fix(dataZoom): data shape distribution for time axis

[andrearoota]

Brief Information

This pull request is in the type of:

• bug fixing
• new feature
• others

What does this PR do?

Fix render data shadow line with data not equally distributed in time axis

Fixed issues

#16924
#15921
#16980

Details

Before: What was the problem?

Data shadow line in bottom dataZoom does not match with serie line on graph

After: How is it fixed in this PR?

If type of axis is set to "time", scale the original axis data range to dataZoom axis range
I used this mathematical formula

Misc

• The API has been changed (apache/echarts-doc#xxx).
• This PR depends on ZRender changes (ecomfe/zrender#xxx).

Related test cases or examples to use the new APIs

N.A.

Others

Merging options

• Please squash the commits into a single one when merging.

Other information

[echarts-bot]

Thanks for your contribution!
The community will review it ASAP. In the meanwhile, please checkout the coding standard and Wiki about How to make a pull request.

[Ovilia]

I'm also fixing related problems at #17311 . I will check the logic more carefully later. Thanks for your contribution!

[piotrbubelE]

Hello! Is there a plan to merge this PR anytime soon?
It also enables workaround for bug with axis.min and axis.max on dataShadow

[pchrzaszczewski]

I am also waiting for this fix. What are your plans?

[kamilapolewczykatos]

Merge would be great. I need this fix for my app.

[zsamotsil]

It would be great to ship it. Need also this fix in my application.

[piotrbubelE]

Here is issue that can use workaround from this PR:
#19957

[grzesikj]

When will this be available for use? Without this I cannot use data zoom slider in my project.

[Bernard-code]

I really could use this fix. When will this be available to use?

[grzesikj]

Add it, please.

[grzesikj]

When can you add it?

[noahhai]

Fix PR pending for two years 😭

[andrearoota]

@pchrzaszczewski Woah easy there. I get the frustration though.

I am going to create a pnpm patch based on the fixes in the PR and just pin the echarts version in my dependencies. Not great, but I assume at some point it will get fixed.

Does my code still work in the current version? I am willing to improve it and update it if necessary

[pchrzaszczewski]

@pchrzaszczewski Woah easy there. I get the frustration though.

I am going to create a pnpm patch based on the fixes in the PR and just pin the echarts version in my dependencies. Not great, but I assume at some point it will get fixed.

@noahhai I'm grateful that you responded. We can see that something is happening ;)

I've been patient for 2 years 💪 It is always better to at least know if there is a plan or not.
I'm keeping my fingers crossed and cheering that the day will finally come... 🤞

[grzesikj]

I'm still waiting

[alinscodes]

Can we just deploy this pleasE?

[alinscodes]

This is good.

[jteez]

Still hoping for this fix too!

[ElicaInc]

@andrearoota fixed the mismatch between the chart and dataZoom. It works well in my setup—thank you, @andrearoota!

[alinscodes]

Lets get this in please

[ElicaInc]

This patch, made possible thanks to @andrearoota, worked perfectly in my setup while waiting for the official merge.

diff --git a/node_modules/echarts/lib/component/dataZoom/SliderZoomView.js b/node_modules/echarts/lib/component/dataZoom/SliderZoomView.js
index be21afa..25e561e 100644
--- a/node_modules/echarts/lib/component/dataZoom/SliderZoomView.js
+++ b/node_modules/echarts/lib/component/dataZoom/SliderZoomView.js
@@ -281,22 +281,35 @@ var SliderZoomView = /** @class */function (_super) {
       var areaPoints_1 = [[size[0], 0], [0, 0]];
       var linePoints_1 = [];
       var step_1 = thisShadowExtent[1] / (data.count() - 1);
-      var thisCoord_1 = 0;
+        // Added: Variables to support time axis
+      var thisDataExtent = data.getDataExtent(info.thisDim);
+      var normalizationConstant = size[0] / (thisDataExtent[1] - thisDataExtent[0]);
+      var isTimeAxis = info.thisAxis.type === 'time';
+      var thisCoord_1 = -step_1;
       // Optimize for large data shadow
       var stride_1 = Math.round(data.count() / size[0]);
       var lastIsEmpty_1;
-      data.each([otherDim], function (value, index) {
-        if (stride_1 > 0 && index % stride_1) {
-          thisCoord_1 += step_1;
+      data.each([info.thisDim, otherDim], function (thisValue, otherValue, index) {
+        if (stride_1 > 0 && (index % stride_1)) {
+          // Added: Only increment step when not using a time axis
+          if (!isTimeAxis) {
+            thisCoord_1 += step_1;
+          }
           return;
         }
+        
+        // Fix: Coordinate calculation split between time axis and non-time axis
+        thisCoord_1 = isTimeAxis
+          ? (+thisValue - thisDataExtent[0]) * normalizationConstant
+          : thisCoord_1 + step_1;
+    
         // FIXME
         // Should consider axis.min/axis.max when drawing dataShadow.
         // FIXME
         // 应该使用统一的空判断？还是在list里进行空判断？
-        var isEmpty = value == null || isNaN(value) || value === '';
+        var isEmpty = otherValue == null || isNaN(otherValue) || otherValue === '';
         // See #4235.
-        var otherCoord = isEmpty ? 0 : linearMap(value, otherDataExtent_1, otherShadowExtent_1, true);
+        var otherCoord = isEmpty ? 0 : linearMap(otherValue, otherDataExtent_1, otherShadowExtent_1, true);
         // Attempt to draw data shadow precisely when there are empty value.
         if (isEmpty && !lastIsEmpty_1 && index) {
           areaPoints_1.push([areaPoints_1[areaPoints_1.length - 1][0], 0]);
@@ -307,7 +320,7 @@ var SliderZoomView = /** @class */function (_super) {
         }
         areaPoints_1.push([thisCoord_1, otherCoord]);
         linePoints_1.push([thisCoord_1, otherCoord]);
-        thisCoord_1 += step_1;
+        
         lastIsEmpty_1 = isEmpty;
       });
       polygonPts = this._shadowPolygonPts = areaPoints_1;

[carlos-manuel-soares-alb]

Any news for this PR to be merged?

[echarts-bot]

To reviewers: If this PR is going to be described in the changelog in the future release, please make sure this PR has one of the following labels: PR: doc ready, PR: awaiting doc, PR: doc unchanged

This message is shown because the PR description doesn't contain the document related template.

[echarts-bot]

Congratulations! Your PR has been merged. Thanks for your contribution! 👍

[Bilge]

Does this address #13408 or nah?

Spoiler: it only half fixes it. It fixes the time series issue but not the step line issue.