Hi Bryan,

I have a question about using different lengths of data in streaming plots via the bokeh server. Here is what I have now

# create the source

source = ColumnDataSource(dict(time=np.zeros(10), x=np.zeros(10)))

# Now create the plot

fig = figure(… details…)

fig.plot(x=‘time’,y=‘x’, source=source)

# now during each update

def update(t):

newData = dict(time=[t],x=“new value”)

source.stream(newData,rollover=10)

However, now I want to add a second plot, a histogram. This time I dont want the data in there to roll, I just want it to accumulate. So in effect:

# create the source

source = ColumnDataSource(dict(time=np.zeros(10), x=np.zeros(10),otherData=np.zeros(arbitrary)))

# Now create the plot

fig = figure(… details…)

fig.plot(x=‘time’,y=‘x’, source=source)

fig2 = figure(… details…)

fig2.histogram(data=‘otherData’, source=source)

# update

def update(t):

newData = dict(time=[t],x=“new value”, otherData=“new value”)

source.stream(newData,rollover=10, dont roll over other data)

Is this possible? If all arrays in the ColumnDataSource must be the same length, can I push 2 ColumnDataSourcess in source.stream()?

Thanks a lot in advance for the help!