Speed up spot reading

ppanggolin/formats/readBinaries.py

@@ -773,23 +773,27 @@ def read_rgp(pangenome: Pangenome, h5f: tables.File, disable_bar: bool = False):
 
 def read_spots(pangenome: Pangenome, h5f: tables.File, disable_bar: bool = False):
     """
-    Read hotspot in pangenome hdf5 file to add in pangenome object
+    Read hotspots in the pangenome HDF5 file and add them to the pangenome object.
 
-    :param pangenome: Pangenome object without spot
-    :param h5f: Pangenome HDF5 file with spot computed
-    :param disable_bar: Disable the progress bar
+    Args:
+        pangenome (Pangenome): Pangenome object.
+        h5f (tables.File): Pangenome HDF5 file with spots computed.
+        disable_bar (bool): Whether to disable the progress bar.
     """
     table = h5f.root.spots
     spots = {}
+    curr_spot_id = None
     for row in tqdm(read_chunks(table, chunk=20000), total=table.nrows, unit="spot", disable=disable_bar):
-        curr_spot = spots.get(int(row["spot"]))
-        if curr_spot is None:
-            curr_spot = Spot(int(row["spot"]))
-            spots[row["spot"]] = curr_spot
+        if curr_spot_id != int(row["spot"]):
+            curr_spot_id = int(row["spot"])
+            curr_spot = spots.get(curr_spot_id)
+            if curr_spot is None:
+                curr_spot = Spot(int(row["spot"]))
+                spots[row["spot"]] = curr_spot
         region = pangenome.get_region(row["RGP"].decode())
         curr_spot.add(region)
-        curr_spot.spot_2_families()
     for spot in spots.values():
+        spot.spot_2_families()
         pangenome.add_spot(spot)
     pangenome.status["spots"] = "Loaded"
 
@@ -1174,9 +1178,8 @@ def read_pangenome(pangenome, annotation: bool = False, gene_families: bool = Fa
             logging.getLogger("PPanGGOLiN").info("Reading the spots...")
             read_spots(pangenome, h5f, disable_bar=disable_bar)
         else:
-            raise Exception(f"The pangenome in file '{filename}' does not have spots information, "
-                            f"or has been improperly filled")
-
+            raise AttributeError(f"The pangenome in file '{pangenome.file}' does not have spots information, "
+                                 f"or has been improperly filled")
     if modules:
         if h5f.root.status._v_attrs.modules:
             logging.getLogger("PPanGGOLiN").info("Reading the modules...")