fix: fix adjusting contrast limits for rgb data #276
Conversation
src/napari_spatialdata/_view.py
Outdated
| has_adata = layer is not None and layer.metadata.get("adata") is not None | ||
|
|
||
| # has_adata is added so we see the channels in the view widget under vars | ||
| if layer is None or not is_image or not has_sdata or not has_adata: |
There was a problem hiding this comment.
maybe it should look like
| if layer is None or not is_image or not has_sdata or not has_adata: | |
| if layer is None or not is_image or not has_sdata or not has_adata or image.rgb: |
| current_point = list(event.value) | ||
| displayed = self._viewer.dims.displayed | ||
|
|
||
| for i, (lo_size, hi_size, cord) in enumerate(zip(layer.data[-1].shape, layer.data[0].shape, displayed)): | ||
| if i in displayed: | ||
| current_point[i] = slice(None) | ||
| else: | ||
| current_point[i] = int(cord * lo_size / hi_size) |
There was a problem hiding this comment.
This is logic for any dimensional data.
If you are 100% sure that data will always be 2D, then this logic may be obsolete.
If contrast limits needs to be calculated for the whole channel, not only the visible portion, then it is also overcomplicated. But then, for 3D data, should be no recalculation when change Z slice.
There was a problem hiding this comment.
In the future we will not be sure about this. SpatialData already does have 3D models. Where bugs occur for 3D napari-spatialdata should be adjusted.
|
I will first add #153 . That PR required some upstream changes that I have now implemented. I will then check how your PR interacts with it. |
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #276 +/- ##
==========================================
- Coverage 63.69% 63.08% -0.62%
==========================================
Files 19 19
Lines 2628 2636 +8
==========================================
- Hits 1674 1663 -11
- Misses 954 973 +19 ☔ View full report in Codecov by Sentry. |
While working on reproducing of performance problem for points and shapes, I found that adjusting contrast limits do not work for data with 3 channels, that are represented as RGB data.
This PR contains the following changes:
channel_changedfrom local function to class method_channel_changedto accept any dimensional data.AnnDatainitialization for rgb data