[kde-edu]: Report Bug action

[Joshua Keel]

Hi, 

On Monday 05 September 2005 07:15 pm, Inge Wallin wrote:
> > Hmmm, have you or anyone else commited the fix? The code i have is
> >
> > void KanagramGame::checkFile()
> > {
> >         if (!QFile::exists(locate("appdata", m_filename))) {
> >                 QString msg = i18n("File %1 cannot be found.\n Please
> > ensure that Kanagram is properly installed.")
> >                   .arg(m_filename);
> >                 KMessageBox::sorry(m_parent, msg, i18n("Error"));
> >                 exit(0);
> >         }
> > }
>
> That's just because I fixed it.  I can add that I find it surprising (in
> the sense that diplomats use the word), that a program that doesn't even
> compile is added to kdeedu and that I fully support annma in her bug
> report.

Honestly, Inge, I don't understand why you're making a big deal of this. "It 
doesn't compile" is an overstatement of the actual situation. It DOES compile 
now that you fixed the problem. So someone (maybe me) made an error when they 
were changing a string. So Kanagram didn't compile until you noticed it and 
fixed it. I don't understand why that's surprising or outrageous. kdebase 
from branches/KDE/3.5 doesn't compile for me right now, but I'm not 
complaining and saying it should be removed from KDE. I really don't see it 
as a big deal, as I know it's alpha-quality code. 

I also don't understand what connection annma's bug report has to whether or 
not an app that doesn't compile gets put into kdeedu. I support annma's (or 
anyone's) bug report as much as anyone. I love bug reports. They help me 
improve Kanagram. But sometimes bug reports are helpful, and sometimes 
they're vague and unhelpful. You're a good programmer, Inge. You should know 
a good bug report from a bad one. Programmers need the details of the problem 
(like a backtrace, for instance) to be able to fix it. 

I'm not criticizing annma, or you, or anybody else. I'm just saying that 
reporting that "x doesn't work" is not the kind of bug report devs can use.

>
> > (From another mail:)
> >
> > annma:
> >> - in kanagramgame.cpp the code for the createAnagram(QString original)
> >> method seems weird. It issues a compilation warning not here
> >> and it should be at least commented a bit to allow a better
> >> understanding.
> >
> > Right seems a bit weird but that's not a problem IMHO
>
> How can the following code not be a problem
> (kanagramgame.cpp:createAnagram())?
>
>        for(int i=0; count = objData.count(); i++)
>         {
>            ...
>         }
>
> There is no stop criterium for the loop, just another assignment, which is
> very likely wrong.  Besides, the application crashes for me when I try to
> start it.

What's interesting about that bit of code is that it's actually already in 
kdeedu. It's part of KMessedWords, in minparser.cpp. There, you will see some 
comments that I removed. I can't remember exactly why, now, except that I 
thought they looked tacky, and weren't very helpful. 

As to why Kanagram crashes, I have no idea. It doesn't crash for me or dannya 
or tsdgeos or annma. As I mentioned above, that's not a good "bug report." I 
need to know details (like a backtrace, for instance). If you can't give me 
details, I can't fix the problem.

I'm beginning to feel like people don't want Kanagram in kdeedu for some 
reason. This is ridiculous. It's quite complete and stable, and it's better 
than KMessedWords in every way. Anyone who doesn't see that just isn't 
looking very hard. I don't understand why such a good application is being 
complained about so much, when we have applications of lesser quality in 
kdeedu already (think KMessedWords). I don't mind "bug reports," but the 
things I'm reading seem much more like complaints than bug reports from 
people who want Kanagram to succeed. 

Welp, that was my two cents... 

-- 
Joshua Keel
Code is poetry.