The Qt Quick 2 bars example shows how to make a simple 3D bar graph using Bars3D and Qt Quick 2.

The interesting thing about this example is switching series and displaying more than one series at once. We'll concentrate on those and skip explaining the basic Bars3D functionality - for more detailed QML example documentation, see Qt Quick 2 Scatter Example.

Running the Example

To run the example from Qt Creator, open the Welcome mode and select the example from Examples. For more information, visit Building and Running an Example.

Data

The example data is monthly income and expenses of a fictional company over several years. The data is defined in a list model in Data.qml like this:

ListModel {
    id: dataModel
    ListElement{ timestamp: "2006-01"; expenses: "-4";  income: "5" }
    ListElement{ timestamp: "2006-02"; expenses: "-5";  income: "6" }
    ListElement{ timestamp: "2006-03"; expenses: "-7";  income: "4" }
    ...

Each data item has three roles: timestamp, income, and expenses. The timestamp value is in format: <four digit year>-<two digit month>. Years and months are natural to map to rows and columns of a bar chart, but we can only show either income or expenses as the value.

Now we need to add the data to the Bars3D graph. We will create two Bar3DSeries inside it, starting with a series for the income:

Bar3DSeries {
    id: barSeries
    itemLabelFormat: "Income, @colLabel, @rowLabel: @valueLabel"
    baseGradient: barGradient

    ItemModelBarDataProxy {
        id: modelProxy
        itemModel: graphData.model
        rowRole: "timestamp"
        columnRole: "timestamp"
        valueRole: "income"
        rowRolePattern: /^(\d\d\d\d).*$/
        columnRolePattern: /^.*-(\d\d)$/
        rowRoleReplace: "\\1"
        columnRoleReplace: "\\1"
        multiMatchBehavior: ItemModelBarDataProxy.MMBCumulative
    }
    ...

The data is attached to the itemModel property of the ItemModelBarDataProxy inside the series. For valueRole we simply specify the income field, as it contains the value we want, but getting the years and months is a bit more complicated, since they are both found in the same field. To extract those values, we specify the timestamp field for both rowRole and columnRole, and additionally specify a search pattern and a replace rule for those roles to extract the correct portion of the field contents for each role. The search pattern is a normal JavaScript regular expression and the replace rule specifies what the field content that matches the regular expression is replaced with. In this case we want to replace the entire field content with just the year or the month, which is the first captured substring for both rows and columns. For more information how the replace using regular expressions works, see QString::replace(const QRegExp &rx, const QString &after) function documentation.

The multiMatchBehavior property specifies what to do in case multiple item model items match the same row/column combination. In this case we want to add their values together. This property has no effect when we are showing values for each month, as there are no duplicate months in our item model, but it becomes relevant later when we want to show the yearly totals.

Then we add another series for the expenses:

Bar3DSeries {
    id: secondarySeries
    visible: false
    itemLabelFormat: "Expenses, @colLabel, @rowLabel: -@valueLabel"
    baseGradient: secondaryGradient

    ItemModelBarDataProxy {
        id: secondaryProxy
        itemModel: graphData.model
        rowRole: "timestamp"
        columnRole: "timestamp"
        valueRole: "expenses"
        rowRolePattern: /^(\d\d\d\d).*$/
        columnRolePattern: /^.*-(\d\d)$/
        valueRolePattern: /-/
        rowRoleReplace: "\\1"
        columnRoleReplace: "\\1"
        multiMatchBehavior: ItemModelBarDataProxy.MMBCumulative
    }
    ...

The model contains expenses as negative values, but we want to show them as positive bars, so that we can easily compare them to income bars. We use valueRolePattern to remove the minus sign to achieve this. No replacement string needs to be specified as the default replacement is an empty string.

We use the visible property of the series to hide the second series for now.

Custom Axis Labels

One interesting tidbit about axes is that we redefine the category labels for column axis in Axes.qml. This is done because the data contains numbers for months, which we don't want to use for our column labels:

CategoryAxis3D {
    id: columnAxis
    labels: ["January", "February", "March", "April", "May", "June",
        "July", "August", "September", "October", "November", "December"]
    labelAutoRotation: 30
}

We also set automatic axis label rotation to make axis labels more readable at low camera angles.

Switching Series

In the main.qml, we set up the graph and various UI elements. There are three interesting code blocks we want to highlight here. The first one shows how to change the visualized data between income, expenses, and both, by simply changing visibility of the two series:

onClicked: {
    if (text === "Show Expenses") {
        barSeries.visible = false
        secondarySeries.visible = true
        barGraph.valueAxis.labelFormat = "-%.2f M\u20AC"
        secondarySeries.itemLabelFormat = "Expenses, @colLabel, @rowLabel: @valueLabel"
        text = "Show Both"
    } else if (text === "Show Both") {
        barSeries.visible = true
        barGraph.valueAxis.labelFormat = "%.2f M\u20AC"
        secondarySeries.itemLabelFormat = "Expenses, @colLabel, @rowLabel: -@valueLabel"
        text = "Show Income"
    } else { // text === "Show Income"
        secondarySeries.visible = false
        text = "Show Expenses"
    }
}

The axis label format and item selection label formats are tweaked to get the negative sign showing properly for expenses, which were actually resolved as positive values.

The second interesting block is where we change the visualized data by adjusting the proxy propertes:

onClicked: {
    if (text === "Show yearly totals") {
        modelProxy.autoRowCategories = true
        secondaryProxy.autoRowCategories = true
        modelProxy.columnRolePattern = /^.*$/
        secondaryProxy.columnRolePattern = /^.*$/
        graphAxes.value.autoAdjustRange = true
        barGraph.columnAxis = graphAxes.total
        text = "Show all years"
    } else if (text === "Show all years") {
        modelProxy.autoRowCategories = true
        secondaryProxy.autoRowCategories = true
        modelProxy.columnRolePattern = /^.*-(\d\d)$/
        secondaryProxy.columnRolePattern = /^.*-(\d\d)$/
        graphAxes.value.min = 0
        graphAxes.value.max = 35
        barGraph.columnAxis = graphAxes.column
        text = "Show 2010 - 2012"
    } else { // text === "Show 2010 - 2012"
        // Explicitly defining row categories, since we do not want to show data for
        // all years in the model, just for the selected ones.
        modelProxy.autoRowCategories = false
        secondaryProxy.autoRowCategories = false
        modelProxy.rowCategories = ["2010", "2011", "2012"]
        secondaryProxy.rowCategories = ["2010", "2011", "2012"]
        text = "Show yearly totals"
    }
}

To show yearly totals, we need to combine the twelve months of each year into a single bar. We achieve this by specifying a columnRolePattern that matches all model items. That way the data proxy will only have a single column. The cumulative multiMatchBehavior we specified earlier for the proxy becomes relevant now, causing the values of all twelve months of each year to be added up into a single bar.

To show just a subset of years, we set autoRowCategories to false on the ItemModelBarDataProxy item and define the row categories explicitly. This way, only the items in specified row categories are visualized.

The third interesting block shows how to get the row and column index of an item if you know the row and column values by using ItemModelBarDataProxy methods rowCategoryIndex() and columnCategoryIndex():

onCurrentRowChanged: {
    var timestamp = graphData.model.get(currentRow).timestamp
    var pattern = /(\d\d\d\d)-(\d\d)/
    var matches = pattern.exec(timestamp)
    var rowIndex = modelProxy.rowCategoryIndex(matches[1])
    var colIndex
    if (barGraph.columnAxis === graphAxes.total)
        colIndex = 0 // Just one column when showing yearly totals
    else
        colIndex = modelProxy.columnCategoryIndex(matches[2])
    if (selectedSeries.visible)
        mainview.selectedSeries.selectedBar = Qt.point(rowIndex, colIndex)
    else if (barSeries.visible)
        barSeries.selectedBar = Qt.point(rowIndex, colIndex)
    else
        secondarySeries.selectedBar = Qt.point(rowIndex, colIndex)
}

Example project @ code.qt.io