Add the 'load_data_index_db' function

beat/backend/python/data.py

@@ -1077,6 +1077,39 @@ def load_data_index(cache_root, hash_path):
 #----------------------------------------------------------
 
 
+def load_data_index_db(cache_root, hash_path):
+    filename = os.path.join(cache_root, hash_path)
+
+    if not os.path.exists(filename):
+        return None
+
+    retval = []
+
+    with open(filename, 'rb') as f:
+        data = json.loads(f.read())
+
+    current = data[0]
+
+    column_names = current.keys()
+
+    for col in range(0, len(column_names)):
+        retval.append([0])
+
+    for row, entry in enumerate(data[1:]):
+        for col, name in enumerate(column_names):
+            if entry[name] != current[name]:
+                retval[col].append(row + 1)
+                current[name] = entry[name]
+
+    for col in range(0, len(column_names)):
+        retval[col].append(len(data))
+
+    return retval
+
+
+#----------------------------------------------------------
+
+
 def _foundCommonIndices(lst):
     """Returns the list of common indices, given a list of list of indices
     """
@@ -1103,9 +1136,10 @@ def foundSplitRanges(lst, n_split):
     c = 0
     s = 0
     for i in range(1, len(ci_lst)):
-        if (ci_lst[i] - ci_lst[s] >= average_length and c < n_split - 1) or \
-                len(ci_lst) - i <= n_split - c:
+        if (ci_lst[i] >= (c + 1) * average_length and c < n_split - 1):
             res.append((ci_lst[s], ci_lst[i] - 1))
             s = i
             c += 1
+        elif i == len(ci_lst) - 1:
+            res.append((ci_lst[s], ci_lst[i] - 1))
     return res