Replies: 1 comment · 4 replies
-
|
I would just make a second figure with one subplot and display the data in the clicked subplot, something like (not tested, just to give you an idea): fig_many = fpl.Figure(shape=(m, n))
fig_detailed = fpl.Figure(shape=(1, 1))
# add graphics to your subplots etc, I recommend naming them for convenience
# initialize the detailed fig with data from the first subplot
fig_detailed[0, 0].add_line_collection(fig_many[0, 0]["line-coll"].data[:], colors=line_stack.data[:], name="line-collection") # set any other properties you want
@fig_many.renderer.add_event_handler("click")
def fig_many_clicked(ev):
for subplot in fig_many:
pos = (ev.x, ev.y)
if subplot.map_screen_to_world(pos) is not None:
# this is the subplot that has been clicked, set the data of the line collection in the detailed subplot
for g in subplot["line-collecection"].graphics:
fig_detailed[0, 0]["line-collection"].data = g.data[:]If your lines are huge and GPU VRAM is a real concern you can share the buffer, but if you're plotting multiple lines to start with I'm assuming that's not an issue here. |
Beta Was this translation helpful? Give feedback.
-
|
Brilliant! If someone comes looking, I got my example to work with some minor changes to the code above: # initialize the detailed fig with data from the first subplot
fig_detailed[0, 0].add_line_stack(fig_many[0, 0]["line-coll"].data[:], cmap="rainbow", name="line-collection") # set any other properties you want
@fig_many.renderer.add_event_handler("click")
def fig_many_clicked(ev):
for subplot in list(fig_many):
pos = (ev.x, ev.y)
if subplot.map_screen_to_world(pos) is not None:
# this is the subplot that has been clicked, set the data of the line stack in the detailed subplot
fig_detailed[0, 0]["line-collection"].data = subplot["line-coll"].data[:]Note that you need to click on the actual subplot graph. Clicking on the subplot title does not work. |
Beta Was this translation helpful? Give feedback.
-
|
Does this actually work? fig_detailed[0, 0]["line-collection"].data = subplot["line-coll"].data[:]I can't remember if we made line collections settable like that. |
Beta Was this translation helpful? Give feedback.
-
|
It does for my simple dummy example above. I ran into some issues with the real data, which I think is related to not all time series having the same length. I solved it this way, which at least works for all my use-cases (might not be the smartest way): @fig_many.renderer.add_event_handler("click")
def fig_many_clicked(ev):
for subplot in list(fig_many):
pos = (ev.x, ev.y)
if subplot.map_screen_to_world(pos) is not None:
fig_detailed[0, 0].delete_graphic(fig_detailed[0, 0]["line-collection"])
fig_detailed[0, 0].add_line_stack(subplot["line-coll"].data[:], cmap="rainbow", name="line-collection") |
Beta Was this translation helpful? Give feedback.
-
|
yes GPU buffers are always of a fixed size so you have to delete the graphic and create a new one. For the future we have ideas for swapping out the buffer under the hood and keeping the graphic: #869 |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
Hi
I am looking at a lot of (stacked) time series grouped by some criteria. I would like to be able to click on one of the graphs in the example below and have a new pop-up window showing a larger version of the graph. Is this possible with current features?
Beta Was this translation helpful? Give feedback.
All reactions