Forums Home / Wijmo / General Discussion
Posted by: Yifan_Zhang on 22 April 2021, 5:53 pm EST
Yifan_Zhang
Posted 22 April 2021, 5:53 pm EST
Is it possible to implement a box plot that takes the statistics directly to create the chart? The data is so large that it won’t be passed to wijmo. The data would be pre-processed instead to generate only box plot statistics like min/max/median etc.
sharad.tomer
Posted 23 April 2021, 8:50 am EST
Hi,
If I understand correctly you need to create candlestick charts. Please refer to the following demos and let us know if this fulfills your requirement:
• https://www.grapecity.com/wijmo/demos/Chart/Financial/CandlestickChart/purejs
• https://www.grapecity.com/wijmo/demos/Chart/Financial/HeikinAshiChart/purejs
If not, please explain more about your use requirement so that we could have a better understanding.
Regards
Sharad
Yifan_Zhang
Posted 29 April 2021, 2:26 pm EST
Hi Sharad,
Sorry for the late reply and thanks for your input. But those two charts are not exactly what we want. For box-whisker plot, there will be a median line and outliers while those two charts don’t have. Wijmo already have this box-whisker plot and we are using it now: https://www.grapecity.com/wijmo/demos/Chart/Special/BoxWhisker/purejs, but for now it only takes the data and draws automatically for us. It could be really slow and limited when the data is huge. So what we want is to supply that box-whisker plot with statistics like min/max/median directly and create the chart. Is it possible to do that? Thanks.
sharad.tomer
Posted 3 May 2021, 4:27 am EST
We are sorry but we cannot pass the computed data to box and whiskers graph. But I’ve prepared a sample to demonstrate how we could add median line and outliers on candlestick chart. Please refer to the following sample and let us know if this fulfill your requirement:
https://codesandbox.io/s/wijmo-starter-forked-kk2em?file=/src/index.js
Yifan_Zhang
Posted 28 October 2024, 5:38 pm EST
Hi, this would be an important feature for our product and it can greatly improve the performance as well as the usability of the box plot. So is it possible to have this request of creating the box plot from statistics implemented in the near future?
vivek.pannu
Posted 29 October 2024, 3:19 am EST
Hi,
We created an enhancement to create a box plot from statistics as per your request. The internal tracking ID for the same is - WJM-35181. We will update you on its progress when we have some updates to share.
Regards
Yifan_Zhang
Posted 29 October 2024, 3:20 am EST
That’s great. Looking forward to the enhancement!
Yifan_Zhang
Posted 17 May 2025, 2:49 am EST
Hi, any updates so far?
vivek.pannu
Posted 20 May 2025, 8:25 am EST
Hi Yifan,
The engineering team is working on the enhancement. For now, they have shared a workaround that can be used to render the chart with processed data by overriding the ‘_parse’ method of the boxplot. Please find attached a sample for your reference, you can test it on your end. In case, you face any issues, you can let us know.
Regards
Yifan_Zhang
Posted 20 May 2025, 4:48 pm EST
Thanks for the reply. One more thing about this request: If we supply the outliers along with the statistics, could you include the functionality of drawing them?
vivek.pannu
Posted 21 May 2025, 7:36 am EST
Hi Yifan,
As per our understanding, you want to pass the outlier points along with the other processed data and want to show the outlier points on the chart as well. If so, then yes, you can pass outlier points data as well, you can verify it in the sample we shared in our previous comment. Just run the sample and check the show outliers checkbox above the chart.
In case, if we are missing something, please let us know.
Regards
Yifan_Zhang
Posted 21 May 2025, 12:18 pm EST
I see. That’s great. The current method would override the original parse method, we want to keep both so we can either pass in the whole data or the statistics. Is there a way of doing that or do we have to wait for the official enhancement?
vivek.pannu
Posted 22 May 2025, 6:02 am EST
Hi Yifan,
We would recommend sticking with the current workaround for now and waiting for the official support for this feature to reduce the complexity of the chart and minimize the chances of code failures. We will let you know when this feature is supported in the FlexChart.
Regards
