2019-01-20 23:15:10 +08:00
# Example (JKQTPlotter): Line Graph with Logarithmic y-axis {#JKQTPlotterLogAxes}
This project (see `./examples/simpletest_logaxes/` ) simply creates a JKQTPlotter widget (as a new window) and several line-graphs of different resonance curves.
2018-12-29 00:46:47 +08:00
2019-02-03 22:54:41 +08:00
The source code of the main application can be found in [`jkqtplotter_simpletest_logaxes.cpp` ](https://github.com/jkriege2/JKQtPlotter/tree/master/examples/simpletest_logaxes/jkqtplotter_simpletest_logaxes.cpp ). Mainly several graphs are generated in a loop and then different line styles are applied to the graphs (set by ``graph->setStyle()`). The colors are set automtically from an internal default palette. The main loop looks like this:
2018-12-29 00:46:47 +08:00
2019-01-19 16:40:52 +08:00
```.cpp
2018-12-29 00:46:47 +08:00
QVector< Qt::PenStyle > pens {Qt::SolidLine, Qt::DashLine, Qt::DotLine, Qt::DashDotLine, Qt::DashDotDotLine };
for (int id=0; id< D.size ( ) ; id + + ) {
// generate some plot data
QVector< double > Y;
for (auto& xx: X) {
Y< < 1.0 / sqrt ( sqr ( 1-sqr ( xx ) ) + sqr ( 2 * xx * D [ id ] ) ) ;
}
2019-01-20 17:49:29 +08:00
JKQTPXYLineGraph* graph=new JKQTPXYLineGraph(&plot);
2018-12-29 00:46:47 +08:00
// copy data into datastore and immediately set the yColumn
2019-01-26 20:00:40 +08:00
graph->setXColumn(columnX);
graph->setYColumn(ds->addCopiedColumn(Y, "y"+QString::number(id)));
2018-12-29 00:46:47 +08:00
// don't use symbols
2019-01-26 20:00:40 +08:00
graph->setSymbol(JKQTPNoSymbol);
2018-12-29 00:46:47 +08:00
// use one of different pens
2019-01-26 20:00:40 +08:00
graph->setStyle(pens[id%pens.size()]);
2018-12-29 00:46:47 +08:00
// set width of graph line
2019-01-26 03:16:04 +08:00
graph->setLineWidth(1.5);
2018-12-29 00:46:47 +08:00
// graph title is made from symbol+penstyle
2019-01-26 20:00:40 +08:00
graph->setTitle(QString("D=\\delta/\\omega_0=%1").arg(D[id]));
2018-12-29 00:46:47 +08:00
// add the graph to the plot, so it is actually displayed
plot.addGraph(graph);
}
```
2019-01-20 17:49:29 +08:00
Then a `JKQTPGeoText` is added to the graph, which shows the function plotted in the plot:
2019-01-19 16:40:52 +08:00
```.cpp
2018-12-29 00:46:47 +08:00
// 4. Also we add a text-element in the plot to show the plotted function
2019-01-20 17:49:29 +08:00
// This element (JKQTPGeoText) is taken from the set of geometric elements
2018-12-29 00:46:47 +08:00
// and is simply parametrized by a position (1.25/10) and the text to display.
// In addition you can also set the font size (here to 15)
// Use '$...$' around the actual math string to ensure rendering with a math font
// (the internal renderer uses XITS fonts by default, which are free and auto-distributed
// and loaded in the library). If you don't use the math-mode modifiers, the default
// font of the other rendering text is used, which might not be suitable for
// high-quality math rendering.
2019-01-20 17:49:29 +08:00
plot.addGraph(new JKQTPGeoText(& plot, 1.25, 10, "$\\frac{A}{A_{stat}}=\\frac{1}{\\sqrt{\\left(1-\\eta^2\\right)^2+\\left(2{\\eta}D\\right)^2}}$", 15));
2018-12-29 00:46:47 +08:00
```
The difference between not using and using `$...$` for the equation can be seen here:
2019-02-03 22:54:41 +08:00
- no $-math-mode: ![](https://raw.githubusercontent.com/jkriege2/JKQtPlotter/master/screenshots/jkqtplotter_simpletest_logaxes_lowqmathrendering.png)
- using $-math-mode: ![](https://raw.githubusercontent.com/jkriege2/JKQtPlotter/master/screenshots/jkqtplotter_simpletest_logaxes_highqmathrendering.png)
2018-12-29 00:46:47 +08:00
Finally the y-axis is switched to logarithmic scaling and the axis labels are set:
2019-01-19 16:40:52 +08:00
```.cpp
2019-01-26 03:16:04 +08:00
// 5. set y-axis to logarithmic (x-axis would be analogous, but using `plot.getXAxis()` )
2019-01-26 20:00:40 +08:00
plot.getYAxis()->setLogAxis(true);
2018-12-29 00:46:47 +08:00
// now we set the number of label ticks to 9 (you give the count if minor between two majors,
// so if you want ticks for 1,2,3,...,10,20,30,...,100... you need to use 9:
2019-01-26 20:00:40 +08:00
plot.getYAxis()->setMinorTicks(9);
2018-12-29 00:46:47 +08:00
// the minor grid is not shown by default. You can switch it on:
2019-01-26 20:00:40 +08:00
plot.getYAxis()->setDrawMinorGrid(true);
2018-12-29 00:46:47 +08:00
// usually axis ticks are shown as numbers 0.01, 0.1, 1, 10, ... You can also force the scientific
// power-of-10 notation, using:
2019-01-26 20:00:40 +08:00
plot.getYAxis()->setLabelType(JKQTPCALTexponent);
2018-12-29 00:46:47 +08:00
// the number of digits in JKQTPCALTexponent determines which labels are drawn in standard-notation,
// as compared to power-notation (e.g. if set to 1, the labels 0.1=10^{-1}, 1, 10 are shown in
// standard notation, the rest in power-notation. This tweak improves readability)
2019-01-26 20:00:40 +08:00
plot.getYAxis()->setLabelDigits(0);
2018-12-29 00:46:47 +08:00
// minor tick labels are usually not displayed, but you can switch them on, using
2019-01-26 20:00:40 +08:00
//plot.getYAxis()->setMinorTickLabelsEnabled(true);
2018-12-29 00:46:47 +08:00
// the axis font sizes can be set with:
2019-01-26 20:00:40 +08:00
plot.getYAxis()->setTickLabelFontSize(10); // axis tick labels
plot.getYAxis()->setMinorTickLabelFontSize(7); // minor axis tick labels
plot.getYAxis()->setLabelFontSize(14); // axis label size
plot.getXAxis()->setTickLabelFontSize(10); // axis tick labels
plot.getXAxis()->setMinorTickLabelFontSize(7); // minor axis tick labels
plot.getXAxis()->setLabelFontSize(14); // axis label size
2018-12-29 00:46:47 +08:00
// ... and finally set axis labels (using LaTeX notation and $...$ to improve rendering)
2019-01-26 03:16:04 +08:00
plot.getYAxis()->setAxisLabel("Amplitude $A/A_{stat}$");
plot.getXAxis()->setAxisLabel("relative driving frequency $\\eta=\\omega/\\omega_0$");
2018-12-29 00:46:47 +08:00
```
As an alternative `JKQTPCALTexponentCharacter` does not use the power-of-10 notation, but uses the usual unit-characters, e.g. 0.001=1m, 0.000001=1µ, 10000=10k, ...
The result looks like this:
2019-02-03 22:54:41 +08:00
![jkqtplotter_simpletest_logaxes ](https://raw.githubusercontent.com/jkriege2/JKQtPlotter/master/screenshots/jkqtplotter_simpletest_logaxes.png )
2018-12-29 00:46:47 +08:00
Without the logarithmic scaling we would have:
2019-02-03 22:54:41 +08:00
![jkqtplotter_simpletest_logaxes_nolog ](https://raw.githubusercontent.com/jkriege2/JKQtPlotter/master/screenshots/jkqtplotter_simpletest_logaxes_nolog.png )
2018-12-29 00:46:47 +08:00
Switching the minor grid off results in a plot like this:
2019-02-03 22:54:41 +08:00
![jkqtplotter_simpletest_logaxes_nominorgrid ](https://raw.githubusercontent.com/jkriege2/JKQtPlotter/master/screenshots/jkqtplotter_simpletest_logaxes_nominorgrid.png )
2018-12-29 00:46:47 +08:00
2019-01-26 20:00:40 +08:00
These examples show the results for different typical values for `setMinorTicks()` :
2018-12-29 00:46:47 +08:00
2019-02-03 22:54:41 +08:00
![logaxis_setMinorticks ](https://raw.githubusercontent.com/jkriege2/JKQtPlotter/master/screenshots/logaxis_setMinorticks.png )
2018-12-29 00:46:47 +08:00
2019-01-26 20:00:40 +08:00
These examples show the results for different typical values for `setLabelType()` :
2018-12-29 00:46:47 +08:00
2019-02-03 22:54:41 +08:00
![logaxis_setLabelType ](https://raw.githubusercontent.com/jkriege2/JKQtPlotter/master/screenshots/logaxis_setLabelType.png )
2018-12-29 00:46:47 +08:00
2019-01-13 01:53:16 +08:00