994 exploration colormap #1107
994 exploration colormap #1107
Conversation
✅ Deploy Preview for veda-ui ready!
To edit notification comments on pull requests, go to your Netlify site configuration. |
|
I had a first look - awesome to see this feature materialize! 🎉 switching between colormaps and reversing them works - the tiles on the map change and use the right colormaps 🐞 - the legend graphics is off - the tiles are shaded the way I would expect (knowing Viridis or Plasma), but the graphics in the drop-down do not match.
|
Yeah, I think this is because I tried to rescale the colors based on the rescale values. I've removed this now since it makes the legend unclear. |
|
Ah, I see. I think we will always want to make use of the full color range. We only limit the data range. So there should not be a situation where the colormap would be rescaled / cropped, as far as I can see. |
| display: none; | ||
| `; | ||
|
|
||
| export const SEQUENTIAL_COLORMAPS = [ |
There was a problem hiding this comment.
🤔 Should we have a step to validate these colormaps are supported color maps?( from https://openveda.cloud/api/raster/colorMaps)? In reality, the supported colormap will change so rarely that I am not sure if we need to add this one more step, but then we will be safe from whichever update from the backend.
There was a problem hiding this comment.
Ah, I was hoping that we were doing that - fetching the supported colormaps from TiTiler instead of relying on another source (D3).
There was a problem hiding this comment.
Ah, I see; I missed that there is even an endpoint for getting the colors of each colormap using each colormap ID - ex. https://openveda.cloud/api/raster/colorMaps/magma. Then it might make sense to have only IDs for colormaps on the client side and get the values from the endpoint.
There was a problem hiding this comment.
I wasn't aware of the endpoint as well, thanks @j08lue!
@hanbyul-here Since we use d3-scale-chromatic already which supports (I think most of) the colorMaps by ID and to avoid sending requests on each selection of a colorMap to fetch the values, I don't think we need to hardcode them at all in the UI. In general, we could fetch the list from titiler, list them in the UI as options to choose from and use d3-scale-chromatic to interpolate the values. What do you think about this?
There was a problem hiding this comment.
Oh, don't we want to curate the colormaps (not all the colormap options from the colormap endpoint)? I meant the IDs for these curated colormaps can be on the front end. (Sorry for not being clear.)
I think it is always best to have one source of truth, so we don't have to think about how the colors from Titiler don't seem to match 100% with what the legend shows. (If that ever happens - So we probably should at least check where D3 colormap values are from and how the values match up with the titiler values.) I also understand your concerns about not making too many requests. Since this is the type of data that doesn't change very often, I wonder if we can make a request while building and save it as a local file? 🤔 (This sounds like a way to make this ticket go out of the scope 😅 )
There was a problem hiding this comment.
So titiler uses colorMaps from rio-tiler. And rio-tiler uses colorMaps (at least some that are currently on the client-side) derived from Matplotlib: https://cogeotiff.github.io/rio-tiler/colormap/#colormaps
d3-scale-chromatic also derives the ones we use from Matplotlib, which is why I think we will add unnecessary overhead if we send requests to fetch hexes per colorMap ID or save them in local files.
Oh, don't we want to curate the colormaps (not all the colormap options from the colormap endpoint
Yes I think this was the idea, to curate a selection of most widely used colors instead of showing all the supported ones. Users could still pass in a valid colorMap name which is not in the list at the moment (e.g. PuRd) and it should show up fine as a default value.
EDIT:
d3-scale-chromatic also derives the ones we use from Matplotlib
Ah, I have no proof for all of the colormaps though, only some. So it might make sense to save the ones which we're unsure of as hex objects in the veda-ui and simply use them. On the other hand, the colors seem to match what the tiles endpoint returns 😄
There was a problem hiding this comment.
I see the pros and cons of all options. I would really like a solution that makes maximum use of the nice TiTiler built-ins. But the decision is up to you!
I lined up the options here, for overview (using an ADR template):
Please feel free to review and let me know if I should open a proper ADR PR.
Related Ticket: {link related ticket here}
Description of Changes
{Update with description of changes made in this pull request}
Notes & Questions About Changes
{Add additonal notes and outstanding questions here related to changes in this pull request}
Validation / Testing
{Update with info on what can be manually validated in the Deploy Preview link for example "Validate style updates to selection modal do NOT affect cards"}