Hfile pycbc version information check

bin/all_sky_search/pycbc_add_statmap

@@ -3,10 +3,11 @@
 with more than one ifo combination available. Cluster to keep foreground
 coincs with the highest stat value.
 """
-import h5py, numpy as np, argparse, logging, pycbc, pycbc.events, pycbc.io
+import numpy as np, argparse, logging, pycbc, pycbc.events, pycbc.io
 import pycbc.version
 import pycbc.conversions as conv
 from pycbc.events import coinc, significance
+from pycbc.io import HFile
 from ligo import segments
 import sys, copy
 
@@ -91,9 +92,9 @@ else :
 
 pycbc.init_logging(args.verbose)
 
-files = [h5py.File(n, 'r') for n in args.statmap_files]
+files = [HFile(n, 'r') for n in args.statmap_files]
 
-f = h5py.File(args.output_file, "w")
+f = HFile(args.output_file, "w")
 
 # Work out the combinations of detectors used by each input file
 all_ifo_combos = []
@@ -308,7 +309,7 @@ if injection_style:
     # if background files are provided, this is being used for injections
     # use provided background files to calculate the FARs
     for bg_fname in args.background_files:
-        bg_f = h5py.File(bg_fname, 'r')
+        bg_f = HFile(bg_fname, 'r')
         ifo_combo_key = bg_f.attrs['ifos'].replace(' ','')
         _, far[ifo_combo_key] = significance.get_far(
                 bg_f['background/stat'][:],

pycbc/io/hdf.py

@@ -24,11 +24,47 @@
 from pycbc import events, conversions, pnutils
 from pycbc.events import ranking, veto
 from pycbc.events import mean_if_greater_than_zero
+from pycbc.version import version as pycbc_version
 
 
 class HFile(h5py.File):
     """ Low level extensions to the capabilities of reading an hdf5 File
     """
+    def __init__(
+        self,
+        filename,
+        permission='r',
+        check_pycbc_version=True,
+        **kwargs
+    ):
+        h5py.File.__init__(self, filename, permission, **kwargs)
+        if permission in ['r','r+', 'a'] and check_pycbc_version:
+            # Check the pycbc version in the file matches the current one
+            if 'pycbc_version' in self.attrs:
+                if not self.attrs['pycbc_version'] == pycbc_version:
+                    logging.info(
+                        "PyCBC version of the file (%s) does not match the "
+                        "one currently being used (%s). Results may not be "
+                        "as expected.",
+                        self.attrs['pycbc_version'],
+                        pycbc_version
+                    )
+        if permission in ['w', 'x', 'w-', 'r+', 'a'] and check_pycbc_version:
+            if permission in ['r+','a']:
+                original_version = self.attrs['pycbc_version'] \
+                    if 'pycbc_version' in self.attrs else 'None'
+                if original_version != 'None' and not \
+                        original_version == pycbc_version:
+                    # Read/write on the file - update the pycbc version,
+                    # but warn
+                    logging.warning(
+                        "File opened with read and write permissions, "
+                        "updating pycbc_version from %s to %s.",
+                        original_version,
+                        pycbc_version
+                    )
+            self.attrs['pycbc_version'] = pycbc_version
+
     def select(self, fcn, *args, chunksize=10**6, derived=None, group='',
                return_data=True, premask=None):
         """ Return arrays from an hdf5 file that satisfy the given function