worker_db: write hdf5 dataset metadata

artiq/master/worker_db.py

@@ -111,11 +111,12 @@ class DatasetManager:
         self._broadcaster = Notifier(dict())
         self.local = dict()
         self.archive = dict()
+        self.metadata = dict()
 
         self.ddb = ddb
         self._broadcaster.publish = ddb.update
 
-    def set(self, key, value, broadcast=False, persist=False, archive=True):
+    def set(self, key, value, metadata, broadcast, persist, archive):
         if persist:
             broadcast = True
 
@@ -123,7 +124,7 @@ class DatasetManager:
             logger.warning(f"Dataset '{key}' will not be stored. Both 'broadcast' and 'archive' are set to False.")
 
         if broadcast:
-            self._broadcaster[key] = persist, value
+            self._broadcaster[key] = persist, value, metadata
         elif key in self._broadcaster.raw_view:
             del self._broadcaster[key]
 
@@ -132,6 +133,8 @@ class DatasetManager:
         elif key in self.local:
             del self.local[key]
         
+        self.metadata[key] = metadata
+
     def _get_mutation_target(self, key):
         target = self.local.get(key, None)
         if key in self._broadcaster.raw_view:
@@ -166,21 +169,30 @@ class DatasetManager:
             self.archive[key] = data
         return data
 
+    def get_metadata(self, key):
+        if key in self.metadata:
+            return self.metadata[key]
+        return self.ddb.get_metadata(key)
+
     def write_hdf5(self, f):
         datasets_group = f.create_group("datasets")
         for k, v in self.local.items():
-            _write(datasets_group, k, v)
+            m = self.metadata.get(k, {})
+            _write(datasets_group, k, v, m)
 
         archive_group = f.create_group("archive")
         for k, v in self.archive.items():
-            _write(archive_group, k, v)
+            m = self.metadata.get(k, {})
+            _write(archive_group, k, v, m)
 
 
-def _write(group, k, v):
+def _write(group, k, v, m):
     # Add context to exception message when the user writes a dataset that is
     # not representable in HDF5.
     try:
         group[k] = v
+        for key, val in m.items():
+            group[k].attrs[key] = val
     except TypeError as e:
         raise TypeError("Error writing dataset '{}' of type '{}': {}".format(
             k, type(v), e))