change: to accommodate renaming in agora

agora.dispatch_metadata_parser to parse_metadata

imageviewer.py

@@ -6,7 +6,7 @@ import h5py
 import napari
 import numpy as np
 from agora.io.cells import Cells
-from agora.io.metadata import dispatch_metadata_parser
+from agora.io.metadata import parse_metadata
 from aliby.io.image import dispatch_image
 from aliby.tile.tiler import Tiler
 
@@ -43,7 +43,8 @@ class BaseImageViewer(ABC):
                 tiles_dict[ch] = tiles
             else:
                 masks = [
-                    self.cells.at_time(tp, kind="mask").get(trap_id, []) for tp in tps
+                    self.cells.at_time(tp, kind="mask").get(trap_id, [])
+                    for tp in tps
                 ]
                 # some masks may be empty
                 default_mask = [np.ones(self.cells.tile_size).astype(bool)]
@@ -65,10 +66,13 @@ class BaseImageViewer(ABC):
         """Get uniquely labelled outlines for each cell time point."""
         # get outlines for each time point
         outlines = [
-            self.cells.at_time(tp, kind="edgemask").get(trap_id, []) for tp in tps
+            self.cells.at_time(tp, kind="edgemask").get(trap_id, [])
+            for tp in tps
         ]
         # get cell labels for each time point
-        cell_labels = [self.cells.labels_at_time(tp).get(trap_id, []) for tp in tps]
+        cell_labels = [
+            self.cells.labels_at_time(tp).get(trap_id, []) for tp in tps
+        ]
         # generate one image with all cell outlines uniquely labelled per tile
         labelled_outlines = [
             (
@@ -114,12 +118,16 @@ class BaseImageViewer(ABC):
         channels = list(tiles_dict.keys())
         # put time series into one array with dimensions TCZYX
         ydim, xdim = tiles_dict[list(tiles_dict.keys())[0]][0].shape
-        ts_images = np.zeros((tps.size, len(tiles_dict), 1, ydim, xdim)).astype(int)
+        ts_images = np.zeros(
+            (tps.size, len(tiles_dict), 1, ydim, xdim)
+        ).astype(int)
         ts_labels = np.zeros((tps.size, 1, ydim, xdim)).astype(int)
         # make array of time series of tiles
         for ch_index, channel in enumerate(tiles_dict):
             for tp_index in range(tps.size):
-                ts_images[tp_index, ch_index, 0, ...] = tiles_dict[channel][tp_index]
+                ts_images[tp_index, ch_index, 0, ...] = tiles_dict[channel][
+                    tp_index
+                ]
         # make array of time series of outlines with no channels dimension
         for tp_index in range(tps.size):
             ts_labels[tp_index, 0, ...] = outlines[tp_index]
@@ -145,7 +153,9 @@ class BaseImageViewer(ABC):
             tps = np.arange(self.cells.ntimepoints)
         elif type(tps) is int:
             tps = np.arange(tps)
-        ts_images, ts_labels, channels = self.get_data_for_viewing(trap_id, tps)
+        ts_images, ts_labels, channels = self.get_data_for_viewing(
+            trap_id, tps
+        )
         # launch napari
         viewer = napari.Viewer()
         viewer.add_image(
@@ -207,7 +217,7 @@ class RemoteImageViewer(BaseImageViewer):
         server_info = server_info or {
             k: self.attrs["parameters"]["general"][k] for k in self.credentials
         }
-        logfiles_meta = dispatch_metadata_parser(h5file_path.parent)
+        logfiles_meta = parse_metadata(h5file_path.parent)
         image_id = logfiles_meta.get("image_id")
         if image_id is None:
             with h5py.File(h5file_path, "r") as f: