Recently the Meson Build System gained some momentum. It is time to stop that.
Not that Meson is a bad piece of software - on the contrary, it is quite well designed.
Still it makes building C/C++ applications worse, by (quoting xkcd) basically creating this:

It sets out to create a cross-platform, more readable and faster alternative to autotools. But there is already CMake that solves this.

You might say that CMake is ugly, but note that the CMake 2.x you might have tried is not the same CMake 3.x that is available today. Many patterns have improved and are now both more logical and more readable.

Nowadays the difference between Meson and CMake is just a matter of syntactic preference. The Meson authors seem to agree here.

The actual criterion for selecting a build system however should be tooling support and community spread. CMake easily wins here:

After the introduction of the server mode it got native support by QtCreator, CLion, Android Studio (NDK) and even Microsofts Visual Studio. Native means that you do not have to generate any intermediate project files, but the CMakeLists.txt is used directly by the IDE.

On the community spread side we got e.g. KDE, OpenCV, zlib, libpng, freetype and as of recently Boost. These projects using CMake not only guarantees that you can easily use them, but that you can also include them in your build via add_subdirectory such that the become part of your project. This is especially useful if you are cross-compiling - for instance to a Raspberry Pi.

On the other hand, reinventing a wheel that is tailored to the needs of a specific community (Gnome), means that it will fall behind and eventually die. This is what is currently happening to the Vala language that had a similar birth to Meson.

The meson devs might object that Meson generates build files that run faster on a Raspberry Pi. However if your cross compiling is working you do not need that. And honestly, that particular improvement could have been also achieved by providing a patch to the CMake Ninja generator..

0 0

25 Jul 2017 2:05pm GMT

Dear Maemo community, I have the great honor of introducing the new Community Council for the upcoming Q2/2017 period.

**The members of the new council are (in alphabetical order):**

• Juiceme (Jussi Ohenoja)
• Mosen (Timo Könnecke)
• Sicelo (Sicelo Mhlongo)

The voting results can be seen on the [voting page]

I want to thank warmly all the members of the community who participated in this most important action of choosing a new council for us!

The new council shall meet on the #maemo-meeting IRC channel next tuesday 18.06 at 20:00 UTC for the formal handover with the passing council.

Jussi Ohenoja, On behalf of the outgoing Maemo Community Council

0 0

11 Jul 2017 8:17pm GMT

I'm at home now. I don't do non-public unpaid work. So let's blog the example I'm making for him.

workplace.h

#ifndef Workplace_H
#define Workplace_H

#include <QObject>
#include <QFuture>
#include <QWaitCondition>
#include <QMutex>
#include <QStack>
#include <QList>
#include <QThread>
#include <QFutureWatcher>

class Workplace;

typedef enum {
    WT_INSERTS,
    WT_QUERY
} WorkplaceWorkType;

typedef struct {
    WorkplaceWorkType type;
    QList<int> values;
    QString query;
    QFutureInterface<bool> insertIface;
    QFutureInterface<QList<QStringList> > queryIface;
} WorkplaceWork;

class WorkplaceWorker: public QThread {
    Q_OBJECT
public:
    WorkplaceWorker(QObject *parent = NULL)
        : QThread(parent), m_running(false) { }
    void run() Q_DECL_OVERRIDE;
    void pushWork(WorkplaceWork *a_work);
private:
    QStack<WorkplaceWork*> m_ongoing;
    QMutex m_mutex;
    QWaitCondition m_waitCondition;
    bool m_running;
};

class Workplace: public QObject {
    Q_OBJECT
public:
    explicit Workplace(QObject *a_parent=0) : QObject (a_parent) {}
    bool insert(QList<int> a_values);
    QList<QStringList> query(const QString &a_param);
    QFuture<bool> insertAsync(QList<int> a_values);
    QFuture<QList<QStringList> > queryAsync(const QString &a_param);
private:
    WorkplaceWorker m_worker;
};

class App: public QObject {
    Q_OBJECT
public slots:
    void perform();
    void onFinished();
private:
    Workplace m_workplace;
};

#endif// Workplace_H

workplace.cpp

#include "workplace.h"

void App::onFinished()
{
    QFutureWatcher<bool> *watcher = static_cast<QFutureWatcher<bool>* > ( sender() );
    delete watcher;
}

void App::perform()
{
    for (int i=0; i<10; i++) {
       QList<int> vals;
       vals.append(1);
       vals.append(2);
       QFutureWatcher<bool> *watcher = new QFutureWatcher<bool>;
       connect (watcher, &QFutureWatcher<bool>::finished, this, &App::onFinished);
       watcher->setFuture( m_workplace.insertAsync( vals ) );
    }

    for (int i=0; i<10; i++) {
       QList<int> vals;
       vals.append(1);
       vals.append(2);
       qWarning() << m_workplace.insert( vals );
       qWarning() << m_workplace.query("test");
    }
}

void WorkplaceWorker::pushWork(WorkplaceWork *a_work)
{
    if (!m_running) {
        start();
    }

    m_mutex.lock();
    switch (a_work->type) {
    case WT_QUERY:
        m_ongoing.push_front( a_work );
    break;
    default:
        m_ongoing.push_back( a_work );
    }
    m_waitCondition.wakeAll();
    m_mutex.unlock();
}

void WorkplaceWorker::run()
{
    m_mutex.lock();
    m_running = true;
    while ( m_running ) {
        m_mutex.unlock();
        m_mutex.lock();
        if ( m_ongoing.isEmpty() ) {
            m_waitCondition.wait(&m_mutex);
        }
        WorkplaceWork *work = m_ongoing.pop();
        m_mutex.unlock();

        // Do work here and report progress
        sleep(1);

        switch (work->type) {
        case WT_QUERY: {
            // Report result here
            QList<QStringList> result;
            QStringList row;
            row.append("abc"); row.append("def");
            result.append(row);
            work->queryIface.reportFinished( &result );
            } break;

        case WT_INSERTS:
        default: {
            // Report result here
            bool result = true;
            work->insertIface.reportFinished( &result );
            } break;
        }

        m_mutex.lock();
        delete work;
    }
    m_mutex.unlock();
}

bool Workplace::insert(QList<int> a_values)
{
    WorkplaceWork *work = new WorkplaceWork;;
    QFutureWatcher<bool> watcher;
    work->type = WT_INSERTS;
    work->values = a_values;
    work->insertIface.reportStarted();
    watcher.setFuture ( work->insertIface.future() );
    m_worker.pushWork( work );
    watcher.waitForFinished();
    return watcher.result();
}

QList<QStringList> Workplace::query(const QString &a_param)
{
    WorkplaceWork *work = new WorkplaceWork;
    QFutureWatcher<QList<QStringList> > watcher;
    work->type = WT_QUERY;
    work->query = a_param;
    work->queryIface.reportStarted();
    watcher.setFuture ( work->queryIface.future() );
    m_worker.pushWork( work );
    watcher.waitForFinished();
    return watcher.result();
}

QFuture<bool> Workplace::insertAsync(QList<int> a_values)
{
    WorkplaceWork *work = new WorkplaceWork;
    work->type = WT_INSERTS;
    work->values = a_values;
    work->insertIface.reportStarted();
    QFuture<bool> future = work->insertIface.future();
    m_worker.pushWork( work );
    return future;
}

QFuture<QList<QStringList> > Workplace::queryAsync(const QString &a_param)
{
    WorkplaceWork *work = new WorkplaceWork;
    work->type = WT_QUERY;
    work->query = a_param;
    work->queryIface.reportStarted();
    QFuture<QList<QStringList> > future = work->queryIface.future();
    m_worker.pushWork( work );
    return future;
}

0 0

06 Jul 2017 7:49pm GMT

Imagine we want an editor that has undo and redo capability. But the operations on the editor are all asynchronous. This implies that also undo and redo are asynchronous operations.

We want all this to be available in QML, we want to use QFuture for the asynchronous stuff and we want to use QUndoCommand for the undo and redo capability.

But how do they do it?

First of all we will make a status object, to put the status of the asynchronous operations in (asyncundoable.h).

class AbstractAsyncStatus: public QObject
{
    Q_OBJECT

    Q_PROPERTY(bool success READ success CONSTANT)
    Q_PROPERTY(int extra READ extra CONSTANT)
public:
    AbstractAsyncStatus(QObject *parent):QObject (parent) {}
    virtual bool success() = 0;
    virtual int extra() = 0;
};

We will be passing it around as a QSharedPointer, so that lifetime management becomes easy. But typing that out is going to give us long APIs. So let's make a typedef for that (asyncundoable.h).

typedef QSharedPointer<AbstractAsyncStatus> AsyncStatusPointer;

Now let's make ourselves an undo command that allows us to wait for asynchronous undo and asynchronous redo. We're combining QUndoCommand and QFutureInterface here (asyncundoable.h).

class AbstractAsyncUndoable: public QUndoCommand
{
public:
    AbstractAsyncUndoable( QUndoCommand *parent = nullptr )
        : QUndoCommand ( parent )
        , m_undoFuture ( new QFutureInterface<AsyncStatusPointer>() )
        , m_redoFuture ( new QFutureInterface<AsyncStatusPointer>() ) {}
    QFuture<AsyncStatusPointer> undoFuture()
        { return m_undoFuture->future(); }
    QFuture<AsyncStatusPointer> redoFuture()
        { return m_redoFuture->future(); }

protected:
    QScopedPointer<QFutureInterface<AsyncStatusPointer> > m_undoFuture;
    QScopedPointer<QFutureInterface<AsyncStatusPointer> > m_redoFuture;

};

Okay, let's implement these with an example operation. First the concrete status object (asyncexample1command.h).

class AsyncExample1Status: public AbstractAsyncStatus
{
    Q_OBJECT
    Q_PROPERTY(bool example1 READ example1 CONSTANT)
public:
    AsyncExample1Status ( bool success, int extra, bool example1,
                          QObject *parent = nullptr )
        : AbstractAsyncStatus(parent)
        , m_example1 ( example1 )
        , m_success ( success )
        , m_extra ( extra ) {}
    bool example1() { return m_example1; }
    bool success() Q_DECL_OVERRIDE { return m_success; }
    int extra() Q_DECL_OVERRIDE { return m_extra; }
private:
    bool m_example1 = false;
    bool m_success = false;
    int m_extra = -1;
};

Let's make a QUndoCommand that uses a timer to simulate asynchronous behavior. We could also use QtConcurrent's run function to use a QThreadPool and QRunnable instances that also implement QFutureInterface, of course. Seasoned Qt developers know what I mean. For the sake of example, I wanted to illustrate that QFuture can also be used for asynchronous things that aren't threads. We'll use the lambda because QUndoCommand isn't a QObject, so no easy slots. That's the only reason (asyncexample1command.h).

class AsyncExample1Command: public AbstractAsyncUndoable
{
public:
    AsyncExample1Command(bool example1, QUndoCommand *parent = nullptr)
        : AbstractAsyncUndoable ( parent ), m_example1(example1) {}
    void undo() Q_DECL_OVERRIDE {
        m_undoFuture->reportStarted();
        QTimer *timer = new QTimer();
        timer->setSingleShot(true);
        QObject::connect(timer, &QTimer::timeout, [=]() {
            QSharedPointer<AbstractAsyncStatus> result;
            result.reset(new AsyncExample1Status ( true, 1, m_example1 ));
            m_undoFuture->reportFinished(&result);
            timer->deleteLater();
        } );
        timer->start(1000);
    }
    void redo() Q_DECL_OVERRIDE {
        m_redoFuture->reportStarted();
        QTimer *timer = new QTimer();
        timer->setSingleShot(true);
        QObject::connect(timer, &QTimer::timeout, [=]() {
            QSharedPointer<AbstractAsyncStatus> result;
            result.reset(new AsyncExample1Status ( true, 2, m_example1 ));
            m_redoFuture->reportFinished(&result);
            timer->deleteLater();
        } );
        timer->start(1000);
    }
private:
    QTimer m_timer;
    bool m_example1;
};

Let's now define something we get from the strategy design pattern; a editor behavior. Implementations provide an editor all its editing behaviors (abtracteditorbehavior.h).

class AbstractEditorBehavior : public QObject
{
    Q_OBJECT
public:
    AbstractEditorBehavior( QObject *parent) : QObject (parent) {}

    virtual QFuture<AsyncStatusPointer> performExample1( bool example1 ) = 0;
    virtual QFuture<AsyncStatusPointer> performUndo() = 0;
    virtual QFuture<AsyncStatusPointer> performRedo() = 0;
    virtual bool canRedo() = 0;
    virtual bool canUndo() = 0;
};

So far so good, so let's make an implementation that has a QUndoStack and that therefor is undoable (undoableeditorbehavior.h).

class UndoableEditorBehavior: public AbstractEditorBehavior
{
public:
    UndoableEditorBehavior(QObject *parent = nullptr)
        : AbstractEditorBehavior (parent)
        , m_undoStack ( new QUndoStack ){}

    QFuture<AsyncStatusPointer> performExample1( bool example1 ) Q_DECL_OVERRIDE {
        AsyncExample1Command *command = new AsyncExample1Command ( example1 );
        m_undoStack->push(command);
        return command->redoFuture();
    }
    QFuture<AsyncStatusPointer> performUndo() {
        const AbstractAsyncUndoable *undoable =
            dynamic_cast<const AbstractAsyncUndoable *>(
                    m_undoStack->command( m_undoStack->index() - 1));
        m_undoStack->undo();
        return const_cast<AbstractAsyncUndoable*>(undoable)->undoFuture();
    }
    QFuture<AsyncStatusPointer> performRedo() {
        const AbstractAsyncUndoable *undoable =
            dynamic_cast<const AbstractAsyncUndoable *>(
                    m_undoStack->command( m_undoStack->index() ));
        m_undoStack->redo();
        return const_cast<AbstractAsyncUndoable*>(undoable)->redoFuture();
    }
    bool canRedo() Q_DECL_OVERRIDE { return m_undoStack->canRedo(); }
    bool canUndo() Q_DECL_OVERRIDE { return m_undoStack->canUndo(); }
private:
    QScopedPointer<QUndoStack> m_undoStack;
};

Now we only need an editor, right (editor.h)?

class Editor: public QObject
{
    Q_OBJECT
    Q_PROPERTY(AbstractEditorBehavior* editorBehavior READ editorBehavior CONSTANT)
public:
    Editor(QObject *parent=nullptr) : QObject(parent)
        , m_editorBehavior ( new UndoableEditorBehavior ) { }
    AbstractEditorBehavior* editorBehavior() { return m_editorBehavior.data(); }
    Q_INVOKABLE void example1Async(bool example1) {
        QFutureWatcher<AsyncStatusPointer> *watcher = new QFutureWatcher<AsyncStatusPointer>(this);
        connect(watcher, &QFutureWatcher<AsyncStatusPointer>::finished,
                this, &Editor::onExample1Finished);
        watcher->setFuture ( m_editorBehavior->performExample1(example1) );
    }
    Q_INVOKABLE void undoAsync() {
        if (m_editorBehavior->canUndo()) {
            QFutureWatcher<AsyncStatusPointer> *watcher = new QFutureWatcher<AsyncStatusPointer>(this);
            connect(watcher, &QFutureWatcher<AsyncStatusPointer>::finished,
                    this, &Editor::onUndoFinished);
            watcher->setFuture ( m_editorBehavior->performUndo() );
        }
    }
    Q_INVOKABLE void redoAsync() {
        if (m_editorBehavior->canRedo()) {
            QFutureWatcher<AsyncStatusPointer> *watcher = new QFutureWatcher<AsyncStatusPointer>(this);
            connect(watcher, &QFutureWatcher<AsyncStatusPointer>::finished,
                    this, &Editor::onRedoFinished);
            watcher->setFuture ( m_editorBehavior->performRedo() );
        }
    }
signals:
    void example1Finished( AsyncExample1Status *status );
    void undoFinished( AbstractAsyncStatus *status );
    void redoFinished( AbstractAsyncStatus *status );
private slots:
    void onExample1Finished() {
        QFutureWatcher<AsyncStatusPointer> *watcher =
                dynamic_cast<QFutureWatcher<AsyncStatusPointer>*> (sender());
        emit example1Finished( watcher->result().objectCast<AsyncExample1Status>().data() );
        watcher->deleteLater();
    }
    void onUndoFinished() {
        QFutureWatcher<AsyncStatusPointer> *watcher =
                dynamic_cast<QFutureWatcher<AsyncStatusPointer>*> (sender());
        emit undoFinished( watcher->result().objectCast<AbstractAsyncStatus>().data() );
        watcher->deleteLater();
    }
    void onRedoFinished() {
        QFutureWatcher<AsyncStatusPointer> *watcher =
                dynamic_cast<QFutureWatcher<AsyncStatusPointer>*> (sender());
        emit redoFinished( watcher->result().objectCast<AbstractAsyncStatus>().data() );
        watcher->deleteLater();
    }
private:
    QScopedPointer<AbstractEditorBehavior> m_editorBehavior;
};

Okay, let's register this up to make it known in QML and make ourselves a main function (main.cpp).

#include <QtQml>
#include <QGuiApplication>
#include <QQmlApplicationEngine>
#include <editor.h>
int main(int argc, char *argv[])
{
    QGuiApplication app(argc, argv);
    QQmlApplicationEngine engine;
    qmlRegisterType<Editor>("be.codeminded.asyncundo", 1, 0, "Editor");
    engine.load(QUrl(QStringLiteral("qrc:/main.qml")));
    return app.exec();
}

Now, let's make ourselves a simple QML UI to use this with (main.qml).

import QtQuick 2.3
import QtQuick.Window 2.2
import QtQuick.Controls 1.2
import be.codeminded.asyncundo 1.0
Window {
    visible: true
    width: 360
    height: 360
    Editor {
        id: editor
        onUndoFinished: text.text = "undo"
        onRedoFinished: text.text = "redo"
        onExample1Finished: text.text = "whoohoo " + status.example1
    }
    Text {
        id: text
        text: qsTr("Hello World")
        anchors.centerIn: parent
    }
    Action {
        shortcut: "Ctrl+z"
        onTriggered: editor.undoAsync()
    }
    Action {
        shortcut: "Ctrl+y"
        onTriggered: editor.redoAsync()
    }
    Button  {
        onClicked: editor.example1Async(99);
    }
}

You can find the sources of this complete example at github. Enjoy!

0 0

11 May 2017 8:09pm GMT

Dear friends and Maemoans. It is again the time for us to elect the new Community Council.

The schedule for the voting process is as follows:

• The nomination period starts next Monday, on the 1st of May 2017 and will continue until the 23rd of May 2017.
• The election starts on Thursday, on the 1st of June 2017 and will continue until the 7th of June 2017. In order for us to keep the community strong, we need to have new people with fresh ideas to carry on the torch. So, please consider volunteering for the position of Maemo Council.

  On behalf of the outgoing Community Council,

  eekkelund

0 0

30 Apr 2017 10:16am GMT

As mentioned in my Working on Android post, I've been using a mechanical keyboard for a couple of years now. Now that I work on Flowhub from home, it was a good time to re-evaluate the whole work setup. As far as regular keyboards go, the MiniLa was nice, but I wanted something more compact and ergonomic.

The Atreus keyboard

Atreus is a 40% ergonomic mechanical keyboard designed by Phil Hagelberg. It is an open hardware design, but he also sells kits for easier construction. From the kit introduction:

The Atreus is a small mechanical keyboard that is based around the shape of the human hand. It combines the comfort of a split ergonomic keyboard with the crisp key action of mechanical switches, all while fitting into a tiny profile.

My use case was also quite travel-oriented. I wanted a small keyboard that would enable me to work with it also on the road. There are many other small-ish DIY keyboard designs like Planck and Gherkin available, but Atreus had the advantage of better ergonomics. I really liked the design of the Ergodox keyboard, and Atreus essentially is that made mobile:

I found the split halves and relatively large size (which are fantastic for stationary use at a desk) make me reluctant to use it on the lap, at a coffee shop, or on the couch, so that's the primary use case I've targeted with the Atreus. It still has most of the other characteristics that make the Ergodox stand out, like mechanical Cherry switches, staggered columns instead of rows, heavy usage of the thumbs, and a hackable microcontroller with flexible firmware, but it's dramatically smaller and lighter

I had the opportunity to try a kit-built Atreus in the Berlin Mechanical Keyboard meetup, and it felt nice. It was time to start the project.

Sourcing the parts

When building an Atreus the first decision is whether to go with the kit or hand-wire it yourself. Building from a kit is certainly easier, but since I'm a member of a hackerspace, doing a hand-wired build seemed like the way to go.

To build a custom keyboard, you need:

• Switches: in my case 37 Cherry MX blues and 5 Cherry MX blacks
• Diodes: one 1N4148 per switch
• Microcontroller: a Arduino Pro Micro on my keyboard
• Keycaps: started with recycled ones and later upgraded to DSA blanks
• Case: got a set of laset-cut steel plates

Even though Cherry - the maker of the most common mechanical key switches - is a German company, it is quite difficult to get switches in retail here. Luckily a fellow hackerspace member had just dismantled some old mechanical keyboards, and so I was able to get the switches I needed via barter.

The Cherry MX blues are tactile clicky switches that feel super-nice to type on, but are quite loud. For modifiers I went with Cherry MX blacks that are linear. This way there is quite a clear difference in feel between keys you typically hold down compared to the ones you just press.

The diodes and the microcontroller I ordered from Amazon for about 20€ total.

At first I used a set of old keycaps that I got with the switches, but once the keyboard was up and running I upgraded to a very nice set of blank DSA-profile keycaps that I ordered from AliExpress for 30€. That set came with enough keycaps that I'll have myself covered if I ever build a second Atreus.

All put together, I think the parts ended up costing me around 100€ total.

Preparations

When I received all the parts, there were some preparation steps to be made. Since the key switches were 2nd hand, I had to start by dismantling them and removing old diodes that had been left inside some of them.

The keycaps I had gotten with the switches were super grimy, and so I ended up sending them to the washing machine. After that you could see that they were not new, but at least they were clean.

With the steel mounting plate there had been a slight misunderstading, and the plates I received were a few millimeters thicker than needed, so the switches wouldn't "click" in place. While this could've been worked around with hot glue, we ended up filing the mounting holes down to the right thickness.

Wiring the keyboard

Once the mounting plate was in the right shape, I clicked the switches in and it was time to solder.

Hand-wiring keyboards is not that tricky. You have to attach a diode to each keyswitch, and then connect each row together via the diodes.

The two thumb keys are wired to be on the same column, but different rows.

Then each column is connected together via the other pin on the switches.

This is how the matrix looks like:

After these are done, connect a wire from each column, and each row to a I/O pin on the microcontroller.

If you haven't done it earlier, this is a good stage to test all connections with a multimeter!

Firmware

After finishing the wiring, I downloaded the QMK firmware, changed the PIN mapping for how my Atreus is wired up, switched the layout to Colemak, and the keyboard was ready to go.

Don't mind the key labels in the picture above. These are the second-hand keycaps I started with. Since then I've switched to blank ones.

USB-C

The default Atreus design has the USB cable connected directly to the microcontroller, meaning that you'll have to open the case to change the cable. To mitigate that I wanted to add a USB breakout board to the project, and this being 2017, it felt right to go with USB-C.

I found some cheap USB-C breakout boards from AliExpress. Once they arrived, it was time to figure out how the spec works. Since USB-C is quite new, there are very few resources available on how to use it with microcontrollers. These tutorials were quite helpful:

• Using USB-C on hobbyist projects
• USB Type C in a Micro-B world

Here is how we ended up wiring the breakout board. After these you only have four wires to connect to the microcontroller: ground, power, and the positive and negative data pins.

This Atreus build log was useful for figuring out where to connect the USB wires on the Pro Micro. Once all was done, I had a custom, USB-C keyboard!

Next steps

Now I have the Atreus working nicely on my new standing desk. Learning Colemak is a bit painful, but the keyboard itself feels super nice!

However, I'd still like to CNC mill a proper wooden case for the keyboard. I may update this post once that happens.

I'm also considering to order an Atreus kit so I'd have a second, always packed for travel keyboard. The kit comes with a PCB, which might work better at airport security checks than the hand-wired build.

Another thing that is quite tempting is to make a custom firmware with MicroFlo. I have no complaints on how QMK works, but it'd be super-cool to use our visual programming tool to tweak the keyboard live.

Big thanks to Technomancy for the Atreus design, and to XenGi for all the help during the build!

0 0

20 Apr 2017 12:00am GMT

Combining QFuture with QUndoCommand made a lot of sense for us. The undo and the redo methods of the QUndoCommand can also be asynchronous, of course. We wanted to use QFuture without involving threads, because our asynchronosity is done through a process and IPC, and not a thread. It's the design mistake of QtConcurrent's run method, in my opinion. That meant using QFutureInterface instead (which is undocumented, but luckily public - so it'll remain with us until at least Qt's 6.y.z releases).

So how do we make a QUndoCommand that has a undo, and that has a redo method that returns a asynchronous QFuture<ResultType>?

We just did that, today. I'm very satisfied with the resulting API and design. It might have helped if QUndoStack would be a QUndoStack<T> and QUndoCommand would have been a QUndoCommand<T> with undo and redo's return type being T. Just an idea for the Qt 6.y.z developers.

0 0

13 Apr 2017 9:32pm GMT

Yes, I'm looking for a job. :-)

These six years I've spent at Canonical have literally been flying. I enjoyed my work from the very first day, when I was assigned to the Unity 2D team, developing a lightweight desktop environment for Ubuntu, though I stayed in that team just for a few weeks. The next task, which I've been carrying on till today, has been implementing the Online Accounts feature in Ubuntu; this project has been especially dear to me, given that I got to reuse and improve much of the work we developed for the Nokia N9 phone. Seeing it being adopted also by Sailfish OS and KDE has been a major satisfaction, and a proof that we were on the right track. And indeed, porting the UI to Qt/QML for running in Unity 8, plus extending and simplifying the APIs and helping with the development of client applications has been a fantastic ride.
In the times where calm was reigning in the project, I reached out to other teams and offered help, mainly for improving the geolocation service and the webapps project.

Unfortunately, with the decision to terminate the development of Unity8 and to set aside the convergence goals, all of the above is no longer relevant for Canonical's future and I, along many other developers, have left the company.

So, here's my CV.

Given that reading is boring, here's a few pictures (and even a video!) of programs I've done, not as part of my daily work but in my spare time; though, to be honest, I do enjoy middleware and logic development (and even kernel, though I got little chances to work on that so far) more than UI development:

Imaginario on the Ubuntu phone

Imaginario for your desktop (under development)

Mappero Geotagger

If you wish to see my code, please have a look at my gitlab, github and launchpad accounts.

0 0

10 Apr 2017 7:07pm GMT

Among the problems we'll face is that we want asynchronous APIs that are undoable and that we want to switch to read only, undoable editing, non-undoable editing and that QML doesn't really work well with QFuture. At least not yet. We want an interface that is easy to talk with from QML. Yet we want to switch between complicated behaviors.

We will also want synchronous mode and asynchronous mode. Because I just invented that requirement out of thin air.

Ok, first the "design". We see a lot of behaviors, for something that can do something. The behaviors will perform for that something, the actions it can do. That is the strategy design pattern, then. It's the one about ducks and wing fly behavior and rocket propelled fly behavior and the ostrich that has a can't fly behavior. For undo and redo, we have the command pattern. We have this neat thing in Qt for that. We'll use it. We don't reinvent the wheel. Reinventing the wheel is stupid.

Let's create the duck. I mean, the thing-editor as I will use "Thing" for the thing that is being edited. We want copy (sync is sufficient), paste (must be aysnc), and edit (must be async). We could also have insert and delete, but those APIs would be just like edit. Paste is usually similar to insert, of course. Except that it can be a combined delete and insert when overwriting content. The command pattern allows you to make such combinations. Not the purpose of this example, though.

Enough explanation. Let's start! The ThingEditor, is like the flying Duck in strategy. This is going to be more or less the API that we will present to the QML world. It could be your ViewModel, for example (ie. you could let your ThingViewModel subclass ThingEditor).

class ThingEditor : public QObject
{
    Q_OBJECT

    Q_PROPERTY ( ThingEditingBehavior* editingBehavior READ editingBehavior
                 WRITE setEditingBehavior NOTIFY editingBehaviorChanged )
    Q_PROPERTY ( Thing* thing READ thing WRITE setThing NOTIFY thingChanged )

public:
    explicit ThingEditor( QSharedPointer<Thing> &a_thing,
            ThingEditingBehavior *a_editBehavior,
            QObject *a_parent = nullptr );

    explicit ThingEditor( QObject *a_parent = nullptr );

    Thing* thing() const { return m_thing.data(); }
    virtual void setThing( QSharedPointer<Thing> &a_thing );
    virtual void setThing( Thing *a_thing );

    ThingEditingBehavior* editingBehavior() const { return m_editingBehavior.data(); }
    virtual void setEditingBehavior ( ThingEditingBehavior *a_editingBehavior );

    Q_INVOKABLE virtual void copyCurrentToClipboard ( );
    Q_INVOKABLE virtual void editCurrentAsync( const QString &a_value );
    Q_INVOKABLE virtual void pasteCurrentFromClipboardAsync( );

signals:
    void editingBehaviorChanged ();
    void thingChanged();
    void editCurrentFinished( EditCurrentCommand *a_command );
    void pasteCurrentFromClipboardFinished( EditCurrentCommand *a_command );

private slots:
    void onEditCurrentFinished();
    void onPasteCurrentFromClipboardFinished();

private:
    QScopedPointer<ThingEditingBehavior> m_editingBehavior;
    QSharedPointer<Thing> m_thing;
    QList<QFutureWatcher<EditCurrentCommand*> *> m_editCurrentFutureWatchers;
    QList<QFutureWatcher<EditCurrentCommand*> *> m_pasteCurrentFromClipboardFutureWatchers;
};

For the implementation of this class, I'll only provide the non-obvious pieces. I'm sure you can do that setThing, setEditingBehavior and the constructor yourself. I'm also providing it only once, and also only for the EditCurrentCommand. The one about paste is going to be exactly the same.

void ThingEditor::copyCurrentToClipboard ( )
{
    m_editingBehavior->copyCurrentToClipboard( );
}

void ThingEditor::onEditCurrentFinished( )
{
    QFutureWatcher<EditCurrentCommand*> *resultWatcher
            = static_cast<QFutureWatcher<EditCurrentCommand*>*> ( sender() );
    emit editCurrentFinished ( resultWatcher->result() );
    if (m_editCurrentFutureWatchers.contains( resultWatcher )) {
        m_editCurrentFutureWatchers.removeAll( resultWatcher );
    }
    delete resultWatcher;
}

void ThingEditor::editCurrentAsync( const QString &a_value )
{
    QFutureWatcher<EditCurrentCommand*> *resultWatcher
            = new QFutureWatcher<EditCurrentCommand*>();
    connect ( resultWatcher, &QFutureWatcher<EditCurrentCommand*>::finished,
              this, &ThingEditor::onEditCurrentFinished, Qt::QueuedConnection );
    resultWatcher->setFuture ( m_editingBehavior->editCurrentAsync( a_value ) );
    m_editCurrentFutureWatchers.append ( resultWatcher );
}

For QUndo we'll need a QUndoCommand. For each undoable action we indeed need to make such a command. You could add more state and pass it to the constructor. It's common, for example, to pass Thing, or the ThingEditor or the behavior (this is why I used QSharedPointer for those: as long as your command lives in the stack, you'll need it to hold a reference to that state).

class EditCurrentCommand: public QUndoCommand
{
public:
    explicit EditCurrentCommand( const QString &a_value,
                                 QUndoCommand *a_parent = nullptr )
        : QUndoCommand ( a_parent )
        , m_value ( a_value ) { }
    void redo() Q_DECL_OVERRIDE {
       // Perform action goes here
    }
    void undo() Q_DECL_OVERRIDE {
      // Undo what got performed goes here
    }
private:
    const QString &m_value;
};

You can (and probably should) also make this one abstract (and/or a so called pure interface), as you'll usually want many implementations of this one (one for every kind of editing behavior). Note that it leaks the QUndoCommand instances unless you handle them (ie. storing them in a QUndoStack). That in itself is a good reason to keep it abstract.

class ThingEditingBehavior : public QObject
{
    Q_OBJECT

    Q_PROPERTY ( ThingEditor* editor READ editor WRITE setEditor NOTIFY editorChanged )
    Q_PROPERTY ( Thing* thing READ thing NOTIFY thingChanged )

public:
    explicit ThingEditingBehavior( ThingEditor *a_editor,
                                   QObject *a_parent = nullptr )
        : QObject ( a_parent )
        , m_editor ( a_editor ) { }

    explicit ThingEditingBehavior( QObject *a_parent = nullptr )
        : QObject ( a_parent ) { }

    ThingEditor* editor() const { return m_editor.data(); }
    virtual void setEditor( ThingEditor *a_editor );
    Thing* thing() const;

    virtual void copyCurrentToClipboard ( );
    virtual QFuture<EditCurrentCommand*> editCurrentAsync( const QString &a_value, bool a_exec = true );
    virtual QFuture<EditCurrentCommand*> pasteCurrentFromClipboardAsync( bool a_exec = true );

protected:
    virtual EditCurrentCommand* editCurrentSync( const QString &a_value, bool a_exec = true );
    virtual EditCurrentCommand* pasteCurrentFromClipboardSync( bool a_exec = true );

signals:
    void editorChanged();
    void thingChanged();

private:
    QPointer<ThingEditor> m_editor;
    bool m_synchronous = true;
};

That setEditor, the constructor, etc: these are too obvious to write here. Here are the non-obvious ones:

void ThingEditingBehavior::copyToClipboard ( )
{
}

EditCurrentCommand* ThingEditingBehavior::editCurrentSync( const QString &a_value, bool a_exec )
{
    EditCurrentCommand *ret = new EditCurrentCommand ( a_value );
    if ( a_exec )
        ret->redo();
    return ret;
}

QFuture<EditCurrentCommand*> ThingEditingBehavior::editCurrentAsync( const QString &a_value, bool a_exec )
{
    QFuture<EditCurrentCommand*> resultFuture =
            QtConcurrent::run( QThreadPool::globalInstance(), this,
                               &ThingEditingBehavior::editCurrentSync,
                               a_value, a_exec );
    if (m_synchronous)
        resultFuture.waitForFinished();
    return resultFuture;
}

And now we can make the whole thing undoable by making a undoable editing behavior. I'll leave a non-undoable editing behavior as an exercise to the reader (ie. just perform redo() on the QUndoCommand, don't store it in the QUndoStack and immediately delete or cmd->deleteLater() the instance).

Note that if m_synchronous is false, that (all access to) m_undoStack, and the undo and redo methods of your QUndoCommands, must be (made) thread-safe. The thread-safety is not the purpose of this example, though.

class UndoableThingEditingBehavior : public ThingEditingBehavior
{
    Q_OBJECT
public:
    explicit UndoableThingEditingBehavior( ThingEditor *a_editor,
                                           QObject *a_parent = nullptr );
protected:
    EditCellCommand* editCurrentSync( const QString &a_value, bool a_exec = true ) Q_DECL_OVERRIDE;
    EditCurrentCommand* pasteCurrentFromClipboardSync( bool a_exec = true ) Q_DECL_OVERRIDE;
private:
    QScopedPointer<QUndoStack> m_undoStack;
};

EditCellCommand* UndoableThingEditingBehavior::editCurrentSync( const QString &a_value, bool a_exec )
{
    Q_UNUSED(a_exec)
    EditCellCommand *undoable = ThingEditingBehavior::editCurrentSync(  a_value, false );
    m_undoStack->push( undoable );
    return undoable;
}

EditCellCommand* UndoableThingEditingBehavior::pasteCurrentFromClipboardSync( bool a_exec )
{
    Q_UNUSED(a_exec)
    EditCellCommand *undoable = ThingEditingBehavior::pasteCurrentFromClipboardSync( false );
    m_undoStack->push( undoable );
    return undoable;
}

0 0

24 Mar 2017 9:42am GMT

Perfection has been reached not when there is nothing left to add, but when there is nothing left to take away.

0 0

23 Mar 2017 12:17am GMT

A lot of good things have happened to the Media Source Extensions support since my last post, almost a year ago.

The most important piece of news is that the code upstreaming has kept going forward at a slow, but steady pace. The amount of code Igalia had to port was pretty big. Calvaris (my favourite reviewer) and I considered that the regular review tools in WebKit bugzilla were not going to be enough for a good exhaustive review. Instead, we did a pre-review in GitHub using a pull request on my own repository. It was an interesting experience, because the change set was so large that it had to be (artificially) divided in smaller commits just to avoid reaching GitHub diff display limits.

394 GitHub comments later, the patches were mature enough to be submitted to bugzilla as child bugs of Bug 157314 - [GStreamer][MSE] Complete backend rework. After some comments more in bugzilla, they were finally committed during Web Engines Hackfest 2016:

Some unforeseen regressions in the layout tests appeared, but after a couple of commits more, all the mediasource WebKit tests were passing. There are also some other tests imported from W3C, but I kept them still skipped because webm support was needed for many of them. I'll focus again on that set of tests at its due time.

Igalia is proud of having brought the MSE support up to date to WebKitGTK+. Eventually, this will improve the browser video experience for a lot of users using Epiphany and other web browsers based on that library. Here's how it enables the usage of YouTube TV at 1080p@30fps on desktop Linux:

Our future roadmap includes bugfixing and webm/vp9+opus support. This support is important for users from countries enforcing patents on H.264. The current implementation can't be included in distros such as Fedora for that reason.

As mentioned before, part of this upstreaming work happened during Web Engines Hackfest 2016. I'd like to thank our sponsors for having made this hackfest possible, as well as Metrological for giving upstreaming the importance it deserves.

Thank you for reading.

0 0

20 Mar 2017 11:55am GMT

Imagine you have a duck. Imagine you have a wall. Now imagine you throw the duck with a lot of force against a wall. Duck typing means that the duck hitting the wall quacks like a duck would.

ps. Replace wall with API and duck with ugly stupid script written by an idiot. You can leave quacks.

0 0

17 Mar 2017 10:34am GMT

First one should ask though: why? My main motivation was that many of the apps I use were easily available in the nextcloud store, while with owncloud I had to manually pull them from github.
Additionally some of the app authors migrated to nextcloud and did not provide further updates for owncloud.

Another reason is this:

the graphs above show the number of commits for owncloud and nextcloud. Owncloud has taken a very noticeable hit here after the fork - even though they deny it.

From the user perspective the lack of contribution is visible for instance in the admin interface where with nextcloud you get a nice log browser and system stats while with owncloud you do not. Furthermore the nextcloud android app handles Auto-Upload much better and generally seems more polished - I think one can expect nextcloud to advance faster in general.

Migrating

For migrating you can follow the excellent instructions of Jos Poortvliet.

In my case owncloud 9.1 was installed on Ubuntu in /var/www/owncloud and I put nextcloud 11 to /var/www/nextcloud. Then the following steps had to be applied:

1. put owncloud in maintenance mode

   sudo -u www-data php occ maintenance:mode --on
   

2. copy over the config.php

   cp /var/www/owncloud/config/config.php /var/www/nextcloud/config/
   

3. adapt the path in config.php

   # from 
   'path' => '/var/www/owncloud/apps',
   # to
   'path' => '/var/www/nextcloud/apps',
   

4. adapt the path in crontab

   sudo crontab -u www-data -e
   

5. adapt the paths in the apache config
6. run the upgrade script which takes care of the actual migration. Then disable the maintanance mode.

   sudo -u www-data php occ upgrade
   sudo -u www-data php occ maintenance:mode --off
   

and thats it.

1 0

10 Feb 2017 11:33pm GMT

Back in 2013 I was working exclusively on an Android tablet. Then with the NoFlo Kickstarter I needed a device with a desktop browser. What followed were brief periods working on a Chromebook, on a 12" MacBook, and even an iPad Pro.

But from April 2016 onwards I've been again working with an Android device. Some people have asked me about my setup, and so here is an update.

Why work on a tablet?

When I started on this path in 2013, using a tablet for "real work" was considered crazy. While every story on tablet productivity still brings out the people claiming it is not a real computer for real work, using tablets for real work is becoming more and more common.

A big contributor to this has been the plethora of work-oriented tablets and convertibles released since then. Microsoft's popular Surface Pro line brought the PC to tablet form factor, and Apple's iPad Pro devices gave the iPad a keyboard.

Here are couple of great posts talking about how it feels to work on an iPad:

• Rediscovering the iPad
• When traveling, my iPad is essential and my Mac is the add-on
• A Computer for Everything: One Year of iPad Pro
• Evolving iPad Desktop Usage
• My Tablet Has Stickers

With all the activity going on, one could claim using a tablet for work has been normalized. But why work on a tablet instead of a "real computer"? Here are some reasons, at least for me:

Free of legacy cruft

Desktop operating systems have become clunky. Window management. File management. Multiple ways to discover, install, and uninstall applications. Broken notification mechanisms.

With a tablet you can bypass pretty much all of that, and jump into a simpler, cleaner interface designed for the modern connected world.

I think this is also the reason driving some developers back to Linux and tiling window managers - cutting manual tweaking and staying focused.

Amazing endurance

Admittedly, laptop battery life has increased a lot since 2013. But with some manufacturers using this an excuse to ship thinner devices, tablets still win the endurance game.

With my current work tablet, I'm customarily getting 12 or more hours of usage. This means I can power through the typical long days of a startup founder without having to plug in. And when traveling, I really don't have to care where power sockets are located on trains, airplanes, and conference centers.

Low power usage also means that I can really get a lot of more runtime by utilizing the mobile battery pack I originally bought to use with my phone. While I've never actually had to try this, back-of-the-envelope math claims I should be able to get a full workweek from the combo without plugging in.

Work and play

The other aspect of using a tablet is that it becomes a very nice content consumption device after I'm done working. Simply disconnect the keyboard and lean back, and the same device you used for writing software becomes a great e-reader, video player, or a gaming machine.

This combined with the battery life has meant that I've actually stopped carrying a Kindle with me. While an e-ink screen is still nicer to read, not needing an extra device has its benefits, especially for a frequent one-bag traveller.

The setup

I'm writing this on a Pixel C, a 10.2" Android tablet made by Google. I got the device last spring when there were developer discounts available at ramp-up to the Android 7 release, and have been using it full-time since.

Software

Surprisingly little has changed in my software use since 2013 - I still spend the most of the time writing software in either Flowhub or terminal. Here are the apps I use on daily basis:

• JuiceSSH for mosh access to my remote development servers
• Termux for local and offline development
• Flowhub for visual programming
• The Grid for updating my various websites
• Slack, Hangouts, and Inbox by Gmail for communications
• Google Drive and the associated applications for budgeting and planning
• Chrome for web

Looking back to the situation in early 2013, the biggest change is that Slack has pretty much killed work email.

Termux is a new app that has done a lot to improve the local development situation. By starting the app you get a very nice Linux chroot environment where a lot of software is only a quick apt install away.

Since much of my non-Flowhub work is done in tmux and vim, I get the exactly same working environment on both local chroot and cloud machines by simply installing my dotfiles on each of them.

Keyboard

When I'm on the road I'm using the Pixel C keyboard. This doubles as a screen protector, and provides a reasonable laptop-like typing environment. It attaches to the tablet with very strong magnets and allows a good amount of flexibility on the screen angles.

However, when stationary, no laptop keyboard compares to a real mechanical keyboard. When I'm in the office I use a Filco MiniLa Air, a bluetooth keyboard with quiet-ish Cherry MX brown switches.

This tenkeyless (60%) keyboard is extremely comfortable to type on. However, the sturdy metal case means that it is a little too big and heavy to carry on a daily basis.

In practice I've only taken the mechanical keyboard with me when there has been a longer trip where I know that I'll be doing a lot of typing. To solve this, I'm actually looking to build a more compact custom mechanical keyboard so I could always have it with me. (Update: here is the keyboard I built).

Comparison with iOS

So, why work on Android instead of getting an iPad Pro? I've actually worked on both, and here are my reasons:

• Communications between apps: while iOS has extensions now, the ability to send data from an app to another is still a hit-or-miss. Android had intents from day one, meaning pretty much any app can talk to any other app
• Standard charging: all of my other devices charge with the same USB-C chargers and cables. iPads still use the proprietary Lightnight plug, requiring custom dongles for everything
• Standard accessories: this boils down to USB-C just like charging. With Android I can plug in a network adapter or even a mouse, and it'll just work
• Ecosystem lock-in: we're moving to a world where everything - from household electronics to cars - is either locked to the Apple ecosystem or following standards. I don't want to be locked to a single vendor for everything digital
• Browser choice: with iOS you only get one web renderer, the rather dated Safari. On Android I can choose between Chrome, Firefox, or any other browser that has been ported to the platform

Of course, iOS has its own benefits. Apple has a stronger stance on privacy than Google. And there is more well-made tablet software available for iPads than Android. But when almost everything I use is available on the web, this doesn't matter that much.

The future

As a software developer working on Android tablets, the weakest point of the platform is still that there are no browser developer tools available. This was a problem in 2013, and it is still a problem now.

From my conversations with some Chrome developers, it seems Google has very little interest in addressing this. However, there is a bright spot: the new breed of convertible Chromebooks being released now. And they run Android apps:

• How Chromebooks Are About to Totally Transform Laptop Design
• Detachable Chromebooks, Pixel C And The Future Of Chrome OS
• Samsung's Chromebook Pro gives me hope in Chrome OS-thanks to Android's help

Chrome OS is another clean, legacy free, modern computing interface. With these new devices you get the combination of a full desktop browser and the ability to run all Android tablet software.

The Samsung Chromebook Pro/Plus mentioned above is definitely interesting. A high-res 12" screen and a digital pen which I see as something very promising for visual programming purposes.

However, given that I already have a great mechanical keyboard, I'd love a device that shipped without an attached keyboard. We'll see what kind of devices get out later this year.

0 0

10 Feb 2017 12:00am GMT

MyPlugin/MyPlugin.cpp:

#include <ViewModels/MyListClass.h>
#include <ViewModels/DisplayViewModel.h>

qmlRegisterUncreatableType<MyListClass>( a_uri, 1, 0, "MyListClass",
         "Use access via DisplayViewModel instead");
qmlRegisterType<DisplayViewModel>( a_uri, 1, 0, "DisplayViewModel");

Utils/MyQMLListUtils.h

#define MY_DECLARE_QML_LIST(type, name, owner, prop) \
QQmlListProperty<type> name(){ \
   return QQmlListProperty<type>( \
               this, 0,&owner::count ## type ## For ## name ## List, \
               &owner::at ## type ## For ## name ## List); \
} \
static int count ## type ## For ## name ## List(QQmlListProperty<type>*property){ \
   owner *m = qobject_cast<owner *>(property->object); \
   return m->prop.size(); \
} \
static type *at ## type ## For ## name ## List( \
        QQmlListProperty<type>*property, int index){ \
   owner *m = qobject_cast<owner *>(property->object); \
   return m->prop[index]; \
}

ViewModels/DisplayViewModel.h

#ifndef DISPLAYVIEWMODEL_H
#define DISPLAYVIEWMODEL_H

#include <QObject>
#include <QtQml>
#include <ViewModels/MyListClass.h>
#include <Utils/MyQMLListUtils.h>

class DisplayViewModel : public QObject
{
    Q_OBJECT

    Q_PROPERTY(constQString title READ title WRITE setTitle NOTIFY titleChanged )
    Q_PROPERTY(constQList<MyListClass*> objects READ objects
                                          NOTIFY objectsChanged ) 
    Q_PROPERTY( QQmlListProperty<MyListClass> objectList READ objectList
                                              NOTIFY objectsChanged )
public:
    explicit DisplayViewModel( QObject *a_parent = nullptr );
    explicit DisplayViewModel( const QString &a_title,
                               QList<MyListClass*> a_objects,
                               QObject *a_parent = nullptr );
    const QString title()
        { return m_title; }
    void setTitle( const QString &a_title ); 
    const QList<MyListClass*> objects ()
        { return m_objects; } 
    Q_INVOKABLE void appendObject( MyListClass *a_object);
    Q_INVOKABLE void deleteObject( MyListClass *a_object);
    Q_INVOKABLE void reset( );

protected:
    MY_DECLARE_QML_LIST(MyListClass, objectList, DisplayViewModel, m_objects)

signals:
    void titleChanged();
    void objectsChanged();

private:
    QString m_title;
    QList<MyListObject*> m_objects;
};

#endif// DISPLAYVIEWMODEL_H

DisplayViewModel.cpp

#include "DisplayViewModel.h"

DisplayViewModel::DisplayViewModel( const QString &a_title,
                                    QList<MyListClass*> a_objects,
                                    QObject *a_parent )
    : QObject ( a_parent )
    , m_title ( a_title )
    , m_objects ( a_objects )
{
    foreach (MyListClass* mobject, m_objects) {
        mobject->setParent (this);
    }
}

void DisplayViewModel::setTitle (const QString &a_title )
{
    if ( m_title != a_title ) {
        m_title = a_title;
        emit titleChanged();
    }
}

void DisplayViewModel::reset( )
{
    foreach ( MyListClass *mobject, m_objects ) {
        mobject->deleteLater();
    }
    m_objects.clear();
    emit objectsChanged();
}

void DisplayViewModel::appendObject( MyListClass *a_object )
{
    a_object->setParent( this );
    m_objects.append( a_object );
    emit objectsChanged();
}

void DisplayViewModel::deleteObject( MyListClass *a_object )
{
    if (m_objects.contains( a_object )) {
        m_objects.removeOne( a_object );
        a_object->deleteLater();
        emit objectsChanged();
    }
}

Tester.cpp

#include <ViewModels/DisplayViewModel.h>
#include <ViewModels/MyListClass.h>

QList<MyListClass*> objectList;
for( int i = 0; i < 100 ; ++i ) {
    objectList.append ( new MyListClass (i) );
}
DisplayViewModel *viewModel = new DisplayViewModel (objectList);
viewModel->appendObject ( new MyListClass (101) );

Display.qml

import QtQuick 2.5
import MyPlugin 1.0

Repeater { 
    property DisplayViewModel viewModel: DisplayViewModel { } 
    model: viewModel.objectList
    delegate: Item {
        property MyListClass object: modelData
        Text {
            text: object.property
        }
    }
}

0 0

09 Feb 2017 8:07pm GMT

When choosing a disk image format for your virtual machine one of the factors to take into considerations is its I/O performance. In this post I'll talk a bit about the internals of qcow2 and about one of the aspects that can affect its performance under QEMU: its consistency checks.

As you probably know, qcow2 is QEMU's native file format. The first thing that I'd like to highlight is that this format is perfectly fine in most cases and its I/O performance is comparable to that of a raw file. When it isn't, chances are that this is due to an insufficiently large L2 cache. In one of my previous blog posts I wrote about the qcow2 L2 cache and how to tune it, so if your virtual disk is too slow, you should go there first.

I also recommend Max Reitz and Kevin Wolf's qcow2: why (not)? talk from KVM Forum 2015, where they talk about a lot of internal details and show some performance tests.

qcow2 clusters: data and metadata

A qcow2 file is organized into units of constant size called clusters. The cluster size defaults to 64KB, but a different value can be set when creating a new image:

qemu-img create -f qcow2 -o cluster_size=128K hd.qcow2 4G

Clusters can contain either data or metadata. A qcow2 file grows dynamically and only allocates space when it is actually needed, so apart from the header there's no fixed location for any of the data and metadata clusters: they can appear mixed anywhere in the file.

Here's an example of what it looks like internally:

In this example we can see the most important types of clusters that a qcow2 file can have:

• Header: this one contains basic information such as the virtual size of the image, the version number, and pointers to where the rest of the metadata is located, among other things.
• Data clusters: the data that the virtual machine sees.
• L1 and L2 tables: a two-level structure that maps the virtual disk that the guest can see to the actual location of the data clusters in the qcow2 file.
• Refcount table and blocks: a two-level structure with a reference count for each data cluster. Internal snapshots use this: a cluster with a reference count >= 2 means that it's used by other snapshots, and therefore any modifications require a copy-on-write operation.

Metadata overlap checks

In order to detect corruption when writing to qcow2 images QEMU (since v1.7) performs several sanity checks. They verify that QEMU does not try to overwrite sections of the file that are already being used for metadata. If this happens, the image is marked as corrupted and further access is prevented.

Although in most cases these checks are innocuous, under certain scenarios they can have a negative impact on disk write performance. This depends a lot on the case, and I want to insist that in most scenarios it doesn't have any effect. When it does, the general rule is that you'll have more chances of noticing it if the storage backend is very fast or if the qcow2 image is very large.

In these cases, and if I/O performance is critical for you, you might want to consider tweaking the images a bit or disabling some of these checks, so let's take a look at them. There are currently eight different checks. They're named after the metadata sections that they check, and can be divided into the following categories:

1. Checks that run in constant time. These are equally fast for all kinds of images and I don't think they're worth disabling.
   • main-header
   • active-l1
   • refcount-table
   • snapshot-table
2. Checks that run in variable time but don't need to read anything from disk.
   • refcount-block
   • active-l2
   • inactive-l1
3. Checks that need to read data from disk. There is just one check here and it's only needed if there are internal snapshots.
   • inactive-l2

By default all tests are enabled except for the last one (inactive-l2), because it needs to read data from disk.

Disabling the overlap checks

Tests can be disabled or enabled from the command line using the following syntax:

-drive file=hd.qcow2,overlap-check.inactive-l2=on
-drive file=hd.qcow2,overlap-check.snapshot-table=off

It's also possible to select the group of checks that you want to enable using the following syntax:

-drive file=hd.qcow2,overlap-check.template=none
-drive file=hd.qcow2,overlap-check.template=constant
-drive file=hd.qcow2,overlap-check.template=cached
-drive file=hd.qcow2,overlap-check.template=all

Here, none means that no tests are enabled, constant enables all tests from group 1, cached enables all tests from groups 1 and 2, and all enables all of them.

As I explained in the previous section, if you're worried about I/O performance then the checks that are probably worth evaluating are refcount-block, active-l2 and inactive-l1. I'm not counting inactive-l2 because it's off by default. Let's look at the other three:

• inactive-l1: This is a variable length check because it depends on the number of internal snapshots in the qcow2 image. However its performance impact is likely to be negligible in all cases so I don't think it's worth bothering with.
• active-l2: This check depends on the virtual size of the image, and on the percentage that has already been allocated. This check might have some impact if the image is very large (several hundred GBs or more). In that case one way to deal with it is to create an image with a larger cluster size. This also has the nice side effect of reducing the amount of memory needed for the L2 cache.
• refcount-block: This check depends on the actual size of the qcow2 file and it's independent from its virtual size. This check is relatively expensive even for small images, so if you notice performance problems chances are that they are due to this one. The good news is that we have been working on optimizing it, so if it's slowing down your VMs the problem might go away completely in QEMU 2.9.

Conclusion

The qcow2 consistency checks are useful to detect data corruption, but they can affect write performance.

If you're unsure and you want to check it quickly, open an image with overlap-check.template=none and see for yourself, but remember again that this will only affect write operations. To obtain more reliable results you should also open the image with cache=none in order to perform direct I/O and bypass the page cache. I've seen performance increases of 50% and more, but whether you'll see them depends a lot on your setup. In many cases you won't notice any difference.

I hope this post was useful to learn a bit more about the qcow2 format. There are other things that can help QEMU perform better, and I'll probably come back to them in future posts, so stay tuned!

Acknowledgments

My work in QEMU is sponsored by Outscale and has been made possible by Igalia and the help of the rest of the QEMU development team.

0 0

08 Feb 2017 8:52am GMT