Fix azimuth bugs

containers/azimuth/context/main.py

@@ -1,7 +1,7 @@
 import logging
 import subprocess
-from pathlib import Path
 import typing as t
+from pathlib import Path
 
 import anndata
 import pandas
@@ -72,19 +72,21 @@ def copy_annotations(
 
     def run_azimuth_scripts(self, matrix_path: Path, reference_data: Path):
         script_command = ["Rscript", "/run_azimuth.R", matrix_path, reference_data]
-        subprocess.run(script_command, capture_output=True, check=True)
+        subprocess.run(script_command, capture_output=True, check=True, text=True)
         return "./result.h5ad"
 
     def find_reference_data(self, organ: str, dir: Path):
         def is_reference_data_candidate(path: Path):
             return path.is_dir() and organ.lower() in path.name.lower()
 
-        return self._find_in_dir(
+        subdir = self._find_in_dir(
             dir,
             is_reference_data_candidate,
             f"Cannot find reference data for organ '{organ}'",
             f"Multiple reference data candidates for organ '{organ}'",
         )
+        # idx.annoy and ref.Rds is always located inside an 'azimuth' subdirectory
+        return subdir / "azimuth"
 
     def _find_in_dir(
         self, dir: Path, cond: t.Callable[[Path], bool], error_msg: str, warn_msg: str

src/algorithm/report.py

@@ -3,6 +3,7 @@
 import io
 import json
 import pprint
+import subprocess
 import traceback
 import typing as t
 from pathlib import Path
@@ -67,6 +68,11 @@ def format_cause(self, result: dict):
         if isinstance(cause, Exception):
             result["cause"] = repr(cause)
             result["traceback"] = traceback.format_tb(cause.__traceback__)
+            if isinstance(cause, subprocess.CalledProcessError):
+                # stdout and stderr does not show in the repr of a CalledProcessError
+                # Add them to the result here instead to give more context on errors
+                result["stdout"] = cause.stdout
+                result["stderr"] = cause.stderr
         else:
             stream = io.StringIO()
             pprint.pprint(cause, stream=stream)