2 messages in com.googlegroups.google-visualization-api[visualization-api] Re: Dealing with ...
FromSent OnAttachments
opticrealm17 Jun 2008 15:11 
Jason17 Jun 2008 19:31 
Subject:[visualization-api] Re: Dealing with missing datapoints in timeseries/annotated timeline
From:Jason (brec@gmail.com)
Date:06/17/2008 07:31:59 PM
List:com.googlegroups.google-visualization-api

Well... I've seen different types of behavior here:

1) Multiple data series, one has no data for a timestamp - chart breaks 2) Single data series, no data for a timestamp - as long as you don't include the data point at all, no problem... no point generated in the chart 3) Any # of data series, no data except for annotation - IIRC, this works as expected, showing no data point, but showing the annotations on the graph.

I ended up settling with using 0 for missing values. I hope a better solution comes along though!

On Jun 17, 5:11 pm, opticrealm <opti@gmail.com> wrote:

I'm using a custom datasource modeled after the one provided by Google Spreadsheets.

I've noticed that if a 'number' cell for one of the lines is missing, the timeseries chart breaks.

I'm going to force my datasource to replace missing entries with zero temporarily, but I would like to request the writers of the timeseries gadget to add a more graceful way to handle this condition.

I would like the chart to understand that a datapoint is missing, and "break" the line between the two datapoints on either side of the missing datapoint.

I've notice the same behavior when the datasource is javascript:

        data.addRows(6);         data.setValue(0, 0, new Date(2008, 1 ,1));         data.setValue(0, 1, 30000);         data.setValue(0, 2, 40645);         data.setValue(1, 0, new Date(2008, 1 ,2));         data.setValue(1, 1, 14045);         data.setValue(1, 2, 20374);         data.setValue(2, 0, new Date(2008, 1 ,3));         data.setValue(2, 1, 55022); // missing datapoint          data.setValue(2, 2, 55555);         data.setValue(3, 0, new Date(2008, 1 ,4));         data.setValue(3, 1, 75284);         data.setValue(3, 2, 14334);         data.setValue(3, 3, 'Out of Stock');         data.setValue(3, 4, 'Ran out of stock on pens at 4pm');         data.setValue(4, 0, new Date(2008, 1 ,5));