.

wcfs/internal/xbtree.py

@@ -160,8 +160,6 @@ class Tree(object):
         return child0.firstbucket()
 
 
-
-
 # Bucket represents a bucket node.
 class Bucket(object):
     #   .keyv   () of keys
@@ -186,7 +184,7 @@ class Bucket(object):
         return Bucket(*b.keyv)
 
 
-# StructureOf returns internal structure of a BTree.
+# StructureOf returns internal structure of a ZODB BTree.
 #
 # The structure is represented as Tree and Bucket nodes.
 def StructureOf(znode):
@@ -235,7 +233,7 @@ def Restructure(ztree, newStructure):
     kv = dict(ztree)
 
     # walk original and new structures level by level.
-    # XXX push buckets till the end
+    # push buckets till the end, then
     # for each level we have A1...An "old" nodes, and B1...Bm "new" nodes.
     # if n == m - map A-B 1-to-1
     # if n < m  - we have to "insert" (m-n) new nodes
@@ -253,9 +251,10 @@ def Restructure(ztree, newStructure):
     #
     #          sum min d(a,b)     sum min d(a,b)
     #          a∈A b∈B            b∈B a∈A
-    # D(A,B) = ──────────────  +  ──────────────        XXX
+    # D(A,B) = ──────────────  +  ──────────────        XXX -> D(A,B) = |A.lo, B.lo| + |A.hi - B.hi|
     #               N(A)              N(B)
     #
+    # XXX ^^^ -> assignment problem
 
     # we will modify nodes from new set:
     # - node.Z              will point to associated znode
@@ -278,6 +277,7 @@ def Restructure(ztree, newStructure):
             assert not hasattr(rn.node, 'Z')
             rn.node.Z = None
 
+        # D(a,b)
         def D(a, b):
             def fin(v): # TeX hack: inf = 10000
                 if v == +inf: return +1E4
@@ -289,7 +289,7 @@ def Restructure(ztree, newStructure):
             return d(a.range.klo, b.range.klo) + \
                    d(a.range.khi, b.range.khi)
 
-        # prepare cost matrix
+        # cost matrix
         C = np.zeros((len(RNv), len(RZv)))
         for j in range(len(RNv)):       # "workers"
             for i in range(len(RZv)):   # "jobs"
@@ -297,9 +297,9 @@ def Restructure(ztree, newStructure):
 
         # find the assignments.
         # TODO try to avoid scipy dependency; we could also probably make
-        # assignments more efficiently taking that Av and Bv are key↑ and the
-        # property of D function so that if B2 > B1 (all keys in B2 > all keys
-        # in B1) and A < B1.hi, then D(B2, A) > D(B1, A).
+        # assignments more efficiently taking into account that Av and Bv are
+        # key↑ and the property of D function so that if B2 > B1 (all keys in
+        # B2 > all keys in B1) and A < B1.hi, then D(B2, A) > D(B1, A).
         """
         print()
         print('COST:')
@@ -342,33 +342,16 @@ def Restructure(ztree, newStructure):
     rbucketv  = [] # of _NodeInRange
     rlevelv_orig = copy.copy(rlevelv)
     while len(rlevelv) > 0 or len(zrlevelv) > 0:
-        if len(rlevelv) == 0:
-            rlevel  = []
-        else:
-            rlevel  = rlevelv.pop(0)
-
-        if len(zrlevelv) == 0:
-            zrlevel = []
-        else:
+        rlevel  = []
+        zrlevel = []
+        if len(rlevelv) > 0:
+            rlevel  = rlevelv .pop(0)
+        if len(zrlevelv) > 0:
             zrlevel = zrlevelv.pop(0)
 
         # filter-out buckets to be processed in the end
-        _ = []  # zrlevel -> zrlevel, zrbucketv
-        for zrn in zrlevel:
-            assert isinstance(zrn.node, (ztreeType, zbucketType))
-            if isinstance(zrn.node, zbucketType):
-                zrbucketv.append(zrn)
-            else:
-                _.append(zrn)
-        zrlevel = _
-        _ = []  # rlevel  ->  rlevel,  rbucketv
-        for rn in rlevel:
-            assert isinstance(rn.node, (Tree, Bucket))
-            if isinstance(rn.node, Bucket):
-                rbucketv.append(rn)
-            else:
-                _.append(rn)
-        rlevel = _
+        zrbucketv, zrlevel = _filter2(zrlevel, lambda zrn: isinstance(zrn.node, zbucketType))
+        rbucketv,   rlevel = _filter2(rlevel,  lambda  rn: isinstance(rn.node,  Bucket))
 
         if len(rlevel) == 0:
             continue
@@ -808,6 +791,16 @@ def _indent(prefix, text): # -> text
     text  = '\n'.join(textv)
     return text
 
+# _filter2(l,pred) = filter(l,pred), filter(l,!pred)
+def _filter2(l, pred):
+    t, f = [], []
+    for _ in l:
+        if pred(_):
+            t.append(_)
+        else:
+            f.append(_)
+    return t,f
+
 # _assertIncv asserts that values of vector v are strictly ↑
 def _assertIncv(v):
     prev = -inf

wcfs/internal/xbtree_test.py

@@ -664,3 +664,4 @@ def test_iterSplitByN():
     assert X(0,5, 3)    == ( [0,1,2,3,5], [0,1,2,4,5], [0,1,3,4,5], [0,2,3,4,5] )
     assert X(0,5, 4)    == ( [0,1,2,3,4,5], )
     assert X(0,5, 5)    == ()
+