Skip to content

gh-149807 : frozendict_hash : hash (key, value) pairs rather than keys and values separately #149808

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
lambda-abstraction wants to merge 3 commits into from
Closed

gh-149807 : frozendict_hash : hash (key, value) pairs rather than keys and values separately #149808

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
745b3b9
`frozendict_hash` : hash entry mixed hashes rather than just keys and…
lambda-abstraction May 13, 2026
22f55bb
`_combine_hashes` better mixing
lambda-abstraction May 13, 2026
bbebec7
reuse the tuple xxhash algorithm for entries in `frozendict_hash` by …
lambda-abstraction May 14, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
31 changes: 23 additions & 8 deletions Objects/dictobject.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -129,7 +129,7 @@ As a consequence of this, split keys have a maximum size of 16.
#include "pycore_pyerrors.h" // _PyErr_GetRaisedException()
#include "pycore_pystate.h" // _PyThreadState_GET()
#include "pycore_setobject.h" // _PySet_NextEntry()
#include "pycore_tuple.h" // _PyTuple_Recycle()
#include "pycore_tuple.h" // _PyTuple_Recycle(), _PyTuple_HASH_XXPRIME1, _PyTuple_HASH_XXPRIME2, _PyTuple_HASH_XXPRIME5, _PyTuple_HASH_XXROTATE
#include "pycore_unicodeobject.h" // _PyUnicode_InternImmortal()

#include "stringlib/eq.h" // unicode_eq()
Expand Down Expand Up @@ -8222,10 +8222,27 @@ frozendict_repr(PyObject *self)
return res;
}

static Py_uhash_t
_shuffle_bits(Py_uhash_t h)
// Code unrolled from tuple_hash() for 2 values
static inline Py_uhash_t
_tuple2_xxhash(Py_uhash_t h1, Py_uhash_t h2)
{
return ((h ^ 89869747UL) ^ (h << 16)) * 3644798167UL;
Py_uhash_t acc = _PyTuple_HASH_XXPRIME_5;

acc += h1 * _PyTuple_HASH_XXPRIME_2;
acc = _PyTuple_HASH_XXROTATE(acc);
acc *= _PyTuple_HASH_XXPRIME_1;

acc += h2 * _PyTuple_HASH_XXPRIME_2;
acc = _PyTuple_HASH_XXROTATE(acc);
acc *= _PyTuple_HASH_XXPRIME_1;

acc += 2 ^ (_PyTuple_HASH_XXPRIME_5 ^ 3527539UL);

if (acc == (Py_uhash_t)-1) {
acc = 1546275796;
}

return acc;
}

// Code copied from frozenset_hash()
Expand All @@ -8239,7 +8256,7 @@ frozendict_hash(PyObject *op)
}

PyDictObject *mp = _PyAnyDict_CAST(op);
Py_uhash_t hash = 0;
Py_uhash_t hash = 0xfd1c74; // start at a different value from frozenset to avoid collision with frozenset(frozendict.items())

PyObject *key, *value; // borrowed refs
Py_ssize_t pos = 0;
Expand All @@ -8248,13 +8265,11 @@ frozendict_hash(PyObject *op)
if (key_hash == -1) {
return -1;
}
hash ^= _shuffle_bits(key_hash);

Py_hash_t value_hash = PyObject_Hash(value);
if (value_hash == -1) {
return -1;
}
hash ^= _shuffle_bits(value_hash);
hash ^= _tuple2_xxhash(key_hash, value_hash);
}

/* Factor in the number of active entries */
Expand Down
Loading