So excited about the new updates to the chart widget in experience builder. I am currently able to reconfigure charts for the new abilities or create new charts that I had to embed from dashboards before. Awesome update!
However, when binning the chart, Parsing the Time category to 1 year and trimming incomplete intervals (not sure if those setting matter), my chart has date ranges that begin in July instead of January. It makes the data look a little strange...
Does anyone know of a way to change the date settings in this new chart display? Could it be the data it is reading? Reading on the Guide, it seems like it is reading and snapping to whenever your first data point is...? So If my first data point is July the entire chart is binned from a July-July basis. This seems problematic, no? We have to report on a year to year basis so that chart outputs are not helpful.... yet!
Any help or discussion, please!
Hi @JustinH thanks for your feedback!
Currently the behavior you described is as expected. Your firs point of data may start at July of 2018, so it group each bin for 1 year interval. That's why you would see July/2018-July/2019, July/2019-July/2020 and so forth.
We are working on another enhancement to allow user decide the starting point of time binning like Jan or each year or first date of each week. It is still in research so we cannot promise the date of the new feature. But we are working on it. Please stay tuned. Thank you!
Thanks for your reply. I am looking forward to the update! That would be very helpful for our reporting needs.
Hi @Wei_Ying,
Are there plans to allow the manual definition of bin ranges in the future, such as for slices in a pie chart or columns in a histogram? I would like to be able to match my symbology class ranges to the bins in a chart so as to represent, say, the size of classes in proportion to one another.
I am not sure which method the chart widget uses when it automatically sets bin ranges, but it effectively renders symbol class ranges created via manual interval method obsolete. Unless I'm just missing something, which I recognize is quite possible.
Hi @JoshBerna unfortunately so far we do not have such plan.
>> I would like to be able to match my symbology class ranges
When you are using the same category type in Pie chart, or using the same transformation method in histogram chart as your symbology setting, don't you see same binning classes ?
Yes, the binning works perfectly for categorical fields. I have a couple layers using symbology from quantitative fields though, such as building condition assessments. The symbology is broken into 4 classes of values for this particular layer (ex: 0-10%, 11-20%, etc.). When brought into a chart, these classes are not honored as bins, but rather each value is treated as discrete. In the example screenshot I posted, you can see that there is a slice made for each value, rather than each symbol class of values, as would be ideal for this use case.
So, to answer your question, I do very well see the same binning classes when using a genuine categorical field, and that works great. When the field values are not a proper list however, it would be nice to be able to define bin ranges so as to match my symbol classes.
I do recognize that I could create a separate text field that would define which bin the value is in (for condition assessments, values could be good, poor, etc.) and calculate it based on the number field value, but that may not always be the case.
Thank you for your reply.