Bug 253398

In dolphin/trunk the search via KFind was replaced by a new searchbar, see http://ppenz.blogspot.com/2010/07/search-improvements.html.

But with this change of the dolphinpart (used by konqueror in filemanager mode) konqueror in trunk does not have an action find_files anymore :-(

i would prefere kfind instead of nepomuk :(

This is also true for kubuntu upgraded to KDE 4.6.1. This was one of the features I used ALL THE TIME in konqueror and now the only way is to use the command line since dolphin's file searching thing ignores *.php files.

i am ok with the service menu that i found here:
http://mageia.linuxtech.net/forum/index.php?t=msg&th=2404&start=0

Now, with right clic i open the kfind in the current folder

*** Bug 269219 has been marked as a duplicate of this bug. ***

Really bad, CTRL+F does also not work in konqueror and searchbars are only helpful for standard-users, nor for powerusers searching inside files with different/combined options

If the searchbar doesn't work in the KPart, then the lack of a Find menu entry qualifies as a regression for Konqueror.

In trunk compiled from sources, which will released as 4.7 in July, the menu entry Tools->Find File (Ctrl+F) is back in konquerors filemanager mode.

This bug seems to be fixed in 4.7, can anyone confirm?

The bug is fixed for Konqueror in 4.7 but it is not planned to add the entry in Dolphin.

Please can this be backported to 4.6.x where it was destroyed and future replacements of working things are done more carefully?

dolphin does not interest me because it is a filemanager for noobs but on the other hand why take them away a useful search? do the developers really think that a search like "recursive in folder a wehere file-extension is .txt and content contains hello" is not widely useful?

compared with kfind the new search in dolphin feels laughable and i would understand replace such a simple search with kfind but not in the other direction, this is the way apple goes - do we really need OSX on the linux-desktop?

> dolphin does not interest me because it is a filemanager
> for noobs but on the other hand why take them away a useful search? 

Most of us "noobs" are happy with the "laughable replacement" ;-)

Many of them if they are new users does it know other

The main question is do we really want the powerful desktop KDE 3.5.x was to reduce like GNOME and OSX instead leave KDE the main desktop for power-users?

(In reply to comment #11)
> > dolphin does not interest me because it is a filemanager
> > for noobs but on the other hand why take them away a useful search? 
> 
> Most of us "noobs" are happy with the "laughable replacement" ;-)

Your source being... ?

>> Most of us "noobs" are happy with the "laughable replacement" ;-)
>
> Your source being... ?

@Mike: From feedback I got per e-mails, by replies on my Dolphin-weblog, by watching users when they try to search something and by watching all bugs assigned to Dolphin and Konqueror. I consider it quite funny that two Konqueror users (you and Harald) seem to be concerned why the KFind-entry has been replaced by another approach in Dolphin.

The removing of the entry in Konqueror has been done by accident and I'm sorry for that. If you'd like to help in a constructive way please consider to test future beta-releases to help detecting regressions like this. Thanks.

> The removing of the entry in Konqueror has been done by accident and I'm sorry
> for that. If you'd like to help in a constructive way please consider to test
> future beta-releases to help detecting regressions like this. Thanks.

AGAIN:

kde-developers should stop changing existing behavior which works fine and force removing bugs as long KDE4 is wide away from the stability and bugfree-state of 3.5.x

Sample:
* Create "newfile" in folder-view
* try to renamed it to "newfile_renamed"
* it works but you get a error-msg that "newfile" does not exist

introduced in 4.6.x and sorry this are basic functions i think a bugreport is not a must because if somebody working on the code and make MINIMAL tests would see this - feels like some peopole are chaniging code on components they are never used by them self

the whole rewrite of sftp-kio as another example was the worst joke i have seen since KDE 4.0 (https://bugs.kde.org/show_bug.cgi?id=236025) and such a crap should NEVER see a release as long working thing replaced with it

@Peter, I am not concerned that kfind was replaced by another reproach. I think that the current reproach works fine for basic searches, however it is buggy when searching within files. This bug report is for a missing kfind menu entry (or shortcut entry) in konqueror/dolphin and that's why I'm here - not to get into an argument. I'm glad to see it has been fixed.

@Reindl, Please keep the comments relative to this bug report or create a new bug report for each issue. If you're just going to insult the developers, they're not going to want to help you. You're always welcome to use KDE 3.5 http://www.trinitydesktop.org

the same in KDE 4.6.5

it is frustrating that such bugs are introduced and only fixed in next major-releases a half year later - and that is why bad feeback comes often with a rough tone

> bugs are introduced and only fixed in next major-releases a half year later

Reindl,

What is your distribution? If you are using Kubuntu, then it was and is possible to install KDE 4.7.0 since july 29th like other Kubuntu users and I have:

http://www.kubuntu.org/news/kde-4.7

Reindl,

Right now, if my information is correct, there is only 1 developer left contributing and involved on KHTML rendering engine. The KHTML rendering engine for Konqueror certainly needs many more developers. [1] It is extremely easy to criticize, you know, with a harsch tone,... It is a very different story (much more challenging and difficult) to actually create and submit patches into complex C++ code. If you can do it, then why not get involved?

There are other ways to get involved too: 
http://techbase.kde.org/Contribute
http://techbase.kde.org/Contribute/Bugsquad
write a tutorial, do a video, pay someone to fix a particular bug, etc.

Generally speaking, I believe that KDE organization never ever had as much money, fund, human resources and technical resources as others (Microsoft,  Apple, Google, Mozilla) have had during the last 12 years. 

regards, Gérard
[1]: 
http://lists.kde.org/?l=kfm-devel&m=130480023503730&w=2
"
The main problem is that "development team" is in fact not existent. If you 
have looked in changelogs, commit digests, feature plans, latest bugs fixed, 
etc.. you have seen that on konqeror active works in last years at most 3-4 
people and this shorten to 1 person lately. 
"
http://lists.kde.org/?l=kfm-devel&m=130465893508494&w=2

[2]:
"
many users have the time to write articles to
complain about how much KDE sucks?!!?! blah blah, but cannot spare
even 1 hour a week to help the very same volunteer project they want
to cut to pieces. It gets really tyring after a while. Really. And I
have been doing this for over 12 years now.
"
http://lists.kde.org/?l=kfm-devel&m=130480062304783&w=2

IF IT AIN'T BROKEN DON'T FIX IT

IF YOU FIX THINGS WHICH WAS NOT BROKEN UNTIL THEY ARE BROKEN
FIX THEM BACK ASAP IF YOU DO NOT WANT COMPLAINTS

this way developemnt works and this way quality looks
independent freeware/paid development
if changes are breaking things which worked they have
to be reverted or improved without spit in the face of the 
users for months and years because nothing happens

(In reply to comment #18)
> > bugs are introduced and only fixed in next major-releases a half year later
> 
> Reindl,
> 
> What is your distribution? If you are using Kubuntu, then it was and is
> possible to install KDE 4.7.0 since july 29th like other Kubuntu users and I

Fedora and only with unofficeial repos i got 4.7 because it will NOT be 
released for F14/F15, so if you make new bugs you should not fix 
them only some major releases later

anyways - you will not try to tell me that i need a new operating
system because some stupid guy broke something in 4.6 and nobody
will fix it in 4.6?

> It is a very different story (much more challenging and difficult) to 
> actually create and submit patches into
> complex C++ code. If you can do it, then why not get involved?

so why do the developers do not left fuck in peace working
things instead fixing working things until they broken?

nobody needed any replacement of kfind - not in dolphin nor in konqueror

> Generally speaking, I believe that KDE organization never ever had as much
> money, fund, human resources and technical resources as others (Microsoft, 
> Apple, Google, Mozilla) have had during the last 12 years. 

breaking things with useless changes has nothing to do with money

> many users have the time to write articles to
> complain about how much KDE sucks?!!?! blah blah, but cannot spare
> even 1 hour a week to help the very same volunteer project they want
> to cut to pieces. It gets really tyring after a while. Really. And I
> have been doing this for over 12 years now

the problem is that 4.0 was released way too soon, communicated
as development release way too late so some distributions had
changed too much (fedora was one of them) and it took YEARS
to call KDE useable again, and now while things start to work 
instead optimizing them some stupid people are replacing
things without any thoughts what they are doing

* WHY is rename in folder-views inline instead as a popup?
  you will never see the full name in this inline crap
  why was this changed and optimized until it is broken

* https://bugs.kde.org/show_bug.cgi?id=270414
  how braindead can a developer act to not see this
  before the relase?

* https://bugs.kde.org/show_bug.cgi?id=236025
  what idiotic discussion because somebody started a rewrite
  and pushed it to stable-versions while the old kIO worked
  like a charm

* widgets can not be placed and resized on their corners
  they are simply a usibilty desaster, EVERYTIME i touch one
  this idiotic mouseover for move/resize of the neighbour
  widget steals my focus and the wrong is removed - nobody
  cares about usability

if there would not be permanently working things fixed until they 
are broken or have degraded usability there would not be so much
complaints like mine

IF IT AIN'T BROKEN DON'T FIX IT

Reindl,

I'm not going to try to convince you that your judgement is wrong, excessively harsh, very unfair and inappropriate. But I hope one day you get to know what is C++ (how fun C++ is!) and how many lines of code there is behind the whole Konqueror software. There's probably 1 million lines of code.

C++ is foremost about reusing code; defining functions once and then reusing those in as many components as possible. It's about shared libraries, reusing modules, defining classes and interfaces, prototyping, etc. So, if you make a change in a block or a module, it may, it can and it usually will have impacts (and unwanted, unknown, unpredicted side effects) elsewhere, in other blocks and modules not targeted.

A developer who is alone (this is the case here for KDE and the KHTML rendering engine) to tackle bugs and fix bugs, who has no experienced reviewer, no reviewer at all, no testers, no team of people to help is going to make mistakes, unfortunate mistakes, honest mistakes. He would make mistakes nevertheless and despite having reviewers, testers helping him. No developer can know very well and intimately 1 million lines of code that have been written and rewritten by dozens of people over the years and how such million lines of code will interact when making a 1 line change in it.

Now, regarding your "IF IT AIN'T BROKEN DON'T FIX IT", that's called regression bugs by professionals. And regression bugs happen often: that's the reality of modern technology. Happens often in professional development software done by IT corporations like Microsoft, Intel, HP, Dell, Apple, Google, etc. and in other technological areas too (eg car manufacturers, airplanes manufacturers, etc). I could name many famous regression bugs which have been widely documented in public. 

Regression happens so often that, for your information, webkit and bugzilla.mozilla even developed 2 keywords just to identify regression bugs: regression and regressionwindow-wanted. Microsoft also uses this too in their internal databases.

Just verify for yourself:

https://bugzilla.mozilla.org/describekeywords.cgi
identifies 27937 bug reports which are regression bugs 
https://bugs.webkit.org/buglist.cgi?keywords=Regression&resolution=---
and 443 regression bugs 
https://bugzilla.mozilla.org/buglist.cgi?keywords=regressionwindow-wanted&resolution=---
where they don't even know within which period of time and between which build numbers the regression occured by change in the code. That's over 28,000 situations (components, functionalities, features, etc) which were working accordingly, which were not broken and then broke.

https://bugs.webkit.org/describekeywords.cgi
identifies 2703 bug with the regression keywords: that is 2703 situations where "something was working accordingly and then it broke"

Gérard

well, i am software-developer and i know what regressions are

BUT i know the root of this regression
a idiotic change in dolphin to take away the extended search in dolphin and replace it with a useless inline-search - and exactly this changes are from the category "if it ain't broken don't fix it" and if there is less manpower why is there enough time to change such things to a bad behavior?

this whole change is simply useless and taking away features from the users

The worst is that this regression bug it is nt the result of some efforts of
developers to konqueror, but the totally missing of respect to konqueror and
the efforts and this crazy focus to dolphin.