Draw/update circles in rect - nested selections confusion - d3.js

I am trying to get a layout that would, for each object in the data array:
append a rect or a g element that will serve as container
inside or on top of this, append a circle for each of the coordinates.
Below is a mock-up of how the data is massaged before I'm trying to append to the DOM (at the top of the update() function in the block below):
[{
label: 'foo',
circles: [
{ x: 0, y: 10 },
{ x: 10, y: 10 }
]
},{
...
}]
The drawing and updating of the rect elements seems to be working fine, but I am getting the selection and joins confused for the circles.
Here's the block: http://blockbuilder.org/basilesimon/91f75ab5209a62981f11d30a81f618b5
With
var dots = rects.selectAll('.dots')
I can select the right data below but can't draw it.
Could you help me getting the selections right so I can draw and update both the rect and the circle, please?
Thank you Gerard for your help. This is my current state, but I've pitted myself into a hole by running a for loop instead of d3 selections.
I wonder if I couldn't nest the circles in g elements after building a new data object like so:
var data = dataset.map(function(d) {
var circles = d3.range(d.amount).map(function(i) {
return {
x: (i % 5)*20,
y: (i / 5 >> 0)*20
}
});
return {
label: d.label,
dots: circles
};
});
From each object in data, we'll append a g, and inside each g we'll append the circles. Any help appreciated, since this will affect the dots + i used by the update pattern...
New question here

Here is the problem:
var dots = svg.selectAll('dots')
You're selecting something that doesn't exist. Because of that, your "enter" selection will always contain all the data, and your "exit" selection will always be empty.
The solution is changing it for something like this:
var dots = svg.selectAll(".dots" + i)
And, in the enter selection, setting the classes:
.attr("class", "dots" + i)
Here is your updated bl.ocks (with some other minor changes): https://bl.ocks.org/anonymous/4c2e1d66f1ab890da983465a4f84ca9b

Related

Legend in deck.gl

I am wondering if it is possible to get access to aggregated data from a deck.gl layer to be able to draw a legend.
Because the colour scheme is supplied I would only require the extent of the aggregated values calculated by the screengrid layer to be able to add this to the legend.
I know there are tooltips, but in some circumstances it would be nice to have access to these values.
I'm using the HexagonLayer and for that one you can find the values for the layers by using a semi custom onSetColorDomain function when initializing your layer. Then save the domain range array to a variable and call a make legend function.
for example:
const hexlayer = new HexagonLayer({
id: 'heatmap',
pickable: true,
colorRange: COLOR_RANGE,
data: feats_obj.coords_array,
elevationScale: 9,
extruded: true,
radius: 300,
getColorValue: points => {
if( points.length > max_points) {
max_points = points.length
}
renderCategories( incident_categories )
return points.length
},
onSetColorDomain: (ecol) => {
console.log('color domain set', ecol)
max_span.innerHTML = ecol[1]
color_domain = ecol;
create_legend()
// console.log('max_points: ', max_points)
},...
})
The hacky way i figured out was to make a max points in a polygon global variable outside initializing the layer and have it update the value anytime there's a polygon with more points in it than that max value. An example of it in the wild is: https://mpmckenna8.github.io/sfviz/?start_date=2020-05-01&end_date=2020-05-31 with a link to the repo u can hack on there.

Adjusting small multiples sparklines

I have an heatmap that show some data and a sparkline for each line of the heatmap.
If the user click on a row label, then the data are ordered in decreasing order, so each rect is placed in the right position.
Viceversa, if the user click on a column label.
Each react is placed in the right way but I'm not able to place the sparkline.
Here the code.
When the user click on a row label, also the path inside the svg containing the sparkline should be updated.
And then, when the user click on a column label, the svg containing the sparkline should be placed in the correct line.
To place the svg in the right place, I try to use the x and y attributes of svg. They are updated but the svg doesn't change its position. Why?
Here is a piece of code related to that:
var t = svg.transition().duration(1000);
var values = [];
var sorted;
sorted = d3.range(numRegions).sort(function(a, b) {
if(sortOrder) {
return values[b] - values[a];
}
else {
return values[a] - values[b];
}
});
t.selectAll('.rowLabel')
.attr('y', function(d, k) {
return sorted.indexOf(k) * cellSize;
});
Also, I don't know how to change the path of every sparkline svg. I could take the data and order them manually, but this is only good for the row on which the user has clicked and not for all the others.
How can I do?
The vertical and horizontal re-positioning/redrawing of those sparklines require different approaches:
Vertical adjustment
For this solution I'm using selection.sort, which:
Returns a new selection that contains a copy of each group in this selection sorted according to the compare function. After sorting, re-inserts elements to match the resulting order.
So, first, we set our selection:
var sortedSVG = d3.selectAll(".data-svg")
Then, since selection.sort deals with data, we bind the datum, which is the index of the SVG regarding your sorted array:
.datum(function(d){
return sorted.indexOf(+this.dataset.r)
})
Finally, we compare them in ascending order:
.sort(function(a,b){
return d3.ascending(a,b)
});
Have in mind that the change is immediate, not a slow and nice transition. This is because the elements are re-positioned in the DOM, and the new structure is painted immediately. For having a slow transition, you'll have to deal with HTML and CSS inside the container div (which may be worth a new specific question).
Horizontal adjustment
The issue here is getting all the relevant data from the selection:
var sel = d3.selectAll('rect[data-r=\'' + k + '\']')
.each(function() {
arr.push({value:+d3.select(this).attr('data-value'),
pos: +d3.select(this).attr('data-c')});
});
And sorting it according to data-c. After that, we map the result to a simple array:
var result = arr.sort(function(a,b){
return sorted.indexOf(a.pos) - sorted.indexOf(b.pos)
}).map(function(d){
return d.value
});
Conclusion
Here is the updated Plunker: http://next.plnkr.co/edit/85fIXWxmX0l42cHx or http://plnkr.co/edit/85fIXWxmX0l42cHx
PS: You'll need to re-position the circles as well.

D3.js removing element doesn't work as expected

I am building a visualization of a popular brain teaser. So far there are 9 balls on SVG with drag behavior, collision detection and forceY.
I am trying to remove them by dragging. Which kinda works... But not the way it is supposed.
One of balls is yellow. When it is dragged, it should get removed. This works, but the bug is that another ball gets rendered yellow.
Here is the function that deals with the functionality and below is fiddle.
function dragended(d, i) {
balls.splice(i, 1);
var update = svgBalls.data(balls)
update.exit().remove()
simulation.alphaTarget(0.1);
}
http://jsfiddle.net/Anuar/21738t3g/
This is not a solution in the narrow sense of the term, but simply an explanation of your problem.
If we slightly change your data array...
var balls = [{id:0, color:false},{id:1, color:false},{id:2, color:false},
{id:3, color:false}, {id:4, color:false}, {id:5, color:false},
{id:6, color:false}, {id:7, color:false},{id:8, color:false}];
... in such a way that we can set a key function...
svgBalls.data(balls, d => d.id)
... you're gonna see that this line:
balls.splice(i, 1);
Is not correctly removing the dragged element. If you change for this:
balls = balls.filter(d => d.id != i);
It will work.
Here is your updated fiddle: http://jsfiddle.net/27sykjxd/

Trouble with filters and triggers in dc.js - re-drawing is out of sync

I have two plots: a line plot and a bubble plot. When I click on the bubble plot, I want the line plot to be updated so that it is drawn with only the data related to that 'bubble'. This is different from the standard implementation whereby clicking would add or remove the data from the existing filter.
If you look at the image you can see that although 'model 0' is selected the plotted hazard (y-scale in plot 1) does not correspond.
And now when I click on 'model 5', I get the opposite.
My current implementation is posted as a jsfiddle here. I can see from the attached data table that I am achieving what I want, but the line plot does not re-draw correctly. In fact, it seems to re-draw with the last filter, not the new one.
This implementation is hacked from here: in particular, the renderlet and on("filtered", function (chart) { lines. However, to make this work, I have had to comment out the plot1.filter(chart.filter()); line for the second plot.
I don't really understand why a renderlet and the on("filtered" ... or on("postRedraw" ... listeners are needed together.
I have been round the houses on this one, so any suggestions would be very gratefully received.
I tried to simplify the jsfiddle to isolate the problem. Here is the adapted jsfiddle: http://jsfiddle.net/djmartin_umich/mKz7A/
Your plot2 keyAccessor accessed the df value from the p.value.df rather than using a dimension on df. My guess is that this is what was causing problems. Here is the adapted code:
dfDim = ndx.dimension(function (d) {return d.df;});
...
plot2.width(300)
.height(250)
.dimension(dfDim)
I also noticed that your plot2 valueAccessor and radiusAccessor were not using a computed average. Your code would overwrite est and estse for each record added or removed from the group. Here is the adapted code that computes the average:
dfGroup = dfDim.group().reduce(
//add
function (p, v) {
++p.count;
p.est += v.est;
p.avg_est = p.est / p.count;
p.estse += v.estse;
p.avg_estse = p.estse / p.count;
return p;
},
//remove
function (p, v) {
--p.count;
p.est -= v.est;
p.avg_est = p.est / p.count;
p.estse -= v.estse;
p.avg_estse = p.estse / p.count;
return p;
},
//init
function (p, v) {
return {
count: 0,
est: 0,
estse: 0,
avg_est: 0,
avg_estse: 0
};
});
After these changes, I believe the code behaves as you wanted.

Is an NVD3 Line Plot with Markers Possible?

I'm making an NVD3 line plot that will have significantly improved clarity if I can get markers to show for each data point instead of just the line itself. Unfortunately, I haven't been able to find an easy way to do this with NVD3 yet. I also considered using a scatter plot, but I couldn't figure out how to show connecting lines between the points. A third option I considered was to overlay a line and scatter plot, but this would show each series twice in the legend and may cause other unnecessary visual complications.
Is there a way to elegantly pull this off yet? Sample code of my formatting technique is listed below, but the 'size' and 'shape' attributes in test_data have no effect on the line plot with the current code.
test_data = [ { key: 'series1',
values: [
{ x: 1, y: 2.33, size:5, shape:"circle" },
{ x: 2, y: 2.34, size:5, shape:"circle" },
{ x: 3, y: 2.03, size:5, shape:"circle" },
] } ];
nv.addGraph(function() {
var test_chart = nv.models.lineChart();
test_chart.xAxis.axisLabel('Sample Number');
test_chart.yAxis
.axisLabel('Voltage (V)')
.tickFormat(d3.format('.02f'));
d3.select('#test_plot')
.datum(test_data)
.transition().duration(500)
.call(test_chart);
nv.utils.windowResize(test_chart.update);
return test_chart;
});
I also wanted to add markers in a project I was working on. Here is a solution my partner and I found.
First, you have to select all of the points in your chart and set the fill-opacity to 1:
#my-chart .nv-lineChart circle.nv-point
{
fill-opacity: 1;
}
Now your points will be visible. To adjust the size of each point you need to modify each one's "r" (for radius) attribute. This isn't a style so you can't do it with css. Here is some jQuery code that does the job. The 500 millisecond delay is so the code will not run before the chart is rendered. This snippet sets the radius to 3.5:
setTimeout(function() {
$('#my-chart .nv-lineChart circle.nv-point').attr("r", "3.5");
}, 500);
This puzzled me until I got help from the community:
css styling of points in figure
So here is my solution, based on css:
.nv-point {
stroke-opacity: 1!important;
stroke-width: 5px!important;
fill-opacity: 1!important;
}
If anyone has come here from rCharts, below is a rmarkdown template to create an nPlot with both lines and markers:
```{r 'Figure'}
require(rCharts)
load("data/df.Rda")
# round data for rChart tooltip display
df$value <- round(df$value, 2)
n <- nPlot(value ~ Year, group = 'variable', data = df, type = 'lineChart')
n$yAxis(axisLabel = 'Labor and capital income (% national income)')
n$chart(margin = list(left = 100)) # margin makes room for label
n$yAxis(tickFormat = "#! function(d) {return Math.round(d*100*100)/100 + '%'} !#")
n$xAxis(axisLabel = 'Year')
n$chart(useInteractiveGuideline=TRUE)
n$chart(color = colorPalette)
n$addParams(height = 500, width = 800)
n$setTemplate(afterScript = '<style>
.nv-point {
stroke-opacity: 1!important;
stroke-width: 6px!important;
fill-opacity: 1!important;
}
</style>'
)
n$save('figures/Figure.html', standalone = TRUE)
```
The current version of nvd3 use path instead of circle to draw markers. Here is a piece of css code that i used to show markers.
#chart g.nv-scatter g.nv-series-0 path.nv-point
{
fill-opacity: 1;
stroke-opacity: 1;
}
And I also write something about this in https://github.com/novus/nvd3/issues/321, you could find that how i change the shape of makers.
I don't know how to change the size of markers. Trying to find a solution.
Selectively enable points to some series using the following logic in nvd3.
//i is the series number; starts with 0
var selector = 'g.nv-series-'+i+' circle';
d3.selectAll(selector).classed("hover",true);
However an additional parameter( like say 'enable_points':'true') in the data would make better sense. I will hopefully push some changes to nvd3 with this idea.
For current version of NVD3 (1.8.x), I use this D3-based solution (scripting only, no CSS file or style block required):
nv.addGraph(function() {
// ...
return chart;
},
function() {
// this function is called after the chart is added to document
d3.selectAll('#myChart .nv-lineChart .nv-point').style("stroke-width",
"7px").style("fill-opacity", ".95").style("stroke-opacity", ".95");
}
);
The styles used are exactly the styles added by NVD3 by applying the "hover" class to each point (when hovered). Adjust them to your needs.

Resources