Review Request 120794: Implement Unique Mode Properly in Rekonq

[David Faure]

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/120794/#review69243
-----------------------------------------------------------



src/application.h
<https://git.reviewboard.kde.org/r/120794/#comment48403>

    should be const QStringList &arguments



src/application.cpp
<https://git.reviewboard.kde.org/r/120794/#comment48404>

    that's definitely in the wrong place, if it ever gets activated. You want session restoration only in main, not every time activateRequested is called.



src/application.cpp
<https://git.reviewboard.kde.org/r/120794/#comment48405>

    It's a bit confusing because QCLP::process() is not the same, and also because of the other processCmdLine method here. Maybe call this one handleCmdLine()? Sorry for nitpicking, but if we are going to replace this in most other apps, might as well be perfect :-)
    
    parse = just the parsing
    process = parse and handle the builtin options (in QLCP)
    handle = handle the app options
    
    As far as the slot is concerned, it's ok to say that Application::processCmdLine is parse + handle the app options, it's "builtin to Application". So I would rename this method, but leave the other one (processCmdLine(arguments)) as it is.



src/application.cpp
<https://git.reviewboard.kde.org/r/120794/#comment48409>

    Note that you might want to use the cwd in order to resolve urls with it, so that "rekonq localfile.html" works.
    This requires QUrl::fromUserInput(two args) from Qt 5.4 though, so probably too early to do right now, but you can at least add a comment, if not a #if QT_VERSION check.



src/main.cpp
<https://git.reviewboard.kde.org/r/120794/#comment48406>

    why not move this to the Application constructor too? i.e. do all the setup of the parser in a single place?
    
    I guess only the interaction with about stays here.
    
    Alternatively (and because not all apps derive from QApplication), I would just leave it all in main (where it is expected, in most apps), and do app.setParser(&parser).
    
    In addition to having all the setup in one place, the thing I'm trying to solve here is the "getCmdLineParser" method, which doesn't look so great in terms of encapsulation. getters/setters are always better than "returning a ref to an internal member variable".



src/main.cpp
<https://git.reviewboard.kde.org/r/120794/#comment48407>

    app.connect --> QObject::connect.
    You're using the 4-args connect, which is a static method.


- David Faure


On Oct. 28, 2014, 5:44 a.m., David Narváez wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/120794/
> -----------------------------------------------------------
> 
> (Updated Oct. 28, 2014, 5:44 a.m.)
> 
> 
> Review request for KDE Frameworks and rekonq.
> 
> 
> Repository: rekonq
> 
> 
> Description
> -------
> 
> This is my humble attempt to implement the Unique Mode properly. I have been trying to do this for the longest time in a way that avoids code duplication but I can't find a way to jump over all the hurdles these API impose. I tried learning from other ports from KUniqueApplication but a quick look at LXR shows there are plenty of applications that blindly ported to Unique Mode but didn't bother implementing activateRequested and the one I found that did was plasmawindowedcorona.cpp which does not need a QCommandLineParser, so the code duplication is less evident. At this point, I would like someone who knows about the QCommandLineParser + KDBusAddons dance to look at this and tell if it is reasonable or not.
> 
> The current patch just makes it possible to open several Rekonq applications. It does not do the right thing when a Rekonq window is already open in the current activity and a user clicks a link elsewhere (step 4 in the Testing Done section) because it starts a brand new Rekonq window, but that's a different patch. It also does some funky thing asking you if you want to restore the previous session when nothing has crashed, I have to check that.
> 
> 
> Diffs
> -----
> 
>   src/application.h 7ccd60d 
>   src/application.cpp c7c297d 
>   src/main.cpp 7592f7a 
> 
> Diff: https://git.reviewboard.kde.org/r/120794/diff/
> 
> 
> Testing
> -------
> 
> 1. Open one Rekonq window
> 2. Try opening Rekonq again
> 3. Try opeining Rekonq from a command line with some URLs
> 4. Assuming Reknoq is your default browser (why wouldn't it be?) click on a link somewhere (I click on the links at the title of the Konversation channels I am in)
> 5. Open rekonq from the console using rekonq --incognito
> 6. Open rekonq from the console using reknoq --webapp twitter.com
> 
> Before this patch, nothing happens in steps 2 - 6. After a first version of this patch that does not avoid the QCommandLine parser if the argument list is not empty, the window opened at 1 crashes because the activateRequested signal passes an empty list of arguments - not even the binary name - so QCommandLine parser dies. With this patch, every step opens a new window properly, step 5 opens an incognito window and step 6 opens a webapp window (simple window).
> 
> 
> Thanks,
> 
> David Narváez
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kde-frameworks-devel/attachments/20141028/9e6e0cb1/attachment-0001.html>