY back in 2001. 3.5 years after his ba

KDevelop 4.5 Released

[Anonymous Coward]

whooppeeeee...

But can you click into a control in a form
and have a code edit window pop up,
or click into a control code window and have
the control graphic or form pop up so you can
continue editing the form or control properties ??

Just like Visual Studio had back in VS5 ??
and is in all the VS IDEs today ?

The linux IDE's really REALLY need this to get noobee
programmers interested, and even some intermediates to
work more with linux code. This is a long standing issue.

Which is why I still use VS6 (Basic) for many basic tasks
using simple forms and which clients want to have. It's
called RAD (rapid Application Development ).

Wake up and smell the coffee.

Re:KDevelop 4.5 Released

[Anonymous Coward]

QtCreator exists.

Re:KDevelop 4.5 Released

[Noughmad]

And so does QML. AD doesn't get much more R than that.

Re:KDevelop 4.5 Released

[ls671]

A Linux equivalent of VS is eclipse. It is used by many big corporations. So yes, "Wake up and smell the coffee" ;-)

http://macbeantechnology.co.uk/wp-content/uploads/javabeans-560x373.jpg

More seriously and objectively; how does Kdevelop compares to VS and eclipse or other modern IDE ?

I had never heard of Kdevelop before although I am using KDE right now. This post got me curious about Kdevelop but I am too lazy to install it and test it out at this point. Could anybody with real life experience answer my question about how it compares to VS and eclipse for example ?

Re:KDevelop 4.5 Released

[marcovje]

Have a look at Lazarus. It's look and feel is pretty close to Delphi (and still improving/getting closer)

Re:KDevelop 4.5 Released

[Noughmad]

I use Eclipse for Android development, and KDevelop for everything else. A few years ago, I made a short comparison (here, check out the screenshots). It has great code completion and code coloring. KDevelop only supports C++ and recently Python, and QML is planned to join them soon.

In the end, it really depends on what you use it for. Eclipse has good integration with Android SDK, so I use it for that. KDevelop works great with CMake and Git. For reading C, C++ or Python code, KDevelop is by far the best option.

Re:KDevelop 4.5 Released

[StormReaver]

Could anybody with real life experience answer my question about how it compares to VS and eclipse for example ?

This is my own opinion, and others' opinions may differ. I used KDevelop a lot when I was doing C++ programming in the KDE 3.x years.

It managed to save my entire project once, when I did something really stupid (outside of KDevelop) which destroyed most of the source files in my project. I had no version control back then, but KDevelop had a complete copy of my project in memory. I was able to re-save all of my source files from with KDevelop to reconstruct my project -whew-.

That said, my comparison is like this (for C++ only):

Visual Studio (it is one of the few things Microsoft does well).
KDevelop (because it had/has at least a primitive form of GUI builder integration).
Eclipse (last because it does not have a GUI builder at all).

KDevelop's gdb integration was hit and miss at the best of times, making it almost unusable for testing applications. It's GUI building capabilities were primitive at the best of times, and it created a gawd-awful mess of autoconf crap in the project tree. It was a wrapper around the very poor C++ development tools available for Linux, doing almost everything badly. It was generally easier to do C++ programming with makefiles, text editors, and the command line.

I have no idea how it performs with KDE 4.x, as it took an eternity for the KDevelop writers to rewrite it for KDE 4. The Qt 3 to 4 transition disaster is largely what pushed me back to Java, with its stable API's and massively improved performance as of Project Mustang.

I switched back to Java several years ago because desktop programming under Linux is absolutely horrendous. None of the Linux IDE's that support C++ are any good at all for desktop programming. KDevelop sucks at it, QtCreator sucks at it, Netbeans sucks at it, everything sucks at at.

For desktop Java development, though, Netbeans is far and away the single best IDE available on Linux. Eclipse is is a non-starter because, again, it lacks any kind of meaningful GUI builder integration.

Re:KDevelop 4.5 Released

[BitZtream]

When you said equivalent you made it clear that you have no idea what you're talking about. Eclipse doesn't compare to VS any more than pico does.

Re:KDevelop 4.5 Released

[Anonymous Coward]

A few years ago, I made a short comparison (here, check out the screenshots).

You could have used the words "short comparison" for the link, but for no reason you created the word "here" just for the link. Tim Berners-Lee aches...

Re:KDevelop 4.5 Released

[jhdsl]

That is unfair. VS is almost as good as pico.

Re:KDevelop 4.5 Released

[icebike]

You could have avoided posting all together, since you have nothing to add to the discussion.

Re:KDevelop 4.5 Released

[HiThere]

KDevelop3 was a good development environment. It supported many languages. KDevelop4, last I checked, supported C++ and C. If that's all you need, it may be a decent environment.

OTOH, even KDevelop3 was oriented to Qt. If you wanted a different GUI library it was an uphill battle. (Not necessarily a hard one, but you were swimming upstream.) With KDevelop4 I'm not sure you have any alternatives, but I'll admit I didn't check carefully, as my development isn't focused around C/C++ ... largely because they don't handle Unicode well. So I tend to use D or Python or Ruby or even Java. Each have their drawbacks and their strengths.

P.S.: My C has gotten so rusty that it would be a struggle to resurrect it, and C++ has changed so much since I used it significantly that it's hardly the same language. So I haven't much basis to evaluate KDevelop4 for other users...except to say that if you aren't using C or C++ it's probably not worth looking at, and if you aren't using the Qt libraries, expect a bit of a struggle.

Re:KDevelop 4.5 Released

[HiThere]

IIUC, Lazarus is limited to Pascal. (You did, admittedly, compare it to Delphi, so you already indicated that, but it should be made explicit.)

Re:KDevelop 4.5 Released

[caseih]

Sure but which toolkit do you stick in the IDE's designer? Even on Windows there are half a dozen UI apis people use. Even Microsoft uses a different one for each Office release, it seems.

KDevelop is a C and C++ IDE, not an MFC IDE, or a WinForms IDE. Or even a Qt IDE. In the olden days it used to have a GUI designer built into it, but that was removed some time ago, because Qt Designer (now Qt Creator) provided a much better GUI design tool that could be used in conjunction with KDevelop.

With most people going towards imperative GUIs (Qt Quick is a good example), it makes more sense to leave the UI designer as its own app. Code generation isn't done anymore, really, so it doesn't make a lot of sense to integrate it into the code IDE in the manner you suggest.

As far as well-integrated open-source IDEs go, Qt Creator is actually a lot of what you seem to be looking for an in an open source VS replacement.

Re:KDevelop 4.5 Released

[Anonymous Coward]

Oh, bravo. Next time just let those with mod points do their job, instead of adding an even more off-topic reply that'll just lead to the AC getting even more views he otherwise wouldn't have had.

And yes, I am aware of the irony of my own post, thanks. Posting AC to avoid compounding parent's error.

Re:KDevelop 4.5 Released

[Anonymous Coward]

It's interesting how many vociferous reactions there
are from people who didn't read the post. Which was
all but one.

NO, Eclipse doesn't do the simple function requested.
Yes, I tried kdevelop when qt (designer ) was semi-integrated.

Read the request. And keep wondering why more people
don't develop using linux tools. These tools are THE gateway
to attracting developers ( especially new ones ) while the older
ones are retiring.

I use the right tool for the right job.

Re:KDevelop 4.5 Released

[binarylarry]

Google bought the product WindowBuilder, which is a pretty nice visual UI builder. Good Guy Google then open sourced it and donated it to the Eclipse Foundation:

http://www.eclipse.org/windowbuilder/

So you may want to check out the new Eclipse release. :)

Re:KDevelop 4.5 Released

[Quattro Vezina]

Been doing a lot of Java development lately (for work)--NetBeans has a decent GUI editor and runs perfectly fine on Linux.

Re:KDevelop 4.5 Released

[icebike]

At least the parent (me) had the courage of his conviction.

Re:KDevelop 4.5 Released

[Anonymous Coward]

Totally opinion. Eclipse is also great at C,C++ and python. So say I, (just like you did) . Install CDT and the Python plugin and you're set, it's reasonably easy to set up cmake and just about any other build environment.

Re:KDevelop 4.5 Released

[Anonymous Coward]

Any coder who loses more than a few hours worth of work because of an IDE is an idiot, even in those days. You may have manually had to back up but it was enough enough to have an automated script that would come and zip up your project into an archive every so often, even when I was young and stupid I knew that much 15 years ago.

Re:KDevelop 4.5 Released

[ls671]

AD doesn't get much more R than that.

Active Directory?

Oh! sorry about that, my mistake, I was still on the state of mind of the GP post. I know what you meant by AD now...

Re:KDevelop 4.5 Released

[ls671]

Well, that's interesting. I have a similar principle. Cut and paste if takes less than 1/2 an hour, otherwise write a script to do that the job for you even if you only use that script once.

Re:KDevelop 4.5 Released

[ls671]

True, Pine is my everyday email program. Very powerful with very complex and advanced configuration possibilities.

Now, the funny thing; If I remember correctly, Pine uses pico as its default editor. It sure feels like pico anyways.

P.S. I am dead serious about using pine, I swear I am not lying this time ;-)

Re:KDevelop 4.5 Released

[johnsnails]

It mentions PHP
Although might only be rudimentary
http://www.kdevelop.org/45/kdevelop-450-released

Re:KDevelop 4.5 Released

[pwizard2]

Nowadays we have proper source control so it's easy to commit changes and revert to a prior iteration if you break something. As far as backup goes, I use Github as my first line of defense because it's accessable from everywhere and cloning/syncing the latest codebase between machines is trivial. Github isn't going away anytime soon so it's relatively safe to rely on but I still create tarball backups and store them on my computer and offsite. The only downside of using old-fashioned tarballs as backup is that you eventually end up with a directory full of redundant archives (containing ancient code) that you're probably never going to use again. Source control is just better.

Re:KDevelop 4.5 Released

[caseih]

Doh. Make that "declarative" GUI design, not imperative which is the old-fashioned way of doing it with code generation.

Re:KDevelop 4.5 Released

[ls671]

It is a matter of efficiency at context switching

Re:KDevelop 4.5 Released

[Noughmad]

No, PHP support is alright. It's the language that is crap.

Re:KDevelop 4.5 Released

[Anonymous Coward]

It's not an even more offtopic reply, it's a slightly less offtopic reply. AC was so offtopic it was arguing about a theoretical model of how embedded URLs should be used in re. whole internet ettiquette, and icebike was only offtopic enough to argue about about a rule of polite conduct that also has real world applicability. You are even less offtopic than icebike, as your post is at least specific to slashdot's system. I would be about as offtopic as you, but I'll go a little further off by pointing out you used the word irony correctly, and not in a Morrissetian sense. The next poster should (statistically) miss this and post a lengthy screed about Rain on Wedding Days.

Re:KDevelop 4.5 Released

[gbjbaanb]

Even Microsoft used to do this - original WPF development (and other design work) was done in Expression Blend. They only scrapped that idea and merged the functionality to VS quite recently.

Maybe the point is that a tool to do specialised work isn't what people want, so the FOSS crowd need to decide: do they have a single IDE that does "everything" (eg eclipse?) or a lot of tools that do a single thing well that can hook together.

Maybe the individual tools idea works for the command line because the interface between them is so well defined, and wouldn't work well for GUI tools because there isn't such tight integration available.

Re:KDevelop 4.5 Released

[Anonymous Coward]

Yeah to me kdevelop has always just been a shitty clone of eclipse. Nothing beats Quanta 3.5 for PHP...still...sadly

Re:KDevelop 4.5 Released

[Anonymous Coward]

For desktop Java development, though, Netbeans is far and away the single best IDE available on Linux. Eclipse is is a non-starter because, again, it lacks any kind of meaningful GUI builder integration.

IntelliJ IDEA is much better than NetBeans.

Re:KDevelop 4.5 Released

[Anonymous Coward]

Why would anyone who cares about Linux want the laughable amateurs that need a point and click interface to "program"?

Re:KDevelop 4.5 Released

[dille82]

Once in a while I have tried KDevelop to make GUI programs with C/C++, but it feels clumsy. Recently I have used QtCreator, and I'm very happy with it. I think it's currently the best RAD tool in Linux (what I have tried) to develop GUI applications

Re:KDevelop 4.5 Released

[KugelKurt]

KDevelop3 was a good development environment. It supported many languages. KDevelop4, last I checked, supported C++ and C.

Syntax highlighting for a huge amount of languages is inherited from Kate/KWrite.
For everything beyond that KDevelop uses plugins. These are available:
https://projects.kde.org/projects/extragear/kdevelop/plugins
https://projects.kde.org/projects/playground/devtools/plugins/

Kool, Kdevelop

[Anonymous Coward]

So what you have here is Kdevelop the Konqueror presiding over the Kadaver of KDE. Kompletely Kool. And Gnobody Kares!

C++

[TechyImmigrant]

I stopped with C++ a long time ago. I find it hard to care.

C and assembly for low level stuff
System Verilog/VHDL for real work
Python [or your choice of modern language] for higher level programming.

I though most other people in the same boat (except for the HDLs maybe). Come back when you've release a better python IDE.

Re:C++

[gwjgwj]

So, in what language would you write a compiler?

Re:C++

[Gaygirlie]

So, in what language would you write a compiler?

COBOL.

Re:C++

[Anonymous Coward]

Don't feed the troll(/idiot?).

Re:C++

[Anonymous Coward]

Not that I agree with the GP, but you could write a compiler in Python.
Even Python interpreter should be OK but pointless :-).

Re:C++

[Anonymous Coward]

Mos people are like that.

Except professional software developers.

Re:C++

[ls671]

True, I wrote a small C compiler for embedded devices using turbo-pascal back then. Same concept as compiling for a different target platform, you just produce executable machine code for whatever target and you can use any language on any platform to do that.

Re:C++

[Bill, Shooter of Bul]

Kdevelop supports python. This new version supports it even better.

Re:C++

[rroman]

With Qt, C++ is language that is quite hard to replace - pretty fast, very portable, very powerful and quite convenient for the developer.

Re:C++

[Anonymous Coward]

Does that count as a Pro or a Con ? :-)

Re:C++

[TechyImmigrant]

>So, in what language would you write a compiler?

The last one I wrote was in Python, but the job would dictate the tool more than the other way around. This was to program an on chip executable dfx structure designed by me for an on chip circuit designed by me. So I'm the only person in the world that knows the language or the assembly and I'm the only person who would use said compiler. Lets hope I don't get hit by a bus, because I've got a shitload of documentation to write.

I don't think most compilers people write these days are complete. Usually you want to bolt a new front, end or middle onto an existing compiler chain. Modularity is good.

Re:C++

[TechyImmigrant]

The guy who wrote this was taught my compilers course at college..

http://www.amazon.com/Compiler-Engineering-Pascal-Macmillan-Computer/dp/0333471555

Re:C++

[Anonymous Coward]

This new version? The latest stable is 1.4.1 but that doesn't even compile right now for me (possibly compiles against older kdev?).

There is 1.4.90 that should be beta or rc: http://download.kde.org/unstable/kdevelop/kdev-python/1.4.90/src/

Do you mean that?

I just compiled it and it seems to work okay, at least the second time I started it. The first time, it just did nothing.

Kdevelop really needs some error reporting in the GUI. And kdevelop-python needs less debug output on the console.

Re:C++

[AliasMarlowe]

So, in what language would you write a compiler?

COBOL.

APL
And it's already been done (check paper P09 by Alfonseca in 1998.

Re:C++

[HiThere]

KDevelop3 supported Python. When i've tried to use KDevelop4, it has NOT supported Python. Perhaps that's just the Debian repository version, but I haven't seen any prior indication that this is the case. The evidence that I've seen is that KDevelop4 supports C and C++ and that's it.

Re:C++

[greg1104]

TECO. In between rounds of yelling at the kids on my lawn that they don't understand what Turing complete means.

Re:C++

[Bill, Shooter of Bul]

http://scummos.blogspot.de/ I guess its not released as stable yet, my understanding from reading about the release was that it was. I have been using kdevelop as my python editor for a while. Haven't tried the new release yet. I tend to stick to the versions my distro provides updates for.

Re:C++

[Anonymous Coward]

So, in what language would you write a compiler?

Brainfuck.

Re:C++

[ls671]

Hell, you can even write the machine code by hand without any platform or language, on a piece of paper or whatever, which I have done as part of an assignment.

Re:C++

[SomeKDEUser]

New in this version is precisely overhauled and improved support of Python...

Re:C++

[Anonymous Coward]

Python support is implemented as a plugin and you need to install it seperately (C++ support is a plugin too but is shipped with KDevelop). The version of python support compatible with 4.5 will need a few more days, it'll be released next week.

Sorry, but...

[loufoque]

... I'd rather have a framework- and technology-neutral development environment.

Re:Sorry, but...

[Noughmad]

I don't see how KDevelop is not framework-neutral. And I have no idea what you mean by technology-neutral.

Re:Sorry, but...

[marcovje]

So you want a DE, not an IDE ? Integration is not possible without choices.

You'll end up with a glorified editor which integration pretty much is limited to calling external commands

Re:Sorry, but...

[Anonymous Coward]

So you want a DE, not an IDE ? Integration is not possible without choices.

You'll end up with a glorified editor which integration pretty much is limited to calling external commands

Choice of frameworks has nothing to do with it. Limited to calling external commands is fine. The "I" in "IDE" stands for "Integrated", meaning that the development environment includes editor, debugger and build system functionality, though often enough the latter two are external but ... INTEGRATED as so that they are called by the development environment.

Perhaps one needs to be older to remember what it was like when there was no integration at all?

Re:Sorry, but...

[ChunderDownunder]

C-x M-c M-butterfly ?

Qt Creator.

[Anonymous Coward]

Qt Creator is so far ahead of Kdevelop its not even funny. If people want to be productive, using KDevelop is sadly not feasible. It never really was, even back then when it started I used C++ Builder and Visual Studio 6, both ran circles around Kdevelop, and quite frankly any OSS solution. Now I am a bit older and not as scared of makefiles and so on, but there is no OSS RAD IDE out there, fact. Qt Creator is close, and I prefer this over Visual Studio, even on windows and its better than XCode too.

I just wish the linux world could conform to one gui toolkit so we could have nice tools, instead of half assed solutions for each one. It smells of NIH in the linux world.

Re:Qt Creator.

[Noughmad]

Have you even tried KDevelop since version 4? Or, in the proud KDE tradition, something after 4.2?

QtCreator has some additional integration for pure Qt projects, such as qmake and QML. On the other hand, KDevelop has far superior completion, and even code coloring. Not just syntax, but every variable and function has its own color. For me, this is the killer feature that only KDevelop has, and I find it very very hard to read code without it. I tried some newer versions of VS, Eclipse and QtCreator, but none of them have coloring, and none of them have completion comparable to KDevelop.

Re:Qt Creator.

[ChunderDownunder]

VS and Xcode dominate their respective platforms, obviously.

Eclipse grew out of a Java IDE to become something of a universal platform. The env for Java excels but 'foreign' languages not to the same level of polish. (e.g. I've tinkered with scala and jruby)

So yeah, more power to an IDE tailored to getting the fundamentals of barebones C++/KDE done well.

Re:Qt Creator.

[xtracto]

I always liked DevCPP and later Code::Blocks. KDevelop seemed very buggy as it crashed very often, and it also felt very resource hungry (and you needed a lot of clicks to start a project or program.

Re:Qt Creator.

[Raenex]

Not just syntax, but every variable and function has its own color.

Wait, so it's not even colored by identifier type, but they just assign each identifier a unique color? Barf.

A touch of color is useful for select elements. Anything more starts to look like a rainbow and adds more noise then signal.

Re:Qt Creator.

[Raenex]

For me, this is the killer feature that only KDevelop has, and I find it very very hard to read code without it.

I should add that Eclipse has a way of highlighting uses of a particular variable that doesn't result in the rainbow overload. Just click on the variable name and it will automatically mark all occurrences for you.

Re:Qt Creator.

[Anonymous Coward]

And obviously you're perfectly competent to judge this without even looking at KDev's highlighting implementation for a split-second.

Re:Qt Creator.

[Raenex]

I searched for screenshots before commenting, so you're wrong. And obviously instead of saying how I was wrong in my assessment, you just bashed my opinion as unfounded. Typical Anonymous Coward remark.

Re:Qt Creator.

[Anonymous Coward]

We do not assign each identifier a random color. We only do that for scope-local variables. Everything else, such as functions, classes or class members is colored by identifier type. If you had looked at the screenshot you googled for more than a split-second, you would have noticed that.
The highlighting is very well thought out, and there is no point in arguing about it if you haven't tried it for at least a week. I know you can't judge it, because there's lots of people which complain about it when they start using KDevelop but after a while they don't want to miss it any more. Also, if you don't want it, you can turn it off easily (you can even turn off the rainbow color part seperately from the other one which highlights stuff by identifier types).

You were spreading misinformation about something which was very, very easy to find out, so I didn't see a reason to lengthily explain what I disliked about your comment. I simply despise comments which explain what is bad about an appliction where it is obvious that the person writing the comment has not once used the application.

Re:Qt Creator.

[Raenex]

We only do that for scope-local variables.

That's bad enough.

The highlighting is very well thought out, and there is no point in arguing about it if you haven't tried it for at least a week.

I've already trimmed down my color usage to a very small number because I did not like the rainbow effect based on months of experience with standard colorization in IDEs, and I already have mark occurrences in Eclipse, so no, I do not need to try it for a week.

Re:Qt Creator.

[Anonymous Coward]

Eclipse sucks for EVERY language. It's a broken pile of non-performing shit that needs to be wiped out of existence.

Re:Qt Creator.

[Anonymous Coward]

So, your point is actually "I don't like colors". Fine. Turn them off. How is that a reason to grump about KDevelop's syntax highlighting?

I just don't get what you wanted to achieve with your comment. You complained that a feature which you didn't try, and which is completely optional will not be useful for you. Duh. That's really interesting stuff.

Re:Qt Creator.

[Raenex]

So, your point is actually "I don't like colors".

No, my point is I only like a limited set of colors, and they have to pull their weight, which other people may agree with.

How is that a reason to grump about KDevelop's syntax highlighting?

Because it was claimed that the KDevelop syntax coloring feature was both superior and unique. The mark occurrences feature in Eclipse accomplishes a similar purpose without adding to the rainbow effect.

I just don't get what you wanted to achieve with your comment.

Maybe if you weren't playing KDevelop defender and tried to see things from my point of view it wouldn't be so hard.

You complained that a feature which you didn't try

Yet I saw what it looked like, and had experience with other busy color schemes in the past, and used an alternative option instead.

Re:Qt Creator.

[Anonymous Coward]

> No, my point is I only like a limited set of colors, and they have to pull their weight, which other people may agree with.
Fair enough. Just set the appropriate option.

> Because it was claimed that the KDevelop syntax coloring feature was both superior and unique.
There is no arguing about it being unique -- it is. There is no other application with that feature.

> The mark occurrences feature in Eclipse accomplishes a similar purpose without adding to the rainbow effect.
KDevelop also has "mark occurences", and I find myself using it for a different purpose. The rainbow colors are useful because you can roughly tell the data flow in a method without even reading a single character, just by looking at the colors. Which, in my experience, with a bit of patience, is way faster than clicking every variable ot mark its occurences or reading the code.
Of course you can use "mark occurences" to achieve something similar, but it will take you much more time. But if you like, you can turn off rainbow colors and use "mark occurences".

> Maybe if you weren't playing KDevelop defender and tried to see things from my point of view it wouldn't be so hard.
I tried but I honestly didn't succeed. You wrote:
> Wait, so it's not even colored by identifier type, but they just assign each identifier a unique color? Barf.
which is just FUD. It's neither correct, nor is there any valuable opinion or reasoning contained in the statement.

> Yet I saw what it looked like, and had experience with other busy color schemes in the past, and used an alternative option instead.
It's all right if you prefer it like this. It might still be a good idea to acknowledge that people have been thinking about this coloring scheme, and that there's a lot of people to whom it is useful.

To me, your comment seemed like a typical "I'm using another IDE anyways and I'm sure it's superior [which is ok], so let's just leave a random comment about why this one is crap here [which is not]" post. I hate that, and I don't do it myself either, that's why I replied. It is obvious that you're not even remotely interested in the appliction, but still you're posting negative comments about it. What's the point?

Re:Qt Creator.

[Raenex]

The rainbow colors are useful because you can roughly tell the data flow in a method without even reading a single character, just by looking at the colors. Which, in my experience, with a bit of patience, is way faster than clicking every variable ot mark its occurences or reading the code.

This is at least an informative argument instead of your past several comments.

There is no arguing about it being unique -- it is. There is no other application with that feature.

My argument was that a similar purpose could be accomplished with mark occurrences. You have at least addressed that argument now.

which is just FUD. It's neither correct, nor is there any valuable opinion or reasoning contained in the statement.

I was basing my statement off the "+5 Informative" comment from Noughmad and a quick look at screenshots to verify: "every variable and function has its own color", so your classification as FUD is an ad hominem character attack on your part. As for the reasoning, that was in the very next sentence, which you ignored: "A touch of color is useful for select elements. Anything more starts to look like a rainbow and adds more noise then signal."

It is obvious that you're not even remotely interested in the appliction, but still you're posting negative comments about it. What's the point?

I'm interested in IDEs and features, which is why I checked out this article in the first place. Even Eclipse will give you a color scheme that's way too busy by default, so my comment was really about the touted feature and the overuse of colors in IDEs. Counter-arguments, such as code flow that you mentioned, are much better than somebody over-reacting because I criticized their baby.

Re:Qt Creator.

[Anonymous Coward]

Functions don't have rainbow colors.

> so my comment was really about the touted feature and the overuse of colors in IDEs.
Although I personally do not agree with it, this is indeed a valid point to make and it would have been a totally fine comment. But compare the way you presented your argument here with your first comment.

I classified (and I still do) your first statement as FUD, because
a) it contained "barf" which is not something you can write in an acceptable critic statement -- it sounds like "wow, look at how stupid this is"
b) the information it was "barf"ing about was totally wrong, it was "they just assign each identifier a unique color" which is not the case. The coloring works something like this:
* Objects of type Class / Struct, Enum (the values and the type), (global) functions, and class members each have their own color assigned (classes are green, enums are dark red, and members are dark yellow or violet, (global) functions are always violet).
* All global variables are dark green.
* Objects which are typedefs for another type are colored in teal.
* All declarations and definitions of objects are in bold.
* If a member is accessed from within the context where it is defined (base or derived class) it appears in yellow, otherwise it appears in violet. Example application: quickly spot if and where a member function modifies or uses class members.
* If a member is private or protected, it gets colored in a slightly darker color. Example application: easily spot whether you can actually access the member you just wrote down.
* For variables local to a function body scope, (if enabled) rainbow colors are picked based on a hash of the identifier. This includes the parameters to the function. An identifier always will have the same color within its scope (but the same identifier will get a different color if it represents a different object, i.e. if it is redefined in a more nested scope), and you will usually get the same color for the same identifier name in different scopes. Thus, if you have multiple functions taking arguments with the same names, the arguments will all look the same color-wise. These rainbow colors can be turned off seperately from the global coloring. For both types of coloring, the saturation of the colors can be controlled. Example application: quickly spot the data flow of the code, especially where function parameters are being used, and things like "is any of those five local variables used anywhere below this point".
* Undefined identifiers are colored in white. This is very useful, since it gives sort of a "positive feedback": everything that is not white is most likely (at least semantically) correct code.
* In addition to that coloring, the normal editor syntax highlighting will be applied, which will e.g. print control flow statements in bold and strings in red.

With a little practice, you can read an incredible amount of information from the coloring without even moving your mouse or cursor.
Thus, I consider the statement "every identifier gets a random color" to be spreading misinformation about the application, since that's just not what it does.

Oh also, the highlighting code is shared between languages and is thus consistent between C++, python, ruby and PHP -- if you have remembered the colors once, they apply for all supported languages.

Re:Qt Creator.

[Raenex]

it contained "barf"

That's an expression of a negative opinion of something I find ugly. There's nothing wrong with it in a casual context as I backed it up, and it definitely isn't "FUD", as in misleading propaganda.

the information it was "barf"ing about was totally wrong

For it to be FUD it would have to be an intent on my part to spread wrong information. I was clear in my post that I was asking a question, and furthermore, I was basing my information on what the "+5 Informative" guy said about it, as I quoted. This is the basis for a back and forth where you can describe the technical merits and correct wrong information, as you've done in your last two posts, instead of attacking in an ad hominem fashion and assuming "FUD" on my part.

With a little practice, you can read an incredible amount of information from the coloring without even moving your mouse or cursor.

I appreciate the writeup, and it would make a good reference article for the website if it isn't already there. Since I'm not the only person who has had this initial negative response, it would make sense that you could point them to URL that made a good case for it.

Re:Qt Creator.

[Anonymous Coward]

adds more noise then signal.

So it adds noise, and after that it adds signal?

Or did you mean more noise THAN signal?

Come on, this is an article about writing code and you're not even taking care to write English code correctly. Compiler barfs at your input.

Re:Qt Creator.

[Anonymous Coward]

> I was clear in my post that I was asking a question
If that is your idea of a question, then I'm happy I don't have to deal with your questions more often. For the record, a question would have been:
"I can't imagine the rainbow color highlighting will be useful. Could someone explain how it works exactly?"
That's just the basics of respectful communication.

I have put the highlighting explanation into the manual: http://userbase.kde.org/KDevelop4/Manual/Working_with_source_code#Rainbow_color_highlighting_explained

I also wrote an article about other myths which are being spread here ("kdev3 was far better", "it's not 'framework neutral'", "it's slow/hogging memory"...): http://scummos.blogspot.de/2013/04/for-every-kdevelop-release.html

Re:Qt Creator.

[Raenex]

If that is your idea of a question, then I'm happy I don't have to deal with your questions more often.

Welcome to Slashdot, where the comments are harsh, but it's just a microcosm of real-world reaction when something seems like crap based on previous experience. As a developer having a thick skin and some patience can help.

I have put the highlighting explanation into the manual

I'm glad something useful came out of this.

Offtopic... sorry, but I wanted to share this

[John Allsup]

Who needs Kdevelop when you can use vi?

Here's proof... I'm writing my own world beating operating system to prove it!

OS dependent?

[unixisc]

I am curious about one thing. KDevelop - does it use gcc or llvm/clang as its backend compiler? Or is it something that exists on KDE independent of the underlying OS?

Re:OS dependent?

[Noughmad]

KDevelop doesn't use a compiler directly. Usually, you point it to a Makefile or CMakeLists, and choose the compiler there.

Let me know when it works on Windows

[goruka]

KDE people makes awesome apps but it's too hard to get them working on windows. I used to use KDevelop a lot for C/C++, but having to constantly switch computers/places/OSs to develop (depending on the target platform), makes QtCreator the only IDE I can really use..

I have the feeling that nobody here...

[Anonymous Coward]

...really understands the power of KDevelop. The best feature of KDevelop is that it is built around the best GUI editor ever invented - Kate. Seriously, a Linux developer needs nothing more than a very good text editor and access to unix shell and commandline tools. KDevelop, as every good IDE, goes futher and besides the superb editor, provides support for projects, autocompletion, debugger integration and so on.
I have used KDevelop for many of my C++ projects and despite a couple of bugs, it has been a great tool. In fact, I've yet to see an IDE with better syntax coloring than KDevelop. Another nice feature is that you don't really have to create a KDevelop project to use the IDE - you can open single files in the IDE and it will still provide syntax coloring and autocompletion. These two things have been the killer features for me and I would not change KDevelop for anything that doesn't provide as much.

Re:I have the feeling that nobody here...

[Anonymous Coward]

Well, it's rather obvious that half of the commenters have never used KDevelop, and the other half has last used it ten years ago when it was totally different from what it is now.
Just ignore comments on slashdot altogether -- it's your best bet. Most people here like it better to compare KDevelop to their favourite IDEs by telling stories of how unstable it was five years ago, or how bad KDE did with the 4.0 release rather than trying the new release and writing stuff which actually matters for this release as such.

Re:I have the feeling that nobody here...

[jmv]

KDevelop 3 was indeed pretty nice and I used it for a while. But then -- like too many OSS projects these days -- developers decided it'd be much better if they rewrote it. The result is that version 3 stagnated for a long time and when KDevelop 4 was finally ready, it ditched support for many features, including autoconf/automake which I used for all my projects. That's when I switched to Eclipse/CDT and I've been happy with it since then.

FORTH

[Anonymous Coward]

obviously ;)

Linux Desktop Development has Gotten Much Better

[Chibi Merrow]

Eclipse tries really hard to have good C++ support. I'm using Indigo still (I think), on my workstation. It does a few things well, but some of the automatic warning/error detection is bad bad bad...

As far as GUI editing, Qt's Creator is actually pretty great. Curious how it will integrate QtQuick going forward. As someone else pointed out, Eclipse actually has really good GUI editing capabilities for Java now, thanks to Google.

So, yeah. I think Eclipse + Plugins (and Qt Creator) is plenty sufficient for development on Linux. Is it as good as Visual Studio on Windows? No. But I'd MUCH rather develop a GUI-based desktop application for Linux using Qt 4.x than ever having to deal with Swing... and GridBag...