Merge branch 'check_virtual_dataset' into 'master'

Check HDF5 virtual sources

See merge request !35

tomoscan/io.py

@@ -27,15 +27,16 @@ __authors__ = ["W. de Nolf"]
 __license__ = "MIT"
 __date__ = "25/08/2020"
 
-
-import h5py
-from tomoscan.utils import SharedLockPool
-import os
 from contextlib import contextmanager
+import logging
+import os
 import traceback
 import errno
+import h5py
+from tomoscan.utils import SharedLockPool
 
 HASSWMR = h5py.version.hdf5_version_tuple >= h5py.get_config().swmr_min_hdf5_version
+_logger = logging.getLogger(__name__)
 
 
 class HDF5File(h5py.File):
@@ -127,3 +128,32 @@ def isErrno(e, errno):
     """
     # Because e.__cause__ is None for chained exceptions
     return "errno = {}".format(errno) in "".join(traceback.format_exc())
+
+
+def check_virtual_sources_exist(fname, data_path):
+    """
+    Check that a virtual dataset points to actual data.
+
+    :param str fname: HDF5 file path
+    :param str data_path: Path within the HDF5 file
+
+    :return bool res: Whether the virtual dataset points to actual data.
+    """
+    with HDF5File(fname, "r") as f:
+        if data_path not in f:
+            _logger.error("No dataset %s in file %s" % (data_path, fname))
+            return False
+        dptr = f[data_path]
+        if not dptr.is_virtual:
+            return True
+        for vsource in dptr.virtual_sources():
+            vsource_fname = os.path.join(
+                os.path.dirname(dptr.file.filename), vsource.file_name
+            )
+            if not os.path.isfile(vsource_fname):
+                _logger.error("No such file: %s" % vsource_fname)
+                return False
+            elif not check_virtual_sources_exist(vsource_fname, vsource.dset_name):
+                _logger.error("Error with virtual source %s" % vsource_fname)
+                return False
+    return True