Highcharts graphs-strange problem

[mm23]

There is a Highcharts graphs on my webpage http://vrijemeusibeniku.com/povijesni-prikaz . My WS has stored data from January 31. to June 18. and then from October 10. to October 14. and from November 4. continuously until now (like here http://vrijemeusibeniku.com/svi-podaci ).
If you look at my graphs eg.Temperatura (Temperature) http://vrijemeusibeniku.com/povijesni-prikaz for one year (one year = "1 godina") you can see the data from January 31, (which is ok), until 13 August (August = Kol.), although in August, July and half of June there is no data in my tables (dayfile, monthly and realtime) on my database! Can anyone tell me what it is and how to fix it?
The same is true for all parameters older than one month….

Many thanks!

[mcrossley]

You have two ways of presenting time related data to Highcharts..

1. You present the data, and tell Highcharts the start time and interval, this is the most compact, a single array of data plus a couple of values. However it requires that you contiguous data.

2. You present the data as a set of pairs of values, each data point has a corresponding date/time value as well, this is the method you have to use if your data is not contiguous. See http://www.highcharts.com/demo/spline-irregular-time though that uses three separate series, you could simplify it down to a single series.

[mm23]

Many thanks Mark. One more question please.......When the data is contiguous, and when not...I mean How much time should pass between the data to become "uncontiguous" from contiguous.....

[mcrossley]

Any missing data will cause the time stamps to 'slip', how much slip you can tolerate is up to you, I wouldn't tolerate much

[mm23]

Thanks Mark....I will not tolerate anything

[mm23]

......hmm, what will happen if I use a graph designed for a non contiguous data to present contiguous data?

Thanks!

[mcrossley]

It will display it fine, the only downside with non-contiguous data is that the required data set is much larger due to including the time stamp with every data value.

[mm23]

Ok, thanks Mark on a quick response