[pypy-commit] pypy virtual-dicts: merged default

alex_gaynor noreply at buildbot.pypy.org
Thu Oct 27 18:40:34 CEST 2011


Author: Alex Gaynor <alex.gaynor at gmail.com>
Branch: virtual-dicts
Changeset: r48511:5e267e0a16b0
Date: 2011-10-27 12:31 -0400
http://bitbucket.org/pypy/pypy/changeset/5e267e0a16b0/

Log:	merged default

diff --git a/lib-python/modified-2.7/heapq.py b/lib-python/modified-2.7/heapq.py
new file mode 100644
--- /dev/null
+++ b/lib-python/modified-2.7/heapq.py
@@ -0,0 +1,442 @@
+# -*- coding: latin-1 -*-
+
+"""Heap queue algorithm (a.k.a. priority queue).
+
+Heaps are arrays for which a[k] <= a[2*k+1] and a[k] <= a[2*k+2] for
+all k, counting elements from 0.  For the sake of comparison,
+non-existing elements are considered to be infinite.  The interesting
+property of a heap is that a[0] is always its smallest element.
+
+Usage:
+
+heap = []            # creates an empty heap
+heappush(heap, item) # pushes a new item on the heap
+item = heappop(heap) # pops the smallest item from the heap
+item = heap[0]       # smallest item on the heap without popping it
+heapify(x)           # transforms list into a heap, in-place, in linear time
+item = heapreplace(heap, item) # pops and returns smallest item, and adds
+                               # new item; the heap size is unchanged
+
+Our API differs from textbook heap algorithms as follows:
+
+- We use 0-based indexing.  This makes the relationship between the
+  index for a node and the indexes for its children slightly less
+  obvious, but is more suitable since Python uses 0-based indexing.
+
+- Our heappop() method returns the smallest item, not the largest.
+
+These two make it possible to view the heap as a regular Python list
+without surprises: heap[0] is the smallest item, and heap.sort()
+maintains the heap invariant!
+"""
+
+# Original code by Kevin O'Connor, augmented by Tim Peters and Raymond Hettinger
+
+__about__ = """Heap queues
+
+[explanation by Fran&#65533;ois Pinard]
+
+Heaps are arrays for which a[k] <= a[2*k+1] and a[k] <= a[2*k+2] for
+all k, counting elements from 0.  For the sake of comparison,
+non-existing elements are considered to be infinite.  The interesting
+property of a heap is that a[0] is always its smallest element.
+
+The strange invariant above is meant to be an efficient memory
+representation for a tournament.  The numbers below are `k', not a[k]:
+
+                                   0
+
+                  1                                 2
+
+          3               4                5               6
+
+      7       8       9       10      11      12      13      14
+
+    15 16   17 18   19 20   21 22   23 24   25 26   27 28   29 30
+
+
+In the tree above, each cell `k' is topping `2*k+1' and `2*k+2'.  In
+an usual binary tournament we see in sports, each cell is the winner
+over the two cells it tops, and we can trace the winner down the tree
+to see all opponents s/he had.  However, in many computer applications
+of such tournaments, we do not need to trace the history of a winner.
+To be more memory efficient, when a winner is promoted, we try to
+replace it by something else at a lower level, and the rule becomes
+that a cell and the two cells it tops contain three different items,
+but the top cell "wins" over the two topped cells.
+
+If this heap invariant is protected at all time, index 0 is clearly
+the overall winner.  The simplest algorithmic way to remove it and
+find the "next" winner is to move some loser (let's say cell 30 in the
+diagram above) into the 0 position, and then percolate this new 0 down
+the tree, exchanging values, until the invariant is re-established.
+This is clearly logarithmic on the total number of items in the tree.
+By iterating over all items, you get an O(n ln n) sort.
+
+A nice feature of this sort is that you can efficiently insert new
+items while the sort is going on, provided that the inserted items are
+not "better" than the last 0'th element you extracted.  This is
+especially useful in simulation contexts, where the tree holds all
+incoming events, and the "win" condition means the smallest scheduled
+time.  When an event schedule other events for execution, they are
+scheduled into the future, so they can easily go into the heap.  So, a
+heap is a good structure for implementing schedulers (this is what I
+used for my MIDI sequencer :-).
+
+Various structures for implementing schedulers have been extensively
+studied, and heaps are good for this, as they are reasonably speedy,
+the speed is almost constant, and the worst case is not much different
+than the average case.  However, there are other representations which
+are more efficient overall, yet the worst cases might be terrible.
+
+Heaps are also very useful in big disk sorts.  You most probably all
+know that a big sort implies producing "runs" (which are pre-sorted
+sequences, which size is usually related to the amount of CPU memory),
+followed by a merging passes for these runs, which merging is often
+very cleverly organised[1].  It is very important that the initial
+sort produces the longest runs possible.  Tournaments are a good way
+to that.  If, using all the memory available to hold a tournament, you
+replace and percolate items that happen to fit the current run, you'll
+produce runs which are twice the size of the memory for random input,
+and much better for input fuzzily ordered.
+
+Moreover, if you output the 0'th item on disk and get an input which
+may not fit in the current tournament (because the value "wins" over
+the last output value), it cannot fit in the heap, so the size of the
+heap decreases.  The freed memory could be cleverly reused immediately
+for progressively building a second heap, which grows at exactly the
+same rate the first heap is melting.  When the first heap completely
+vanishes, you switch heaps and start a new run.  Clever and quite
+effective!
+
+In a word, heaps are useful memory structures to know.  I use them in
+a few applications, and I think it is good to keep a `heap' module
+around. :-)
+
+--------------------
+[1] The disk balancing algorithms which are current, nowadays, are
+more annoying than clever, and this is a consequence of the seeking
+capabilities of the disks.  On devices which cannot seek, like big
+tape drives, the story was quite different, and one had to be very
+clever to ensure (far in advance) that each tape movement will be the
+most effective possible (that is, will best participate at
+"progressing" the merge).  Some tapes were even able to read
+backwards, and this was also used to avoid the rewinding time.
+Believe me, real good tape sorts were quite spectacular to watch!
+From all times, sorting has always been a Great Art! :-)
+"""
+
+__all__ = ['heappush', 'heappop', 'heapify', 'heapreplace', 'merge',
+           'nlargest', 'nsmallest', 'heappushpop']
+
+from itertools import islice, repeat, count, imap, izip, tee, chain
+from operator import itemgetter
+import bisect
+
+def heappush(heap, item):
+    """Push item onto heap, maintaining the heap invariant."""
+    heap.append(item)
+    _siftdown(heap, 0, len(heap)-1)
+
+def heappop(heap):
+    """Pop the smallest item off the heap, maintaining the heap invariant."""
+    lastelt = heap.pop()    # raises appropriate IndexError if heap is empty
+    if heap:
+        returnitem = heap[0]
+        heap[0] = lastelt
+        _siftup(heap, 0)
+    else:
+        returnitem = lastelt
+    return returnitem
+
+def heapreplace(heap, item):
+    """Pop and return the current smallest value, and add the new item.
+
+    This is more efficient than heappop() followed by heappush(), and can be
+    more appropriate when using a fixed-size heap.  Note that the value
+    returned may be larger than item!  That constrains reasonable uses of
+    this routine unless written as part of a conditional replacement:
+
+        if item > heap[0]:
+            item = heapreplace(heap, item)
+    """
+    returnitem = heap[0]    # raises appropriate IndexError if heap is empty
+    heap[0] = item
+    _siftup(heap, 0)
+    return returnitem
+
+def heappushpop(heap, item):
+    """Fast version of a heappush followed by a heappop."""
+    if heap and heap[0] < item:
+        item, heap[0] = heap[0], item
+        _siftup(heap, 0)
+    return item
+
+def heapify(x):
+    """Transform list into a heap, in-place, in O(len(heap)) time."""
+    n = len(x)
+    # Transform bottom-up.  The largest index there's any point to looking at
+    # is the largest with a child index in-range, so must have 2*i + 1 < n,
+    # or i < (n-1)/2.  If n is even = 2*j, this is (2*j-1)/2 = j-1/2 so
+    # j-1 is the largest, which is n//2 - 1.  If n is odd = 2*j+1, this is
+    # (2*j+1-1)/2 = j so j-1 is the largest, and that's again n//2-1.
+    for i in reversed(xrange(n//2)):
+        _siftup(x, i)
+
+def nlargest(n, iterable):
+    """Find the n largest elements in a dataset.
+
+    Equivalent to:  sorted(iterable, reverse=True)[:n]
+    """
+    if n < 0: # for consistency with the c impl
+        return []
+    it = iter(iterable)
+    result = list(islice(it, n))
+    if not result:
+        return result
+    heapify(result)
+    _heappushpop = heappushpop
+    for elem in it:
+        _heappushpop(result, elem)
+    result.sort(reverse=True)
+    return result
+
+def nsmallest(n, iterable):
+    """Find the n smallest elements in a dataset.
+
+    Equivalent to:  sorted(iterable)[:n]
+    """
+    if n < 0: # for consistency with the c impl
+        return []
+    if hasattr(iterable, '__len__') and n * 10 <= len(iterable):
+        # For smaller values of n, the bisect method is faster than a minheap.
+        # It is also memory efficient, consuming only n elements of space.
+        it = iter(iterable)
+        result = sorted(islice(it, 0, n))
+        if not result:
+            return result
+        insort = bisect.insort
+        pop = result.pop
+        los = result[-1]    # los --> Largest of the nsmallest
+        for elem in it:
+            if los <= elem:
+                continue
+            insort(result, elem)
+            pop()
+            los = result[-1]
+        return result
+    # An alternative approach manifests the whole iterable in memory but
+    # saves comparisons by heapifying all at once.  Also, saves time
+    # over bisect.insort() which has O(n) data movement time for every
+    # insertion.  Finding the n smallest of an m length iterable requires
+    #    O(m) + O(n log m) comparisons.
+    h = list(iterable)
+    heapify(h)
+    return map(heappop, repeat(h, min(n, len(h))))
+
+# 'heap' is a heap at all indices >= startpos, except possibly for pos.  pos
+# is the index of a leaf with a possibly out-of-order value.  Restore the
+# heap invariant.
+def _siftdown(heap, startpos, pos):
+    newitem = heap[pos]
+    # Follow the path to the root, moving parents down until finding a place
+    # newitem fits.
+    while pos > startpos:
+        parentpos = (pos - 1) >> 1
+        parent = heap[parentpos]
+        if newitem < parent:
+            heap[pos] = parent
+            pos = parentpos
+            continue
+        break
+    heap[pos] = newitem
+
+# The child indices of heap index pos are already heaps, and we want to make
+# a heap at index pos too.  We do this by bubbling the smaller child of
+# pos up (and so on with that child's children, etc) until hitting a leaf,
+# then using _siftdown to move the oddball originally at index pos into place.
+#
+# We *could* break out of the loop as soon as we find a pos where newitem <=
+# both its children, but turns out that's not a good idea, and despite that
+# many books write the algorithm that way.  During a heap pop, the last array
+# element is sifted in, and that tends to be large, so that comparing it
+# against values starting from the root usually doesn't pay (= usually doesn't
+# get us out of the loop early).  See Knuth, Volume 3, where this is
+# explained and quantified in an exercise.
+#
+# Cutting the # of comparisons is important, since these routines have no
+# way to extract "the priority" from an array element, so that intelligence
+# is likely to be hiding in custom __cmp__ methods, or in array elements
+# storing (priority, record) tuples.  Comparisons are thus potentially
+# expensive.
+#
+# On random arrays of length 1000, making this change cut the number of
+# comparisons made by heapify() a little, and those made by exhaustive
+# heappop() a lot, in accord with theory.  Here are typical results from 3
+# runs (3 just to demonstrate how small the variance is):
+#
+# Compares needed by heapify     Compares needed by 1000 heappops
+# --------------------------     --------------------------------
+# 1837 cut to 1663               14996 cut to 8680
+# 1855 cut to 1659               14966 cut to 8678
+# 1847 cut to 1660               15024 cut to 8703
+#
+# Building the heap by using heappush() 1000 times instead required
+# 2198, 2148, and 2219 compares:  heapify() is more efficient, when
+# you can use it.
+#
+# The total compares needed by list.sort() on the same lists were 8627,
+# 8627, and 8632 (this should be compared to the sum of heapify() and
+# heappop() compares):  list.sort() is (unsurprisingly!) more efficient
+# for sorting.
+
+def _siftup(heap, pos):
+    endpos = len(heap)
+    startpos = pos
+    newitem = heap[pos]
+    # Bubble up the smaller child until hitting a leaf.
+    childpos = 2*pos + 1    # leftmost child position
+    while childpos < endpos:
+        # Set childpos to index of smaller child.
+        rightpos = childpos + 1
+        if rightpos < endpos and not heap[childpos] < heap[rightpos]:
+            childpos = rightpos
+        # Move the smaller child up.
+        heap[pos] = heap[childpos]
+        pos = childpos
+        childpos = 2*pos + 1
+    # The leaf at pos is empty now.  Put newitem there, and bubble it up
+    # to its final resting place (by sifting its parents down).
+    heap[pos] = newitem
+    _siftdown(heap, startpos, pos)
+
+# If available, use C implementation
+try:
+    from _heapq import *
+except ImportError:
+    pass
+
+def merge(*iterables):
+    '''Merge multiple sorted inputs into a single sorted output.
+
+    Similar to sorted(itertools.chain(*iterables)) but returns a generator,
+    does not pull the data into memory all at once, and assumes that each of
+    the input streams is already sorted (smallest to largest).
+
+    >>> list(merge([1,3,5,7], [0,2,4,8], [5,10,15,20], [], [25]))
+    [0, 1, 2, 3, 4, 5, 5, 7, 8, 10, 15, 20, 25]
+
+    '''
+    _heappop, _heapreplace, _StopIteration = heappop, heapreplace, StopIteration
+
+    h = []
+    h_append = h.append
+    for itnum, it in enumerate(map(iter, iterables)):
+        try:
+            next = it.next
+            h_append([next(), itnum, next])
+        except _StopIteration:
+            pass
+    heapify(h)
+
+    while 1:
+        try:
+            while 1:
+                v, itnum, next = s = h[0]   # raises IndexError when h is empty
+                yield v
+                s[0] = next()               # raises StopIteration when exhausted
+                _heapreplace(h, s)          # restore heap condition
+        except _StopIteration:
+            _heappop(h)                     # remove empty iterator
+        except IndexError:
+            return
+
+# Extend the implementations of nsmallest and nlargest to use a key= argument
+_nsmallest = nsmallest
+def nsmallest(n, iterable, key=None):
+    """Find the n smallest elements in a dataset.
+
+    Equivalent to:  sorted(iterable, key=key)[:n]
+    """
+    # Short-cut for n==1 is to use min() when len(iterable)>0
+    if n == 1:
+        it = iter(iterable)
+        head = list(islice(it, 1))
+        if not head:
+            return []
+        if key is None:
+            return [min(chain(head, it))]
+        return [min(chain(head, it), key=key)]
+
+    # When n>=size, it's faster to use sort()
+    try:
+        size = len(iterable)
+    except (TypeError, AttributeError):
+        pass
+    else:
+        if n >= size:
+            return sorted(iterable, key=key)[:n]
+
+    # When key is none, use simpler decoration
+    if key is None:
+        it = izip(iterable, count())                        # decorate
+        result = _nsmallest(n, it)
+        return map(itemgetter(0), result)                   # undecorate
+
+    # General case, slowest method
+    in1, in2 = tee(iterable)
+    it = izip(imap(key, in1), count(), in2)                 # decorate
+    result = _nsmallest(n, it)
+    return map(itemgetter(2), result)                       # undecorate
+
+_nlargest = nlargest
+def nlargest(n, iterable, key=None):
+    """Find the n largest elements in a dataset.
+
+    Equivalent to:  sorted(iterable, key=key, reverse=True)[:n]
+    """
+
+    # Short-cut for n==1 is to use max() when len(iterable)>0
+    if n == 1:
+        it = iter(iterable)
+        head = list(islice(it, 1))
+        if not head:
+            return []
+        if key is None:
+            return [max(chain(head, it))]
+        return [max(chain(head, it), key=key)]
+
+    # When n>=size, it's faster to use sort()
+    try:
+        size = len(iterable)
+    except (TypeError, AttributeError):
+        pass
+    else:
+        if n >= size:
+            return sorted(iterable, key=key, reverse=True)[:n]
+
+    # When key is none, use simpler decoration
+    if key is None:
+        it = izip(iterable, count(0,-1))                    # decorate
+        result = _nlargest(n, it)
+        return map(itemgetter(0), result)                   # undecorate
+
+    # General case, slowest method
+    in1, in2 = tee(iterable)
+    it = izip(imap(key, in1), count(0,-1), in2)             # decorate
+    result = _nlargest(n, it)
+    return map(itemgetter(2), result)                       # undecorate
+
+if __name__ == "__main__":
+    # Simple sanity test
+    heap = []
+    data = [1, 3, 5, 7, 9, 2, 4, 6, 8, 0]
+    for item in data:
+        heappush(heap, item)
+    sort = []
+    while heap:
+        sort.append(heappop(heap))
+    print sort
+
+    import doctest
+    doctest.testmod()
diff --git a/lib-python/modified-2.7/test/test_heapq.py b/lib-python/modified-2.7/test/test_heapq.py
--- a/lib-python/modified-2.7/test/test_heapq.py
+++ b/lib-python/modified-2.7/test/test_heapq.py
@@ -186,6 +186,11 @@
         self.assertFalse(sys.modules['heapq'] is self.module)
         self.assertTrue(hasattr(self.module.heapify, 'func_code'))
 
+    def test_islice_protection(self):
+        m = self.module
+        self.assertFalse(m.nsmallest(-1, [1]))
+        self.assertFalse(m.nlargest(-1, [1]))
+
 
 class TestHeapC(TestHeap):
     module = c_heapq
diff --git a/pypy/jit/backend/llsupport/regalloc.py b/pypy/jit/backend/llsupport/regalloc.py
--- a/pypy/jit/backend/llsupport/regalloc.py
+++ b/pypy/jit/backend/llsupport/regalloc.py
@@ -178,8 +178,6 @@
         cur_max_age = -1
         candidate = None
         for next in self.reg_bindings:
-            if isinstance(next, TempBox):
-                continue
             reg = self.reg_bindings[next]
             if next in forbidden_vars:
                 continue
diff --git a/pypy/jit/backend/x86/assembler.py b/pypy/jit/backend/x86/assembler.py
--- a/pypy/jit/backend/x86/assembler.py
+++ b/pypy/jit/backend/x86/assembler.py
@@ -1596,11 +1596,26 @@
     genop_getarrayitem_gc_pure = genop_getarrayitem_gc
     genop_getarrayitem_raw = genop_getarrayitem_gc
 
+    def _get_interiorfield_addr(self, temp_loc, index_loc, itemsize_loc,
+                                base_loc, ofs_loc):
+        assert isinstance(itemsize_loc, ImmedLoc)
+        if isinstance(index_loc, ImmedLoc):
+            temp_loc = imm(index_loc.value * itemsize_loc.value)
+        else:
+            # XXX should not use IMUL in most cases
+            assert isinstance(temp_loc, RegLoc)
+            assert isinstance(index_loc, RegLoc)
+            self.mc.IMUL_rri(temp_loc.value, index_loc.value,
+                             itemsize_loc.value)
+        assert isinstance(ofs_loc, ImmedLoc)
+        return AddressLoc(base_loc, temp_loc, 0, ofs_loc.value)
+
     def genop_getinteriorfield_gc(self, op, arglocs, resloc):
-        base_loc, ofs_loc, itemsize_loc, fieldsize_loc, index_loc, sign_loc = arglocs
-        # XXX should not use IMUL in most cases
-        self.mc.IMUL(index_loc, itemsize_loc)
-        src_addr = AddressLoc(base_loc, index_loc, 0, ofs_loc.value)
+        (base_loc, ofs_loc, itemsize_loc, fieldsize_loc,
+            index_loc, sign_loc) = arglocs
+        src_addr = self._get_interiorfield_addr(resloc, index_loc,
+                                                itemsize_loc, base_loc,
+                                                ofs_loc)
         self.load_from_mem(resloc, src_addr, fieldsize_loc, sign_loc)
 
 
@@ -1611,13 +1626,11 @@
         self.save_into_mem(dest_addr, value_loc, size_loc)
 
     def genop_discard_setinteriorfield_gc(self, op, arglocs):
-        base_loc, ofs_loc, itemsize_loc, fieldsize_loc, index_loc, value_loc = arglocs
-        # XXX should not use IMUL in most cases
-        if isinstance(index_loc, ImmedLoc):
-            index_loc = imm(index_loc.value * itemsize_loc.value)
-        else:
-            self.mc.IMUL(index_loc, itemsize_loc)
-        dest_addr = AddressLoc(base_loc, index_loc, 0, ofs_loc.value)
+        (base_loc, ofs_loc, itemsize_loc, fieldsize_loc,
+            index_loc, temp_loc, value_loc) = arglocs
+        dest_addr = self._get_interiorfield_addr(temp_loc, index_loc,
+                                                 itemsize_loc, base_loc,
+                                                 ofs_loc)
         self.save_into_mem(dest_addr, value_loc, fieldsize_loc)
 
     def genop_discard_setarrayitem_gc(self, op, arglocs):
diff --git a/pypy/jit/backend/x86/regalloc.py b/pypy/jit/backend/x86/regalloc.py
--- a/pypy/jit/backend/x86/regalloc.py
+++ b/pypy/jit/backend/x86/regalloc.py
@@ -1046,16 +1046,26 @@
             need_lower_byte = True
         else:
             need_lower_byte = False
-        base_loc = self.rm.make_sure_var_in_reg(op.getarg(0), args)
-        tempvar = TempBox()
-        index_loc = self.rm.force_result_in_reg(tempvar, op.getarg(1), args)
-        # we're free to modify index now
-        value_loc = self.make_sure_var_in_reg(op.getarg(2), args,
+        box_base, box_index, box_value = args
+        base_loc = self.rm.make_sure_var_in_reg(box_base, args)
+        index_loc = self.rm.make_sure_var_in_reg(box_index, args)
+        value_loc = self.make_sure_var_in_reg(box_value, args,
                                               need_lower_byte=need_lower_byte)
-        self.rm.possibly_free_var(tempvar)
-        self.possibly_free_vars(args)
+        # If 'index_loc' is not an immediate, then we need a 'temp_loc' that
+        # is a register whose value will be destroyed.  It's fine to destroy
+        # the same register as 'index_loc', but not the other ones.
+        self.rm.possibly_free_var(box_index)
+        if not isinstance(index_loc, ImmedLoc):
+            tempvar = TempBox()
+            temp_loc = self.rm.force_allocate_reg(tempvar, [box_base,
+                                                            box_value])
+            self.rm.possibly_free_var(tempvar)
+        else:
+            temp_loc = None
+        self.rm.possibly_free_var(box_base)
+        self.possibly_free_var(box_value)
         self.PerformDiscard(op, [base_loc, ofs, itemsize, fieldsize,
-                                 index_loc, value_loc])
+                                 index_loc, temp_loc, value_loc])
 
     def consider_strsetitem(self, op):
         args = op.getarglist()
@@ -1126,13 +1136,14 @@
         else:
             sign_loc = imm0
         args = op.getarglist()
-        tmpvar = TempBox()
         base_loc = self.rm.make_sure_var_in_reg(op.getarg(0), args)
-        index_loc = self.rm.force_result_in_reg(tmpvar, op.getarg(1),
-                                                args)
-        self.rm.possibly_free_vars_for_op(op)
-        self.rm.possibly_free_var(tmpvar)
-        result_loc = self.force_allocate_reg(op.result)
+        index_loc = self.rm.make_sure_var_in_reg(op.getarg(1), args)
+        # 'base' and 'index' are put in two registers (or one if 'index'
+        # is an immediate).  'result' can be in the same register as
+        # 'index' but must be in a different register than 'base'.
+        self.rm.possibly_free_var(op.getarg(1))
+        result_loc = self.force_allocate_reg(op.result, [op.getarg(0)])
+        self.rm.possibly_free_var(op.getarg(0))
         self.Perform(op, [base_loc, ofs, itemsize, fieldsize,
                           index_loc, sign_loc], result_loc)
 


More information about the pypy-commit mailing list