[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-devel
Subject:    closing/identifing bugs [Re: Why kde 2.2.2 ?]
From:       Vadim Plessky <lucy-ples () mtu-net ! ru>
Date:       2001-10-12 16:02:49
[Download RAW message or body]

On Thursday 11 October 2001 23:37, aleXXX wrote:
|   On Fri 12 Oct 01 01:51, Vadim Plessky wrote:
[...]
| Worse is the situation with lanbrowsing/lisa. Everyday I get private mails
| asking me how to configure lisa. In this case a lot changed in the last KDE
| versions, but of course not everybody uses the most recent version, so I
| get mails for different KDE versions.
|   And the same applies to the bug reports for konqy, khtml and kjs.
|   Not all users use the most current version, but an older one and report
| bugs which are fixed in the meantime, whether in cvs HEAD or in a release.
| Then if a developer fixes a problem (which he doesn't necesarrily knows
| from a bug report), well, in theory, he would then have to go through
| approx. 1000 bug reports and check which ones he fixed with his change.

Bug report usually indicates what's wrong. (At least my bug reports do it.)
Most crashes reported against Konq/KHTML/KJS are in fact DOM bug reports. So, 
after fixing some DOM methods/properties developer can close some 
crash-related reports.
"non-working-menu" on site is also related to this (and only partially to CSS)

The biggest issue is table layout and BR related to this.
There is no standard algorithm in HTML4 for table layout, and apparently MS 
IE and NN 4.7 do it in different way. You can just *guess* how NN will layout 
table in this case.
Anyway,  analyzing Mozilla/Bugzilla reports I found that they face similar 
problem.

|   Not really possible.
|   So I'd say, the 900 bug reports for khtml describe probably less than
| maybe 100 different bugs (if at all).

Unfortunately, it's not the case.
Many of KHTML/Konq bugs are reported as "broken layout (rendering)", and most 
of these sites (if not to say all) have invalide HTML, wrong CSS rules, etc.
Yes, there are probbaly duplicate bugs in database. But I would be surprised 
if it is more than 15% of total (in case of KHTML/KJS)

|
|   What we'd really need are some guys who go through the bugs of konqy,
| khtml and kjs to find out which bugs still exist and which are fixed in the
| meantime, or even better which would try to build minimal testcases for the
| bugreports.

That's what I was doing.
But at a moment I don't have time to review 400 bugs, like I did in Jan.-Feb.
I looked through around 40 bugs last days, and probably can take another 50.
But for the rest (1190-90 = 1100) some help would be highly appreciated. 

I can add that some number of bugs alreade have testcases.
So it's possible to fix those bugs right now :-)
|
|   Bye
|   Alex

Bye!
-- 

Vadim Plessky
http://kde2.newmail.ru  (English)
33 Window Decorations and 6 Widget Styles for KDE
http://kde2.newmail.ru/kde_themes.html
KDE mini-Themes
http://kde2.newmail.ru/themes/

 
>> Visit http://mail.kde.org/mailman/listinfo/kde-devel#unsub to unsubscribe <<

[prev in list] [next in list] [prev in thread] [next in thread]