diff options
| -rw-r--r-- | LICENSE | 7 | ||||
| -rw-r--r-- | LICENSE-APACHE | 200 | ||||
| -rw-r--r-- | MANIFEST.in | 4 | ||||
| -rw-r--r-- | Makefile | 33 | ||||
| -rw-r--r-- | NOTICE | 1 | ||||
| -rw-r--r-- | README.rst | 123 | ||||
| -rw-r--r-- | bench.png | bin | 0 -> 36273 bytes | |||
| -rw-r--r-- | immutables/__init__.py | 25 | ||||
| -rw-r--r-- | immutables/_map.c | 4175 | ||||
| -rw-r--r-- | immutables/_map.h | 107 | ||||
| -rw-r--r-- | immutables/_map.pyi | 73 | ||||
| -rw-r--r-- | immutables/_protocols.py | 85 | ||||
| -rw-r--r-- | immutables/_testutils.py | 80 | ||||
| -rw-r--r-- | immutables/_version.py | 13 | ||||
| -rw-r--r-- | immutables/map.py | 855 | ||||
| -rw-r--r-- | immutables/py.typed | 1 | ||||
| -rw-r--r-- | immutables/pythoncapi_compat.h | 446 | ||||
| -rw-r--r-- | pyproject.toml | 25 | ||||
| -rw-r--r-- | setup.py | 93 | ||||
| -rw-r--r-- | tests/__init__.py | 16 | ||||
| -rw-r--r-- | tests/conftest.py | 13 | ||||
| -rw-r--r-- | tests/test-data/check-immu.test | 73 | ||||
| -rw-r--r-- | tests/test_issue24.py | 156 | ||||
| -rw-r--r-- | tests/test_map.py | 1417 | ||||
| -rw-r--r-- | tests/test_mypy.py | 26 | ||||
| -rw-r--r-- | tests/test_none_keys.py | 515 |
26 files changed, 8562 insertions, 0 deletions
@@ -0,0 +1,7 @@ +The Immutables project is provided under the Apache 2.0 license. See +LICENSE-APACHE for the full text of the license. + +Additionally, this software contains the following code distributed a +different license (refer to the specific files for details): + + immutables/pythoncapi_compat.h (MIT License) diff --git a/LICENSE-APACHE b/LICENSE-APACHE new file mode 100644 index 0000000..bf911cf --- /dev/null +++ b/LICENSE-APACHE @@ -0,0 +1,200 @@ + Version 2.0, January 2004 + http://www.apache.org/licenses/ + + TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION + + 1. Definitions. + + "License" shall mean the terms and conditions for use, reproduction, + and distribution as defined by Sections 1 through 9 of this document. + + "Licensor" shall mean the copyright owner or entity authorized by + the copyright owner that is granting the License. + + "Legal Entity" shall mean the union of the acting entity and all + other entities that control, are controlled by, or are under common + control with that entity. For the purposes of this definition, + "control" means (i) the power, direct or indirect, to cause the + direction or management of such entity, whether by contract or + otherwise, or (ii) ownership of fifty percent (50%) or more of the + outstanding shares, or (iii) beneficial ownership of such entity. + + "You" (or "Your") shall mean an individual or Legal Entity + exercising permissions granted by this License. + + "Source" form shall mean the preferred form for making modifications, + including but not limited to software source code, documentation + source, and configuration files. + + "Object" form shall mean any form resulting from mechanical + transformation or translation of a Source form, including but + not limited to compiled object code, generated documentation, + and conversions to other media types. + + "Work" shall mean the work of authorship, whether in Source or + Object form, made available under the License, as indicated by a + copyright notice that is included in or attached to the work + (an example is provided in the Appendix below). + + "Derivative Works" shall mean any work, whether in Source or Object + form, that is based on (or derived from) the Work and for which the + editorial revisions, annotations, elaborations, or other modifications + represent, as a whole, an original work of authorship. For the purposes + of this License, Derivative Works shall not include works that remain + separable from, or merely link (or bind by name) to the interfaces of, + the Work and Derivative Works thereof. + + "Contribution" shall mean any work of authorship, including + the original version of the Work and any modifications or additions + to that Work or Derivative Works thereof, that is intentionally + submitted to Licensor for inclusion in the Work by the copyright owner + or by an individual or Legal Entity authorized to submit on behalf of + the copyright owner. For the purposes of this definition, "submitted" + means any form of electronic, verbal, or written communication sent + to the Licensor or its representatives, including but not limited to + communication on electronic mailing lists, source code control systems, + and issue tracking systems that are managed by, or on behalf of, the + Licensor for the purpose of discussing and improving the Work, but + excluding communication that is conspicuously marked or otherwise + designated in writing by the copyright owner as "Not a Contribution." + + "Contributor" shall mean Licensor and any individual or Legal Entity + on behalf of whom a Contribution has been received by Licensor and + subsequently incorporated within the Work. + + 2. Grant of Copyright License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + copyright license to reproduce, prepare Derivative Works of, + publicly display, publicly perform, sublicense, and distribute the + Work and such Derivative Works in Source or Object form. + + 3. Grant of Patent License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + (except as stated in this section) patent license to make, have made, + use, offer to sell, sell, import, and otherwise transfer the Work, + where such license applies only to those patent claims licensable + by such Contributor that are necessarily infringed by their + Contribution(s) alone or by combination of their Contribution(s) + with the Work to which such Contribution(s) was submitted. If You + institute patent litigation against any entity (including a + cross-claim or counterclaim in a lawsuit) alleging that the Work + or a Contribution incorporated within the Work constitutes direct + or contributory patent infringement, then any patent licenses + granted to You under this License for that Work shall terminate + as of the date such litigation is filed. + + 4. Redistribution. You may reproduce and distribute copies of the + Work or Derivative Works thereof in any medium, with or without + modifications, and in Source or Object form, provided that You + meet the following conditions: + + (a) You must give any other recipients of the Work or + Derivative Works a copy of this License; and + + (b) You must cause any modified files to carry prominent notices + stating that You changed the files; and + + (c) You must retain, in the Source form of any Derivative Works + that You distribute, all copyright, patent, trademark, and + attribution notices from the Source form of the Work, + excluding those notices that do not pertain to any part of + the Derivative Works; and + + (d) If the Work includes a "NOTICE" text file as part of its + distribution, then any Derivative Works that You distribute must + include a readable copy of the attribution notices contained + within such NOTICE file, excluding those notices that do not + pertain to any part of the Derivative Works, in at least one + of the following places: within a NOTICE text file distributed + as part of the Derivative Works; within the Source form or + documentation, if provided along with the Derivative Works; or, + within a display generated by the Derivative Works, if and + wherever such third-party notices normally appear. The contents + of the NOTICE file are for informational purposes only and + do not modify the License. You may add Your own attribution + notices within Derivative Works that You distribute, alongside + or as an addendum to the NOTICE text from the Work, provided + that such additional attribution notices cannot be construed + as modifying the License. + + You may add Your own copyright statement to Your modifications and + may provide additional or different license terms and conditions + for use, reproduction, or distribution of Your modifications, or + for any such Derivative Works as a whole, provided Your use, + reproduction, and distribution of the Work otherwise complies with + the conditions stated in this License. + + 5. Submission of Contributions. Unless You explicitly state otherwise, + any Contribution intentionally submitted for inclusion in the Work + by You to the Licensor shall be under the terms and conditions of + this License, without any additional terms or conditions. + Notwithstanding the above, nothing herein shall supersede or modify + the terms of any separate license agreement you may have executed + with Licensor regarding such Contributions. + + 6. Trademarks. This License does not grant permission to use the trade + names, trademarks, service marks, or product names of the Licensor, + except as required for reasonable and customary use in describing the + origin of the Work and reproducing the content of the NOTICE file. + + 7. Disclaimer of Warranty. Unless required by applicable law or + agreed to in writing, Licensor provides the Work (and each + Contributor provides its Contributions) on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or + implied, including, without limitation, any warranties or conditions + of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A + PARTICULAR PURPOSE. You are solely responsible for determining the + appropriateness of using or redistributing the Work and assume any + risks associated with Your exercise of permissions under this License. + + 8. Limitation of Liability. In no event and under no legal theory, + whether in tort (including negligence), contract, or otherwise, + unless required by applicable law (such as deliberate and grossly + negligent acts) or agreed to in writing, shall any Contributor be + liable to You for damages, including any direct, indirect, special, + incidental, or consequential damages of any character arising as a + result of this License or out of the use or inability to use the + Work (including but not limited to damages for loss of goodwill, + work stoppage, computer failure or malfunction, or any and all + other commercial damages or losses), even if such Contributor + has been advised of the possibility of such damages. + + 9. Accepting Warranty or Additional Liability. While redistributing + the Work or Derivative Works thereof, You may choose to offer, + and charge a fee for, acceptance of support, warranty, indemnity, + or other liability obligations and/or rights consistent with this + License. However, in accepting such obligations, You may act only + on Your own behalf and on Your sole responsibility, not on behalf + of any other Contributor, and only if You agree to indemnify, + defend, and hold each Contributor harmless for any liability + incurred by, or claims asserted against, such Contributor by reason + of your accepting any such warranty or additional liability. + + END OF TERMS AND CONDITIONS + + APPENDIX: How to apply the Apache License to your work. + + To apply the Apache License to your work, attach the following + boilerplate notice, with the fields enclosed by brackets "[]" + replaced with your own identifying information. (Don't include + the brackets!) The text should be enclosed in the appropriate + comment syntax for the file format. We also recommend that a + file or class name and description of purpose be included on the + same "printed page" as the copyright notice for easier + identification within third-party archives. + + Copyright [yyyy] [name of copyright owner] + + Licensed under the Apache License, Version 2.0 (the "License"); + you may not use this file except in compliance with the License. + You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + + Unless required by applicable law or agreed to in writing, software + distributed under the License is distributed on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + See the License for the specific language governing permissions and + limitations under the License. diff --git a/MANIFEST.in b/MANIFEST.in new file mode 100644 index 0000000..d3f372b --- /dev/null +++ b/MANIFEST.in @@ -0,0 +1,4 @@ +recursive-include tests *.py +recursive-include immutables *.py *.c *.h *.pyi +include LICENSE* NOTICE README.rst bench.png +include immutables/py.typed diff --git a/Makefile b/Makefile new file mode 100644 index 0000000..481d190 --- /dev/null +++ b/Makefile @@ -0,0 +1,33 @@ +.PHONY: rtest build test clean all + + +PYTHON ?= python +ROOT = $(dir $(realpath $(firstword $(MAKEFILE_LIST)))) + + +all: build + +build: + $(PYTHON) setup.py build_ext --inplace + +debug: + DEBUG_IMMUTABLES=1 $(PYTHON) setup.py build_ext --inplace + +test: + $(PYTHON) -m pytest -v + +rtest: + ~/dev/venvs/36-debug/bin/python setup.py build_ext --inplace + + env PYTHONPATH=. \ + ~/dev/venvs/36-debug/bin/python -m test.regrtest -R3:3 --testdir tests/ + +clean: + find . -name '*.pyc' | xargs rm -f + find . -name '*.so' | xargs rm -f + rm -rf ./build + rm -rf ./dist + rm -rf ./*.egg-info + +testinstalled: + cd /tmp && $(PYTHON) $(ROOT)/tests/__init__.py @@ -0,0 +1 @@ +Copyright 2018-present Contributors to the immutables project. diff --git a/README.rst b/README.rst new file mode 100644 index 0000000..e14d321 --- /dev/null +++ b/README.rst @@ -0,0 +1,123 @@ +immutables +========== + +.. image:: https://github.com/MagicStack/immutables/workflows/Tests/badge.svg?branch=master + :target: https://github.com/MagicStack/immutables/actions?query=workflow%3ATests+branch%3Amaster+event%3Apush + +.. image:: https://img.shields.io/pypi/v/immutables.svg + :target: https://pypi.python.org/pypi/immutables + +An immutable mapping type for Python. + +The underlying datastructure is a Hash Array Mapped Trie (HAMT) +used in Clojure, Scala, Haskell, and other functional languages. +This implementation is used in CPython 3.7 in the ``contextvars`` +module (see `PEP 550 <https://www.python.org/dev/peps/pep-0550/>`_ and +`PEP 567 <https://www.python.org/dev/peps/pep-0567/>`_ for more details). + +Immutable mappings based on HAMT have O(log N) performance for both +``set()`` and ``get()`` operations, which is essentially O(1) for +relatively small mappings. + +Below is a visualization of a simple get/set benchmark comparing +HAMT to an immutable mapping implemented with a Python dict +copy-on-write approach (the benchmark code is available +`here <https://gist.github.com/1st1/292e3f0bbe43bd65ff3256f80aa2637d>`_): + +.. image:: bench.png + + +Installation +------------ + +``immutables`` requires Python 3.6+ and is available on PyPI:: + + $ pip install immutables + + +API +--- + +``immutables.Map`` is an unordered immutable mapping. ``Map`` objects +are hashable, comparable, and pickleable. + +The ``Map`` object implements the ``collections.abc.Mapping`` ABC +so working with it is very similar to working with Python dicts: + +.. code-block:: python + + import immutables + + map = immutables.Map(a=1, b=2) + + print(map['a']) + # will print '1' + + print(map.get('z', 100)) + # will print '100' + + print('z' in map) + # will print 'False' + +Since Maps are immutable, there is a special API for mutations that +allow apply changes to the Map object and create new (derived) Maps: + +.. code-block:: python + + map2 = map.set('a', 10) + print(map, map2) + # will print: + # <immutables.Map({'a': 1, 'b': 2})> + # <immutables.Map({'a': 10, 'b': 2})> + + map3 = map2.delete('b') + print(map, map2, map3) + # will print: + # <immutables.Map({'a': 1, 'b': 2})> + # <immutables.Map({'a': 10, 'b': 2})> + # <immutables.Map({'a': 10})> + +Maps also implement APIs for bulk updates: ``MapMutation`` objects: + +.. code-block:: python + + map_mutation = map.mutate() + map_mutation['a'] = 100 + del map_mutation['b'] + map_mutation.set('y', 'y') + + map2 = map_mutation.finish() + + print(map, map2) + # will print: + # <immutables.Map({'a': 1, 'b': 2})> + # <immutables.Map({'a': 100, 'y': 'y'})> + +``MapMutation`` objects are context managers. Here's the above example +rewritten in a more idiomatic way: + +.. code-block:: python + + with map.mutate() as mm: + mm['a'] = 100 + del mm['b'] + mm.set('y', 'y') + map2 = mm.finish() + + print(map, map2) + # will print: + # <immutables.Map({'a': 1, 'b': 2})> + # <immutables.Map({'a': 100, 'y': 'y'})> + + +Further development +------------------- + +* An immutable version of Python ``set`` type with efficient + ``add()`` and ``discard()`` operations. + + +License +------- + +Apache 2.0 diff --git a/bench.png b/bench.png Binary files differnew file mode 100644 index 0000000..e77ff5d --- /dev/null +++ b/bench.png diff --git a/immutables/__init__.py b/immutables/__init__.py new file mode 100644 index 0000000..b8565b0 --- /dev/null +++ b/immutables/__init__.py @@ -0,0 +1,25 @@ +# flake8: noqa + +import sys + +from typing import TYPE_CHECKING + +if TYPE_CHECKING: + from ._map import Map +else: + try: + from ._map import Map + except ImportError: + from .map import Map + else: + import collections.abc as _abc + _abc.Mapping.register(Map) + +from ._protocols import MapKeys as MapKeys +from ._protocols import MapValues as MapValues +from ._protocols import MapItems as MapItems +from ._protocols import MapMutation as MapMutation + +from ._version import __version__ + +__all__ = 'Map', diff --git a/immutables/_map.c b/immutables/_map.c new file mode 100644 index 0000000..7e510fd --- /dev/null +++ b/immutables/_map.c @@ -0,0 +1,4175 @@ +#include <stddef.h> /* For offsetof */ +#include "pythoncapi_compat.h" +#include "_map.h" + + +/* +This file provides an implemention of an immutable mapping using the +Hash Array Mapped Trie (or HAMT) datastructure. + +This design allows to have: + +1. Efficient copy: immutable mappings can be copied by reference, + making it an O(1) operation. + +2. Efficient mutations: due to structural sharing, only a portion of + the trie needs to be copied when the collection is mutated. The + cost of set/delete operations is O(log N). + +3. Efficient lookups: O(log N). + +(where N is number of key/value items in the immutable mapping.) + + +HAMT +==== + +The core idea of HAMT is that the shape of the trie is encoded into the +hashes of keys. + +Say we want to store a K/V pair in our mapping. First, we calculate the +hash of K, let's say it's 19830128, or in binary: + + 0b1001011101001010101110000 = 19830128 + +Now let's partition this bit representation of the hash into blocks of +5 bits each: + + 0b00_00000_10010_11101_00101_01011_10000 = 19830128 + (6) (5) (4) (3) (2) (1) + +Each block of 5 bits represents a number between 0 and 31. So if we have +a tree that consists of nodes, each of which is an array of 32 pointers, +those 5-bit blocks will encode a position on a single tree level. + +For example, storing the key K with hash 19830128, results in the following +tree structure: + + (array of 32 pointers) + +---+ -- +----+----+----+ -- +----+ + root node | 0 | .. | 15 | 16 | 17 | .. | 31 | 0b10000 = 16 (1) + (level 1) +---+ -- +----+----+----+ -- +----+ + | + +---+ -- +----+----+----+ -- +----+ + a 2nd level node | 0 | .. | 10 | 11 | 12 | .. | 31 | 0b01011 = 11 (2) + +---+ -- +----+----+----+ -- +----+ + | + +---+ -- +----+----+----+ -- +----+ + a 3rd level node | 0 | .. | 04 | 05 | 06 | .. | 31 | 0b00101 = 5 (3) + +---+ -- +----+----+----+ -- +----+ + | + +---+ -- +----+----+----+----+ + a 4th level node | 0 | .. | 04 | 29 | 30 | 31 | 0b11101 = 29 (4) + +---+ -- +----+----+----+----+ + | + +---+ -- +----+----+----+ -- +----+ + a 5th level node | 0 | .. | 17 | 18 | 19 | .. | 31 | 0b10010 = 18 (5) + +---+ -- +----+----+----+ -- +----+ + | + +--------------+ + | + +---+ -- +----+----+----+ -- +----+ + a 6th level node | 0 | .. | 15 | 16 | 17 | .. | 31 | 0b00000 = 0 (6) + +---+ -- +----+----+----+ -- +----+ + | + V -- our value (or collision) + +To rehash: for a K/V pair, the hash of K encodes where in the tree V will +be stored. + +To optimize memory footprint and handle hash collisions, our implementation +uses three different types of nodes: + + * A Bitmap node; + * An Array node; + * A Collision node. + +Because we implement an immutable dictionary, our nodes are also +immutable. Therefore, when we need to modify a node, we copy it, and +do that modification to the copy. + + +Array Nodes +----------- + +These nodes are very simple. Essentially they are arrays of 32 pointers +we used to illustrate the high-level idea in the previous section. + +We use Array nodes only when we need to store more than 16 pointers +in a single node. + +Array nodes do not store key objects or value objects. They are used +only as an indirection level - their pointers point to other nodes in +the tree. + + +Bitmap Node +----------- + +Allocating a new 32-pointers array for every node of our tree would be +very expensive. Unless we store millions of keys, most of tree nodes would +be very sparse. + +When we have less than 16 elements in a node, we don't want to use the +Array node, that would mean that we waste a lot of memory. Instead, +we can use bitmap compression and can have just as many pointers +as we need! + +Bitmap nodes consist of two fields: + +1. An array of pointers. If a Bitmap node holds N elements, the + array will be of N pointers. + +2. A 32bit integer -- a bitmap field. If an N-th bit is set in the + bitmap, it means that the node has an N-th element. + +For example, say we need to store a 3 elements sparse array: + + +---+ -- +---+ -- +----+ -- +----+ + | 0 | .. | 4 | .. | 11 | .. | 17 | + +---+ -- +---+ -- +----+ -- +----+ + | | | + o1 o2 o3 + +We allocate a three-pointer Bitmap node. Its bitmap field will be +then set to: + + 0b_00100_00010_00000_10000 == (1 << 17) | (1 << 11) | (1 << 4) + +To check if our Bitmap node has an I-th element we can do: + + bitmap & (1 << I) + + +And here's a formula to calculate a position in our pointer array +which would correspond to an I-th element: + + popcount(bitmap & ((1 << I) - 1)) + + +Let's break it down: + + * `popcount` is a function that returns a number of bits set to 1; + + * `((1 << I) - 1)` is a mask to filter the bitmask to contain bits + set to the *right* of our bit. + + +So for our 17, 11, and 4 indexes: + + * bitmap & ((1 << 17) - 1) == 0b100000010000 => 2 bits are set => index is 2. + + * bitmap & ((1 << 11) - 1) == 0b10000 => 1 bit is set => index is 1. + + * bitmap & ((1 << 4) - 1) == 0b0 => 0 bits are set => index is 0. + + +To conclude: Bitmap nodes are just like Array nodes -- they can store +a number of pointers, but use bitmap compression to eliminate unused +pointers. + + +Bitmap nodes have two pointers for each item: + + +----+----+----+----+ -- +----+----+ + | k1 | v1 | k2 | v2 | .. | kN | vN | + +----+----+----+----+ -- +----+----+ + +When kI == NULL, vI points to another tree level. + +When kI != NULL, the actual key object is stored in kI, and its +value is stored in vI. + + +Collision Nodes +--------------- + +Collision nodes are simple arrays of pointers -- two pointers per +key/value. When there's a hash collision, say for k1/v1 and k2/v2 +we have `hash(k1)==hash(k2)`. Then our collision node will be: + + +----+----+----+----+ + | k1 | v1 | k2 | v2 | + +----+----+----+----+ + + +Tree Structure +-------------- + +All nodes are PyObjects. + +The `MapObject` object has a pointer to the root node (h_root), +and has a length field (h_count). + +High-level functions accept a MapObject object and dispatch to +lower-level functions depending on what kind of node h_root points to. + + +Operations +========== + +There are three fundamental operations on an immutable dictionary: + +1. "o.assoc(k, v)" will return a new immutable dictionary, that will be + a copy of "o", but with the "k/v" item set. + + Functions in this file: + + map_node_assoc, map_node_bitmap_assoc, + map_node_array_assoc, map_node_collision_assoc + + `map_node_assoc` function accepts a node object, and calls + other functions depending on its actual type. + +2. "o.find(k)" will lookup key "k" in "o". + + Functions: + + map_node_find, map_node_bitmap_find, + map_node_array_find, map_node_collision_find + +3. "o.without(k)" will return a new immutable dictionary, that will be + a copy of "o", buth without the "k" key. + + Functions: + + map_node_without, map_node_bitmap_without, + map_node_array_without, map_node_collision_without + + +Further Reading +=============== + +1. http://blog.higher-order.net/2009/09/08/understanding-clojures-persistenthashmap-deftwice.html + +2. http://blog.higher-order.net/2010/08/16/assoc-and-clojures-persistenthashmap-part-ii.html + +3. Clojure's PersistentHashMap implementation: + https://github.com/clojure/clojure/blob/master/src/jvm/clojure/lang/PersistentHashMap.java +*/ + + +#define IS_ARRAY_NODE(node) (Py_TYPE(node) == &_Map_ArrayNode_Type) +#define IS_BITMAP_NODE(node) (Py_TYPE(node) == &_Map_BitmapNode_Type) +#define IS_COLLISION_NODE(node) (Py_TYPE(node) == &_Map_CollisionNode_Type) + + +/* Return type for 'find' (lookup a key) functions. + + * F_ERROR - an error occurred; + * F_NOT_FOUND - the key was not found; + * F_FOUND - the key was found. +*/ +typedef enum {F_ERROR, F_NOT_FOUND, F_FOUND} map_find_t; + + +/* Return type for 'without' (delete a key) functions. + + * W_ERROR - an error occurred; + * W_NOT_FOUND - the key was not found: there's nothing to delete; + * W_EMPTY - the key was found: the node/tree would be empty + if the key is deleted; + * W_NEWNODE - the key was found: a new node/tree is returned + without that key. +*/ +typedef enum {W_ERROR, W_NOT_FOUND, W_EMPTY, W_NEWNODE} map_without_t; + + +/* Low-level iterator protocol type. + + * I_ITEM - a new item has been yielded; + * I_END - the whole tree was visited (similar to StopIteration). +*/ +typedef enum {I_ITEM, I_END} map_iter_t; + + +#define HAMT_ARRAY_NODE_SIZE 32 + + +typedef struct { + PyObject_HEAD + MapNode *a_array[HAMT_ARRAY_NODE_SIZE]; + Py_ssize_t a_count; + uint64_t a_mutid; +} MapNode_Array; + + +typedef struct { + PyObject_VAR_HEAD + uint64_t b_mutid; + uint32_t b_bitmap; + PyObject *b_array[1]; +} MapNode_Bitmap; + + +typedef struct { + PyObject_VAR_HEAD + uint64_t c_mutid; + int32_t c_hash; + PyObject *c_array[1]; +} MapNode_Collision; + + +static volatile uint64_t mutid_counter = 1; + +static MapNode_Bitmap *_empty_bitmap_node; + + +/* Create a new HAMT immutable mapping. */ +static MapObject * +map_new(void); + +/* Return a new collection based on "o", but with an additional + key/val pair. */ +static MapObject * +map_assoc(MapObject *o, PyObject *key, PyObject *val); + +/* Return a new collection based on "o", but without "key". */ +static MapObject * +map_without(MapObject *o, PyObject *key); + +/* Check if "v" is equal to "w". + + Return: + - 0: v != w + - 1: v == w + - -1: An error occurred. +*/ +static int +map_eq(BaseMapObject *v, BaseMapObject *w); + +static map_find_t +map_find(BaseMapObject *o, PyObject *key, PyObject **val); + +/* Return the size of "o"; equivalent of "len(o)". */ +static Py_ssize_t +map_len(BaseMapObject *o); + + +static MapObject * +map_alloc(void); + +static MapNode * +map_node_assoc(MapNode *node, + uint32_t shift, int32_t hash, + PyObject *key, PyObject *val, int* added_leaf, + uint64_t mutid); + +static map_without_t +map_node_without(MapNode *node, + uint32_t shift, int32_t hash, + PyObject *key, + MapNode **new_node, + uint64_t mutid); + +static map_find_t +map_node_find(MapNode *node, + uint32_t shift, int32_t hash, + PyObject *key, PyObject **val); + +static int +map_node_dump(MapNode *node, + _PyUnicodeWriter *writer, int level); + +static MapNode * +map_node_array_new(Py_ssize_t, uint64_t mutid); + +static MapNode * +map_node_collision_new(int32_t hash, Py_ssize_t size, uint64_t mutid); + +static inline Py_ssize_t +map_node_collision_count(MapNode_Collision *node); + +static int +map_node_update(uint64_t mutid, + PyObject *seq, + MapNode *root, Py_ssize_t count, + MapNode **new_root, Py_ssize_t *new_count); + + +static int +map_update_inplace(uint64_t mutid, BaseMapObject *o, PyObject *src); + +static MapObject * +map_update(uint64_t mutid, MapObject *o, PyObject *src); + + +#if !defined(NDEBUG) +static void +_map_node_array_validate(void *o) +{ + assert(IS_ARRAY_NODE(o)); + MapNode_Array *node = (MapNode_Array*)(o); + assert(node->a_count <= HAMT_ARRAY_NODE_SIZE); + Py_ssize_t i = 0, count = 0; + for (; i < HAMT_ARRAY_NODE_SIZE; i++) { + if (node->a_array[i] != NULL) { + count++; + } + } + assert(count == node->a_count); +} + +#define VALIDATE_ARRAY_NODE(NODE) \ + do { _map_node_array_validate(NODE); } while (0); +#else +#define VALIDATE_ARRAY_NODE(NODE) +#endif + + +/* Returns -1 on error */ +static inline int32_t +map_hash(PyObject *o) +{ + Py_hash_t hash = PyObject_Hash(o); + +#if SIZEOF_PY_HASH_T <= 4 + return hash; +#else + if (hash == -1) { + /* exception */ + return -1; + } + + /* While it's suboptimal to reduce Python's 64 bit hash to + 32 bits via XOR, it seems that the resulting hash function + is good enough (this is also how Long type is hashed in Java.) + Storing 10, 100, 1000 Python strings results in a relatively + shallow and uniform tree structure. + + Please don't change this hashing algorithm, as there are many + tests that test some exact tree shape to cover all code paths. + */ + int32_t xored = (int32_t)(hash & 0xffffffffl) ^ (int32_t)(hash >> 32); + return xored == -1 ? -2 : xored; +#endif +} + +static inline uint32_t +map_mask(int32_t hash, uint32_t shift) +{ + return (((uint32_t)hash >> shift) & 0x01f); +} + +static inline uint32_t +map_bitpos(int32_t hash, uint32_t shift) +{ + return (uint32_t)1 << map_mask(hash, shift); +} + +static inline uint32_t +map_bitcount(uint32_t i) +{ + /* We could use native popcount instruction but that would + require to either add configure flags to enable SSE4.2 + support or to detect it dynamically. Otherwise, we have + a risk of CPython not working properly on older hardware. + + In practice, there's no observable difference in + performance between using a popcount instruction or the + following fallback code. + + The algorithm is copied from: + https://graphics.stanford.edu/~seander/bithacks.html + */ + i = i - ((i >> 1) & 0x55555555); + i = (i & 0x33333333) + ((i >> 2) & 0x33333333); + return (((i + (i >> 4)) & 0xF0F0F0F) * 0x1010101) >> 24; +} + +static inline uint32_t +map_bitindex(uint32_t bitmap, uint32_t bit) +{ + return map_bitcount(bitmap & (bit - 1)); +} + + +/////////////////////////////////// Dump Helpers + +static int +_map_dump_ident(_PyUnicodeWriter *writer, int level) +{ + /* Write `' ' * level` to the `writer` */ + PyObject *str = NULL; + PyObject *num = NULL; + PyObject *res = NULL; + int ret = -1; + + str = PyUnicode_FromString(" "); + if (str == NULL) { + goto error; + } + + num = PyLong_FromLong((long)level); + if (num == NULL) { + goto error; + } + + res = PyNumber_Multiply(str, num); + if (res == NULL) { + goto error; + } + + ret = _PyUnicodeWriter_WriteStr(writer, res); + +error: + Py_XDECREF(res); + Py_XDECREF(str); + Py_XDECREF(num); + return ret; +} + +static int +_map_dump_format(_PyUnicodeWriter *writer, const char *format, ...) +{ + /* A convenient helper combining _PyUnicodeWriter_WriteStr and + PyUnicode_FromFormatV. + */ + PyObject* msg; + int ret; + + va_list vargs; +#ifdef HAVE_STDARG_PROTOTYPES + va_start(vargs, format); +#else + va_start(vargs); +#endif + msg = PyUnicode_FromFormatV(format, vargs); + va_end(vargs); + + if (msg == NULL) { + return -1; + } + + ret = _PyUnicodeWriter_WriteStr(writer, msg); + Py_DECREF(msg); + return ret; +} + +/////////////////////////////////// Bitmap Node + + +static MapNode * +map_node_bitmap_new(Py_ssize_t size, uint64_t mutid) +{ + /* Create a new bitmap node of size 'size' */ + + MapNode_Bitmap *node; + Py_ssize_t i; + + assert(size >= 0); + assert(size % 2 == 0); + + if (size == 0 && _empty_bitmap_node != NULL && mutid == 0) { + Py_INCREF(_empty_bitmap_node); + return (MapNode *)_empty_bitmap_node; + } + + /* No freelist; allocate a new bitmap node */ + node = PyObject_GC_NewVar( + MapNode_Bitmap, &_Map_BitmapNode_Type, size); + if (node == NULL) { + return NULL; + } + + Py_SET_SIZE(node, size); + + for (i = 0; i < size; i++) { + node->b_array[i] = NULL; + } + + node->b_bitmap = 0; + node->b_mutid = mutid; + + PyObject_GC_Track(node); + + if (size == 0 && _empty_bitmap_node == NULL && mutid == 0) { + /* Since bitmap nodes are immutable, we can cache the instance + for size=0 and reuse it whenever we need an empty bitmap node. + */ + _empty_bitmap_node = node; + Py_INCREF(_empty_bitmap_node); + } + + return (MapNode *)node; +} + +static inline Py_ssize_t +map_node_bitmap_count(MapNode_Bitmap *node) +{ + return Py_SIZE(node) / 2; +} + +static MapNode_Bitmap * +map_node_bitmap_clone(MapNode_Bitmap *node, uint64_t mutid) +{ + /* Clone a bitmap node; return a new one with the same child notes. */ + + MapNode_Bitmap *clone; + Py_ssize_t i; + + clone = (MapNode_Bitmap *)map_node_bitmap_new( + Py_SIZE(node), mutid); + if (clone == NULL) { + return NULL; + } + + for (i = 0; i < Py_SIZE(node); i++) { + Py_XINCREF(node->b_array[i]); + clone->b_array[i] = node->b_array[i]; + } + + clone->b_bitmap = node->b_bitmap; + return clone; +} + +static MapNode_Bitmap * +map_node_bitmap_clone_without(MapNode_Bitmap *o, uint32_t bit, uint64_t mutid) +{ + assert(bit & o->b_bitmap); + assert(map_node_bitmap_count(o) > 1); + + MapNode_Bitmap *new = (MapNode_Bitmap *)map_node_bitmap_new( + Py_SIZE(o) - 2, mutid); + if (new == NULL) { + return NULL; + } + + uint32_t idx = map_bitindex(o->b_bitmap, bit); + uint32_t key_idx = 2 * idx; + uint32_t val_idx = key_idx + 1; + uint32_t i; + + for (i = 0; i < key_idx; i++) { + Py_XINCREF(o->b_array[i]); + new->b_array[i] = o->b_array[i]; + } + + assert(Py_SIZE(o) >= 0 && Py_SIZE(o) <= 32); + for (i = val_idx + 1; i < (uint32_t)Py_SIZE(o); i++) { + Py_XINCREF(o->b_array[i]); + new->b_array[i - 2] = o->b_array[i]; + } + + new->b_bitmap = o->b_bitmap & ~bit; + return new; +} + +static MapNode * +map_node_new_bitmap_or_collision(uint32_t shift, + PyObject *key1, PyObject *val1, + int32_t key2_hash, + PyObject *key2, PyObject *val2, + uint64_t mutid) +{ + /* Helper method. Creates a new node for key1/val and key2/val2 + pairs. + + If key1 hash is equal to the hash of key2, a Collision node + will be created. If they are not equal, a Bitmap node is + created. + */ + + int32_t key1_hash = map_hash(key1); + if (key1_hash == -1) { + return NULL; + } + + if (key1_hash == key2_hash) { + MapNode_Collision *n; + n = (MapNode_Collision *)map_node_collision_new(key1_hash, 4, mutid); + if (n == NULL) { + return NULL; + } + + Py_INCREF(key1); + n->c_array[0] = key1; + Py_INCREF(val1); + n->c_array[1] = val1; + + Py_INCREF(key2); + n->c_array[2] = key2; + Py_INCREF(val2); + n->c_array[3] = val2; + + return (MapNode *)n; + } + else { + int added_leaf = 0; + MapNode *n = map_node_bitmap_new(0, mutid); + if (n == NULL) { + return NULL; + } + + MapNode *n2 = map_node_assoc( + n, shift, key1_hash, key1, val1, &added_leaf, mutid); + Py_DECREF(n); + if (n2 == NULL) { + return NULL; + } + + n = map_node_assoc( + n2, shift, key2_hash, key2, val2, &added_leaf, mutid); + Py_DECREF(n2); + if (n == NULL) { + return NULL; + } + + return n; + } +} + +static MapNode * +map_node_bitmap_assoc(MapNode_Bitmap *self, + uint32_t shift, int32_t hash, + PyObject *key, PyObject *val, int* added_leaf, + uint64_t mutid) +{ + /* assoc operation for bitmap nodes. + + Return: a new node, or self if key/val already is in the + collection. + + 'added_leaf' is later used in 'map_assoc' to determine if + `map.set(key, val)` increased the size of the collection. + */ + + uint32_t bit = map_bitpos(hash, shift); + uint32_t idx = map_bitindex(self->b_bitmap, bit); + + /* Bitmap node layout: + + +------+------+------+------+ --- +------+------+ + | key1 | val1 | key2 | val2 | ... | keyN | valN | + +------+------+------+------+ --- +------+------+ + where `N < Py_SIZE(node)`. + + The `node->b_bitmap` field is a bitmap. For a given + `(shift, hash)` pair we can determine: + + - If this node has the corresponding key/val slots. + - The index of key/val slots. + */ + + if (self->b_bitmap & bit) { + /* The key is set in this node */ + + uint32_t key_idx = 2 * idx; + uint32_t val_idx = key_idx + 1; + + assert(val_idx < (size_t)Py_SIZE(self)); + + PyObject *key_or_null = self->b_array[key_idx]; + PyObject *val_or_node = self->b_array[val_idx]; + + if (key_or_null == NULL) { + /* key is NULL. This means that we have a few keys + that have the same (hash, shift) pair. */ + + assert(val_or_node != NULL); + + MapNode *sub_node = map_node_assoc( + (MapNode *)val_or_node, + shift + 5, hash, key, val, added_leaf, + mutid); + if (sub_node == NULL) { + return NULL; + } + + if (val_or_node == (PyObject *)sub_node) { + Py_DECREF(sub_node); + Py_INCREF(self); + return (MapNode *)self; + } + + if (mutid != 0 && self->b_mutid == mutid) { + Py_SETREF(self->b_array[val_idx], (PyObject*)sub_node); + Py_INCREF(self); + return (MapNode *)self; + } + else { + MapNode_Bitmap *ret = map_node_bitmap_clone(self, mutid); + if (ret == NULL) { + return NULL; + } + Py_SETREF(ret->b_array[val_idx], (PyObject*)sub_node); + return (MapNode *)ret; + } + } + + assert(key != NULL); + /* key is not NULL. This means that we have only one other + key in this collection that matches our hash for this shift. */ + + int comp_err = PyObject_RichCompareBool(key, key_or_null, Py_EQ); + if (comp_err < 0) { /* exception in __eq__ */ + return NULL; + } + if (comp_err == 1) { /* key == key_or_null */ + if (val == val_or_node) { + /* we already have the same key/val pair; return self. */ + Py_INCREF(self); + return (MapNode *)self; + } + + /* We're setting a new value for the key we had before. */ + if (mutid != 0 && self->b_mutid == mutid) { + /* We've been mutating this node before: update inplace. */ + Py_INCREF(val); + Py_SETREF(self->b_array[val_idx], val); + Py_INCREF(self); + return (MapNode *)self; + } + else { + /* Make a new bitmap node with a replaced value, + and return it. */ + MapNode_Bitmap *ret = map_node_bitmap_clone(self, mutid); + if (ret == NULL) { + return NULL; + } + Py_INCREF(val); + Py_SETREF(ret->b_array[val_idx], val); + return (MapNode *)ret; + } + } + + /* It's a new key, and it has the same index as *one* another key. + We have a collision. We need to create a new node which will + combine the existing key and the key we're adding. + + `map_node_new_bitmap_or_collision` will either create a new + Collision node if the keys have identical hashes, or + a new Bitmap node. + */ + MapNode *sub_node = map_node_new_bitmap_or_collision( + shift + 5, + key_or_null, val_or_node, /* existing key/val */ + hash, + key, val, /* new key/val */ + self->b_mutid + ); + if (sub_node == NULL) { + return NULL; + } + + if (mutid != 0 && self->b_mutid == mutid) { + Py_SETREF(self->b_array[key_idx], NULL); + Py_SETREF(self->b_array[val_idx], (PyObject *)sub_node); + Py_INCREF(self); + + *added_leaf = 1; + return (MapNode *)self; + } + else { + MapNode_Bitmap *ret = map_node_bitmap_clone(self, mutid); + if (ret == NULL) { + Py_DECREF(sub_node); + return NULL; + } + Py_SETREF(ret->b_array[key_idx], NULL); + Py_SETREF(ret->b_array[val_idx], (PyObject *)sub_node); + + *added_leaf = 1; + return (MapNode *)ret; + } + } + else { + /* There was no key before with the same (shift,hash). */ + + uint32_t n = map_bitcount(self->b_bitmap); + + if (n >= 16) { + /* When we have a situation where we want to store more + than 16 nodes at one level of the tree, we no longer + want to use the Bitmap node with bitmap encoding. + + Instead we start using an Array node, which has + simpler (faster) implementation at the expense of + having prealocated 32 pointers for its keys/values + pairs. + + Small map objects (<30 keys) usually don't have any + Array nodes at all. Between ~30 and ~400 keys map + objects usually have one Array node, and usually it's + a root node. + */ + + uint32_t jdx = map_mask(hash, shift); + /* 'jdx' is the index of where the new key should be added + in the new Array node we're about to create. */ + + MapNode *empty = NULL; + MapNode_Array *new_node = NULL; + MapNode *res = NULL; + + /* Create a new Array node. */ + new_node = (MapNode_Array *)map_node_array_new(n + 1, mutid); + if (new_node == NULL) { + goto fin; + } + + /* Create an empty bitmap node for the next + map_node_assoc call. */ + empty = map_node_bitmap_new(0, mutid); + if (empty == NULL) { + goto fin; + } + + /* Make a new bitmap node for the key/val we're adding. + Set that bitmap node to new-array-node[jdx]. */ + new_node->a_array[jdx] = map_node_assoc( + empty, shift + 5, hash, key, val, added_leaf, mutid); + if (new_node->a_array[jdx] == NULL) { + goto fin; + } + + /* Copy existing key/value pairs from the current Bitmap + node to the new Array node we've just created. */ + Py_ssize_t i, j; + for (i = 0, j = 0; i < HAMT_ARRAY_NODE_SIZE; i++) { + if (((self->b_bitmap >> i) & 1) != 0) { + /* Ensure we don't accidentally override `jdx` element + we set few lines above. + */ + assert(new_node->a_array[i] == NULL); + + if (self->b_array[j] == NULL) { + new_node->a_array[i] = + (MapNode *)self->b_array[j + 1]; + Py_INCREF(new_node->a_array[i]); + } + else { + int32_t rehash = map_hash(self->b_array[j]); + if (rehash == -1) { + goto fin; + } + + new_node->a_array[i] = map_node_assoc( + empty, shift + 5, + rehash, + self->b_array[j], + self->b_array[j + 1], + added_leaf, + mutid); + + if (new_node->a_array[i] == NULL) { + goto fin; + } + } + j += 2; + } + } + + VALIDATE_ARRAY_NODE(new_node) + + /* That's it! */ + res = (MapNode *)new_node; + + fin: + Py_XDECREF(empty); + if (res == NULL) { + Py_XDECREF(new_node); + } + return res; + } + else { + /* We have less than 16 keys at this level; let's just + create a new bitmap node out of this node with the + new key/val pair added. */ + + uint32_t key_idx = 2 * idx; + uint32_t val_idx = key_idx + 1; + uint32_t i; + + *added_leaf = 1; + + /* Allocate new Bitmap node which can have one more key/val + pair in addition to what we have already. */ + MapNode_Bitmap *new_node = + (MapNode_Bitmap *)map_node_bitmap_new(2 * (n + 1), mutid); + if (new_node == NULL) { + return NULL; + } + + /* Copy all keys/values that will be before the new key/value + we are adding. */ + for (i = 0; i < key_idx; i++) { + Py_XINCREF(self->b_array[i]); + new_node->b_array[i] = self->b_array[i]; + } + + /* Set the new key/value to the new Bitmap node. */ + Py_INCREF(key); + new_node->b_array[key_idx] = key; + Py_INCREF(val); + new_node->b_array[val_idx] = val; + + /* Copy all keys/values that will be after the new key/value + we are adding. */ + assert(Py_SIZE(self) >= 0 && Py_SIZE(self) <= 32); + for (i = key_idx; i < (uint32_t)Py_SIZE(self); i++) { + Py_XINCREF(self->b_array[i]); + new_node->b_array[i + 2] = self->b_array[i]; + } + + new_node->b_bitmap = self->b_bitmap | bit; + return (MapNode *)new_node; + } + } +} + +static map_without_t +map_node_bitmap_without(MapNode_Bitmap *self, + uint32_t shift, int32_t hash, + PyObject *key, + MapNode **new_node, + uint64_t mutid) +{ + uint32_t bit = map_bitpos(hash, shift); + if ((self->b_bitmap & bit) == 0) { + return W_NOT_FOUND; + } + + uint32_t idx = map_bitindex(self->b_bitmap, bit); + + uint32_t key_idx = 2 * idx; + uint32_t val_idx = key_idx + 1; + + PyObject *key_or_null = self->b_array[key_idx]; + PyObject *val_or_node = self->b_array[val_idx]; + + if (key_or_null == NULL) { + /* key == NULL means that 'value' is another tree node. */ + + MapNode *sub_node = NULL; + MapNode_Bitmap *target = NULL; + + map_without_t res = map_node_without( + (MapNode *)val_or_node, + shift + 5, hash, key, &sub_node, + mutid); + + switch (res) { + case W_EMPTY: + /* It's impossible for us to receive a W_EMPTY here: + + - Array nodes are converted to Bitmap nodes when + we delete 16th item from them; + + - Collision nodes are converted to Bitmap when + there is one item in them; + + - Bitmap node's without() inlines single-item + sub-nodes. + + So in no situation we can have a single-item + Bitmap child of another Bitmap node. + */ + abort(); + + case W_NEWNODE: { + assert(sub_node != NULL); + + if (IS_BITMAP_NODE(sub_node)) { + MapNode_Bitmap *sub_tree = (MapNode_Bitmap *)sub_node; + if (map_node_bitmap_count(sub_tree) == 1 && + sub_tree->b_array[0] != NULL) + { + /* A bitmap node with one key/value pair. Just + merge it into this node. + + Note that we don't inline Bitmap nodes that + have a NULL key -- those nodes point to another + tree level, and we cannot simply move tree levels + up or down. + */ + + if (mutid != 0 && self->b_mutid == mutid) { + target = self; + Py_INCREF(target); + } + else { + target = map_node_bitmap_clone(self, mutid); + if (target == NULL) { + Py_DECREF(sub_node); + return W_ERROR; + } + } + + PyObject *key = sub_tree->b_array[0]; + PyObject *val = sub_tree->b_array[1]; + + Py_INCREF(key); + Py_XSETREF(target->b_array[key_idx], key); + Py_INCREF(val); + Py_SETREF(target->b_array[val_idx], val); + + Py_DECREF(sub_tree); + + *new_node = (MapNode *)target; + return W_NEWNODE; + } + } + +#if !defined(NDEBUG) + /* Ensure that Collision.without implementation + converts to Bitmap nodes itself. + */ + if (IS_COLLISION_NODE(sub_node)) { + assert(map_node_collision_count( + (MapNode_Collision*)sub_node) > 1); + } +#endif + + if (mutid != 0 && self->b_mutid == mutid) { + target = self; + Py_INCREF(target); + } + else { + target = map_node_bitmap_clone(self, mutid); + if (target == NULL) { + return W_ERROR; + } + } + + Py_SETREF(target->b_array[val_idx], + (PyObject *)sub_node); /* borrow */ + + *new_node = (MapNode *)target; + return W_NEWNODE; + } + + case W_ERROR: + case W_NOT_FOUND: + assert(sub_node == NULL); + return res; + + default: + abort(); + } + } + else { + /* We have a regular key/value pair */ + + int cmp = PyObject_RichCompareBool(key_or_null, key, Py_EQ); + if (cmp < 0) { + return W_ERROR; + } + if (cmp == 0) { + return W_NOT_FOUND; + } + + if (map_node_bitmap_count(self) == 1) { + return W_EMPTY; + } + + *new_node = (MapNode *) + map_node_bitmap_clone_without(self, bit, mutid); + if (*new_node == NULL) { + return W_ERROR; + } + + return W_NEWNODE; + } +} + +static map_find_t +map_node_bitmap_find(MapNode_Bitmap *self, + uint32_t shift, int32_t hash, + PyObject *key, PyObject **val) +{ + /* Lookup a key in a Bitmap node. */ + + uint32_t bit = map_bitpos(hash, shift); + uint32_t idx; + uint32_t key_idx; + uint32_t val_idx; + PyObject *key_or_null; + PyObject *val_or_node; + int comp_err; + + if ((self->b_bitmap & bit) == 0) { + return F_NOT_FOUND; + } + + idx = map_bitindex(self->b_bitmap, bit); + key_idx = idx * 2; + val_idx = key_idx + 1; + + assert(val_idx < (size_t)Py_SIZE(self)); + + key_or_null = self->b_array[key_idx]; + val_or_node = self->b_array[val_idx]; + + if (key_or_null == NULL) { + /* There are a few keys that have the same hash at the current shift + that match our key. Dispatch the lookup further down the tree. */ + assert(val_or_node != NULL); + return map_node_find((MapNode *)val_or_node, + shift + 5, hash, key, val); + } + + /* We have only one key -- a potential match. Let's compare if the + key we are looking at is equal to the key we are looking for. */ + assert(key != NULL); + comp_err = PyObject_RichCompareBool(key, key_or_null, Py_EQ); + if (comp_err < 0) { /* exception in __eq__ */ + return F_ERROR; + } + if (comp_err == 1) { /* key == key_or_null */ + *val = val_or_node; + return F_FOUND; + } + + return F_NOT_FOUND; +} + +static int +map_node_bitmap_traverse(MapNode_Bitmap *self, visitproc visit, void *arg) +{ + /* Bitmap's tp_traverse */ + + Py_ssize_t i; + + for (i = Py_SIZE(self); --i >= 0; ) { + Py_VISIT(self->b_array[i]); + } + + return 0; +} + +static void +map_node_bitmap_dealloc(MapNode_Bitmap *self) +{ + /* Bitmap's tp_dealloc */ + + Py_ssize_t len = Py_SIZE(self); + Py_ssize_t i; + + PyObject_GC_UnTrack(self); + Py_TRASHCAN_SAFE_BEGIN(self) + + if (len > 0) { + i = len; + while (--i >= 0) { + Py_XDECREF(self->b_array[i]); + } + } + + Py_TYPE(self)->tp_free((PyObject *)self); + Py_TRASHCAN_SAFE_END(self) +} + +static int +map_node_bitmap_dump(MapNode_Bitmap *node, + _PyUnicodeWriter *writer, int level) +{ + /* Debug build: __dump__() method implementation for Bitmap nodes. */ + + Py_ssize_t i; + PyObject *tmp1; + PyObject *tmp2; + + if (_map_dump_ident(writer, level + 1)) { + goto error; + } + + if (_map_dump_format(writer, "BitmapNode(size=%zd count=%zd ", + Py_SIZE(node), Py_SIZE(node) / 2)) + { + goto error; + } + + tmp1 = PyLong_FromUnsignedLong(node->b_bitmap); + if (tmp1 == NULL) { + goto error; + } + tmp2 = _PyLong_Format(tmp1, 2); + Py_DECREF(tmp1); + if (tmp2 == NULL) { + goto error; + } + if (_map_dump_format(writer, "bitmap=%S id=%p):\n", tmp2, node)) { + Py_DECREF(tmp2); + goto error; + } + Py_DECREF(tmp2); + + for (i = 0; i < Py_SIZE(node); i += 2) { + PyObject *key_or_null = node->b_array[i]; + PyObject *val_or_node = node->b_array[i + 1]; + + if (_map_dump_ident(writer, level + 2)) { + goto error; + } + + if (key_or_null == NULL) { + if (_map_dump_format(writer, "NULL:\n")) { + goto error; + } + + if (map_node_dump((MapNode *)val_or_node, + writer, level + 2)) + { + goto error; + } + } + else { + if (_map_dump_format(writer, "%R: %R", key_or_null, + val_or_node)) + { + goto error; + } + } + + if (_map_dump_format(writer, "\n")) { + goto error; + } + } + + return 0; +error: + return -1; +} + + +/////////////////////////////////// Collision Node + + +static MapNode * +map_node_collision_new(int32_t hash, Py_ssize_t size, uint64_t mutid) +{ + /* Create a new Collision node. */ + + MapNode_Collision *node; + Py_ssize_t i; + + assert(size >= 4); + assert(size % 2 == 0); + + node = PyObject_GC_NewVar( + MapNode_Collision, &_Map_CollisionNode_Type, size); + if (node == NULL) { + return NULL; + } + + for (i = 0; i < size; i++) { + node->c_array[i] = NULL; + } + + Py_SET_SIZE(node, size); + node->c_hash = hash; + + node->c_mutid = mutid; + + PyObject_GC_Track(node); + return (MapNode *)node; +} + +static map_find_t +map_node_collision_find_index(MapNode_Collision *self, PyObject *key, + Py_ssize_t *idx) +{ + /* Lookup `key` in the Collision node `self`. Set the index of the + found key to 'idx'. */ + + Py_ssize_t i; + PyObject *el; + + for (i = 0; i < Py_SIZE(self); i += 2) { + el = self->c_array[i]; + + assert(el != NULL); + int cmp = PyObject_RichCompareBool(key, el, Py_EQ); + if (cmp < 0) { + return F_ERROR; + } + if (cmp == 1) { + *idx = i; + return F_FOUND; + } + } + + return F_NOT_FOUND; +} + +static MapNode * +map_node_collision_assoc(MapNode_Collision *self, + uint32_t shift, int32_t hash, + PyObject *key, PyObject *val, int* added_leaf, + uint64_t mutid) +{ + /* Set a new key to this level (currently a Collision node) + of the tree. */ + + if (hash == self->c_hash) { + /* The hash of the 'key' we are adding matches the hash of + other keys in this Collision node. */ + + Py_ssize_t key_idx = -1; + map_find_t found; + MapNode_Collision *new_node; + Py_ssize_t i; + + /* Let's try to lookup the new 'key', maybe we already have it. */ + found = map_node_collision_find_index(self, key, &key_idx); + switch (found) { + case F_ERROR: + /* Exception. */ + return NULL; + + case F_NOT_FOUND: + /* This is a totally new key. Clone the current node, + add a new key/value to the cloned node. */ + + new_node = (MapNode_Collision *)map_node_collision_new( + self->c_hash, Py_SIZE(self) + 2, mutid); + if (new_node == NULL) { + return NULL; + } + + for (i = 0; i < Py_SIZE(self); i++) { + Py_INCREF(self->c_array[i]); + new_node->c_array[i] = self->c_array[i]; + } + + Py_INCREF(key); + new_node->c_array[i] = key; + Py_INCREF(val); + new_node->c_array[i + 1] = val; + + *added_leaf = 1; + return (MapNode *)new_node; + + case F_FOUND: + /* There's a key which is equal to the key we are adding. */ + + assert(key_idx >= 0); + assert(key_idx < Py_SIZE(self)); + Py_ssize_t val_idx = key_idx + 1; + + if (self->c_array[val_idx] == val) { + /* We're setting a key/value pair that's already set. */ + Py_INCREF(self); + return (MapNode *)self; + } + + /* We need to replace old value for the key with + a new value. */ + + if (mutid != 0 && self->c_mutid == mutid) { + new_node = self; + Py_INCREF(self); + } + else { + /* Create a new Collision node.*/ + new_node = (MapNode_Collision *)map_node_collision_new( + self->c_hash, Py_SIZE(self), mutid); + if (new_node == NULL) { + return NULL; + } + + /* Copy all elements of the old node to the new one. */ + for (i = 0; i < Py_SIZE(self); i++) { + Py_INCREF(self->c_array[i]); + new_node->c_array[i] = self->c_array[i]; + } + } + + /* Replace the old value with the new value for the our key. */ + Py_DECREF(new_node->c_array[val_idx]); + Py_INCREF(val); + new_node->c_array[val_idx] = val; + + return (MapNode *)new_node; + + default: + abort(); + } + } + else { + /* The hash of the new key is different from the hash that + all keys of this Collision node have. + + Create a Bitmap node inplace with two children: + key/value pair that we're adding, and the Collision node + we're replacing on this tree level. + */ + + MapNode_Bitmap *new_node; + MapNode *assoc_res; + + new_node = (MapNode_Bitmap *)map_node_bitmap_new(2, mutid); + if (new_node == NULL) { + return NULL; + } + new_node->b_bitmap = map_bitpos(self->c_hash, shift); + Py_INCREF(self); + new_node->b_array[1] = (PyObject*) self; + + assoc_res = map_node_bitmap_assoc( + new_node, shift, hash, key, val, added_leaf, mutid); + Py_DECREF(new_node); + return assoc_res; + } +} + +static inline Py_ssize_t +map_node_collision_count(MapNode_Collision *node) +{ + return Py_SIZE(node) / 2; +} + +static map_without_t +map_node_collision_without(MapNode_Collision *self, + uint32_t shift, int32_t hash, + PyObject *key, + MapNode **new_node, + uint64_t mutid) +{ + if (hash != self->c_hash) { + return W_NOT_FOUND; + } + + Py_ssize_t key_idx = -1; + map_find_t found = map_node_collision_find_index(self, key, &key_idx); + + switch (found) { + case F_ERROR: + return W_ERROR; + + case F_NOT_FOUND: + return W_NOT_FOUND; + + case F_FOUND: + assert(key_idx >= 0); + assert(key_idx < Py_SIZE(self)); + + Py_ssize_t new_count = map_node_collision_count(self) - 1; + + if (new_count == 0) { + /* The node has only one key/value pair and it's for the + key we're trying to delete. So a new node will be empty + after the removal. + */ + return W_EMPTY; + } + + if (new_count == 1) { + /* The node has two keys, and after deletion the + new Collision node would have one. Collision nodes + with one key shouldn't exist, so convert it to a + Bitmap node. + */ + MapNode_Bitmap *node = (MapNode_Bitmap *) + map_node_bitmap_new(2, mutid); + if (node == NULL) { + return W_ERROR; + } + + if (key_idx == 0) { + Py_INCREF(self->c_array[2]); + node->b_array[0] = self->c_array[2]; + Py_INCREF(self->c_array[3]); + node->b_array[1] = self->c_array[3]; + } + else { + assert(key_idx == 2); + Py_INCREF(self->c_array[0]); + node->b_array[0] = self->c_array[0]; + Py_INCREF(self->c_array[1]); + node->b_array[1] = self->c_array[1]; + } + + node->b_bitmap = map_bitpos(hash, shift); + + *new_node = (MapNode *)node; + return W_NEWNODE; + } + + /* Allocate a new Collision node with capacity for one + less key/value pair */ + MapNode_Collision *new = (MapNode_Collision *) + map_node_collision_new( + self->c_hash, Py_SIZE(self) - 2, mutid); + if (new == NULL) { + return W_ERROR; + } + + /* Copy all other keys from `self` to `new` */ + Py_ssize_t i; + for (i = 0; i < key_idx; i++) { + Py_INCREF(self->c_array[i]); + new->c_array[i] = self->c_array[i]; + } + for (i = key_idx + 2; i < Py_SIZE(self); i++) { + Py_INCREF(self->c_array[i]); + new->c_array[i - 2] = self->c_array[i]; + } + + *new_node = (MapNode*)new; + return W_NEWNODE; + + default: + abort(); + } +} + +static map_find_t +map_node_collision_find(MapNode_Collision *self, + uint32_t shift, int32_t hash, + PyObject *key, PyObject **val) +{ + /* Lookup `key` in the Collision node `self`. Set the value + for the found key to 'val'. */ + + Py_ssize_t idx = -1; + map_find_t res; + + res = map_node_collision_find_index(self, key, &idx); + if (res == F_ERROR || res == F_NOT_FOUND) { + return res; + } + + assert(idx >= 0); + assert(idx + 1 < Py_SIZE(self)); + + *val = self->c_array[idx + 1]; + assert(*val != NULL); + + return F_FOUND; +} + + +static int +map_node_collision_traverse(MapNode_Collision *self, + visitproc visit, void *arg) +{ + /* Collision's tp_traverse */ + + Py_ssize_t i; + + for (i = Py_SIZE(self); --i >= 0; ) { + Py_VISIT(self->c_array[i]); + } + + return 0; +} + +static void +map_node_collision_dealloc(MapNode_Collision *self) +{ + /* Collision's tp_dealloc */ + + Py_ssize_t len = Py_SIZE(self); + + PyObject_GC_UnTrack(self); + Py_TRASHCAN_SAFE_BEGIN(self) + + if (len > 0) { + + while (--len >= 0) { + Py_XDECREF(self->c_array[len]); + } + } + + Py_TYPE(self)->tp_free((PyObject *)self); + Py_TRASHCAN_SAFE_END(self) +} + +static int +map_node_collision_dump(MapNode_Collision *node, + _PyUnicodeWriter *writer, int level) +{ + /* Debug build: __dump__() method implementation for Collision nodes. */ + + Py_ssize_t i; + + if (_map_dump_ident(writer, level + 1)) { + goto error; + } + + if (_map_dump_format(writer, "CollisionNode(size=%zd id=%p):\n", + Py_SIZE(node), node)) + { + goto error; + } + + for (i = 0; i < Py_SIZE(node); i += 2) { + PyObject *key = node->c_array[i]; + PyObject *val = node->c_array[i + 1]; + + if (_map_dump_ident(writer, level + 2)) { + goto error; + } + + if (_map_dump_format(writer, "%R: %R\n", key, val)) { + goto error; + } + } + + return 0; +error: + return -1; +} + + +/////////////////////////////////// Array Node + + +static MapNode * +map_node_array_new(Py_ssize_t count, uint64_t mutid) +{ + Py_ssize_t i; + + MapNode_Array *node = PyObject_GC_New( + MapNode_Array, &_Map_ArrayNode_Type); + if (node == NULL) { + return NULL; + } + + for (i = 0; i < HAMT_ARRAY_NODE_SIZE; i++) { + node->a_array[i] = NULL; + } + + node->a_count = count; + node->a_mutid = mutid; + + PyObject_GC_Track(node); + return (MapNode *)node; +} + +static MapNode_Array * +map_node_array_clone(MapNode_Array *node, uint64_t mutid) +{ + MapNode_Array *clone; + Py_ssize_t i; + + VALIDATE_ARRAY_NODE(node) + assert(node->a_count <= HAMT_ARRAY_NODE_SIZE); + + /* Create a new Array node. */ + clone = (MapNode_Array *)map_node_array_new(node->a_count, mutid); + if (clone == NULL) { + return NULL; + } + + /* Copy all elements from the current Array node to the new one. */ + for (i = 0; i < HAMT_ARRAY_NODE_SIZE; i++) { + Py_XINCREF(node->a_array[i]); + clone->a_array[i] = node->a_array[i]; + } + + clone->a_mutid = mutid; + + VALIDATE_ARRAY_NODE(clone) + return clone; +} + +static MapNode * +map_node_array_assoc(MapNode_Array *self, + uint32_t shift, int32_t hash, + PyObject *key, PyObject *val, int* added_leaf, + uint64_t mutid) +{ + /* Set a new key to this level (currently a Collision node) + of the tree. + + Array nodes don't store values, they can only point to + other nodes. They are simple arrays of 32 BaseNode pointers/ + */ + + uint32_t idx = map_mask(hash, shift); + MapNode *node = self->a_array[idx]; + MapNode *child_node; + MapNode_Array *new_node; + Py_ssize_t i; + + if (node == NULL) { + /* There's no child node for the given hash. Create a new + Bitmap node for this key. */ + + MapNode_Bitmap *empty = NULL; + + /* Get an empty Bitmap node to work with. */ + empty = (MapNode_Bitmap *)map_node_bitmap_new(0, mutid); + if (empty == NULL) { + return NULL; + } + + /* Set key/val to the newly created empty Bitmap, thus + creating a new Bitmap node with our key/value pair. */ + child_node = map_node_bitmap_assoc( + empty, + shift + 5, hash, key, val, added_leaf, mutid); + Py_DECREF(empty); + if (child_node == NULL) { + return NULL; + } + + if (mutid != 0 && self->a_mutid == mutid) { + new_node = self; + self->a_count++; + Py_INCREF(self); + } + else { + /* Create a new Array node. */ + new_node = (MapNode_Array *)map_node_array_new( + self->a_count + 1, mutid); + if (new_node == NULL) { + Py_DECREF(child_node); + return NULL; + } + + /* Copy all elements from the current Array node to the + new one. */ + for (i = 0; i < HAMT_ARRAY_NODE_SIZE; i++) { + Py_XINCREF(self->a_array[i]); + new_node->a_array[i] = self->a_array[i]; + } + } + + assert(new_node->a_array[idx] == NULL); + new_node->a_array[idx] = child_node; /* borrow */ + VALIDATE_ARRAY_NODE(new_node) + } + else { + /* There's a child node for the given hash. + Set the key to it./ */ + + child_node = map_node_assoc( + node, shift + 5, hash, key, val, added_leaf, mutid); + if (child_node == NULL) { + return NULL; + } + else if (child_node == (MapNode *)self) { + Py_DECREF(child_node); + return (MapNode *)self; + } + + if (mutid != 0 && self->a_mutid == mutid) { + new_node = self; + Py_INCREF(self); + } + else { + new_node = map_node_array_clone(self, mutid); + } + + if (new_node == NULL) { + Py_DECREF(child_node); + return NULL; + } + + Py_SETREF(new_node->a_array[idx], child_node); /* borrow */ + VALIDATE_ARRAY_NODE(new_node) + } + + return (MapNode *)new_node; +} + +static map_without_t +map_node_array_without(MapNode_Array *self, + uint32_t shift, int32_t hash, + PyObject *key, + MapNode **new_node, + uint64_t mutid) +{ + uint32_t idx = map_mask(hash, shift); + MapNode *node = self->a_array[idx]; + + if (node == NULL) { + return W_NOT_FOUND; + } + + MapNode *sub_node = NULL; + MapNode_Array *target = NULL; + map_without_t res = map_node_without( + (MapNode *)node, + shift + 5, hash, key, &sub_node, mutid); + + switch (res) { + case W_NOT_FOUND: + case W_ERROR: + assert(sub_node == NULL); + return res; + + case W_NEWNODE: { + /* We need to replace a node at the `idx` index. + Clone this node and replace. + */ + assert(sub_node != NULL); + + if (mutid != 0 && self->a_mutid == mutid) { + target = self; + Py_INCREF(self); + } + else { + target = map_node_array_clone(self, mutid); + if (target == NULL) { + Py_DECREF(sub_node); + return W_ERROR; + } + } + + Py_SETREF(target->a_array[idx], sub_node); /* borrow */ + *new_node = (MapNode*)target; /* borrow */ + return W_NEWNODE; + } + + case W_EMPTY: { + assert(sub_node == NULL); + /* We need to remove a node at the `idx` index. + Calculate the size of the replacement Array node. + */ + Py_ssize_t new_count = self->a_count - 1; + + if (new_count == 0) { + return W_EMPTY; + } + + if (new_count >= 16) { + /* We convert Bitmap nodes to Array nodes, when a + Bitmap node needs to store more than 15 key/value + pairs. So we will create a new Array node if we + the number of key/values after deletion is still + greater than 15. + */ + + if (mutid != 0 && self->a_mutid == mutid) { + target = self; + Py_INCREF(self); + } + else { + target = map_node_array_clone(self, mutid); + if (target == NULL) { + return W_ERROR; + } + } + + target->a_count = new_count; + Py_CLEAR(target->a_array[idx]); + + *new_node = (MapNode*)target; /* borrow */ + return W_NEWNODE; + } + + /* New Array node would have less than 16 key/value + pairs. We need to create a replacement Bitmap node. */ + + Py_ssize_t bitmap_size = new_count * 2; + uint32_t bitmap = 0; + + MapNode_Bitmap *new = (MapNode_Bitmap *) + map_node_bitmap_new(bitmap_size, mutid); + if (new == NULL) { + return W_ERROR; + } + + Py_ssize_t new_i = 0; + for (uint32_t i = 0; i < HAMT_ARRAY_NODE_SIZE; i++) { + if (i == idx) { + /* Skip the node we are deleting. */ + continue; + } + + MapNode *node = self->a_array[i]; + if (node == NULL) { + /* Skip any missing nodes. */ + continue; + } + + bitmap |= 1u << i; + + if (IS_BITMAP_NODE(node)) { + MapNode_Bitmap *child = (MapNode_Bitmap *)node; + + if (map_node_bitmap_count(child) == 1 && + child->b_array[0] != NULL) + { + /* node is a Bitmap with one key/value pair, just + merge it into the new Bitmap node we're building. + + Note that we don't inline Bitmap nodes that + have a NULL key -- those nodes point to another + tree level, and we cannot simply move tree levels + up or down. + */ + PyObject *key = child->b_array[0]; + PyObject *val = child->b_array[1]; + + Py_INCREF(key); + new->b_array[new_i] = key; + Py_INCREF(val); + new->b_array[new_i + 1] = val; + } + else { + new->b_array[new_i] = NULL; + Py_INCREF(node); + new->b_array[new_i + 1] = (PyObject*)node; + } + } + else { + +#if !defined(NDEBUG) + if (IS_COLLISION_NODE(node)) { + assert( + (map_node_collision_count( + (MapNode_Collision*)node)) > 1); + } + else if (IS_ARRAY_NODE(node)) { + assert(((MapNode_Array*)node)->a_count >= 16); + } +#endif + + /* Just copy the node into our new Bitmap */ + new->b_array[new_i] = NULL; + Py_INCREF(node); + new->b_array[new_i + 1] = (PyObject*)node; + } + + new_i += 2; + } + + new->b_bitmap = bitmap; + *new_node = (MapNode*)new; /* borrow */ + return W_NEWNODE; + } + + default: + abort(); + } +} + +static map_find_t +map_node_array_find(MapNode_Array *self, + uint32_t shift, int32_t hash, + PyObject *key, PyObject **val) +{ + /* Lookup `key` in the Array node `self`. Set the value + for the found key to 'val'. */ + + uint32_t idx = map_mask(hash, shift); + MapNode *node; + + node = self->a_array[idx]; + if (node == NULL) { + return F_NOT_FOUND; + } + + /* Dispatch to the generic map_node_find */ + return map_node_find(node, shift + 5, hash, key, val); +} + +static int +map_node_array_traverse(MapNode_Array *self, + visitproc visit, void *arg) +{ + /* Array's tp_traverse */ + + Py_ssize_t i; + + for (i = 0; i < HAMT_ARRAY_NODE_SIZE; i++) { + Py_VISIT(self->a_array[i]); + } + + return 0; +} + +static void +map_node_array_dealloc(MapNode_Array *self) +{ + /* Array's tp_dealloc */ + + Py_ssize_t i; + + PyObject_GC_UnTrack(self); + Py_TRASHCAN_SAFE_BEGIN(self) + + for (i = 0; i < HAMT_ARRAY_NODE_SIZE; i++) { + Py_XDECREF(self->a_array[i]); + } + + Py_TYPE(self)->tp_free((PyObject *)self); + Py_TRASHCAN_SAFE_END(self) +} + +static int +map_node_array_dump(MapNode_Array *node, + _PyUnicodeWriter *writer, int level) +{ + /* Debug build: __dump__() method implementation for Array nodes. */ + + Py_ssize_t i; + + if (_map_dump_ident(writer, level + 1)) { + goto error; + } + + if (_map_dump_format(writer, "ArrayNode(id=%p count=%zd):\n", + node, node->a_count) + ) { + goto error; + } + + for (i = 0; i < HAMT_ARRAY_NODE_SIZE; i++) { + if (node->a_array[i] == NULL) { + continue; + } + + if (_map_dump_ident(writer, level + 2)) { + goto error; + } + + if (_map_dump_format(writer, "%d::\n", i)) { + goto error; + } + + if (map_node_dump(node->a_array[i], writer, level + 1)) { + goto error; + } + + if (_map_dump_format(writer, "\n")) { + goto error; + } + } + + return 0; +error: + return -1; +} + + +/////////////////////////////////// Node Dispatch + + +static MapNode * +map_node_assoc(MapNode *node, + uint32_t shift, int32_t hash, + PyObject *key, PyObject *val, int* added_leaf, + uint64_t mutid) +{ + /* Set key/value to the 'node' starting with the given shift/hash. + Return a new node, or the same node if key/value already + set. + + added_leaf will be set to 1 if key/value wasn't in the + tree before. + + This method automatically dispatches to the suitable + map_node_{nodetype}_assoc method. + */ + + *added_leaf = 0; + + if (IS_BITMAP_NODE(node)) { + return map_node_bitmap_assoc( + (MapNode_Bitmap *)node, + shift, hash, key, val, added_leaf, mutid); + } + else if (IS_ARRAY_NODE(node)) { + return map_node_array_assoc( + (MapNode_Array *)node, + shift, hash, key, val, added_leaf, mutid); + } + else { + assert(IS_COLLISION_NODE(node)); + return map_node_collision_assoc( + (MapNode_Collision *)node, + shift, hash, key, val, added_leaf, mutid); + } +} + +static map_without_t +map_node_without(MapNode *node, + uint32_t shift, int32_t hash, + PyObject *key, + MapNode **new_node, + uint64_t mutid) +{ + if (IS_BITMAP_NODE(node)) { + return map_node_bitmap_without( + (MapNode_Bitmap *)node, + shift, hash, key, + new_node, + mutid); + } + else if (IS_ARRAY_NODE(node)) { + return map_node_array_without( + (MapNode_Array *)node, + shift, hash, key, + new_node, + mutid); + } + else { + assert(IS_COLLISION_NODE(node)); + return map_node_collision_without( + (MapNode_Collision *)node, + shift, hash, key, + new_node, + mutid); + } +} + +static map_find_t +map_node_find(MapNode *node, + uint32_t shift, int32_t hash, + PyObject *key, PyObject **val) +{ + /* Find the key in the node starting with the given shift/hash. + + If a value is found, the result will be set to F_FOUND, and + *val will point to the found value object. + + If a value wasn't found, the result will be set to F_NOT_FOUND. + + If an exception occurs during the call, the result will be F_ERROR. + + This method automatically dispatches to the suitable + map_node_{nodetype}_find method. + */ + + if (IS_BITMAP_NODE(node)) { + return map_node_bitmap_find( + (MapNode_Bitmap *)node, + shift, hash, key, val); + + } + else if (IS_ARRAY_NODE(node)) { + return map_node_array_find( + (MapNode_Array *)node, + shift, hash, key, val); + } + else { + assert(IS_COLLISION_NODE(node)); + return map_node_collision_find( + (MapNode_Collision *)node, + shift, hash, key, val); + } +} + +static int +map_node_dump(MapNode *node, + _PyUnicodeWriter *writer, int level) +{ + /* Debug build: __dump__() method implementation for a node. + + This method automatically dispatches to the suitable + map_node_{nodetype})_dump method. + */ + + if (IS_BITMAP_NODE(node)) { + return map_node_bitmap_dump( + (MapNode_Bitmap *)node, writer, level); + } + else if (IS_ARRAY_NODE(node)) { + return map_node_array_dump( + (MapNode_Array *)node, writer, level); + } + else { + assert(IS_COLLISION_NODE(node)); + return map_node_collision_dump( + (MapNode_Collision *)node, writer, level); + } +} + + +/////////////////////////////////// Iterators: Machinery + + +static map_iter_t +map_iterator_next(MapIteratorState *iter, PyObject **key, PyObject **val); + + +static void +map_iterator_init(MapIteratorState *iter, MapNode *root) +{ + for (uint32_t i = 0; i < _Py_HAMT_MAX_TREE_DEPTH; i++) { + iter->i_nodes[i] = NULL; + iter->i_pos[i] = 0; + } + + iter->i_level = 0; + + /* Note: we don't incref/decref nodes in i_nodes. */ + iter->i_nodes[0] = root; +} + +static map_iter_t +map_iterator_bitmap_next(MapIteratorState *iter, + PyObject **key, PyObject **val) +{ + int8_t level = iter->i_level; + + MapNode_Bitmap *node = (MapNode_Bitmap *)(iter->i_nodes[level]); + Py_ssize_t pos = iter->i_pos[level]; + + if (pos + 1 >= Py_SIZE(node)) { +#if !defined(NDEBUG) + assert(iter->i_level >= 0); + iter->i_nodes[iter->i_level] = NULL; +#endif + iter->i_level--; + return map_iterator_next(iter, key, val); + } + + if (node->b_array[pos] == NULL) { + iter->i_pos[level] = pos + 2; + + assert(level + 1 < _Py_HAMT_MAX_TREE_DEPTH); + int8_t next_level = (int8_t)(level + 1); + iter->i_level = next_level; + iter->i_pos[next_level] = 0; + iter->i_nodes[next_level] = (MapNode *) + node->b_array[pos + 1]; + + return map_iterator_next(iter, key, val); + } + + *key = node->b_array[pos]; + *val = node->b_array[pos + 1]; + iter->i_pos[level] = pos + 2; + return I_ITEM; +} + +static map_iter_t +map_iterator_collision_next(MapIteratorState *iter, + PyObject **key, PyObject **val) +{ + int8_t level = iter->i_level; + + MapNode_Collision *node = (MapNode_Collision *)(iter->i_nodes[level]); + Py_ssize_t pos = iter->i_pos[level]; + + if (pos + 1 >= Py_SIZE(node)) { +#if !defined(NDEBUG) + assert(iter->i_level >= 0); + iter->i_nodes[iter->i_level] = NULL; +#endif + iter->i_level--; + return map_iterator_next(iter, key, val); + } + + *key = node->c_array[pos]; + *val = node->c_array[pos + 1]; + iter->i_pos[level] = pos + 2; + return I_ITEM; +} + +static map_iter_t +map_iterator_array_next(MapIteratorState *iter, + PyObject **key, PyObject **val) +{ + int8_t level = iter->i_level; + + MapNode_Array *node = (MapNode_Array *)(iter->i_nodes[level]); + Py_ssize_t pos = iter->i_pos[level]; + + if (pos >= HAMT_ARRAY_NODE_SIZE) { +#if !defined(NDEBUG) + assert(iter->i_level >= 0); + iter->i_nodes[iter->i_level] = NULL; +#endif + iter->i_level--; + return map_iterator_next(iter, key, val); + } + + for (Py_ssize_t i = pos; i < HAMT_ARRAY_NODE_SIZE; i++) { + if (node->a_array[i] != NULL) { + iter->i_pos[level] = i + 1; + + assert((level + 1) < _Py_HAMT_MAX_TREE_DEPTH); + int8_t next_level = (int8_t)(level + 1); + iter->i_pos[next_level] = 0; + iter->i_nodes[next_level] = node->a_array[i]; + iter->i_level = next_level; + + return map_iterator_next(iter, key, val); + } + } + +#if !defined(NDEBUG) + assert(iter->i_level >= 0); + iter->i_nodes[iter->i_level] = NULL; +#endif + + iter->i_level--; + return map_iterator_next(iter, key, val); +} + +static map_iter_t +map_iterator_next(MapIteratorState *iter, PyObject **key, PyObject **val) +{ + if (iter->i_level < 0) { + return I_END; + } + + assert(iter->i_level < _Py_HAMT_MAX_TREE_DEPTH); + + MapNode *current = iter->i_nodes[iter->i_level]; + + if (IS_BITMAP_NODE(current)) { + return map_iterator_bitmap_next(iter, key, val); + } + else if (IS_ARRAY_NODE(current)) { + return map_iterator_array_next(iter, key, val); + } + else { + assert(IS_COLLISION_NODE(current)); + return map_iterator_collision_next(iter, key, val); + } +} + + +/////////////////////////////////// HAMT high-level functions + + +static MapObject * +map_assoc(MapObject *o, PyObject *key, PyObject *val) +{ + int32_t key_hash; + int added_leaf = 0; + MapNode *new_root; + MapObject *new_o; + + key_hash = map_hash(key); + if (key_hash == -1) { + return NULL; + } + + new_root = map_node_assoc( + (MapNode *)(o->h_root), + 0, key_hash, key, val, &added_leaf, + 0); + if (new_root == NULL) { + return NULL; + } + + if (new_root == o->h_root) { + Py_DECREF(new_root); + Py_INCREF(o); + return o; + } + + new_o = map_alloc(); + if (new_o == NULL) { + Py_DECREF(new_root); + return NULL; + } + + new_o->h_root = new_root; /* borrow */ + new_o->h_count = added_leaf ? o->h_count + 1 : o->h_count; + + return new_o; +} + +static MapObject * +map_without(MapObject *o, PyObject *key) +{ + int32_t key_hash = map_hash(key); + if (key_hash == -1) { + return NULL; + } + + MapNode *new_root = NULL; + + map_without_t res = map_node_without( + (MapNode *)(o->h_root), + 0, key_hash, key, + &new_root, + 0); + + switch (res) { + case W_ERROR: + return NULL; + case W_EMPTY: + return map_new(); + case W_NOT_FOUND: + PyErr_SetObject(PyExc_KeyError, key); + return NULL; + case W_NEWNODE: { + assert(new_root != NULL); + + MapObject *new_o = map_alloc(); + if (new_o == NULL) { + Py_DECREF(new_root); + return NULL; + } + + new_o->h_root = new_root; /* borrow */ + new_o->h_count = o->h_count - 1; + assert(new_o->h_count >= 0); + return new_o; + } + default: + abort(); + } +} + +static map_find_t +map_find(BaseMapObject *o, PyObject *key, PyObject **val) +{ + if (o->b_count == 0) { + return F_NOT_FOUND; + } + + int32_t key_hash = map_hash(key); + if (key_hash == -1) { + return F_ERROR; + } + + return map_node_find(o->b_root, 0, key_hash, key, val); +} + +static int +map_eq(BaseMapObject *v, BaseMapObject *w) +{ + if (v == w) { + return 1; + } + + if (v->b_count != w->b_count) { + return 0; + } + + MapIteratorState iter; + map_iter_t iter_res; + map_find_t find_res; + PyObject *v_key; + PyObject *v_val; + PyObject *w_val; + + map_iterator_init(&iter, v->b_root); + + do { + iter_res = map_iterator_next(&iter, &v_key, &v_val); + if (iter_res == I_ITEM) { + find_res = map_find(w, v_key, &w_val); + switch (find_res) { + case F_ERROR: + return -1; + + case F_NOT_FOUND: + return 0; + + case F_FOUND: { + int cmp = PyObject_RichCompareBool(v_val, w_val, Py_EQ); + if (cmp < 0) { + return -1; + } + if (cmp == 0) { + return 0; + } + } + } + } + } while (iter_res != I_END); + + return 1; +} + +static Py_ssize_t +map_len(BaseMapObject *o) +{ + return o->b_count; +} + +static MapObject * +map_alloc(void) +{ + MapObject *o; + o = PyObject_GC_New(MapObject, &_Map_Type); + if (o == NULL) { + return NULL; + } + o->h_weakreflist = NULL; + o->h_hash = -1; + o->h_count = 0; + o->h_root = NULL; + PyObject_GC_Track(o); + return o; +} + +static MapObject * +map_new(void) +{ + MapObject *o = map_alloc(); + if (o == NULL) { + return NULL; + } + + o->h_root = map_node_bitmap_new(0, 0); + if (o->h_root == NULL) { + Py_DECREF(o); + return NULL; + } + + return o; +} + +static PyObject * +map_dump(MapObject *self) +{ + _PyUnicodeWriter writer; + + _PyUnicodeWriter_Init(&writer); + + if (_map_dump_format(&writer, "HAMT(len=%zd):\n", self->h_count)) { + goto error; + } + + if (map_node_dump(self->h_root, &writer, 0)) { + goto error; + } + + return _PyUnicodeWriter_Finish(&writer); + +error: + _PyUnicodeWriter_Dealloc(&writer); + return NULL; +} + + +/////////////////////////////////// Iterators: Shared Iterator Implementation + + +static int +map_baseiter_tp_clear(MapIterator *it) +{ + Py_CLEAR(it->mi_obj); + return 0; +} + +static void +map_baseiter_tp_dealloc(MapIterator *it) +{ + PyObject_GC_UnTrack(it); + (void)map_baseiter_tp_clear(it); + PyObject_GC_Del(it); +} + +static int +map_baseiter_tp_traverse(MapIterator *it, visitproc visit, void *arg) +{ + Py_VISIT(it->mi_obj); + return 0; +} + +static PyObject * +map_baseiter_tp_iternext(MapIterator *it) +{ + PyObject *key; + PyObject *val; + map_iter_t res = map_iterator_next(&it->mi_iter, &key, &val); + + switch (res) { + case I_END: + PyErr_SetNone(PyExc_StopIteration); + return NULL; + + case I_ITEM: { + return (*(it->mi_yield))(key, val); + } + + default: { + abort(); + } + } +} + +static int +map_baseview_tp_clear(MapView *view) +{ + Py_CLEAR(view->mv_obj); + Py_CLEAR(view->mv_itertype); + return 0; +} + +static void +map_baseview_tp_dealloc(MapView *view) +{ + PyObject_GC_UnTrack(view); + (void)map_baseview_tp_clear(view); + PyObject_GC_Del(view); +} + +static int +map_baseview_tp_traverse(MapView *view, visitproc visit, void *arg) +{ + Py_VISIT(view->mv_obj); + return 0; +} + +static Py_ssize_t +map_baseview_tp_len(MapView *view) +{ + return view->mv_obj->h_count; +} + +static PyMappingMethods MapView_as_mapping = { + (lenfunc)map_baseview_tp_len, +}; + +static PyObject * +map_baseview_newiter(PyTypeObject *type, binaryfunc yield, MapObject *map) +{ + MapIterator *iter = PyObject_GC_New(MapIterator, type); + if (iter == NULL) { + return NULL; + } + + Py_INCREF(map); + iter->mi_obj = map; + iter->mi_yield = yield; + map_iterator_init(&iter->mi_iter, map->h_root); + + PyObject_GC_Track(iter); + return (PyObject *)iter; +} + +static PyObject * +map_baseview_iter(MapView *view) +{ + return map_baseview_newiter( + view->mv_itertype, view->mv_yield, view->mv_obj); +} + +static PyObject * +map_baseview_new(PyTypeObject *type, binaryfunc yield, + MapObject *o, PyTypeObject *itertype) +{ + MapView *view = PyObject_GC_New(MapView, type); + if (view == NULL) { + return NULL; + } + + Py_INCREF(o); + view->mv_obj = o; + view->mv_yield = yield; + + Py_INCREF(itertype); + view->mv_itertype = itertype; + + PyObject_GC_Track(view); + return (PyObject *)view; +} + +#define ITERATOR_TYPE_SHARED_SLOTS \ + .tp_basicsize = sizeof(MapIterator), \ + .tp_itemsize = 0, \ + .tp_dealloc = (destructor)map_baseiter_tp_dealloc, \ + .tp_getattro = PyObject_GenericGetAttr, \ + .tp_flags = Py_TPFLAGS_DEFAULT | Py_TPFLAGS_HAVE_GC, \ + .tp_traverse = (traverseproc)map_baseiter_tp_traverse, \ + .tp_clear = (inquiry)map_baseiter_tp_clear, \ + .tp_iter = PyObject_SelfIter, \ + .tp_iternext = (iternextfunc)map_baseiter_tp_iternext, + + +#define VIEW_TYPE_SHARED_SLOTS \ + .tp_basicsize = sizeof(MapView), \ + .tp_itemsize = 0, \ + .tp_as_mapping = &MapView_as_mapping, \ + .tp_dealloc = (destructor)map_baseview_tp_dealloc, \ + .tp_getattro = PyObject_GenericGetAttr, \ + .tp_flags = Py_TPFLAGS_DEFAULT | Py_TPFLAGS_HAVE_GC, \ + .tp_traverse = (traverseproc)map_baseview_tp_traverse, \ + .tp_clear = (inquiry)map_baseview_tp_clear, \ + .tp_iter = (getiterfunc)map_baseview_iter, \ + + +/////////////////////////////////// _MapItems_Type + + +PyTypeObject _MapItems_Type = { + PyVarObject_HEAD_INIT(NULL, 0) + "items", + VIEW_TYPE_SHARED_SLOTS +}; + +PyTypeObject _MapItemsIter_Type = { + PyVarObject_HEAD_INIT(NULL, 0) + "items_iterator", + ITERATOR_TYPE_SHARED_SLOTS +}; + +static PyObject * +map_iter_yield_items(PyObject *key, PyObject *val) +{ + return PyTuple_Pack(2, key, val); +} + +static PyObject * +map_new_items_view(MapObject *o) +{ + return map_baseview_new( + &_MapItems_Type, map_iter_yield_items, o, + &_MapItemsIter_Type); +} + + +/////////////////////////////////// _MapKeys_Type + + +PyTypeObject _MapKeys_Type = { + PyVarObject_HEAD_INIT(NULL, 0) + "keys", + VIEW_TYPE_SHARED_SLOTS +}; + +PyTypeObject _MapKeysIter_Type = { + PyVarObject_HEAD_INIT(NULL, 0) + "keys_iterator", + ITERATOR_TYPE_SHARED_SLOTS +}; + +static PyObject * +map_iter_yield_keys(PyObject *key, PyObject *val) +{ + Py_INCREF(key); + return key; +} + +static PyObject * +map_new_keys_iter(MapObject *o) +{ + return map_baseview_newiter( + &_MapKeysIter_Type, map_iter_yield_keys, o); +} + +static PyObject * +map_new_keys_view(MapObject *o) +{ + return map_baseview_new( + &_MapKeys_Type, map_iter_yield_keys, o, + &_MapKeysIter_Type); +} + +/////////////////////////////////// _MapValues_Type + + +PyTypeObject _MapValues_Type = { + PyVarObject_HEAD_INIT(NULL, 0) + "values", + VIEW_TYPE_SHARED_SLOTS +}; + +PyTypeObject _MapValuesIter_Type = { + PyVarObject_HEAD_INIT(NULL, 0) + "values_iterator", + ITERATOR_TYPE_SHARED_SLOTS +}; + +static PyObject * +map_iter_yield_values(PyObject *key, PyObject *val) +{ + Py_INCREF(val); + return val; +} + +static PyObject * +map_new_values_view(MapObject *o) +{ + return map_baseview_new( + &_MapValues_Type, map_iter_yield_values, o, + &_MapValuesIter_Type); +} + + +/////////////////////////////////// _Map_Type + + +static PyObject * +map_dump(MapObject *self); + + +static PyObject * +map_tp_new(PyTypeObject *type, PyObject *args, PyObject *kwds) +{ + return (PyObject*)map_new(); +} + + +static int +map_tp_init(MapObject *self, PyObject *args, PyObject *kwds) +{ + PyObject *arg = NULL; + uint64_t mutid = 0; + + if (!PyArg_UnpackTuple(args, "immutables.Map", 0, 1, &arg)) { + return -1; + } + + if (arg != NULL) { + if (Map_Check(arg)) { + MapObject *other = (MapObject *)arg; + + Py_INCREF(other->h_root); + Py_SETREF(self->h_root, other->h_root); + + self->h_count = other->h_count; + self->h_hash = other->h_hash; + } + else if (MapMutation_Check(arg)) { + PyErr_Format( + PyExc_TypeError, + "cannot create Maps from MapMutations"); + return -1; + } + else { + mutid = mutid_counter++; + if (map_update_inplace(mutid, (BaseMapObject *)self, arg)) { + return -1; + } + } + } + + if (kwds != NULL) { + if (!PyArg_ValidateKeywordArguments(kwds)) { + return -1; + } + + if (!mutid) { + mutid = mutid_counter++; + } + + if (map_update_inplace(mutid, (BaseMapObject *)self, kwds)) { + return -1; + } + } + + return 0; +} + + +static int +map_tp_clear(BaseMapObject *self) +{ + Py_CLEAR(self->b_root); + return 0; +} + + +static int +map_tp_traverse(BaseMapObject *self, visitproc visit, void *arg) +{ + Py_VISIT(self->b_root); + return 0; +} + +static void +map_tp_dealloc(BaseMapObject *self) +{ + PyObject_GC_UnTrack(self); + if (self->b_weakreflist != NULL) { + PyObject_ClearWeakRefs((PyObject*)self); + } + (void)map_tp_clear(self); + Py_TYPE(self)->tp_free(self); +} + + +static PyObject * +map_tp_richcompare(PyObject *v, PyObject *w, int op) +{ + if (!Map_Check(v) || !Map_Check(w) || (op != Py_EQ && op != Py_NE)) { + Py_RETURN_NOTIMPLEMENTED; + } + + int res = map_eq((BaseMapObject *)v, (BaseMapObject *)w); + if (res < 0) { + return NULL; + } + + if (op == Py_NE) { + res = !res; + } + + if (res) { + Py_RETURN_TRUE; + } + else { + Py_RETURN_FALSE; + } +} + +static int +map_tp_contains(BaseMapObject *self, PyObject *key) +{ + PyObject *val; + map_find_t res = map_find(self, key, &val); + switch (res) { + case F_ERROR: + return -1; + case F_NOT_FOUND: + return 0; + case F_FOUND: + return 1; + default: + abort(); + } +} + +static PyObject * +map_tp_subscript(BaseMapObject *self, PyObject *key) +{ + PyObject *val; + map_find_t res = map_find(self, key, &val); + switch (res) { + case F_ERROR: + return NULL; + case F_FOUND: + Py_INCREF(val); + return val; + case F_NOT_FOUND: + PyErr_SetObject(PyExc_KeyError, key); + return NULL; + default: + abort(); + } +} + +static Py_ssize_t +map_tp_len(BaseMapObject *self) +{ + return map_len(self); +} + +static PyObject * +map_tp_iter(MapObject *self) +{ + return map_new_keys_iter(self); +} + +static PyObject * +map_py_set(MapObject *self, PyObject *args) +{ + PyObject *key; + PyObject *val; + + if (!PyArg_UnpackTuple(args, "set", 2, 2, &key, &val)) { + return NULL; + } + + return (PyObject *)map_assoc(self, key, val); +} + +static PyObject * +map_py_get(BaseMapObject *self, PyObject *args) +{ + PyObject *key; + PyObject *def = NULL; + + if (!PyArg_UnpackTuple(args, "get", 1, 2, &key, &def)) { + return NULL; + } + + PyObject *val = NULL; + map_find_t res = map_find(self, key, &val); + switch (res) { + case F_ERROR: + return NULL; + case F_FOUND: + Py_INCREF(val); + return val; + case F_NOT_FOUND: + if (def == NULL) { + Py_RETURN_NONE; + } + Py_INCREF(def); + return def; + default: + abort(); + } +} + +static PyObject * +map_py_delete(MapObject *self, PyObject *key) +{ + return (PyObject *)map_without(self, key); +} + +static PyObject * +map_py_mutate(MapObject *self, PyObject *args) +{ + + MapMutationObject *o; + o = PyObject_GC_New(MapMutationObject, &_MapMutation_Type); + if (o == NULL) { + return NULL; + } + o->m_weakreflist = NULL; + o->m_count = self->h_count; + + Py_INCREF(self->h_root); + o->m_root = self->h_root; + + o->m_mutid = mutid_counter++; + + PyObject_GC_Track(o); + return (PyObject *)o; +} + +static PyObject * +map_py_update(MapObject *self, PyObject *args, PyObject *kwds) +{ + PyObject *arg = NULL; + MapObject *new = NULL; + uint64_t mutid = 0; + + if (!PyArg_UnpackTuple(args, "update", 0, 1, &arg)) { + return NULL; + } + + if (arg != NULL) { + mutid = mutid_counter++; + new = map_update(mutid, self, arg); + if (new == NULL) { + return NULL; + } + } + else { + Py_INCREF(self); + new = self; + } + + if (kwds != NULL) { + if (!PyArg_ValidateKeywordArguments(kwds)) { + Py_DECREF(new); + return NULL; + } + + if (!mutid) { + mutid = mutid_counter++; + } + + MapObject *new2 = map_update(mutid, new, kwds); + Py_DECREF(new); + if (new2 == NULL) { + return NULL; + } + new = new2; + } + + return (PyObject *)new; +} + +static PyObject * +map_py_items(MapObject *self, PyObject *args) +{ + return map_new_items_view(self); +} + +static PyObject * +map_py_values(MapObject *self, PyObject *args) +{ + return map_new_values_view(self); +} + +static PyObject * +map_py_keys(MapObject *self, PyObject *args) +{ + return map_new_keys_view(self); +} + +static PyObject * +map_py_dump(MapObject *self, PyObject *args) +{ + return map_dump(self); +} + + +static PyObject * +map_py_repr(BaseMapObject *m) +{ + Py_ssize_t i; + _PyUnicodeWriter writer; + + + i = Py_ReprEnter((PyObject *)m); + if (i != 0) { + return i > 0 ? PyUnicode_FromString("{...}") : NULL; + } + + _PyUnicodeWriter_Init(&writer); + + if (MapMutation_Check(m)) { + if (_PyUnicodeWriter_WriteASCIIString( + &writer, "immutables.MapMutation({", 24) < 0) + { + goto error; + } + } + else { + if (_PyUnicodeWriter_WriteASCIIString( + &writer, "immutables.Map({", 16) < 0) + { + goto error; + } + } + + MapIteratorState iter; + map_iter_t iter_res; + map_iterator_init(&iter, m->b_root); + int second = 0; + do { + PyObject *v_key; + PyObject *v_val; + + iter_res = map_iterator_next(&iter, &v_key, &v_val); + if (iter_res == I_ITEM) { + if (second) { + if (_PyUnicodeWriter_WriteASCIIString(&writer, ", ", 2) < 0) { + goto error; + } + } + + PyObject *s = PyObject_Repr(v_key); + if (s == NULL) { + goto error; + } + if (_PyUnicodeWriter_WriteStr(&writer, s) < 0) { + Py_DECREF(s); + goto error; + } + Py_DECREF(s); + + if (_PyUnicodeWriter_WriteASCIIString(&writer, ": ", 2) < 0) { + goto error; + } + + s = PyObject_Repr(v_val); + if (s == NULL) { + goto error; + } + if (_PyUnicodeWriter_WriteStr(&writer, s) < 0) { + Py_DECREF(s); + goto error; + } + Py_DECREF(s); + } + + second = 1; + } while (iter_res != I_END); + + if (_PyUnicodeWriter_WriteASCIIString(&writer, "})", 2) < 0) { + goto error; + } + + Py_ReprLeave((PyObject *)m); + return _PyUnicodeWriter_Finish(&writer); + +error: + _PyUnicodeWriter_Dealloc(&writer); + Py_ReprLeave((PyObject *)m); + return NULL; +} + + +static Py_uhash_t +_shuffle_bits(Py_uhash_t h) +{ + return ((h ^ 89869747UL) ^ (h << 16)) * 3644798167UL; +} + + +static Py_hash_t +map_py_hash(MapObject *self) +{ + /* Adapted version of frozenset.__hash__: it's important + that Map.__hash__ is independant of key/values order. + + Optimization idea: compute and memoize intermediate + hash values for HAMT nodes. + */ + + if (self->h_hash != -1) { + return self->h_hash; + } + + Py_uhash_t hash = 0; + + MapIteratorState iter; + map_iter_t iter_res; + map_iterator_init(&iter, self->h_root); + do { + PyObject *v_key; + PyObject *v_val; + + iter_res = map_iterator_next(&iter, &v_key, &v_val); + if (iter_res == I_ITEM) { + Py_hash_t vh = PyObject_Hash(v_key); + if (vh == -1) { + return -1; + } + hash ^= _shuffle_bits((Py_uhash_t)vh); + + vh = PyObject_Hash(v_val); + if (vh == -1) { + return -1; + } + hash ^= _shuffle_bits((Py_uhash_t)vh); + } + } while (iter_res != I_END); + + hash ^= ((Py_uhash_t)self->h_count * 2 + 1) * 1927868237UL; + + hash ^= (hash >> 11) ^ (hash >> 25); + hash = hash * 69069U + 907133923UL; + + self->h_hash = (Py_hash_t)hash; + if (self->h_hash == -1) { + self->h_hash = 1; + } + return self->h_hash; +} + +static PyObject * +map_reduce(MapObject *self) +{ + MapIteratorState iter; + map_iter_t iter_res; + + PyObject *dict = PyDict_New(); + if (dict == NULL) { + return NULL; + } + + map_iterator_init(&iter, self->h_root); + do { + PyObject *key; + PyObject *val; + + iter_res = map_iterator_next(&iter, &key, &val); + if (iter_res == I_ITEM) { + if (PyDict_SetItem(dict, key, val) < 0) { + Py_DECREF(dict); + return NULL; + } + } + } while (iter_res != I_END); + + PyObject *args = PyTuple_Pack(1, dict); + Py_DECREF(dict); + if (args == NULL) { + return NULL; + } + + PyObject *tup = PyTuple_Pack(2, Py_TYPE(self), args); + Py_DECREF(args); + return tup; +} + +static PyObject * +map_py_class_getitem(PyObject *type, PyObject *item) +{ + Py_INCREF(type); + return type; +} + +static PyMethodDef Map_methods[] = { + {"set", (PyCFunction)map_py_set, METH_VARARGS, NULL}, + {"get", (PyCFunction)map_py_get, METH_VARARGS, NULL}, + {"delete", (PyCFunction)map_py_delete, METH_O, NULL}, + {"mutate", (PyCFunction)map_py_mutate, METH_NOARGS, NULL}, + {"items", (PyCFunction)map_py_items, METH_NOARGS, NULL}, + {"keys", (PyCFunction)map_py_keys, METH_NOARGS, NULL}, + {"values", (PyCFunction)map_py_values, METH_NOARGS, NULL}, + {"update", (PyCFunction)map_py_update, METH_VARARGS | METH_KEYWORDS, NULL}, + {"__reduce__", (PyCFunction)map_reduce, METH_NOARGS, NULL}, + {"__dump__", (PyCFunction)map_py_dump, METH_NOARGS, NULL}, + { + "__class_getitem__", + (PyCFunction)map_py_class_getitem, + METH_O|METH_CLASS, + NULL + }, + {NULL, NULL} +}; + +static PySequenceMethods Map_as_sequence = { + 0, /* sq_length */ + 0, /* sq_concat */ + 0, /* sq_repeat */ + 0, /* sq_item */ + 0, /* sq_slice */ + 0, /* sq_ass_item */ + 0, /* sq_ass_slice */ + (objobjproc)map_tp_contains, /* sq_contains */ + 0, /* sq_inplace_concat */ + 0, /* sq_inplace_repeat */ +}; + +static PyMappingMethods Map_as_mapping = { + (lenfunc)map_tp_len, /* mp_length */ + (binaryfunc)map_tp_subscript, /* mp_subscript */ +}; + +PyTypeObject _Map_Type = { + PyVarObject_HEAD_INIT(NULL, 0) + "immutables._map.Map", + sizeof(MapObject), + .tp_methods = Map_methods, + .tp_as_mapping = &Map_as_mapping, + .tp_as_sequence = &Map_as_sequence, + .tp_iter = (getiterfunc)map_tp_iter, + .tp_dealloc = (destructor)map_tp_dealloc, + .tp_getattro = PyObject_GenericGetAttr, + .tp_flags = Py_TPFLAGS_DEFAULT | Py_TPFLAGS_HAVE_GC, + .tp_richcompare = map_tp_richcompare, + .tp_traverse = (traverseproc)map_tp_traverse, + .tp_clear = (inquiry)map_tp_clear, + .tp_new = map_tp_new, + .tp_init = (initproc)map_tp_init, + .tp_weaklistoffset = offsetof(MapObject, h_weakreflist), + .tp_hash = (hashfunc)map_py_hash, + .tp_repr = (reprfunc)map_py_repr, +}; + + +/////////////////////////////////// MapMutation + + +static int +map_node_update_from_map(uint64_t mutid, + MapObject *map, + MapNode *root, Py_ssize_t count, + MapNode **new_root, Py_ssize_t *new_count) +{ + assert(Map_Check(map)); + + MapIteratorState iter; + map_iter_t iter_res; + + MapNode *last_root; + Py_ssize_t last_count; + + Py_INCREF(root); + last_root = root; + last_count = count; + + map_iterator_init(&iter, map->h_root); + do { + PyObject *key; + PyObject *val; + int32_t key_hash; + int added_leaf; + + iter_res = map_iterator_next(&iter, &key, &val); + if (iter_res == I_ITEM) { + key_hash = map_hash(key); + if (key_hash == -1) { + goto err; + } + + MapNode *iter_root = map_node_assoc( + last_root, + 0, key_hash, key, val, &added_leaf, + mutid); + + if (iter_root == NULL) { + goto err; + } + + if (added_leaf) { + last_count++; + } + + Py_SETREF(last_root, iter_root); + } + } while (iter_res != I_END); + + *new_root = last_root; + *new_count = last_count; + + return 0; + +err: + Py_DECREF(last_root); + return -1; +} + + +static int +map_node_update_from_dict(uint64_t mutid, + PyObject *dct, + MapNode *root, Py_ssize_t count, + MapNode **new_root, Py_ssize_t *new_count) +{ + assert(PyDict_Check(dct)); + + PyObject *it = PyObject_GetIter(dct); + if (it == NULL) { + return -1; + } + + MapNode *last_root; + Py_ssize_t last_count; + + Py_INCREF(root); + last_root = root; + last_count = count; + + PyObject *key; + + while ((key = PyIter_Next(it))) { + PyObject *val; + int added_leaf; + int32_t key_hash; + + key_hash = map_hash(key); + if (key_hash == -1) { + Py_DECREF(key); + goto err; + } + + val = PyDict_GetItemWithError(dct, key); + if (val == NULL) { + Py_DECREF(key); + goto err; + } + + MapNode *iter_root = map_node_assoc( + last_root, + 0, key_hash, key, val, &added_leaf, + mutid); + + Py_DECREF(key); + + if (iter_root == NULL) { + goto err; + } + + if (added_leaf) { + last_count++; + } + + Py_SETREF(last_root, iter_root); + } + + if (key == NULL && PyErr_Occurred()) { + goto err; + } + + Py_DECREF(it); + + *new_root = last_root; + *new_count = last_count; + + return 0; + +err: + Py_DECREF(it); + Py_DECREF(last_root); + return -1; +} + + +static int +map_node_update_from_seq(uint64_t mutid, + PyObject *seq, + MapNode *root, Py_ssize_t count, + MapNode **new_root, Py_ssize_t *new_count) +{ + PyObject *it; + Py_ssize_t i; + PyObject *item = NULL; + PyObject *fast = NULL; + + MapNode *last_root; + Py_ssize_t last_count; + + it = PyObject_GetIter(seq); + if (it == NULL) { + return -1; + } + + Py_INCREF(root); + last_root = root; + last_count = count; + + for (i = 0; ; i++) { + PyObject *key, *val; + Py_ssize_t n; + int32_t key_hash; + int added_leaf; + + item = PyIter_Next(it); + if (item == NULL) { + if (PyErr_Occurred()) { + goto err; + } + break; + } + + fast = PySequence_Fast(item, ""); + if (fast == NULL) { + if (PyErr_ExceptionMatches(PyExc_TypeError)) + PyErr_Format(PyExc_TypeError, + "cannot convert map update " + "sequence element #%zd to a sequence", + i); + goto err; + } + + n = PySequence_Fast_GET_SIZE(fast); + if (n != 2) { + PyErr_Format(PyExc_ValueError, + "map update sequence element #%zd " + "has length %zd; 2 is required", + i, n); + goto err; + } + + key = PySequence_Fast_GET_ITEM(fast, 0); + val = PySequence_Fast_GET_ITEM(fast, 1); + Py_INCREF(key); + Py_INCREF(val); + + key_hash = map_hash(key); + if (key_hash == -1) { + Py_DECREF(key); + Py_DECREF(val); + goto err; + } + + MapNode *iter_root = map_node_assoc( + last_root, + 0, key_hash, key, val, &added_leaf, + mutid); + + Py_DECREF(key); + Py_DECREF(val); + + if (iter_root == NULL) { + goto err; + } + + if (added_leaf) { + last_count++; + } + + Py_SETREF(last_root, iter_root); + + Py_DECREF(fast); + Py_DECREF(item); + } + + Py_DECREF(it); + + *new_root = last_root; + *new_count = last_count; + + return 0; + +err: + Py_DECREF(last_root); + Py_XDECREF(item); + Py_XDECREF(fast); + Py_DECREF(it); + return -1; +} + + +static int +map_node_update(uint64_t mutid, + PyObject *src, + MapNode *root, Py_ssize_t count, + MapNode **new_root, Py_ssize_t *new_count) +{ + if (Map_Check(src)) { + return map_node_update_from_map( + mutid, (MapObject *)src, root, count, new_root, new_count); + } + else if (PyDict_Check(src)) { + return map_node_update_from_dict( + mutid, src, root, count, new_root, new_count); + } + else { + return map_node_update_from_seq( + mutid, src, root, count, new_root, new_count); + } +} + + +static int +map_update_inplace(uint64_t mutid, BaseMapObject *o, PyObject *src) +{ + MapNode *new_root = NULL; + Py_ssize_t new_count; + + int ret = map_node_update( + mutid, src, + o->b_root, o->b_count, + &new_root, &new_count); + + if (ret) { + return -1; + } + + assert(new_root); + + Py_SETREF(o->b_root, new_root); + o->b_count = new_count; + + return 0; +} + + +static MapObject * +map_update(uint64_t mutid, MapObject *o, PyObject *src) +{ + MapNode *new_root = NULL; + Py_ssize_t new_count; + + int ret = map_node_update( + mutid, src, + o->h_root, o->h_count, + &new_root, &new_count); + + if (ret) { + return NULL; + } + + assert(new_root); + + MapObject *new = map_alloc(); + if (new == NULL) { + Py_DECREF(new_root); + return NULL; + } + + Py_XSETREF(new->h_root, new_root); + new->h_count = new_count; + + return new; +} + +static int +mapmut_check_finalized(MapMutationObject *o) +{ + if (o->m_mutid == 0) { + PyErr_Format( + PyExc_ValueError, + "mutation %R has been finished", + o, NULL); + return -1; + } + + return 0; +} + +static int +mapmut_delete(MapMutationObject *o, PyObject *key, int32_t key_hash) +{ + MapNode *new_root = NULL; + + assert(key_hash != -1); + map_without_t res = map_node_without( + (MapNode *)(o->m_root), + 0, key_hash, key, + &new_root, + o->m_mutid); + + switch (res) { + case W_ERROR: + return -1; + + case W_EMPTY: + new_root = map_node_bitmap_new(0, o->m_mutid); + if (new_root == NULL) { + return -1; + } + Py_SETREF(o->m_root, new_root); + o->m_count = 0; + return 0; + + case W_NOT_FOUND: + PyErr_SetObject(PyExc_KeyError, key); + return -1; + + case W_NEWNODE: { + assert(new_root != NULL); + Py_SETREF(o->m_root, new_root); + o->m_count--; + return 0; + } + + default: + abort(); + } +} + +static int +mapmut_set(MapMutationObject *o, PyObject *key, int32_t key_hash, + PyObject *val) +{ + int added_leaf = 0; + + assert(key_hash != -1); + MapNode *new_root = map_node_assoc( + (MapNode *)(o->m_root), + 0, key_hash, key, val, &added_leaf, + o->m_mutid); + if (new_root == NULL) { + return -1; + } + + if (added_leaf) { + o->m_count++; + } + + if (new_root == o->m_root) { + Py_DECREF(new_root); + return 0; + } + + Py_SETREF(o->m_root, new_root); + return 0; +} + +static int +mapmut_finish(MapMutationObject *o) +{ + o->m_mutid = 0; + return 0; +} + +static PyObject * +mapmut_py_set(MapMutationObject *o, PyObject *args) +{ + PyObject *key; + PyObject *val; + + if (!PyArg_UnpackTuple(args, "set", 2, 2, &key, &val)) { + return NULL; + } + + if (mapmut_check_finalized(o)) { + return NULL; + } + + int32_t key_hash = map_hash(key); + if (key_hash == -1) { + return NULL; + } + + if (mapmut_set(o, key, key_hash, val)) { + return NULL; + } + + Py_RETURN_NONE; +} + +static PyObject * +mapmut_tp_richcompare(PyObject *v, PyObject *w, int op) +{ + if (!MapMutation_Check(v) || !MapMutation_Check(w) || + (op != Py_EQ && op != Py_NE)) + { + Py_RETURN_NOTIMPLEMENTED; + } + + int res = map_eq((BaseMapObject *)v, (BaseMapObject *)w); + if (res < 0) { + return NULL; + } + + if (op == Py_NE) { + res = !res; + } + + if (res) { + Py_RETURN_TRUE; + } + else { + Py_RETURN_FALSE; + } +} + +static PyObject * +mapmut_py_update(MapMutationObject *self, PyObject *args, PyObject *kwds) +{ + PyObject *arg = NULL; + + if (!PyArg_UnpackTuple(args, "update", 0, 1, &arg)) { + return NULL; + } + + if (mapmut_check_finalized(self)) { + return NULL; + } + + if (arg != NULL) { + if (map_update_inplace(self->m_mutid, (BaseMapObject *)self, arg)) { + return NULL; + } + } + + if (kwds != NULL) { + if (!PyArg_ValidateKeywordArguments(kwds)) { + return NULL; + } + + if (map_update_inplace(self->m_mutid, (BaseMapObject *)self, kwds)) { + return NULL; + } + } + + Py_RETURN_NONE; +} + + +static PyObject * +mapmut_py_finish(MapMutationObject *self, PyObject *args) +{ + if (mapmut_finish(self)) { + return NULL; + } + + MapObject *o = map_alloc(); + if (o == NULL) { + return NULL; + } + + Py_INCREF(self->m_root); + o->h_root = self->m_root; + o->h_count = self->m_count; + + return (PyObject *)o; +} + +static PyObject * +mapmut_py_enter(MapMutationObject *self, PyObject *args) +{ + Py_INCREF(self); + return (PyObject *)self; +} + +static PyObject * +mapmut_py_exit(MapMutationObject *self, PyObject *args) +{ + if (mapmut_finish(self)) { + return NULL; + } + Py_RETURN_FALSE; +} + +static int +mapmut_tp_ass_sub(MapMutationObject *self, PyObject *key, PyObject *val) +{ + if (mapmut_check_finalized(self)) { + return -1; + } + + int32_t key_hash = map_hash(key); + if (key_hash == -1) { + return -1; + } + + if (val == NULL) { + return mapmut_delete(self, key, key_hash); + } + else { + return mapmut_set(self, key, key_hash, val); + } +} + +static PyObject * +mapmut_py_pop(MapMutationObject *self, PyObject *args) +{ + PyObject *key, *deflt = NULL, *val = NULL; + + if(!PyArg_UnpackTuple(args, "pop", 1, 2, &key, &deflt)) { + return NULL; + } + + if (mapmut_check_finalized(self)) { + return NULL; + } + + if (!self->m_count) { + goto not_found; + } + + int32_t key_hash = map_hash(key); + if (key_hash == -1) { + return NULL; + } + + map_find_t find_res = map_node_find(self->m_root, 0, key_hash, key, &val); + + switch (find_res) { + case F_ERROR: + return NULL; + + case F_NOT_FOUND: + goto not_found; + + case F_FOUND: + break; + + default: + abort(); + } + + Py_INCREF(val); + + if (mapmut_delete(self, key, key_hash)) { + Py_DECREF(val); + return NULL; + } + + return val; + +not_found: + if (deflt) { + Py_INCREF(deflt); + return deflt; + } + + PyErr_SetObject(PyExc_KeyError, key); + return NULL; +} + + +static PyMethodDef MapMutation_methods[] = { + {"set", (PyCFunction)mapmut_py_set, METH_VARARGS, NULL}, + {"get", (PyCFunction)map_py_get, METH_VARARGS, NULL}, + {"pop", (PyCFunction)mapmut_py_pop, METH_VARARGS, NULL}, + {"finish", (PyCFunction)mapmut_py_finish, METH_NOARGS, NULL}, + {"update", (PyCFunction)mapmut_py_update, + METH_VARARGS | METH_KEYWORDS, NULL}, + {"__enter__", (PyCFunction)mapmut_py_enter, METH_NOARGS, NULL}, + {"__exit__", (PyCFunction)mapmut_py_exit, METH_VARARGS, NULL}, + {NULL, NULL} +}; + +static PySequenceMethods MapMutation_as_sequence = { + 0, /* sq_length */ + 0, /* sq_concat */ + 0, /* sq_repeat */ + 0, /* sq_item */ + 0, /* sq_slice */ + 0, /* sq_ass_item */ + 0, /* sq_ass_slice */ + (objobjproc)map_tp_contains, /* sq_contains */ + 0, /* sq_inplace_concat */ + 0, /* sq_inplace_repeat */ +}; + +static PyMappingMethods MapMutation_as_mapping = { + (lenfunc)map_tp_len, /* mp_length */ + (binaryfunc)map_tp_subscript, /* mp_subscript */ + (objobjargproc)mapmut_tp_ass_sub, /* mp_subscript */ +}; + +PyTypeObject _MapMutation_Type = { + PyVarObject_HEAD_INIT(NULL, 0) + "immutables._map.MapMutation", + sizeof(MapMutationObject), + .tp_methods = MapMutation_methods, + .tp_as_mapping = &MapMutation_as_mapping, + .tp_as_sequence = &MapMutation_as_sequence, + .tp_dealloc = (destructor)map_tp_dealloc, + .tp_getattro = PyObject_GenericGetAttr, + .tp_flags = Py_TPFLAGS_DEFAULT | Py_TPFLAGS_HAVE_GC, + .tp_traverse = (traverseproc)map_tp_traverse, + .tp_richcompare = mapmut_tp_richcompare, + .tp_clear = (inquiry)map_tp_clear, + .tp_weaklistoffset = offsetof(MapMutationObject, m_weakreflist), + .tp_repr = (reprfunc)map_py_repr, + .tp_hash = PyObject_HashNotImplemented, +}; + + +/////////////////////////////////// Tree Node Types + + +PyTypeObject _Map_ArrayNode_Type = { + PyVarObject_HEAD_INIT(NULL, 0) + "map_array_node", + sizeof(MapNode_Array), + 0, + .tp_dealloc = (destructor)map_node_array_dealloc, + .tp_getattro = PyObject_GenericGetAttr, + .tp_flags = Py_TPFLAGS_DEFAULT | Py_TPFLAGS_HAVE_GC, + .tp_traverse = (traverseproc)map_node_array_traverse, + .tp_free = PyObject_GC_Del, + .tp_hash = PyObject_HashNotImplemented, +}; + +PyTypeObject _Map_BitmapNode_Type = { + PyVarObject_HEAD_INIT(NULL, 0) + "map_bitmap_node", + sizeof(MapNode_Bitmap) - sizeof(PyObject *), + sizeof(PyObject *), + .tp_dealloc = (destructor)map_node_bitmap_dealloc, + .tp_getattro = PyObject_GenericGetAttr, + .tp_flags = Py_TPFLAGS_DEFAULT | Py_TPFLAGS_HAVE_GC, + .tp_traverse = (traverseproc)map_node_bitmap_traverse, + .tp_free = PyObject_GC_Del, + .tp_hash = PyObject_HashNotImplemented, +}; + +PyTypeObject _Map_CollisionNode_Type = { + PyVarObject_HEAD_INIT(NULL, 0) + "map_collision_node", + sizeof(MapNode_Collision) - sizeof(PyObject *), + sizeof(PyObject *), + .tp_dealloc = (destructor)map_node_collision_dealloc, + .tp_getattro = PyObject_GenericGetAttr, + .tp_flags = Py_TPFLAGS_DEFAULT | Py_TPFLAGS_HAVE_GC, + .tp_traverse = (traverseproc)map_node_collision_traverse, + .tp_free = PyObject_GC_Del, + .tp_hash = PyObject_HashNotImplemented, +}; + + +static void +module_free(void *m) +{ + Py_CLEAR(_empty_bitmap_node); +} + + +static struct PyModuleDef _mapmodule = { + PyModuleDef_HEAD_INIT, /* m_base */ + "_map", /* m_name */ + NULL, /* m_doc */ + -1, /* m_size */ + NULL, /* m_methods */ + NULL, /* m_slots */ + NULL, /* m_traverse */ + NULL, /* m_clear */ + module_free, /* m_free */ +}; + + +PyMODINIT_FUNC +PyInit__map(void) +{ + PyObject *m = PyModule_Create(&_mapmodule); + + if ((PyType_Ready(&_Map_Type) < 0) || + (PyType_Ready(&_MapMutation_Type) < 0) || + (PyType_Ready(&_Map_ArrayNode_Type) < 0) || + (PyType_Ready(&_Map_BitmapNode_Type) < 0) || + (PyType_Ready(&_Map_CollisionNode_Type) < 0) || + (PyType_Ready(&_MapKeys_Type) < 0) || + (PyType_Ready(&_MapValues_Type) < 0) || + (PyType_Ready(&_MapItems_Type) < 0) || + (PyType_Ready(&_MapKeysIter_Type) < 0) || + (PyType_Ready(&_MapValuesIter_Type) < 0) || + (PyType_Ready(&_MapItemsIter_Type) < 0)) + { + return 0; + } + + Py_INCREF(&_Map_Type); + if (PyModule_AddObject(m, "Map", (PyObject *)&_Map_Type) < 0) { + Py_DECREF(&_Map_Type); + return NULL; + } + + return m; +} diff --git a/immutables/_map.h b/immutables/_map.h new file mode 100644 index 0000000..28142ce --- /dev/null +++ b/immutables/_map.h @@ -0,0 +1,107 @@ +#ifndef IMMUTABLES_MAP_H +#define IMMUTABLES_MAP_H + +#include <stdint.h> +#include "Python.h" + +#define _Py_HAMT_MAX_TREE_DEPTH 8 + + +#define Map_Check(o) (Py_TYPE(o) == &_Map_Type) +#define MapMutation_Check(o) (Py_TYPE(o) == &_MapMutation_Type) + + +/* Abstract tree node. */ +typedef struct { + PyObject_HEAD +} MapNode; + + +#define _MapCommonFields(pref) \ + PyObject_HEAD \ + MapNode *pref##_root; \ + PyObject *pref##_weakreflist; \ + Py_ssize_t pref##_count; + + +/* Base mapping struct; used in methods shared between + MapObject and MapMutationObject types. */ +typedef struct { + _MapCommonFields(b) +} BaseMapObject; + + +/* An HAMT immutable mapping collection. */ +typedef struct { + _MapCommonFields(h) + Py_hash_t h_hash; +} MapObject; + + +/* MapMutation object (returned from `map.mutate()`.) */ +typedef struct { + _MapCommonFields(m) + uint64_t m_mutid; +} MapMutationObject; + + +/* A struct to hold the state of depth-first traverse of the tree. + + HAMT is an immutable collection. Iterators will hold a strong reference + to it, and every node in the HAMT has strong references to its children. + + So for iterators, we can implement zero allocations and zero reference + inc/dec depth-first iteration. + + - i_nodes: an array of seven pointers to tree nodes + - i_level: the current node in i_nodes + - i_pos: an array of positions within nodes in i_nodes. +*/ +typedef struct { + MapNode *i_nodes[_Py_HAMT_MAX_TREE_DEPTH]; + Py_ssize_t i_pos[_Py_HAMT_MAX_TREE_DEPTH]; + int8_t i_level; +} MapIteratorState; + + +/* Base iterator object. + + Contains the iteration state, a pointer to the HAMT tree, + and a pointer to the 'yield function'. The latter is a simple + function that returns a key/value tuple for the 'Items' iterator, + just a key for the 'Keys' iterator, and a value for the 'Values' + iterator. +*/ + +typedef struct { + PyObject_HEAD + MapObject *mv_obj; + binaryfunc mv_yield; + PyTypeObject *mv_itertype; +} MapView; + +typedef struct { + PyObject_HEAD + MapObject *mi_obj; + binaryfunc mi_yield; + MapIteratorState mi_iter; +} MapIterator; + + +/* PyTypes */ + + +PyTypeObject _Map_Type; +PyTypeObject _MapMutation_Type; +PyTypeObject _Map_ArrayNode_Type; +PyTypeObject _Map_BitmapNode_Type; +PyTypeObject _Map_CollisionNode_Type; +PyTypeObject _MapKeys_Type; +PyTypeObject _MapValues_Type; +PyTypeObject _MapItems_Type; +PyTypeObject _MapKeysIter_Type; +PyTypeObject _MapValuesIter_Type; +PyTypeObject _MapItemsIter_Type; + + +#endif diff --git a/immutables/_map.pyi b/immutables/_map.pyi new file mode 100644 index 0000000..4590cd4 --- /dev/null +++ b/immutables/_map.pyi @@ -0,0 +1,73 @@ +from typing import Any +from typing import Dict +from typing import Generic +from typing import Iterable +from typing import Iterator +from typing import Mapping +from typing import Optional +from typing import Tuple +from typing import Type +from typing import Union +from typing import overload + +from ._protocols import IterableItems +from ._protocols import MapItems +from ._protocols import MapKeys +from ._protocols import MapMutation +from ._protocols import MapValues +from ._protocols import HT +from ._protocols import KT +from ._protocols import T +from ._protocols import VT_co + + +class Map(Mapping[KT, VT_co]): + @overload + def __init__(self) -> None: ... + @overload + def __init__(self: Map[str, VT_co], **kw: VT_co) -> None: ... + @overload + def __init__( + self, __col: Union[IterableItems[KT, VT_co], Iterable[Tuple[KT, VT_co]]] + ) -> None: ... + @overload + def __init__( + self: Map[Union[KT, str], VT_co], + __col: Union[IterableItems[KT, VT_co], Iterable[Tuple[KT, VT_co]]], + **kw: VT_co + ) -> None: ... + def __reduce__(self) -> Tuple[Type[Map[KT, VT_co]], Tuple[Dict[KT, VT_co]]]: ... + def __len__(self) -> int: ... + def __eq__(self, other: Any) -> bool: ... + @overload + def update( + self, + __col: Union[IterableItems[KT, VT_co], Iterable[Tuple[KT, VT_co]]] + ) -> Map[KT, VT_co]: ... + @overload + def update( + self: Map[Union[HT, str], Any], + __col: Union[IterableItems[KT, VT_co], Iterable[Tuple[KT, VT_co]]], + **kw: VT_co # type: ignore[misc] + ) -> Map[KT, VT_co]: ... + @overload + def update( + self: Map[Union[HT, str], Any], + **kw: VT_co # type: ignore[misc] + ) -> Map[KT, VT_co]: ... + def mutate(self) -> MapMutation[KT, VT_co]: ... + def set(self, key: KT, val: VT_co) -> Map[KT, VT_co]: ... # type: ignore[misc] + def delete(self, key: KT) -> Map[KT, VT_co]: ... + @overload + def get(self, key: KT) -> Optional[VT_co]: ... + @overload + def get(self, key: KT, default: Union[VT_co, T]) -> Union[VT_co, T]: ... + def __getitem__(self, key: KT) -> VT_co: ... + def __contains__(self, key: Any) -> bool: ... + def __iter__(self) -> Iterator[KT]: ... + def keys(self) -> MapKeys[KT]: ... # type: ignore[override] + def values(self) -> MapValues[VT_co]: ... # type: ignore[override] + def items(self) -> MapItems[KT, VT_co]: ... # type: ignore[override] + def __hash__(self) -> int: ... + def __dump__(self) -> str: ... + def __class_getitem__(cls, item: Any) -> Type[Map[Any, Any]]: ... diff --git a/immutables/_protocols.py b/immutables/_protocols.py new file mode 100644 index 0000000..de87d23 --- /dev/null +++ b/immutables/_protocols.py @@ -0,0 +1,85 @@ +import sys +from typing import Any +from typing import Hashable +from typing import Iterable +from typing import Iterator +from typing import NoReturn +from typing import Optional +from typing import Tuple +from typing import TypeVar +from typing import Union +from typing import overload + +if sys.version_info >= (3, 8): + from typing import Protocol + from typing import TYPE_CHECKING +else: + from typing_extensions import Protocol + from typing_extensions import TYPE_CHECKING + +if TYPE_CHECKING: + from ._map import Map + +HT = TypeVar('HT', bound=Hashable) +KT = TypeVar('KT', bound=Hashable) +KT_co = TypeVar('KT_co', covariant=True) +MM = TypeVar('MM', bound='MapMutation[Any, Any]') +T = TypeVar('T') +VT = TypeVar('VT') +VT_co = TypeVar('VT_co', covariant=True) + + +class MapKeys(Protocol[KT_co]): + def __len__(self) -> int: ... + def __iter__(self) -> Iterator[KT_co]: ... + + +class MapValues(Protocol[VT_co]): + def __len__(self) -> int: ... + def __iter__(self) -> Iterator[VT_co]: ... + + +class MapItems(Protocol[KT_co, VT_co]): + def __len__(self) -> int: ... + def __iter__(self) -> Iterator[Tuple[KT_co, VT_co]]: ... + + +class IterableItems(Protocol[KT_co, VT_co]): + def items(self) -> Iterable[Tuple[KT_co, VT_co]]: ... + + +class MapMutation(Protocol[KT, VT]): + def set(self, key: KT, val: VT) -> None: ... + def __enter__(self: MM) -> MM: ... + def __exit__(self, *exc: Any) -> bool: ... + def __iter__(self) -> NoReturn: ... + def __delitem__(self, key: KT) -> None: ... + def __setitem__(self, key: KT, val: VT) -> None: ... + @overload + def pop(self, __key: KT) -> VT: ... + @overload + def pop(self, __key: KT, __default: T) -> Union[VT, T]: ... + @overload + def get(self, key: KT) -> Optional[VT]: ... + @overload + def get(self, key: KT, default: Union[VT, T]) -> Union[VT, T]: ... + def __getitem__(self, key: KT) -> VT: ... + def __contains__(self, key: object) -> bool: ... + + @overload + def update( + self, + __col: Union[IterableItems[KT, VT], Iterable[Tuple[KT, VT]]] + ) -> None: ... + + @overload + def update( + self: 'MapMutation[Union[HT, str], Any]', + __col: Union[IterableItems[KT, VT], Iterable[Tuple[KT, VT]]], + **kw: VT + ) -> None: ... + @overload + def update(self: 'MapMutation[Union[HT, str], Any]', **kw: VT) -> None: ... + def finish(self) -> 'Map[KT, VT]': ... + def __len__(self) -> int: ... + def __eq__(self, other: Any) -> bool: ... diff --git a/immutables/_testutils.py b/immutables/_testutils.py new file mode 100644 index 0000000..3f174b2 --- /dev/null +++ b/immutables/_testutils.py @@ -0,0 +1,80 @@ +class HashKey: + _crasher = None + + def __init__(self, hash, name, *, error_on_eq_to=None): + assert hash != -1 + self.name = name + self.hash = hash + self.error_on_eq_to = error_on_eq_to + + def __repr__(self): + if self._crasher is not None and self._crasher.error_on_repr: + raise ReprError + return '<Key name:{} hash:{}>'.format(self.name, self.hash) + + def __hash__(self): + if self._crasher is not None and self._crasher.error_on_hash: + raise HashingError + + return self.hash + + def __eq__(self, other): + if not isinstance(other, HashKey): + return NotImplemented + + if self._crasher is not None and self._crasher.error_on_eq: + raise EqError + + if self.error_on_eq_to is not None and self.error_on_eq_to is other: + raise ValueError('cannot compare {!r} to {!r}'.format(self, other)) + if other.error_on_eq_to is not None and other.error_on_eq_to is self: + raise ValueError('cannot compare {!r} to {!r}'.format(other, self)) + + return (self.name, self.hash) == (other.name, other.hash) + + +class KeyStr(str): + + def __hash__(self): + if HashKey._crasher is not None and HashKey._crasher.error_on_hash: + raise HashingError + return super().__hash__() + + def __eq__(self, other): + if HashKey._crasher is not None and HashKey._crasher.error_on_eq: + raise EqError + return super().__eq__(other) + + def __repr__(self, other): + if HashKey._crasher is not None and HashKey._crasher.error_on_repr: + raise ReprError + return super().__eq__(other) + + +class HashKeyCrasher: + + def __init__(self, *, error_on_hash=False, error_on_eq=False, + error_on_repr=False): + self.error_on_hash = error_on_hash + self.error_on_eq = error_on_eq + self.error_on_repr = error_on_repr + + def __enter__(self): + if HashKey._crasher is not None: + raise RuntimeError('cannot nest crashers') + HashKey._crasher = self + + def __exit__(self, *exc): + HashKey._crasher = None + + +class HashingError(Exception): + pass + + +class EqError(Exception): + pass + + +class ReprError(Exception): + pass diff --git a/immutables/_version.py b/immutables/_version.py new file mode 100644 index 0000000..b8cc76f --- /dev/null +++ b/immutables/_version.py @@ -0,0 +1,13 @@ +# This file MUST NOT contain anything but the __version__ assignment. +# +# When making a release, change the value of __version__ +# to an appropriate value, and open a pull request against +# the correct branch (master if making a new feature release). +# The commit message MUST contain a properly formatted release +# log, and the commit must be signed. +# +# The release automation will: build and test the packages for the +# supported platforms, publish the packages on PyPI, merge the PR +# to the target branch, create a Git tag pointing to the commit. + +__version__ = '0.18' diff --git a/immutables/map.py b/immutables/map.py new file mode 100644 index 0000000..0ad2858 --- /dev/null +++ b/immutables/map.py @@ -0,0 +1,855 @@ +import collections.abc +import itertools +import reprlib +import sys + + +__all__ = ('Map',) + + +# Thread-safe counter. +_mut_id = itertools.count(1).__next__ + + +# Python version of _map.c. The topmost comment there explains +# all datastructures and algorithms. +# The code here follows C code closely on purpose to make +# debugging and testing easier. + + +def map_hash(o): + x = hash(o) + if sys.hash_info.width > 32: + return (x & 0xffffffff) ^ ((x >> 32) & 0xffffffff) + else: + return x + + +def map_mask(hash, shift): + return (hash >> shift) & 0x01f + + +def map_bitpos(hash, shift): + return 1 << map_mask(hash, shift) + + +def map_bitcount(v): + v = v - ((v >> 1) & 0x55555555) + v = (v & 0x33333333) + ((v >> 2) & 0x33333333) + v = (v & 0x0F0F0F0F) + ((v >> 4) & 0x0F0F0F0F) + v = v + (v >> 8) + v = (v + (v >> 16)) & 0x3F + return v + + +def map_bitindex(bitmap, bit): + return map_bitcount(bitmap & (bit - 1)) + + +W_EMPTY, W_NEWNODE, W_NOT_FOUND = range(3) +void = object() + + +class _Unhashable: + __slots__ = () + __hash__ = None + + +_NULL = _Unhashable() +del _Unhashable + + +class BitmapNode: + + def __init__(self, size, bitmap, array, mutid): + self.size = size + self.bitmap = bitmap + assert isinstance(array, list) and len(array) == size + self.array = array + self.mutid = mutid + + def clone(self, mutid): + return BitmapNode(self.size, self.bitmap, self.array.copy(), mutid) + + def assoc(self, shift, hash, key, val, mutid): + bit = map_bitpos(hash, shift) + idx = map_bitindex(self.bitmap, bit) + + if self.bitmap & bit: + key_idx = 2 * idx + val_idx = key_idx + 1 + + key_or_null = self.array[key_idx] + val_or_node = self.array[val_idx] + + if key_or_null is _NULL: + sub_node, added = val_or_node.assoc( + shift + 5, hash, key, val, mutid) + if val_or_node is sub_node: + return self, added + + if mutid and mutid == self.mutid: + self.array[val_idx] = sub_node + return self, added + else: + ret = self.clone(mutid) + ret.array[val_idx] = sub_node + return ret, added + + if key == key_or_null: + if val is val_or_node: + return self, False + + if mutid and mutid == self.mutid: + self.array[val_idx] = val + return self, False + else: + ret = self.clone(mutid) + ret.array[val_idx] = val + return ret, False + + existing_key_hash = map_hash(key_or_null) + if existing_key_hash == hash: + sub_node = CollisionNode( + 4, hash, [key_or_null, val_or_node, key, val], mutid) + else: + sub_node = BitmapNode(0, 0, [], mutid) + sub_node, _ = sub_node.assoc( + shift + 5, existing_key_hash, + key_or_null, val_or_node, + mutid) + sub_node, _ = sub_node.assoc( + shift + 5, hash, key, val, + mutid) + + if mutid and mutid == self.mutid: + self.array[key_idx] = _NULL + self.array[val_idx] = sub_node + return self, True + else: + ret = self.clone(mutid) + ret.array[key_idx] = _NULL + ret.array[val_idx] = sub_node + return ret, True + + else: + key_idx = 2 * idx + val_idx = key_idx + 1 + + n = map_bitcount(self.bitmap) + + new_array = self.array[:key_idx] + new_array.append(key) + new_array.append(val) + new_array.extend(self.array[key_idx:]) + + if mutid and mutid == self.mutid: + self.size = 2 * (n + 1) + self.bitmap |= bit + self.array = new_array + return self, True + else: + return BitmapNode( + 2 * (n + 1), self.bitmap | bit, new_array, mutid), True + + def find(self, shift, hash, key): + bit = map_bitpos(hash, shift) + + if not (self.bitmap & bit): + raise KeyError + + idx = map_bitindex(self.bitmap, bit) + key_idx = idx * 2 + val_idx = key_idx + 1 + + key_or_null = self.array[key_idx] + val_or_node = self.array[val_idx] + + if key_or_null is _NULL: + return val_or_node.find(shift + 5, hash, key) + + if key == key_or_null: + return val_or_node + + raise KeyError(key) + + def without(self, shift, hash, key, mutid): + bit = map_bitpos(hash, shift) + if not (self.bitmap & bit): + return W_NOT_FOUND, None + + idx = map_bitindex(self.bitmap, bit) + key_idx = 2 * idx + val_idx = key_idx + 1 + + key_or_null = self.array[key_idx] + val_or_node = self.array[val_idx] + + if key_or_null is _NULL: + res, sub_node = val_or_node.without(shift + 5, hash, key, mutid) + + if res is W_EMPTY: + raise RuntimeError('unreachable code') # pragma: no cover + + elif res is W_NEWNODE: + if (type(sub_node) is BitmapNode and + sub_node.size == 2 and + sub_node.array[0] is not _NULL): + + if mutid and mutid == self.mutid: + self.array[key_idx] = sub_node.array[0] + self.array[val_idx] = sub_node.array[1] + return W_NEWNODE, self + else: + clone = self.clone(mutid) + clone.array[key_idx] = sub_node.array[0] + clone.array[val_idx] = sub_node.array[1] + return W_NEWNODE, clone + + if mutid and mutid == self.mutid: + self.array[val_idx] = sub_node + return W_NEWNODE, self + else: + clone = self.clone(mutid) + clone.array[val_idx] = sub_node + return W_NEWNODE, clone + + else: + assert sub_node is None + return res, None + + else: + if key == key_or_null: + if self.size == 2: + return W_EMPTY, None + + new_array = self.array[:key_idx] + new_array.extend(self.array[val_idx + 1:]) + + if mutid and mutid == self.mutid: + self.size -= 2 + self.bitmap &= ~bit + self.array = new_array + return W_NEWNODE, self + else: + new_node = BitmapNode( + self.size - 2, self.bitmap & ~bit, new_array, mutid) + return W_NEWNODE, new_node + + else: + return W_NOT_FOUND, None + + def keys(self): + for i in range(0, self.size, 2): + key_or_null = self.array[i] + + if key_or_null is _NULL: + val_or_node = self.array[i + 1] + yield from val_or_node.keys() + else: + yield key_or_null + + def values(self): + for i in range(0, self.size, 2): + key_or_null = self.array[i] + val_or_node = self.array[i + 1] + + if key_or_null is _NULL: + yield from val_or_node.values() + else: + yield val_or_node + + def items(self): + for i in range(0, self.size, 2): + key_or_null = self.array[i] + val_or_node = self.array[i + 1] + + if key_or_null is _NULL: + yield from val_or_node.items() + else: + yield key_or_null, val_or_node + + def dump(self, buf, level): # pragma: no cover + buf.append( + ' ' * (level + 1) + + 'BitmapNode(size={} count={} bitmap={} id={:0x}):'.format( + self.size, self.size / 2, bin(self.bitmap), id(self))) + + for i in range(0, self.size, 2): + key_or_null = self.array[i] + val_or_node = self.array[i + 1] + + pad = ' ' * (level + 2) + + if key_or_null is _NULL: + buf.append(pad + 'NULL:') + val_or_node.dump(buf, level + 2) + else: + buf.append(pad + '{!r}: {!r}'.format(key_or_null, val_or_node)) + + +class CollisionNode: + + def __init__(self, size, hash, array, mutid): + self.size = size + self.hash = hash + self.array = array + self.mutid = mutid + + def find_index(self, key): + for i in range(0, self.size, 2): + if self.array[i] == key: + return i + return -1 + + def find(self, shift, hash, key): + for i in range(0, self.size, 2): + if self.array[i] == key: + return self.array[i + 1] + raise KeyError(key) + + def assoc(self, shift, hash, key, val, mutid): + if hash == self.hash: + key_idx = self.find_index(key) + + if key_idx == -1: + new_array = self.array.copy() + new_array.append(key) + new_array.append(val) + + if mutid and mutid == self.mutid: + self.size += 2 + self.array = new_array + return self, True + else: + new_node = CollisionNode( + self.size + 2, hash, new_array, mutid) + return new_node, True + + val_idx = key_idx + 1 + if self.array[val_idx] is val: + return self, False + + if mutid and mutid == self.mutid: + self.array[val_idx] = val + return self, False + else: + new_array = self.array.copy() + new_array[val_idx] = val + return CollisionNode(self.size, hash, new_array, mutid), False + + else: + new_node = BitmapNode( + 2, map_bitpos(self.hash, shift), [_NULL, self], mutid) + return new_node.assoc(shift, hash, key, val, mutid) + + def without(self, shift, hash, key, mutid): + if hash != self.hash: + return W_NOT_FOUND, None + + key_idx = self.find_index(key) + if key_idx == -1: + return W_NOT_FOUND, None + + new_size = self.size - 2 + if new_size == 0: + # Shouldn't be ever reachable + return W_EMPTY, None # pragma: no cover + + if new_size == 2: + if key_idx == 0: + new_array = [self.array[2], self.array[3]] + else: + assert key_idx == 2 + new_array = [self.array[0], self.array[1]] + + new_node = BitmapNode( + 2, map_bitpos(hash, shift), new_array, mutid) + return W_NEWNODE, new_node + + new_array = self.array[:key_idx] + new_array.extend(self.array[key_idx + 2:]) + if mutid and mutid == self.mutid: + self.array = new_array + self.size -= 2 + return W_NEWNODE, self + else: + new_node = CollisionNode( + self.size - 2, self.hash, new_array, mutid) + return W_NEWNODE, new_node + + def keys(self): + for i in range(0, self.size, 2): + yield self.array[i] + + def values(self): + for i in range(1, self.size, 2): + yield self.array[i] + + def items(self): + for i in range(0, self.size, 2): + yield self.array[i], self.array[i + 1] + + def dump(self, buf, level): # pragma: no cover + pad = ' ' * (level + 1) + buf.append( + pad + 'CollisionNode(size={} id={:0x}):'.format( + self.size, id(self))) + + pad = ' ' * (level + 2) + for i in range(0, self.size, 2): + key = self.array[i] + val = self.array[i + 1] + + buf.append('{}{!r}: {!r}'.format(pad, key, val)) + + +class MapKeys: + + def __init__(self, c, m): + self.__count = c + self.__root = m + + def __len__(self): + return self.__count + + def __iter__(self): + return iter(self.__root.keys()) + + +class MapValues: + + def __init__(self, c, m): + self.__count = c + self.__root = m + + def __len__(self): + return self.__count + + def __iter__(self): + return iter(self.__root.values()) + + +class MapItems: + + def __init__(self, c, m): + self.__count = c + self.__root = m + + def __len__(self): + return self.__count + + def __iter__(self): + return iter(self.__root.items()) + + +class Map: + + def __init__(self, *args, **kw): + if not args: + col = None + elif len(args) == 1: + col = args[0] + else: + raise TypeError( + "immutables.Map expected at most 1 arguments, " + "got {}".format(len(args)) + ) + + self.__count = 0 + self.__root = BitmapNode(0, 0, [], 0) + self.__hash = -1 + + if isinstance(col, Map): + self.__count = col.__count + self.__root = col.__root + self.__hash = col.__hash + col = None + elif isinstance(col, MapMutation): + raise TypeError('cannot create Maps from MapMutations') + + if col or kw: + init = self.update(col, **kw) + self.__count = init.__count + self.__root = init.__root + + @classmethod + def _new(cls, count, root): + m = Map.__new__(Map) + m.__count = count + m.__root = root + m.__hash = -1 + return m + + def __reduce__(self): + return (type(self), (dict(self.items()),)) + + def __len__(self): + return self.__count + + def __eq__(self, other): + if not isinstance(other, Map): + return NotImplemented + + if len(self) != len(other): + return False + + for key, val in self.__root.items(): + try: + oval = other.__root.find(0, map_hash(key), key) + except KeyError: + return False + else: + if oval != val: + return False + + return True + + def update(self, *args, **kw): + if not args: + col = None + elif len(args) == 1: + col = args[0] + else: + raise TypeError( + "update expected at most 1 arguments, got {}".format(len(args)) + ) + + it = None + + if col is not None: + if hasattr(col, 'items'): + it = iter(col.items()) + else: + it = iter(col) + + if it is not None: + if kw: + it = iter(itertools.chain(it, kw.items())) + else: + if kw: + it = iter(kw.items()) + + if it is None: + + return self + + mutid = _mut_id() + root = self.__root + count = self.__count + + i = 0 + while True: + try: + tup = next(it) + except StopIteration: + break + + try: + tup = tuple(tup) + except TypeError: + raise TypeError( + 'cannot convert map update ' + 'sequence element #{} to a sequence'.format(i)) from None + key, val, *r = tup + if r: + raise ValueError( + 'map update sequence element #{} has length ' + '{}; 2 is required'.format(i, len(r) + 2)) + + root, added = root.assoc(0, map_hash(key), key, val, mutid) + if added: + count += 1 + + i += 1 + + return Map._new(count, root) + + def mutate(self): + return MapMutation(self.__count, self.__root) + + def set(self, key, val): + new_count = self.__count + new_root, added = self.__root.assoc(0, map_hash(key), key, val, 0) + + if new_root is self.__root: + assert not added + return self + + if added: + new_count += 1 + + return Map._new(new_count, new_root) + + def delete(self, key): + res, node = self.__root.without(0, map_hash(key), key, 0) + if res is W_EMPTY: + return Map() + elif res is W_NOT_FOUND: + raise KeyError(key) + else: + return Map._new(self.__count - 1, node) + + def get(self, key, default=None): + try: + return self.__root.find(0, map_hash(key), key) + except KeyError: + return default + + def __getitem__(self, key): + return self.__root.find(0, map_hash(key), key) + + def __contains__(self, key): + try: + self.__root.find(0, map_hash(key), key) + except KeyError: + return False + else: + return True + + def __iter__(self): + yield from self.__root.keys() + + def keys(self): + return MapKeys(self.__count, self.__root) + + def values(self): + return MapValues(self.__count, self.__root) + + def items(self): + return MapItems(self.__count, self.__root) + + def __hash__(self): + if self.__hash != -1: + return self.__hash + + MAX = sys.maxsize + MASK = 2 * MAX + 1 + + h = 1927868237 * (self.__count * 2 + 1) + h &= MASK + + for key, value in self.__root.items(): + hx = hash(key) + h ^= (hx ^ (hx << 16) ^ 89869747) * 3644798167 + h &= MASK + + hx = hash(value) + h ^= (hx ^ (hx << 16) ^ 89869747) * 3644798167 + h &= MASK + + h = h * 69069 + 907133923 + h &= MASK + + if h > MAX: + h -= MASK + 1 # pragma: no cover + if h == -1: + h = 590923713 # pragma: no cover + + self.__hash = h + return h + + @reprlib.recursive_repr("{...}") + def __repr__(self): + items = [] + for key, val in self.items(): + items.append("{!r}: {!r}".format(key, val)) + return 'immutables.Map({{{}}})'.format(', '.join(items)) + + def __dump__(self): # pragma: no cover + buf = [] + self.__root.dump(buf, 0) + return '\n'.join(buf) + + def __class_getitem__(cls, item): + return cls + + +class MapMutation: + + def __init__(self, count, root): + self.__count = count + self.__root = root + self.__mutid = _mut_id() + + def set(self, key, val): + self[key] = val + + def __enter__(self): + return self + + def __exit__(self, *exc): + self.finish() + return False + + def __iter__(self): + raise TypeError('{} is not iterable'.format(type(self))) + + def __delitem__(self, key): + if self.__mutid == 0: + raise ValueError('mutation {!r} has been finished'.format(self)) + + res, new_root = self.__root.without( + 0, map_hash(key), key, self.__mutid) + if res is W_EMPTY: + self.__count = 0 + self.__root = BitmapNode(0, 0, [], self.__mutid) + elif res is W_NOT_FOUND: + raise KeyError(key) + else: + self.__root = new_root + self.__count -= 1 + + def __setitem__(self, key, val): + if self.__mutid == 0: + raise ValueError('mutation {!r} has been finished'.format(self)) + + self.__root, added = self.__root.assoc( + 0, map_hash(key), key, val, self.__mutid) + + if added: + self.__count += 1 + + def pop(self, key, *args): + if self.__mutid == 0: + raise ValueError('mutation {!r} has been finished'.format(self)) + + if len(args) > 1: + raise TypeError( + 'pop() accepts 1 to 2 positional arguments, ' + 'got {}'.format(len(args) + 1)) + elif len(args) == 1: + default = args[0] + else: + default = void + + val = self.get(key, default) + + try: + del self[key] + except KeyError: + if val is void: + raise + return val + else: + assert val is not void + return val + + def get(self, key, default=None): + try: + return self.__root.find(0, map_hash(key), key) + except KeyError: + return default + + def __getitem__(self, key): + return self.__root.find(0, map_hash(key), key) + + def __contains__(self, key): + try: + self.__root.find(0, map_hash(key), key) + except KeyError: + return False + else: + return True + + def update(self, *args, **kw): + if not args: + col = None + elif len(args) == 1: + col = args[0] + else: + raise TypeError( + "update expected at most 1 arguments, got {}".format(len(args)) + ) + + if self.__mutid == 0: + raise ValueError('mutation {!r} has been finished'.format(self)) + + it = None + if col is not None: + if hasattr(col, 'items'): + it = iter(col.items()) + else: + it = iter(col) + + if it is not None: + if kw: + it = iter(itertools.chain(it, kw.items())) + else: + if kw: + it = iter(kw.items()) + + if it is None: + return + + root = self.__root + count = self.__count + + i = 0 + while True: + try: + tup = next(it) + except StopIteration: + break + + try: + tup = tuple(tup) + except TypeError: + raise TypeError( + 'cannot convert map update ' + 'sequence element #{} to a sequence'.format(i)) from None + key, val, *r = tup + if r: + raise ValueError( + 'map update sequence element #{} has length ' + '{}; 2 is required'.format(i, len(r) + 2)) + + root, added = root.assoc(0, map_hash(key), key, val, self.__mutid) + if added: + count += 1 + + i += 1 + + self.__root = root + self.__count = count + + def finish(self): + self.__mutid = 0 + return Map._new(self.__count, self.__root) + + @reprlib.recursive_repr("{...}") + def __repr__(self): + items = [] + for key, val in self.__root.items(): + items.append("{!r}: {!r}".format(key, val)) + return 'immutables.MapMutation({{{}}})'.format(', '.join(items)) + + def __len__(self): + return self.__count + + def __reduce__(self): + raise TypeError("can't pickle {} objects".format(type(self).__name__)) + + def __hash__(self): + raise TypeError('unhashable type: {}'.format(type(self).__name__)) + + def __eq__(self, other): + if not isinstance(other, MapMutation): + return NotImplemented + + if len(self) != len(other): + return False + + for key, val in self.__root.items(): + try: + oval = other.__root.find(0, map_hash(key), key) + except KeyError: + return False + else: + if oval != val: + return False + + return True + + +collections.abc.Mapping.register(Map) diff --git a/immutables/py.typed b/immutables/py.typed new file mode 100644 index 0000000..b648ac9 --- /dev/null +++ b/immutables/py.typed @@ -0,0 +1 @@ +partial diff --git a/immutables/pythoncapi_compat.h b/immutables/pythoncapi_compat.h new file mode 100644 index 0000000..2837467 --- /dev/null +++ b/immutables/pythoncapi_compat.h @@ -0,0 +1,446 @@ +// Header file providing new C API functions to old Python versions. +// +// File distributed under the Zero Clause BSD (0BSD) license. +// Copyright Contributors to the pythoncapi_compat project. +// +// Homepage: +// https://github.com/python/pythoncapi_compat +// +// Latest version: +// https://raw.githubusercontent.com/python/pythoncapi_compat/master/pythoncapi_compat.h +// +// SPDX-License-Identifier: 0BSD + +#ifndef PYTHONCAPI_COMPAT +#define PYTHONCAPI_COMPAT + +#ifdef __cplusplus +extern "C" { +#endif + +#include <Python.h> +#include "frameobject.h" // PyFrameObject, PyFrame_GetBack() + + +// Compatibility with Visual Studio 2013 and older which don't support +// the inline keyword in C (only in C++): use __inline instead. +#if (defined(_MSC_VER) && _MSC_VER < 1900 \ + && !defined(__cplusplus) && !defined(inline)) +# define PYCAPI_COMPAT_INLINE(TYPE static __inline TYPE +#else +# define PYCAPI_COMPAT_STATIC_INLINE(TYPE) static inline TYPE +#endif + + +// C++ compatibility +#ifdef __cplusplus +# define PYCAPI_COMPAT_CAST(TYPE, EXPR) reinterpret_cast<TYPE>(EXPR) +# define PYCAPI_COMPAT_NULL nullptr +#else +# define PYCAPI_COMPAT_CAST(TYPE, EXPR) ((TYPE)(EXPR)) +# define PYCAPI_COMPAT_NULL NULL +#endif + +// Cast argument to PyObject* type. +#ifndef _PyObject_CAST +# define _PyObject_CAST(op) PYCAPI_COMPAT_CAST(PyObject*, op) +#endif +#ifndef _PyObject_CAST_CONST +# define _PyObject_CAST_CONST(op) PYCAPI_COMPAT_CAST(const PyObject*, op) +#endif + + +// bpo-42262 added Py_NewRef() to Python 3.10.0a3 +#if PY_VERSION_HEX < 0x030A00A3 && !defined(Py_NewRef) +PYCAPI_COMPAT_STATIC_INLINE(PyObject*) +_Py_NewRef(PyObject *obj) +{ + Py_INCREF(obj); + return obj; +} +#define Py_NewRef(obj) _Py_NewRef(_PyObject_CAST(obj)) +#endif + + +// bpo-42262 added Py_XNewRef() to Python 3.10.0a3 +#if PY_VERSION_HEX < 0x030A00A3 && !defined(Py_XNewRef) +PYCAPI_COMPAT_STATIC_INLINE(PyObject*) +_Py_XNewRef(PyObject *obj) +{ + Py_XINCREF(obj); + return obj; +} +#define Py_XNewRef(obj) _Py_XNewRef(_PyObject_CAST(obj)) +#endif + + +// See https://bugs.python.org/issue42522 +#if !defined(_Py_StealRef) +PYCAPI_COMPAT_STATIC_INLINE(PyObject*) +__Py_StealRef(PyObject *obj) +{ + Py_DECREF(obj); + return obj; +} +#define _Py_StealRef(obj) __Py_StealRef(_PyObject_CAST(obj)) +#endif + + +// See https://bugs.python.org/issue42522 +#if !defined(_Py_XStealRef) +PYCAPI_COMPAT_STATIC_INLINE(PyObject*) +__Py_XStealRef(PyObject *obj) +{ + Py_XDECREF(obj); + return obj; +} +#define _Py_XStealRef(obj) __Py_XStealRef(_PyObject_CAST(obj)) +#endif + + +// bpo-39573 added Py_SET_REFCNT() to Python 3.9.0a4 +#if PY_VERSION_HEX < 0x030900A4 && !defined(Py_SET_REFCNT) +PYCAPI_COMPAT_STATIC_INLINE(void) +_Py_SET_REFCNT(PyObject *ob, Py_ssize_t refcnt) +{ + ob->ob_refcnt = refcnt; +} +#define Py_SET_REFCNT(ob, refcnt) _Py_SET_REFCNT(_PyObject_CAST(ob), refcnt) +#endif + + +// Py_SETREF() and Py_XSETREF() were added to Python 3.5.2. +// It is excluded from the limited C API. +#if (PY_VERSION_HEX < 0x03050200 && !defined(Py_SETREF)) && !defined(Py_LIMITED_API) +#define Py_SETREF(op, op2) \ + do { \ + PyObject *_py_tmp = _PyObject_CAST(op); \ + (op) = (op2); \ + Py_DECREF(_py_tmp); \ + } while (0) + +#define Py_XSETREF(op, op2) \ + do { \ + PyObject *_py_tmp = _PyObject_CAST(op); \ + (op) = (op2); \ + Py_XDECREF(_py_tmp); \ + } while (0) +#endif + + +// bpo-43753 added Py_Is(), Py_IsNone(), Py_IsTrue() and Py_IsFalse() +// to Python 3.10.0b1. +#if PY_VERSION_HEX < 0x030A00B1 && !defined(Py_Is) +# define Py_Is(x, y) ((x) == (y)) +#endif +#if PY_VERSION_HEX < 0x030A00B1 && !defined(Py_IsNone) +# define Py_IsNone(x) Py_Is(x, Py_None) +#endif +#if PY_VERSION_HEX < 0x030A00B1 && !defined(Py_IsTrue) +# define Py_IsTrue(x) Py_Is(x, Py_True) +#endif +#if PY_VERSION_HEX < 0x030A00B1 && !defined(Py_IsFalse) +# define Py_IsFalse(x) Py_Is(x, Py_False) +#endif + + +// bpo-39573 added Py_SET_TYPE() to Python 3.9.0a4 +#if PY_VERSION_HEX < 0x030900A4 && !defined(Py_SET_TYPE) +PYCAPI_COMPAT_STATIC_INLINE(void) +_Py_SET_TYPE(PyObject *ob, PyTypeObject *type) +{ + ob->ob_type = type; +} +#define Py_SET_TYPE(ob, type) _Py_SET_TYPE(_PyObject_CAST(ob), type) +#endif + + +// bpo-39573 added Py_SET_SIZE() to Python 3.9.0a4 +#if PY_VERSION_HEX < 0x030900A4 && !defined(Py_SET_SIZE) +PYCAPI_COMPAT_STATIC_INLINE(void) +_Py_SET_SIZE(PyVarObject *ob, Py_ssize_t size) +{ + ob->ob_size = size; +} +#define Py_SET_SIZE(ob, size) _Py_SET_SIZE((PyVarObject*)(ob), size) +#endif + + +// bpo-40421 added PyFrame_GetCode() to Python 3.9.0b1 +#if PY_VERSION_HEX < 0x030900B1 +PYCAPI_COMPAT_STATIC_INLINE(PyCodeObject*) +PyFrame_GetCode(PyFrameObject *frame) +{ + assert(frame != PYCAPI_COMPAT_NULL); + assert(frame->f_code != PYCAPI_COMPAT_NULL); + return PYCAPI_COMPAT_CAST(PyCodeObject*, Py_NewRef(frame->f_code)); +} +#endif + +PYCAPI_COMPAT_STATIC_INLINE(PyCodeObject*) +_PyFrame_GetCodeBorrow(PyFrameObject *frame) +{ + return PYCAPI_COMPAT_CAST(PyCodeObject *, + _Py_StealRef(PyFrame_GetCode(frame))); +} + + +// bpo-40421 added PyFrame_GetCode() to Python 3.9.0b1 +#if PY_VERSION_HEX < 0x030900B1 && !defined(PYPY_VERSION) +PYCAPI_COMPAT_STATIC_INLINE(PyFrameObject*) +PyFrame_GetBack(PyFrameObject *frame) +{ + assert(frame != PYCAPI_COMPAT_NULL); + return PYCAPI_COMPAT_CAST(PyFrameObject*, Py_XNewRef(frame->f_back)); +} +#endif + +#if !defined(PYPY_VERSION) +PYCAPI_COMPAT_STATIC_INLINE(PyFrameObject*) +_PyFrame_GetBackBorrow(PyFrameObject *frame) +{ + return PYCAPI_COMPAT_CAST(PyFrameObject *, + _Py_XStealRef(PyFrame_GetBack(frame))); +} +#endif + + +// bpo-39947 added PyThreadState_GetInterpreter() to Python 3.9.0a5 +#if PY_VERSION_HEX < 0x030900A5 +PYCAPI_COMPAT_STATIC_INLINE(PyInterpreterState *) +PyThreadState_GetInterpreter(PyThreadState *tstate) +{ + assert(tstate != PYCAPI_COMPAT_NULL); + return tstate->interp; +} +#endif + + +// bpo-40429 added PyThreadState_GetFrame() to Python 3.9.0b1 +#if PY_VERSION_HEX < 0x030900B1 && !defined(PYPY_VERSION) +PYCAPI_COMPAT_STATIC_INLINE(PyFrameObject*) +PyThreadState_GetFrame(PyThreadState *tstate) +{ + assert(tstate != PYCAPI_COMPAT_NULL); + return PYCAPI_COMPAT_CAST(PyFrameObject *, Py_XNewRef(tstate->frame)); +} +#endif + +#if !defined(PYPY_VERSION) +PYCAPI_COMPAT_STATIC_INLINE(PyFrameObject*) +_PyThreadState_GetFrameBorrow(PyThreadState *tstate) +{ + return PYCAPI_COMPAT_CAST(PyFrameObject*, + _Py_XStealRef(PyThreadState_GetFrame(tstate))); +} +#endif + + +// bpo-39947 added PyInterpreterState_Get() to Python 3.9.0a5 +#if PY_VERSION_HEX < 0x030900A5 +PYCAPI_COMPAT_STATIC_INLINE(PyInterpreterState*) +PyInterpreterState_Get(void) +{ + PyThreadState *tstate; + PyInterpreterState *interp; + + tstate = PyThreadState_GET(); + if (tstate == PYCAPI_COMPAT_NULL) { + Py_FatalError("GIL released (tstate is NULL)"); + } + interp = tstate->interp; + if (interp == PYCAPI_COMPAT_NULL) { + Py_FatalError("no current interpreter"); + } + return interp; +} +#endif + + +// bpo-39947 added PyInterpreterState_Get() to Python 3.9.0a6 +#if 0x030700A1 <= PY_VERSION_HEX && PY_VERSION_HEX < 0x030900A6 && !defined(PYPY_VERSION) +PYCAPI_COMPAT_STATIC_INLINE(uint64_t) +PyThreadState_GetID(PyThreadState *tstate) +{ + assert(tstate != PYCAPI_COMPAT_NULL); + return tstate->id; +} +#endif + +// bpo-43760 added PyThreadState_EnterTracing() to Python 3.11.0a2 +#if PY_VERSION_HEX < 0x030B00A2 && !defined(PYPY_VERSION) +PYCAPI_COMPAT_STATIC_INLINE(void) +PyThreadState_EnterTracing(PyThreadState *tstate) +{ + tstate->tracing++; +#if PY_VERSION_HEX >= 0x030A00A1 + tstate->cframe->use_tracing = 0; +#else + tstate->use_tracing = 0; +#endif +} +#endif + +// bpo-43760 added PyThreadState_LeaveTracing() to Python 3.11.0a2 +#if PY_VERSION_HEX < 0x030B00A2 && !defined(PYPY_VERSION) +PYCAPI_COMPAT_STATIC_INLINE(void) +PyThreadState_LeaveTracing(PyThreadState *tstate) +{ + int use_tracing = (tstate->c_tracefunc != PYCAPI_COMPAT_NULL + || tstate->c_profilefunc != PYCAPI_COMPAT_NULL); + tstate->tracing--; +#if PY_VERSION_HEX >= 0x030A00A1 + tstate->cframe->use_tracing = use_tracing; +#else + tstate->use_tracing = use_tracing; +#endif +} +#endif + + +// bpo-37194 added PyObject_CallNoArgs() to Python 3.9.0a1 +#if PY_VERSION_HEX < 0x030900A1 +PYCAPI_COMPAT_STATIC_INLINE(PyObject*) +PyObject_CallNoArgs(PyObject *func) +{ + return PyObject_CallFunctionObjArgs(func, NULL); +} +#endif + + +// bpo-39245 made PyObject_CallOneArg() public (previously called +// _PyObject_CallOneArg) in Python 3.9.0a4 +#if PY_VERSION_HEX < 0x030900A4 +PYCAPI_COMPAT_STATIC_INLINE(PyObject*) +PyObject_CallOneArg(PyObject *func, PyObject *arg) +{ + return PyObject_CallFunctionObjArgs(func, arg, NULL); +} +#endif + + +// bpo-1635741 added PyModule_AddObjectRef() to Python 3.10.0a3 +#if PY_VERSION_HEX < 0x030A00A3 +PYCAPI_COMPAT_STATIC_INLINE(int) +PyModule_AddObjectRef(PyObject *mod, const char *name, PyObject *value) +{ + int res; + Py_XINCREF(value); + res = PyModule_AddObject(mod, name, value); + if (res < 0) { + Py_XDECREF(value); + } + return res; +} +#endif + + +// bpo-40024 added PyModule_AddType() to Python 3.9.0a5 +#if PY_VERSION_HEX < 0x030900A5 +PYCAPI_COMPAT_STATIC_INLINE(int) +PyModule_AddType(PyObject *mod, PyTypeObject *type) +{ + const char *name, *dot; + + if (PyType_Ready(type) < 0) { + return -1; + } + + // inline _PyType_Name() + name = type->tp_name; + assert(name != PYCAPI_COMPAT_NULL); + dot = strrchr(name, '.'); + if (dot != PYCAPI_COMPAT_NULL) { + name = dot + 1; + } + + return PyModule_AddObjectRef(mod, name, _PyObject_CAST(type)); +} +#endif + + +// bpo-40241 added PyObject_GC_IsTracked() to Python 3.9.0a6. +// bpo-4688 added _PyObject_GC_IS_TRACKED() to Python 2.7.0a2. +#if PY_VERSION_HEX < 0x030900A6 && !defined(PYPY_VERSION) +PYCAPI_COMPAT_STATIC_INLINE(int) +PyObject_GC_IsTracked(PyObject* obj) +{ + return (PyObject_IS_GC(obj) && _PyObject_GC_IS_TRACKED(obj)); +} +#endif + +// bpo-40241 added PyObject_GC_IsFinalized() to Python 3.9.0a6. +// bpo-18112 added _PyGCHead_FINALIZED() to Python 3.4.0 final. +#if PY_VERSION_HEX < 0x030900A6 && PY_VERSION_HEX >= 0x030400F0 && !defined(PYPY_VERSION) +PYCAPI_COMPAT_STATIC_INLINE(int) +PyObject_GC_IsFinalized(PyObject *obj) +{ + PyGC_Head *gc = PYCAPI_COMPAT_CAST(PyGC_Head *, obj) - 1; + return (PyObject_IS_GC(obj) && _PyGCHead_FINALIZED(gc)); +} +#endif + + +// bpo-39573 added Py_IS_TYPE() to Python 3.9.0a4 +#if PY_VERSION_HEX < 0x030900A4 && !defined(Py_IS_TYPE) +PYCAPI_COMPAT_STATIC_INLINE(int) +_Py_IS_TYPE(const PyObject *ob, const PyTypeObject *type) { + return ob->ob_type == type; +} +#define Py_IS_TYPE(ob, type) _Py_IS_TYPE(_PyObject_CAST_CONST(ob), type) +#endif + + +// bpo-46906 added PyFloat_Pack2() and PyFloat_Unpack2() to Python 3.11a7. +// bpo-11734 added _PyFloat_Pack2() and _PyFloat_Unpack2() to Python 3.6.0b1. +// Python 3.11a2 moved _PyFloat_Pack2() and _PyFloat_Unpack2() to the internal +// C API: Python 3.11a2-3.11a6 versions are not supported. +#if 0x030600B1 <= PY_VERSION_HEX && PY_VERSION_HEX <= 0x030B00A1 && !defined(PYPY_VERSION) +PYCAPI_COMPAT_STATIC_INLINE(int) +PyFloat_Pack2(double x, char *p, int le) +{ return _PyFloat_Pack2(x, (unsigned char*)p, le); } + +PYCAPI_COMPAT_STATIC_INLINE(double) +PyFloat_Unpack2(const char *p, int le) +{ return _PyFloat_Unpack2((const unsigned char *)p, le); } +#endif + + +// bpo-46906 added PyFloat_Pack4(), PyFloat_Pack8(), PyFloat_Unpack4() and +// PyFloat_Unpack8() to Python 3.11a7. +// Python 3.11a2 moved _PyFloat_Pack4(), _PyFloat_Pack8(), _PyFloat_Unpack4() +// and _PyFloat_Unpack8() to the internal C API: Python 3.11a2-3.11a6 versions +// are not supported. +#if PY_VERSION_HEX <= 0x030B00A1 && !defined(PYPY_VERSION) +PYCAPI_COMPAT_STATIC_INLINE(int) +PyFloat_Pack4(double x, char *p, int le) +{ return _PyFloat_Pack4(x, (unsigned char*)p, le); } + +PYCAPI_COMPAT_STATIC_INLINE(int) +PyFloat_Pack8(double x, char *p, int le) +{ return _PyFloat_Pack8(x, (unsigned char*)p, le); } + +PYCAPI_COMPAT_STATIC_INLINE(double) +PyFloat_Unpack4(const char *p, int le) +{ return _PyFloat_Unpack4((const unsigned char *)p, le); } + +PYCAPI_COMPAT_STATIC_INLINE(double) +PyFloat_Unpack8(const char *p, int le) +{ return _PyFloat_Unpack8((const unsigned char *)p, le); } +#endif + + +// Py_UNUSED() was added to Python 3.4.0b2. +#if PY_VERSION_HEX < 0x030400B2 && !defined(Py_UNUSED) +# if defined(__GNUC__) || defined(__clang__) +# define Py_UNUSED(name) _unused_ ## name __attribute__((unused)) +# else +# define Py_UNUSED(name) _unused_ ## name +# endif +#endif + + +#ifdef __cplusplus +} +#endif +#endif // PYTHONCAPI_COMPAT diff --git a/pyproject.toml b/pyproject.toml new file mode 100644 index 0000000..e9018e7 --- /dev/null +++ b/pyproject.toml @@ -0,0 +1,25 @@ +[project] +name = 'immutables' +requires-python = ">=3.6" + +[build-system] +requires = ["setuptools>=42", "wheel"] +build-backend = "setuptools.build_meta" + +[tool.pytest.ini_options] +minversion = "6.0" +addopts = "--capture=no --assert=plain --strict-markers --tb=native --import-mode=importlib" +testpaths = "tests" +filterwarnings = "default" + +[tool.mypy] +incremental = true +strict = true + +[[tool.mypy.overrides]] +module = "immutables.map" +ignore_errors = true + +[[tool.mypy.overrides]] +module = "immutables._testutils" +ignore_errors = true diff --git a/setup.py b/setup.py new file mode 100644 index 0000000..09b8de4 --- /dev/null +++ b/setup.py @@ -0,0 +1,93 @@ +import os +import platform +import setuptools + + +# Minimal dependencies required to test immutables. +TEST_DEPENDENCIES = [ + # pycodestyle is a dependency of flake8, but it must be frozen because + # their combination breaks too often + # (example breakage: https://gitlab.com/pycqa/flake8/issues/427) + 'flake8~=3.8.4', + 'pycodestyle~=2.6.0', + 'mypy==0.942', + 'pytest~=6.2.4', +] + +EXTRA_DEPENDENCIES = { + 'test': TEST_DEPENDENCIES, +} + +CFLAGS = ['-O2'] +if platform.uname().system != 'Windows': + CFLAGS.extend(['-std=c99', '-fsigned-char', '-Wall', + '-Wsign-compare', '-Wconversion']) + + +with open(os.path.join( + os.path.dirname(__file__), 'immutables', '_version.py')) as f: + for line in f: + if line.startswith('__version__ ='): + _, _, version = line.partition('=') + VERSION = version.strip(" \n'\"") + break + else: + raise RuntimeError( + 'unable to read the version from immutables/_version.py') + + +if platform.python_implementation() == 'CPython': + if os.environ.get("DEBUG_IMMUTABLES") == '1': + define_macros = [] + undef_macros = ['NDEBUG'] + else: + define_macros = [('NDEBUG', '1')] + undef_macros = [] + + ext_modules = [ + setuptools.Extension( + "immutables._map", + ["immutables/_map.c"], + extra_compile_args=CFLAGS, + define_macros=define_macros, + undef_macros=undef_macros) + ] +else: + ext_modules = [] + + +with open(os.path.join(os.path.dirname(__file__), 'README.rst')) as f: + readme = f.read() + + +setuptools.setup( + name='immutables', + version=VERSION, + description='Immutable Collections', + long_description=readme, + python_requires='>=3.6', + classifiers=[ + 'License :: OSI Approved :: Apache Software License', + 'Intended Audience :: Developers', + 'Programming Language :: Python :: 3 :: Only', + 'Programming Language :: Python :: 3.6', + 'Programming Language :: Python :: 3.7', + 'Programming Language :: Python :: 3.8', + 'Programming Language :: Python :: 3.9', + 'Programming Language :: Python :: 3.10', + 'Operating System :: POSIX', + 'Operating System :: MacOS :: MacOS X', + 'Operating System :: Microsoft :: Windows', + ], + author='MagicStack Inc', + author_email='hello@magic.io', + url='https://github.com/MagicStack/immutables', + license='Apache License, Version 2.0', + packages=['immutables'], + package_data={"immutables": ["py.typed", "*.pyi"]}, + provides=['immutables'], + include_package_data=True, + ext_modules=ext_modules, + install_requires=['typing-extensions>=3.7.4.3;python_version<"3.8"'], + extras_require=EXTRA_DEPENDENCIES, +) diff --git a/tests/__init__.py b/tests/__init__.py new file mode 100644 index 0000000..450c573 --- /dev/null +++ b/tests/__init__.py @@ -0,0 +1,16 @@ +import os.path +import sys +import unittest + + +def suite(): + test_loader = unittest.TestLoader() + test_suite = test_loader.discover(os.path.dirname(__file__), + pattern='test_*.py') + return test_suite + + +if __name__ == '__main__': + runner = unittest.runner.TextTestRunner() + result = runner.run(suite()) + sys.exit(not result.wasSuccessful()) diff --git a/tests/conftest.py b/tests/conftest.py new file mode 100644 index 0000000..9eb9e61 --- /dev/null +++ b/tests/conftest.py @@ -0,0 +1,13 @@ +# We need the mypy pytest plugin to do the test collection for our +# typing tests. + +# mypy demands that its test-data be present for mypy.test.config to be +# imported, so thwart that check. mypy PR #10919 fixes this. +import unittest.mock +with unittest.mock.patch('os.path.isdir') as isdir: + isdir.return_value = True + import mypy.test.config # noqa + +pytest_plugins = [ + 'mypy.test.data', +] diff --git a/tests/test-data/check-immu.test b/tests/test-data/check-immu.test new file mode 100644 index 0000000..4998542 --- /dev/null +++ b/tests/test-data/check-immu.test @@ -0,0 +1,73 @@ +[case testMypyImmu] +# cmd: mypy test.py +[file test.py] +from immutables import Map +from typing import Dict, Union, Any, cast + +def init() -> None: + def thing(m: Map[str, Union[str, int]]) -> None: + ... + + thing(Map(foo=1)) + thing(Map(foo='bar', baz=1)) + thing(Map([('foo', 'bar'), ('bar', 1)])) + thing(Map(Map(foo=1), bar='foo')) + m = Map({1: 2}) + thing(m) # E: Argument 1 to "thing" has incompatible type "Map[int, int]"; expected "Map[str, Union[str, int]]" + +def assignments() -> None: + m_int__str = Map[int, str]() + m_str__str = Map[str, str]() + m_int_str__str = Map[Union[int, str], str]() + m_str__int_str = Map[str, Union[int, str]]() + + m_int__str = m_str__str # E: Incompatible types in assignment (expression has type "Map[str, str]", variable has type "Map[int, str]") + m_int__str = m_int_str__str # E: Incompatible types in assignment (expression has type "Map[Union[int, str], str]", variable has type "Map[int, str]") + m_int__str = m_str__int_str # E: Incompatible types in assignment (expression has type "Map[str, Union[int, str]]", variable has type "Map[int, str]") + + m_str__str = m_int__str # E: Incompatible types in assignment (expression has type "Map[int, str]", variable has type "Map[str, str]") + m_str__str = m_int_str__str # E: Incompatible types in assignment (expression has type "Map[Union[int, str], str]", variable has type "Map[str, str]") + m_str__str = m_str__int_str # E: Incompatible types in assignment (expression has type "Map[str, Union[int, str]]", variable has type "Map[str, str]") + + m_int_str__str = m_int__str # E: Incompatible types in assignment (expression has type "Map[int, str]", variable has type "Map[Union[int, str], str]") + m_int_str__str = m_str__str # E: Incompatible types in assignment (expression has type "Map[str, str]", variable has type "Map[Union[int, str], str]") + m_int_str__str = m_str__int_str # E: Incompatible types in assignment (expression has type "Map[str, Union[int, str]]", variable has type "Map[Union[int, str], str]") + + m_str__int_str = m_int__str # E: Incompatible types in assignment (expression has type "Map[int, str]", variable has type "Map[str, Union[int, str]]") + m_str__int_str = m_int_str__str # E: Incompatible types in assignment (expression has type "Map[Union[int, str], str]", variable has type "Map[str, Union[int, str]]") + m_str__int_str = m_str__str + +def update() -> None: + m_int__str: Map[int, str] = Map() + m_str__str: Map[str, str] = Map() + m_int_str__str: Map[Union[int, str], str] = Map() + m_str__int_str: Map[str, Union[int, str]] = Map() + + m_int__str.update({1: '2'}) + m_int__str.update({1: '2'}, three='4') # E: Unexpected keyword argument "three" for "update" of "Map" + m_int__str.update({1: 2}) # E: Argument 1 to "update" of "Map" has incompatible type "Dict[int, int]"; expected "Union[IterableItems[int, str], Iterable[Tuple[int, str]]]" + + m_str__str.update({'1': '2'}) + m_str__str.update({'1': '2'}, three='4') + m_str__str.update({'1': 2}) # E: Argument 1 to "update" of "Map" has incompatible type "Dict[str, int]"; expected "Union[IterableItems[str, str], Iterable[Tuple[str, str]]]" + + m_int_str__str.update(cast(Dict[Union[int, str], str], {1: '2', '3': '4'})) + m_int_str__str.update({1: '2'}, three='4') + m_int_str__str.update({'1': 2}) # E: Argument 1 to "update" of "Map" has incompatible type "Dict[str, int]"; expected "Union[IterableItems[Union[int, str], str], Iterable[Tuple[Union[int, str], str]]]" + + m_str__int_str.update({'1': 2, '2': 3}) + m_str__int_str.update({'1': 2, '2': 3}, four='5') + m_str__int_str.update({1: 2}) # E: Argument 1 to "update" of "Map" has incompatible type "Dict[int, int]"; expected "Union[IterableItems[str, Union[int, str]], Iterable[Tuple[str, Union[int, str]]]]" + +def mutate() -> None: + m = Map[str, str]() + + with m.mutate() as mm: + mm[0] = '1' # E: Invalid index type "int" for "MapMutation[str, str]"; expected type "str" + mm['1'] = 0 # E: Incompatible types in assignment (expression has type "int", target has type "str") + mm['1'] = '2' + del mm['1'] + mm.set('3', '4') + m2 = mm.finish() + + reveal_type(m2) # N: Revealed type is "immutables._map.Map[builtins.str*, builtins.str*]" diff --git a/tests/test_issue24.py b/tests/test_issue24.py new file mode 100644 index 0000000..7d51e34 --- /dev/null +++ b/tests/test_issue24.py @@ -0,0 +1,156 @@ +import unittest + +from immutables.map import Map as PyMap, map_bitcount + + +class CollisionKey: + def __hash__(self): + return 0 + + +class Issue24Base: + Map = None + + def test_issue24(self): + keys = range(27) + new_entries = dict.fromkeys(keys, True) + m = self.Map(new_entries) + self.assertTrue(17 in m) + with m.mutate() as mm: + for i in keys: + del mm[i] + self.assertEqual(len(mm), 0) + + def dump_check_node_kind(self, header, kind): + header = header.strip() + self.assertTrue(header.strip().startswith(kind)) + + def dump_check_node_size(self, header, size): + node_size = header.split('size=', 1)[1] + node_size = int(node_size.split(maxsplit=1)[0]) + self.assertEqual(node_size, size) + + def dump_check_bitmap_count(self, header, count): + header = header.split('bitmap=')[1] + bitmap = int(header.split(maxsplit=1)[0], 0) + self.assertEqual(map_bitcount(bitmap), count) + + def dump_check_bitmap_node_count(self, header, count): + self.dump_check_node_kind(header, 'Bitmap') + self.dump_check_node_size(header, count * 2) + self.dump_check_bitmap_count(header, count) + + def dump_check_collision_node_count(self, header, count): + self.dump_check_node_kind(header, 'Collision') + self.dump_check_node_size(header, 2 * count) + + def test_bitmap_node_update_in_place_count(self): + keys = range(7) + new_entries = dict.fromkeys(keys, True) + m = self.Map(new_entries) + d = m.__dump__().splitlines() + self.assertTrue(d) + if d[0].startswith('HAMT'): + header = d[1] # skip _map.Map.__dump__() header + else: + header = d[0] + self.dump_check_bitmap_node_count(header, 7) + + def test_bitmap_node_delete_in_place_count(self): + keys = range(7) + new_entries = dict.fromkeys(keys, True) + m = self.Map(new_entries) + with m.mutate() as mm: + del mm[0], mm[2], mm[3] + m2 = mm.finish() + d = m2.__dump__().splitlines() + self.assertTrue(d) + if d[0].startswith('HAMT'): + header = d[1] # skip _map.Map.__dump__() header + else: + header = d[0] + self.dump_check_bitmap_node_count(header, 4) + + def test_collision_node_update_in_place_count(self): + keys = (CollisionKey() for i in range(7)) + new_entries = dict.fromkeys(keys, True) + m = self.Map(new_entries) + d = m.__dump__().splitlines() + self.assertTrue(len(d) > 3) + # get node headers + if d[0].startswith('HAMT'): + h1, h2 = d[1], d[3] # skip _map.Map.__dump__() header + else: + h1, h2 = d[0], d[2] + self.dump_check_node_kind(h1, 'Bitmap') + self.dump_check_collision_node_count(h2, 7) + + def test_collision_node_delete_in_place_count(self): + keys = [CollisionKey() for i in range(7)] + new_entries = dict.fromkeys(keys, True) + m = self.Map(new_entries) + with m.mutate() as mm: + del mm[keys[0]], mm[keys[2]], mm[keys[3]] + m2 = mm.finish() + d = m2.__dump__().splitlines() + self.assertTrue(len(d) > 3) + # get node headers + if d[0].startswith('HAMT'): + h1, h2 = d[1], d[3] # skip _map.Map.__dump__() header + else: + h1, h2 = d[0], d[2] + self.dump_check_node_kind(h1, 'Bitmap') + self.dump_check_collision_node_count(h2, 4) + + +try: + from immutables._map import Map as CMap +except ImportError: + CMap = None + + +class Issue24PyTest(Issue24Base, unittest.TestCase): + Map = PyMap + + +@unittest.skipIf(CMap is None, 'C Map is not available') +class Issue24CTest(Issue24Base, unittest.TestCase): + Map = CMap + + def hamt_dump_check_first_return_second(self, m): + d = m.__dump__().splitlines() + self.assertTrue(len(d) > 2) + self.assertTrue(d[0].startswith('HAMT')) + return d[1] + + def test_array_node_update_in_place_count(self): + keys = range(27) + new_entries = dict.fromkeys(keys, True) + m = self.Map(new_entries) + header = self.hamt_dump_check_first_return_second(m) + self.dump_check_node_kind(header, 'Array') + for i in range(2, 18): + m = m.delete(i) + header = self.hamt_dump_check_first_return_second(m) + self.dump_check_bitmap_node_count(header, 11) + + def test_array_node_delete_in_place_count(self): + keys = range(27) + new_entries = dict.fromkeys(keys, True) + m = self.Map(new_entries) + header = self.hamt_dump_check_first_return_second(m) + self.dump_check_node_kind(header, 'Array') + with m.mutate() as mm: + for i in range(5): + del mm[i] + m2 = mm.finish() + header = self.hamt_dump_check_first_return_second(m2) + self.dump_check_node_kind(header, 'Array') + for i in range(6, 17): + m2 = m2.delete(i) + header = self.hamt_dump_check_first_return_second(m2) + self.dump_check_bitmap_node_count(header, 11) + + +if __name__ == '__main__': + unittest.main() diff --git a/tests/test_map.py b/tests/test_map.py new file mode 100644 index 0000000..4640029 --- /dev/null +++ b/tests/test_map.py @@ -0,0 +1,1417 @@ +import collections.abc +import gc +import pickle +import random +import sys +import unittest +import weakref + +from immutables.map import Map as PyMap +from immutables._testutils import EqError +from immutables._testutils import HashKey +from immutables._testutils import HashKeyCrasher +from immutables._testutils import HashingError +from immutables._testutils import KeyStr +from immutables._testutils import ReprError + + +class BaseMapTest: + + Map = None + + def test_hashkey_helper_1(self): + k1 = HashKey(10, 'aaa') + k2 = HashKey(10, 'bbb') + + self.assertNotEqual(k1, k2) + self.assertEqual(hash(k1), hash(k2)) + + d = dict() + d[k1] = 'a' + d[k2] = 'b' + + self.assertEqual(d[k1], 'a') + self.assertEqual(d[k2], 'b') + + def test_map_basics_1(self): + h = self.Map() + h = None # NoQA + + def test_map_basics_2(self): + h = self.Map() + self.assertEqual(len(h), 0) + + h2 = h.set('a', 'b') + self.assertIsNot(h, h2) + self.assertEqual(len(h), 0) + self.assertEqual(len(h2), 1) + + self.assertIsNone(h.get('a')) + self.assertEqual(h.get('a', 42), 42) + + self.assertEqual(h2.get('a'), 'b') + + h3 = h2.set('b', 10) + self.assertIsNot(h2, h3) + self.assertEqual(len(h), 0) + self.assertEqual(len(h2), 1) + self.assertEqual(len(h3), 2) + self.assertEqual(h3.get('a'), 'b') + self.assertEqual(h3.get('b'), 10) + + self.assertIsNone(h.get('b')) + self.assertIsNone(h2.get('b')) + + self.assertIsNone(h.get('a')) + self.assertEqual(h2.get('a'), 'b') + + h = h2 = h3 = None + + def test_map_basics_3(self): + h = self.Map() + o = object() + h1 = h.set('1', o) + h2 = h1.set('1', o) + self.assertIs(h1, h2) + + def test_map_basics_4(self): + h = self.Map() + h1 = h.set('key', []) + h2 = h1.set('key', []) + self.assertIsNot(h1, h2) + self.assertEqual(len(h1), 1) + self.assertEqual(len(h2), 1) + self.assertIsNot(h1.get('key'), h2.get('key')) + + def test_map_collision_1(self): + k1 = HashKey(10, 'aaa') + k2 = HashKey(10, 'bbb') + k3 = HashKey(10, 'ccc') + + h = self.Map() + h2 = h.set(k1, 'a') + h3 = h2.set(k2, 'b') + + self.assertEqual(h.get(k1), None) + self.assertEqual(h.get(k2), None) + + self.assertEqual(h2.get(k1), 'a') + self.assertEqual(h2.get(k2), None) + + self.assertEqual(h3.get(k1), 'a') + self.assertEqual(h3.get(k2), 'b') + + h4 = h3.set(k2, 'cc') + h5 = h4.set(k3, 'aa') + + self.assertEqual(h3.get(k1), 'a') + self.assertEqual(h3.get(k2), 'b') + self.assertEqual(h4.get(k1), 'a') + self.assertEqual(h4.get(k2), 'cc') + self.assertEqual(h4.get(k3), None) + self.assertEqual(h5.get(k1), 'a') + self.assertEqual(h5.get(k2), 'cc') + self.assertEqual(h5.get(k2), 'cc') + self.assertEqual(h5.get(k3), 'aa') + + self.assertEqual(len(h), 0) + self.assertEqual(len(h2), 1) + self.assertEqual(len(h3), 2) + self.assertEqual(len(h4), 2) + self.assertEqual(len(h5), 3) + + def test_map_collision_2(self): + A = HashKey(100, 'A') + B = HashKey(101, 'B') + C = HashKey(0b011000011100000100, 'C') + D = HashKey(0b011000011100000100, 'D') + E = HashKey(0b1011000011100000100, 'E') + + h = self.Map() + h = h.set(A, 'a') + h = h.set(B, 'b') + h = h.set(C, 'c') + h = h.set(D, 'd') + + # BitmapNode(size=6 bitmap=0b100110000): + # NULL: + # BitmapNode(size=4 bitmap=0b1000000000000000000001000): + # <Key name:A hash:100>: 'a' + # NULL: + # CollisionNode(size=4 id=0x108572410): + # <Key name:C hash:100100>: 'c' + # <Key name:D hash:100100>: 'd' + # <Key name:B hash:101>: 'b' + + h = h.set(E, 'e') + + # BitmapNode(size=4 count=2.0 bitmap=0b110000 id=10b8ea5c0): + # None: + # BitmapNode(size=4 count=2.0 + # bitmap=0b1000000000000000000001000 id=10b8ea518): + # <Key name:A hash:100>: 'a' + # None: + # BitmapNode(size=2 count=1.0 bitmap=0b10 + # id=10b8ea4a8): + # None: + # BitmapNode(size=4 count=2.0 + # bitmap=0b100000001000 + # id=10b8ea4e0): + # None: + # CollisionNode(size=4 id=10b8ea470): + # <Key name:C hash:100100>: 'c' + # <Key name:D hash:100100>: 'd' + # <Key name:E hash:362244>: 'e' + # <Key name:B hash:101>: 'b' + + def test_map_stress_01(self): + COLLECTION_SIZE = 7000 + TEST_ITERS_EVERY = 647 + CRASH_HASH_EVERY = 97 + CRASH_EQ_EVERY = 11 + RUN_XTIMES = 3 + + for _ in range(RUN_XTIMES): + h = self.Map() + d = dict() + + for i in range(COLLECTION_SIZE): + key = KeyStr(i) + + if not (i % CRASH_HASH_EVERY): + with HashKeyCrasher(error_on_hash=True): + with self.assertRaises(HashingError): + h.set(key, i) + + h = h.set(key, i) + + if not (i % CRASH_EQ_EVERY): + with HashKeyCrasher(error_on_eq=True): + with self.assertRaises(EqError): + h.get(KeyStr(i)) # really trigger __eq__ + + d[key] = i + self.assertEqual(len(d), len(h)) + + if not (i % TEST_ITERS_EVERY): + self.assertEqual(set(h.items()), set(d.items())) + self.assertEqual(len(h.items()), len(d.items())) + + self.assertEqual(len(h), COLLECTION_SIZE) + + for key in range(COLLECTION_SIZE): + self.assertEqual(h.get(KeyStr(key), 'not found'), key) + + keys_to_delete = list(range(COLLECTION_SIZE)) + random.shuffle(keys_to_delete) + for iter_i, i in enumerate(keys_to_delete): + key = KeyStr(i) + + if not (iter_i % CRASH_HASH_EVERY): + with HashKeyCrasher(error_on_hash=True): + with self.assertRaises(HashingError): + h.delete(key) + + if not (iter_i % CRASH_EQ_EVERY): + with HashKeyCrasher(error_on_eq=True): + with self.assertRaises(EqError): + h.delete(KeyStr(i)) + + h = h.delete(key) + self.assertEqual(h.get(key, 'not found'), 'not found') + del d[key] + self.assertEqual(len(d), len(h)) + + if iter_i == COLLECTION_SIZE // 2: + hm = h + dm = d.copy() + + if not (iter_i % TEST_ITERS_EVERY): + self.assertEqual(set(h.keys()), set(d.keys())) + self.assertEqual(len(h.keys()), len(d.keys())) + + self.assertEqual(len(d), 0) + self.assertEqual(len(h), 0) + + # ============ + + for key in dm: + self.assertEqual(hm.get(str(key)), dm[key]) + self.assertEqual(len(dm), len(hm)) + + for i, key in enumerate(keys_to_delete): + if str(key) in dm: + hm = hm.delete(str(key)) + dm.pop(str(key)) + self.assertEqual(hm.get(str(key), 'not found'), 'not found') + self.assertEqual(len(d), len(h)) + + if not (i % TEST_ITERS_EVERY): + self.assertEqual(set(h.values()), set(d.values())) + self.assertEqual(len(h.values()), len(d.values())) + + self.assertEqual(len(d), 0) + self.assertEqual(len(h), 0) + self.assertEqual(list(h.items()), []) + + def test_map_collision_3(self): + # Test that iteration works with the deepest tree possible. + + C = HashKey(0b10000000_00000000_00000000_00000000, 'C') + D = HashKey(0b10000000_00000000_00000000_00000000, 'D') + + E = HashKey(0b00000000_00000000_00000000_00000000, 'E') + + h = self.Map() + h = h.set(C, 'C') + h = h.set(D, 'D') + h = h.set(E, 'E') + + # BitmapNode(size=2 count=1 bitmap=0b1): + # NULL: + # BitmapNode(size=2 count=1 bitmap=0b1): + # NULL: + # BitmapNode(size=2 count=1 bitmap=0b1): + # NULL: + # BitmapNode(size=2 count=1 bitmap=0b1): + # NULL: + # BitmapNode(size=2 count=1 bitmap=0b1): + # NULL: + # BitmapNode(size=2 count=1 bitmap=0b1): + # NULL: + # BitmapNode(size=4 count=2 bitmap=0b101): + # <Key name:E hash:0>: 'E' + # NULL: + # CollisionNode(size=4 id=0x107a24520): + # <Key name:C hash:2147483648>: 'C' + # <Key name:D hash:2147483648>: 'D' + + self.assertEqual({k.name for k in h.keys()}, {'C', 'D', 'E'}) + + def test_map_stress_02(self): + COLLECTION_SIZE = 20000 + TEST_ITERS_EVERY = 647 + CRASH_HASH_EVERY = 97 + DELETE_EVERY = 3 + CRASH_EQ_EVERY = 11 + + h = self.Map() + d = dict() + + for i in range(COLLECTION_SIZE // 2): + key = KeyStr(i) + + if not (i % CRASH_HASH_EVERY): + with HashKeyCrasher(error_on_hash=True): + with self.assertRaises(HashingError): + h.set(key, i) + + h = h.set(key, i) + + if not (i % CRASH_EQ_EVERY): + with HashKeyCrasher(error_on_eq=True): + with self.assertRaises(EqError): + h.get(KeyStr(i)) # really trigger __eq__ + + d[key] = i + self.assertEqual(len(d), len(h)) + + if not (i % TEST_ITERS_EVERY): + self.assertEqual(set(h.items()), set(d.items())) + self.assertEqual(len(h.items()), len(d.items())) + + with h.mutate() as m: + for i in range(COLLECTION_SIZE // 2, COLLECTION_SIZE): + key = KeyStr(i) + + if not (i % CRASH_HASH_EVERY): + with HashKeyCrasher(error_on_hash=True): + with self.assertRaises(HashingError): + m[key] = i + + m[key] = i + + if not (i % CRASH_EQ_EVERY): + with HashKeyCrasher(error_on_eq=True): + with self.assertRaises(EqError): + m[KeyStr(i)] + + d[key] = i + self.assertEqual(len(d), len(m)) + + if not (i % DELETE_EVERY): + del m[key] + del d[key] + + self.assertEqual(len(d), len(m)) + + h = m.finish() + + self.assertEqual(len(h), len(d)) + self.assertEqual(set(h.items()), set(d.items())) + + with h.mutate() as m: + for key in list(d): + del d[key] + del m[key] + self.assertEqual(len(m), len(d)) + h = m.finish() + + self.assertEqual(len(h), len(d)) + self.assertEqual(set(h.items()), set(d.items())) + + def test_map_delete_1(self): + A = HashKey(100, 'A') + B = HashKey(101, 'B') + C = HashKey(102, 'C') + D = HashKey(103, 'D') + E = HashKey(104, 'E') + Z = HashKey(-100, 'Z') + + Er = HashKey(103, 'Er', error_on_eq_to=D) + + h = self.Map() + h = h.set(A, 'a') + h = h.set(A, 'a') + h = h.set(B, 'b') + h = h.set(C, 'c') + h = h.set(D, 'd') + h = h.set(E, 'e') + + orig_len = len(h) + + # BitmapNode(size=10 bitmap=0b111110000 id=0x10eadc618): + # <Key name:A hash:100>: 'a' + # <Key name:B hash:101>: 'b' + # <Key name:C hash:102>: 'c' + # <Key name:D hash:103>: 'd' + # <Key name:E hash:104>: 'e' + + h = h.delete(C) + self.assertEqual(len(h), orig_len - 1) + + with self.assertRaisesRegex(ValueError, 'cannot compare'): + h.delete(Er) + + h = h.delete(D) + self.assertEqual(len(h), orig_len - 2) + + with self.assertRaises(KeyError) as ex: + h.delete(Z) + self.assertIs(ex.exception.args[0], Z) + + h = h.delete(A) + self.assertEqual(len(h), orig_len - 3) + + self.assertEqual(h.get(A, 42), 42) + self.assertEqual(h.get(B), 'b') + self.assertEqual(h.get(E), 'e') + + def test_map_delete_2(self): + A = HashKey(100, 'A') + B = HashKey(201001, 'B') + C = HashKey(101001, 'C') + BLike = HashKey(201001, 'B-like') + D = HashKey(103, 'D') + E = HashKey(104, 'E') + Z = HashKey(-100, 'Z') + + Er = HashKey(201001, 'Er', error_on_eq_to=B) + + h = self.Map() + h = h.set(A, 'a') + h = h.set(B, 'b') + h = h.set(C, 'c') + h = h.set(D, 'd') + h = h.set(E, 'e') + + h = h.set(B, 'b') # trigger branch in BitmapNode.assoc + + with self.assertRaises(KeyError): + h.delete(BLike) # trigger branch in BitmapNode.without + + orig_len = len(h) + + # BitmapNode(size=8 bitmap=0b1110010000): + # <Key name:A hash:100>: 'a' + # <Key name:D hash:103>: 'd' + # <Key name:E hash:104>: 'e' + # NULL: + # BitmapNode(size=4 bitmap=0b100000000001000000000): + # <Key name:B hash:201001>: 'b' + # <Key name:C hash:101001>: 'c' + + with self.assertRaisesRegex(ValueError, 'cannot compare'): + h.delete(Er) + + with self.assertRaises(KeyError) as ex: + h.delete(Z) + self.assertIs(ex.exception.args[0], Z) + self.assertEqual(len(h), orig_len) + + h = h.delete(C) + self.assertEqual(len(h), orig_len - 1) + + h = h.delete(B) + self.assertEqual(len(h), orig_len - 2) + + h = h.delete(A) + self.assertEqual(len(h), orig_len - 3) + + self.assertEqual(h.get(D), 'd') + self.assertEqual(h.get(E), 'e') + + with self.assertRaises(KeyError): + h = h.delete(A) + with self.assertRaises(KeyError): + h = h.delete(B) + h = h.delete(D) + h = h.delete(E) + self.assertEqual(len(h), 0) + + def test_map_delete_3(self): + A = HashKey(0b00000000001100100, 'A') + B = HashKey(0b00000000001100101, 'B') + + C = HashKey(0b11000011100000100, 'C') + D = HashKey(0b11000011100000100, 'D') + X = HashKey(0b01000011100000100, 'Z') + Y = HashKey(0b11000011100000100, 'Y') + + E = HashKey(0b00000000001101000, 'E') + + h = self.Map() + h = h.set(A, 'a') + h = h.set(B, 'b') + h = h.set(C, 'c') + h = h.set(D, 'd') + h = h.set(E, 'e') + + self.assertEqual(len(h), 5) + h = h.set(C, 'c') # trigger branch in CollisionNode.assoc + self.assertEqual(len(h), 5) + + orig_len = len(h) + + with self.assertRaises(KeyError): + h.delete(X) + with self.assertRaises(KeyError): + h.delete(Y) + + # BitmapNode(size=6 bitmap=0b100110000): + # NULL: + # BitmapNode(size=4 bitmap=0b1000000000000000000001000): + # <Key name:A hash:100>: 'a' + # NULL: + # CollisionNode(size=4 id=0x108572410): + # <Key name:C hash:100100>: 'c' + # <Key name:D hash:100100>: 'd' + # <Key name:B hash:101>: 'b' + # <Key name:E hash:104>: 'e' + + h = h.delete(A) + self.assertEqual(len(h), orig_len - 1) + + h = h.delete(E) + self.assertEqual(len(h), orig_len - 2) + + self.assertEqual(h.get(C), 'c') + self.assertEqual(h.get(B), 'b') + + h2 = h.delete(C) + self.assertEqual(len(h2), orig_len - 3) + + h2 = h.delete(D) + self.assertEqual(len(h2), orig_len - 3) + + self.assertEqual(len(h), orig_len - 2) + + def test_map_delete_4(self): + A = HashKey(100, 'A') + B = HashKey(101, 'B') + C = HashKey(100100, 'C') + D = HashKey(100100, 'D') + E = HashKey(100100, 'E') + + h = self.Map() + h = h.set(A, 'a') + h = h.set(B, 'b') + h = h.set(C, 'c') + h = h.set(D, 'd') + h = h.set(E, 'e') + + orig_len = len(h) + + # BitmapNode(size=4 bitmap=0b110000): + # NULL: + # BitmapNode(size=4 bitmap=0b1000000000000000000001000): + # <Key name:A hash:100>: 'a' + # NULL: + # CollisionNode(size=6 id=0x10515ef30): + # <Key name:C hash:100100>: 'c' + # <Key name:D hash:100100>: 'd' + # <Key name:E hash:100100>: 'e' + # <Key name:B hash:101>: 'b' + + h = h.delete(D) + self.assertEqual(len(h), orig_len - 1) + + h = h.delete(E) + self.assertEqual(len(h), orig_len - 2) + + h = h.delete(C) + self.assertEqual(len(h), orig_len - 3) + + h = h.delete(A) + self.assertEqual(len(h), orig_len - 4) + + h = h.delete(B) + self.assertEqual(len(h), 0) + + def test_map_delete_5(self): + h = self.Map() + + keys = [] + for i in range(17): + key = HashKey(i, str(i)) + keys.append(key) + h = h.set(key, 'val-{}'.format(i)) + + collision_key16 = HashKey(16, '18') + h = h.set(collision_key16, 'collision') + + # ArrayNode(id=0x10f8b9318): + # 0:: + # BitmapNode(size=2 count=1 bitmap=0b1): + # <Key name:0 hash:0>: 'val-0' + # + # ... 14 more BitmapNodes ... + # + # 15:: + # BitmapNode(size=2 count=1 bitmap=0b1): + # <Key name:15 hash:15>: 'val-15' + # + # 16:: + # BitmapNode(size=2 count=1 bitmap=0b1): + # NULL: + # CollisionNode(size=4 id=0x10f2f5af8): + # <Key name:16 hash:16>: 'val-16' + # <Key name:18 hash:16>: 'collision' + + self.assertEqual(len(h), 18) + + h = h.delete(keys[2]) + self.assertEqual(len(h), 17) + + h = h.delete(collision_key16) + self.assertEqual(len(h), 16) + h = h.delete(keys[16]) + self.assertEqual(len(h), 15) + + h = h.delete(keys[1]) + self.assertEqual(len(h), 14) + with self.assertRaises(KeyError) as ex: + h.delete(keys[1]) + self.assertIs(ex.exception.args[0], keys[1]) + self.assertEqual(len(h), 14) + + for key in keys: + if key in h: + h = h.delete(key) + self.assertEqual(len(h), 0) + + def test_map_delete_6(self): + h = self.Map() + h = h.set(1, 1) + h = h.delete(1) + self.assertEqual(len(h), 0) + self.assertEqual(h, self.Map()) + + def test_map_items_1(self): + A = HashKey(100, 'A') + B = HashKey(201001, 'B') + C = HashKey(101001, 'C') + D = HashKey(103, 'D') + E = HashKey(104, 'E') + F = HashKey(110, 'F') + + h = self.Map() + h = h.set(A, 'a') + h = h.set(B, 'b') + h = h.set(C, 'c') + h = h.set(D, 'd') + h = h.set(E, 'e') + h = h.set(F, 'f') + + it = h.items() + self.assertEqual( + set(list(it)), + {(A, 'a'), (B, 'b'), (C, 'c'), (D, 'd'), (E, 'e'), (F, 'f')}) + + def test_map_items_2(self): + A = HashKey(100, 'A') + B = HashKey(101, 'B') + C = HashKey(100100, 'C') + D = HashKey(100100, 'D') + E = HashKey(100100, 'E') + F = HashKey(110, 'F') + + h = self.Map() + h = h.set(A, 'a') + h = h.set(B, 'b') + h = h.set(C, 'c') + h = h.set(D, 'd') + h = h.set(E, 'e') + h = h.set(F, 'f') + + it = h.items() + self.assertEqual( + set(list(it)), + {(A, 'a'), (B, 'b'), (C, 'c'), (D, 'd'), (E, 'e'), (F, 'f')}) + + def test_map_items_3(self): + h = self.Map() + self.assertEqual(len(h.items()), 0) + self.assertEqual(list(h.items()), []) + + def test_map_items_4(self): + h = self.Map(a=1, b=2, c=3) + k = h.items() + self.assertEqual(set(k), {('a', 1), ('b', 2), ('c', 3)}) + self.assertEqual(set(k), {('a', 1), ('b', 2), ('c', 3)}) + + def test_map_keys_1(self): + A = HashKey(100, 'A') + B = HashKey(101, 'B') + C = HashKey(100100, 'C') + D = HashKey(100100, 'D') + E = HashKey(100100, 'E') + F = HashKey(110, 'F') + + h = self.Map() + h = h.set(A, 'a') + h = h.set(B, 'b') + h = h.set(C, 'c') + h = h.set(D, 'd') + h = h.set(E, 'e') + h = h.set(F, 'f') + + self.assertEqual(set(list(h.keys())), {A, B, C, D, E, F}) + self.assertEqual(set(list(h)), {A, B, C, D, E, F}) + + def test_map_keys_2(self): + h = self.Map(a=1, b=2, c=3) + k = h.keys() + self.assertEqual(set(k), {'a', 'b', 'c'}) + self.assertEqual(set(k), {'a', 'b', 'c'}) + + def test_map_values_1(self): + A = HashKey(100, 'A') + B = HashKey(101, 'B') + C = HashKey(100100, 'C') + D = HashKey(100100, 'D') + E = HashKey(100100, 'E') + F = HashKey(110, 'F') + + h = self.Map() + h = h.set(A, 'a') + h = h.set(B, 'b') + h = h.set(C, 'c') + h = h.set(D, 'd') + h = h.set(E, 'e') + h = h.set(F, 'f') + + self.assertEqual(set(list(h.values())), {'a', 'b', 'c', 'd', 'e', 'f'}) + + def test_map_values_2(self): + h = self.Map(a=1, b=2, c=3) + k = h.values() + self.assertEqual(set(k), {1, 2, 3}) + self.assertEqual(set(k), {1, 2, 3}) + + def test_map_eq_1(self): + A = HashKey(100, 'A') + B = HashKey(101, 'B') + C = HashKey(100100, 'C') + D = HashKey(100100, 'D') + E = HashKey(120, 'E') + + h1 = self.Map() + h1 = h1.set(A, 'a') + h1 = h1.set(B, 'b') + h1 = h1.set(C, 'c') + h1 = h1.set(D, 'd') + + h2 = self.Map() + h2 = h2.set(A, 'a') + + self.assertFalse(h1 == h2) + self.assertTrue(h1 != h2) + + h2 = h2.set(B, 'b') + self.assertFalse(h1 == h2) + self.assertTrue(h1 != h2) + + h2 = h2.set(C, 'c') + self.assertFalse(h1 == h2) + self.assertTrue(h1 != h2) + + h2 = h2.set(D, 'd2') + self.assertFalse(h1 == h2) + self.assertTrue(h1 != h2) + + h2 = h2.set(D, 'd') + self.assertTrue(h1 == h2) + self.assertFalse(h1 != h2) + + h2 = h2.set(E, 'e') + self.assertFalse(h1 == h2) + self.assertTrue(h1 != h2) + + h2 = h2.delete(D) + self.assertFalse(h1 == h2) + self.assertTrue(h1 != h2) + + h2 = h2.set(E, 'd') + self.assertFalse(h1 == h2) + self.assertTrue(h1 != h2) + + def test_map_eq_2(self): + A = HashKey(100, 'A') + Er = HashKey(100, 'Er', error_on_eq_to=A) + + h1 = self.Map() + h1 = h1.set(A, 'a') + + h2 = self.Map() + h2 = h2.set(Er, 'a') + + with self.assertRaisesRegex(ValueError, 'cannot compare'): + h1 == h2 + + with self.assertRaisesRegex(ValueError, 'cannot compare'): + h1 != h2 + + def test_map_eq_3(self): + self.assertNotEqual(self.Map(), 1) + + def test_map_gc_1(self): + A = HashKey(100, 'A') + + h = self.Map() + h = h.set(0, 0) # empty Map node is memoized in _map.c + ref = weakref.ref(h) + + a = [] + a.append(a) + a.append(h) + b = [] + a.append(b) + b.append(a) + h = h.set(A, b) + + del h, a, b + + gc.collect() + gc.collect() + gc.collect() + + self.assertIsNone(ref()) + + def test_map_gc_2(self): + A = HashKey(100, 'A') + + h = self.Map() + h = h.set(A, 'a') + h = h.set(A, h) + + ref = weakref.ref(h) + hi = iter(h.items()) + next(hi) + + del h, hi + + gc.collect() + gc.collect() + gc.collect() + + self.assertIsNone(ref()) + + def test_map_in_1(self): + A = HashKey(100, 'A') + AA = HashKey(100, 'A') + + B = HashKey(101, 'B') + + h = self.Map() + h = h.set(A, 1) + + self.assertTrue(A in h) + self.assertFalse(B in h) + + with self.assertRaises(EqError): + with HashKeyCrasher(error_on_eq=True): + AA in h + + with self.assertRaises(HashingError): + with HashKeyCrasher(error_on_hash=True): + AA in h + + def test_map_getitem_1(self): + A = HashKey(100, 'A') + AA = HashKey(100, 'A') + + B = HashKey(101, 'B') + + h = self.Map() + h = h.set(A, 1) + + self.assertEqual(h[A], 1) + self.assertEqual(h[AA], 1) + + with self.assertRaises(KeyError): + h[B] + + with self.assertRaises(EqError): + with HashKeyCrasher(error_on_eq=True): + h[AA] + + with self.assertRaises(HashingError): + with HashKeyCrasher(error_on_hash=True): + h[AA] + + def test_repr_1(self): + h = self.Map() + self.assertEqual(repr(h), 'immutables.Map({})') + + h = h.set(1, 2).set(2, 3).set(3, 4) + self.assertEqual(repr(h), 'immutables.Map({1: 2, 2: 3, 3: 4})') + + def test_repr_2(self): + h = self.Map() + A = HashKey(100, 'A') + + with self.assertRaises(ReprError): + with HashKeyCrasher(error_on_repr=True): + repr(h.set(1, 2).set(A, 3).set(3, 4)) + + with self.assertRaises(ReprError): + with HashKeyCrasher(error_on_repr=True): + repr(h.set(1, 2).set(2, A).set(3, 4)) + + def test_repr_3(self): + class Key: + def __init__(self): + self.val = None + + def __hash__(self): + return 123 + + def __repr__(self): + return repr(self.val) + + h = self.Map() + k = Key() + h = h.set(k, 1) + k.val = h + + self.assertEqual(repr(h), 'immutables.Map({{...}: 1})') + + def test_hash_1(self): + h = self.Map() + self.assertNotEqual(hash(h), -1) + self.assertEqual(hash(h), hash(h)) + + h = h.set(1, 2).set('a', 'b') + self.assertNotEqual(hash(h), -1) + self.assertEqual(hash(h), hash(h)) + + self.assertEqual( + hash(h.set(1, 2).set('a', 'b')), + hash(h.set('a', 'b').set(1, 2))) + + def test_hash_2(self): + h = self.Map() + A = HashKey(100, 'A') + + m = h.set(1, 2).set(A, 3).set(3, 4) + with self.assertRaises(HashingError): + with HashKeyCrasher(error_on_hash=True): + hash(m) + + m = h.set(1, 2).set(2, A).set(3, 4) + with self.assertRaises(HashingError): + with HashKeyCrasher(error_on_hash=True): + hash(m) + + def test_abc_1(self): + self.assertTrue(issubclass(self.Map, collections.abc.Mapping)) + + def test_map_mut_1(self): + h = self.Map() + h = h.set('a', 1) + + hm1 = h.mutate() + hm2 = h.mutate() + + self.assertFalse(isinstance(hm1, self.Map)) + + self.assertIsNot(hm1, hm2) + self.assertEqual(hm1['a'], 1) + self.assertEqual(hm2['a'], 1) + + hm1.set('b', 2) + hm1.set('c', 3) + + hm2.set('x', 100) + hm2.set('a', 1000) + + self.assertEqual(hm1['a'], 1) + self.assertEqual(hm1.get('x', -1), -1) + + self.assertEqual(hm2['a'], 1000) + self.assertTrue('x' in hm2) + + h1 = hm1.finish() + h2 = hm2.finish() + + self.assertTrue(isinstance(h1, self.Map)) + + self.assertEqual(dict(h.items()), {'a': 1}) + self.assertEqual(dict(h1.items()), {'a': 1, 'b': 2, 'c': 3}) + self.assertEqual(dict(h2.items()), {'a': 1000, 'x': 100}) + + def test_map_mut_2(self): + h = self.Map() + h = h.set('a', 1) + + hm1 = h.mutate() + hm1.set('a', 2) + hm1.set('a', 3) + hm1.set('a', 4) + h2 = hm1.finish() + + self.assertEqual(dict(h.items()), {'a': 1}) + self.assertEqual(dict(h2.items()), {'a': 4}) + + def test_map_mut_3(self): + h = self.Map() + h = h.set('a', 1) + hm1 = h.mutate() + + self.assertEqual(repr(hm1), "immutables.MapMutation({'a': 1})") + + with self.assertRaisesRegex(TypeError, 'unhashable type'): + hash(hm1) + + def test_map_mut_4(self): + h = self.Map() + h = h.set('a', 1) + h = h.set('b', 2) + + hm1 = h.mutate() + hm2 = h.mutate() + + self.assertEqual(hm1, hm2) + + hm1.set('a', 10) + self.assertNotEqual(hm1, hm2) + + hm2.set('a', 10) + self.assertEqual(hm1, hm2) + + self.assertEqual(hm2.pop('a'), 10) + self.assertNotEqual(hm1, hm2) + + def test_map_mut_5(self): + h = self.Map({'a': 1, 'b': 2}, z=100) + self.assertTrue(isinstance(h, self.Map)) + self.assertEqual(dict(h.items()), {'a': 1, 'b': 2, 'z': 100}) + + h2 = h.update(z=200, y=-1) + self.assertEqual(dict(h.items()), {'a': 1, 'b': 2, 'z': 100}) + self.assertEqual(dict(h2.items()), {'a': 1, 'b': 2, 'z': 200, 'y': -1}) + + h3 = h2.update([(1, 2), (3, 4)]) + self.assertEqual(dict(h.items()), {'a': 1, 'b': 2, 'z': 100}) + self.assertEqual(dict(h2.items()), {'a': 1, 'b': 2, 'z': 200, 'y': -1}) + self.assertEqual(dict(h3.items()), + {'a': 1, 'b': 2, 'z': 200, 'y': -1, 1: 2, 3: 4}) + + h4 = h3.update() + self.assertIs(h4, h3) + + h5 = h4.update(self.Map({'zzz': 'yyz'})) + + self.assertEqual(dict(h5.items()), + {'a': 1, 'b': 2, 'z': 200, 'y': -1, 1: 2, 3: 4, + 'zzz': 'yyz'}) + + def test_map_mut_6(self): + h = self.Map({'a': 1, 'b': 2}, z=100) + self.assertEqual(dict(h.items()), {'a': 1, 'b': 2, 'z': 100}) + + with self.assertRaisesRegex(TypeError, 'not iterable'): + h.update(1) + + with self.assertRaisesRegex(ValueError, 'map update sequence element'): + h.update([(1, 2), (3, 4, 5)]) + + with self.assertRaisesRegex(TypeError, 'cannot convert map update'): + h.update([(1, 2), 1]) + + self.assertEqual(dict(h.items()), {'a': 1, 'b': 2, 'z': 100}) + + def test_map_mut_7(self): + key = HashKey(123, 'aaa') + + h = self.Map({'a': 1, 'b': 2}, z=100) + self.assertEqual(dict(h.items()), {'a': 1, 'b': 2, 'z': 100}) + + upd = {key: 1} + with HashKeyCrasher(error_on_hash=True): + with self.assertRaises(HashingError): + h.update(upd) + + upd = self.Map({key: 'zzz'}) + with HashKeyCrasher(error_on_hash=True): + with self.assertRaises(HashingError): + h.update(upd) + + upd = [(1, 2), (key, 'zzz')] + with HashKeyCrasher(error_on_hash=True): + with self.assertRaises(HashingError): + h.update(upd) + + self.assertEqual(dict(h.items()), {'a': 1, 'b': 2, 'z': 100}) + + def test_map_mut_8(self): + key1 = HashKey(123, 'aaa') + key2 = HashKey(123, 'bbb') + + h = self.Map({key1: 123}) + self.assertEqual(dict(h.items()), {key1: 123}) + + upd = {key2: 1} + with HashKeyCrasher(error_on_eq=True): + with self.assertRaises(EqError): + h.update(upd) + + upd = self.Map({key2: 'zzz'}) + with HashKeyCrasher(error_on_eq=True): + with self.assertRaises(EqError): + h.update(upd) + + upd = [(1, 2), (key2, 'zzz')] + with HashKeyCrasher(error_on_eq=True): + with self.assertRaises(EqError): + h.update(upd) + + self.assertEqual(dict(h.items()), {key1: 123}) + + def test_map_mut_9(self): + key1 = HashKey(123, 'aaa') + + src = {key1: 123} + with HashKeyCrasher(error_on_hash=True): + with self.assertRaises(HashingError): + self.Map(src) + + src = [(1, 2), (key1, 123)] + with HashKeyCrasher(error_on_hash=True): + with self.assertRaises(HashingError): + self.Map(src) + + def test_map_mut_10(self): + key1 = HashKey(123, 'aaa') + + m = self.Map({key1: 123}) + + mm = m.mutate() + with HashKeyCrasher(error_on_hash=True): + with self.assertRaises(HashingError): + del mm[key1] + + mm = m.mutate() + with HashKeyCrasher(error_on_hash=True): + with self.assertRaises(HashingError): + mm.pop(key1, None) + + mm = m.mutate() + with HashKeyCrasher(error_on_hash=True): + with self.assertRaises(HashingError): + mm.set(key1, 123) + + def test_map_mut_11(self): + m = self.Map({'a': 1, 'b': 2}) + + mm = m.mutate() + self.assertEqual(mm.pop('a', 1), 1) + self.assertEqual(mm.finish(), self.Map({'b': 2})) + + mm = m.mutate() + self.assertEqual(mm.pop('b', 1), 2) + self.assertEqual(mm.finish(), self.Map({'a': 1})) + + mm = m.mutate() + self.assertEqual(mm.pop('b', 1), 2) + del mm['a'] + self.assertEqual(mm.finish(), self.Map()) + + def test_map_mut_12(self): + m = self.Map({'a': 1, 'b': 2}) + + mm = m.mutate() + mm.finish() + + with self.assertRaisesRegex(ValueError, 'has been finished'): + mm.pop('a') + + with self.assertRaisesRegex(ValueError, 'has been finished'): + del mm['a'] + + with self.assertRaisesRegex(ValueError, 'has been finished'): + mm.set('a', 'b') + + with self.assertRaisesRegex(ValueError, 'has been finished'): + mm['a'] = 'b' + + with self.assertRaisesRegex(ValueError, 'has been finished'): + mm.update(a='b') + + def test_map_mut_13(self): + key1 = HashKey(123, 'aaa') + key2 = HashKey(123, 'aaa') + + m = self.Map({key1: 123}) + + mm = m.mutate() + with HashKeyCrasher(error_on_eq=True): + with self.assertRaises(EqError): + del mm[key2] + + mm = m.mutate() + with HashKeyCrasher(error_on_eq=True): + with self.assertRaises(EqError): + mm.pop(key2, None) + + mm = m.mutate() + with HashKeyCrasher(error_on_eq=True): + with self.assertRaises(EqError): + mm.set(key2, 123) + + def test_map_mut_14(self): + m = self.Map(a=1, b=2) + + with m.mutate() as mm: + mm['z'] = 100 + del mm['a'] + + self.assertEqual(mm.finish(), self.Map(z=100, b=2)) + + def test_map_mut_15(self): + m = self.Map(a=1, b=2) + + with self.assertRaises(ZeroDivisionError): + with m.mutate() as mm: + mm['z'] = 100 + del mm['a'] + 1 / 0 + + self.assertEqual(mm.finish(), self.Map(z=100, b=2)) + self.assertEqual(m, self.Map(a=1, b=2)) + + def test_map_mut_16(self): + m = self.Map(a=1, b=2) + hash(m) + + m2 = self.Map(m) + m3 = self.Map(m, c=3) + + self.assertEqual(m, m2) + self.assertEqual(len(m), len(m2)) + self.assertEqual(hash(m), hash(m2)) + + self.assertIsNot(m, m2) + self.assertEqual(m3, self.Map(a=1, b=2, c=3)) + + def test_map_mut_17(self): + m = self.Map(a=1) + with m.mutate() as mm: + with self.assertRaisesRegex( + TypeError, 'cannot create Maps from MapMutations'): + self.Map(mm) + + def test_map_mut_18(self): + m = self.Map(a=1, b=2) + with m.mutate() as mm: + mm.update(self.Map(x=1), z=2) + mm.update(c=3) + mm.update({'n': 100, 'a': 20}) + m2 = mm.finish() + + expected = self.Map( + {'b': 2, 'c': 3, 'n': 100, 'z': 2, 'x': 1, 'a': 20}) + + self.assertEqual(len(m2), 6) + self.assertEqual(m2, expected) + self.assertEqual(m, self.Map(a=1, b=2)) + + def test_map_mut_19(self): + m = self.Map(a=1, b=2) + m2 = m.update({'a': 20}) + self.assertEqual(len(m2), 2) + + def test_map_mut_20(self): + # Issue 24: + + h = self.Map() + + for i in range(19): + # Create more than 16 keys to trigger the root bitmap + # node to be converted into an array node + h = h.set(HashKey(i, i), i) + + h = h.set(HashKey(18, '18-collision'), 18) + + with h.mutate() as m: + del m[HashKey(18, 18)] + del m[HashKey(18, '18-collision')] + + # The pre-issue-24 code failed to update the number of array + # node element, so at this point it would be greater than it + # actually is. + h = m.finish() + + # Any of the below operations shouldn't crash the debug build. + with h.mutate() as m: + for i in range(18): + del m[HashKey(i, i)] + h = m.finish() + h = h.set(HashKey(21, 21), 21) + h = h.set(HashKey(22, 22), 22) + + def test_map_mut_21(self): + # Issue 24: + # Array nodes, while in mutation, failed to increment the + # internal count of elements when adding a new key to it. + # Because the internal count + + h = self.Map() + + for i in range(18): + # Create more than 16 keys to trigger the root bitmap + # node to be converted into an array node + h = h.set(HashKey(i, i), i) + + with h.mutate() as m: + # Add one new key to the array node + m[HashKey(18, 18)] = 18 + # Add another key -- after this the old code failed + # to increment the number of elements in the mutated + # array node. + m[HashKey(19, 19)] = 19 + h = m.finish() + + for i in range(20): + # Start deleting keys one by one. Because array node + # element count was accounted incorrectly (smaller by 1 + # than it actually is, the mutation for "del h[18]" would + # create an empty array node, clipping the "19" key). + # Before the issue #24 fix, the below line would crash + # on i=19. + h = h.delete(HashKey(i, i)) + + def test_map_mut_stress(self): + COLLECTION_SIZE = 7000 + TEST_ITERS_EVERY = 647 + RUN_XTIMES = 3 + + for _ in range(RUN_XTIMES): + h = self.Map() + d = dict() + + for i in range(COLLECTION_SIZE // TEST_ITERS_EVERY): + + hm = h.mutate() + for j in range(TEST_ITERS_EVERY): + key = random.randint(1, 100000) + key = HashKey(key % 271, str(key)) + + hm.set(key, key) + d[key] = key + + self.assertEqual(len(hm), len(d)) + + h2 = hm.finish() + self.assertEqual(dict(h2.items()), d) + h = h2 + + self.assertEqual(dict(h.items()), d) + self.assertEqual(len(h), len(d)) + + it = iter(tuple(d.keys())) + for i in range(COLLECTION_SIZE // TEST_ITERS_EVERY): + + hm = h.mutate() + for j in range(TEST_ITERS_EVERY): + try: + key = next(it) + except StopIteration: + break + + del d[key] + del hm[key] + + self.assertEqual(len(hm), len(d)) + + h2 = hm.finish() + self.assertEqual(dict(h2.items()), d) + h = h2 + + self.assertEqual(dict(h.items()), d) + self.assertEqual(len(h), len(d)) + + def test_map_pickle(self): + h = self.Map(a=1, b=2) + for proto in range(pickle.HIGHEST_PROTOCOL): + p = pickle.dumps(h, proto) + uh = pickle.loads(p) + + self.assertTrue(isinstance(uh, self.Map)) + self.assertEqual(h, uh) + + with self.assertRaisesRegex(TypeError, "can('t|not) pickle"): + pickle.dumps(h.mutate()) + + @unittest.skipIf( + sys.version_info < (3, 7, 0), "__class_getitem__ is not available" + ) + def test_map_is_subscriptable(self): + self.assertIs(self.Map[int, str], self.Map) + + def test_kwarg_named_col(self): + self.assertEqual(dict(self.Map(col=0)), {"col": 0}) + self.assertEqual(dict(self.Map(a=0, col=1)), {"a": 0, "col": 1}) + self.assertEqual(dict(self.Map({"a": 0}, col=1)), {"a": 0, "col": 1}) + + +class PyMapTest(BaseMapTest, unittest.TestCase): + + Map = PyMap + + +try: + from immutables._map import Map as CMap +except ImportError: + CMap = None + + +@unittest.skipIf(CMap is None, 'C Map is not available') +class CMapTest(BaseMapTest, unittest.TestCase): + + Map = CMap + + +if __name__ == "__main__": + unittest.main() diff --git a/tests/test_mypy.py b/tests/test_mypy.py new file mode 100644 index 0000000..fcdc11c --- /dev/null +++ b/tests/test_mypy.py @@ -0,0 +1,26 @@ +import os + +try: + import mypy.test.testcmdline + from mypy.test.helpers import normalize_error_messages +except (ImportError, AssertionError): + if os.environ.get('IMMU_SKIP_MYPY_TESTS'): + pass + else: + raise +else: + # I'm upset. There's no other way to deal with the little 'defined here' + # notes that mypy emits when passing an unexpected keyword argument + # and at no other time. + def renormalize_error_messages(messages): + messages = [x for x in messages if not x.endswith(' defined here')] + return normalize_error_messages(messages) + + mypy.test.testcmdline.normalize_error_messages = renormalize_error_messages + + this_file_dir = os.path.dirname(os.path.realpath(__file__)) + test_data_prefix = os.path.join(this_file_dir, 'test-data') + + class ImmuMypyTest(mypy.test.testcmdline.PythonCmdlineSuite): + data_prefix = test_data_prefix + files = ['check-immu.test'] diff --git a/tests/test_none_keys.py b/tests/test_none_keys.py new file mode 100644 index 0000000..26d4220 --- /dev/null +++ b/tests/test_none_keys.py @@ -0,0 +1,515 @@ +import ctypes +import unittest + +from immutables.map import map_hash, map_mask, Map as PyMap +from immutables._testutils import HashKey + + +none_hash = map_hash(None) +assert(none_hash != 1) +assert(none_hash.bit_length() <= 32) + +none_hash_u = ctypes.c_size_t(none_hash).value +not_collision = 0xffffffff & (~none_hash_u) + +mask = 0x7ffffffff +none_collisions = [none_hash_u & (mask >> shift) + for shift in reversed(range(0, 32, 5))] +assert(len(none_collisions) == 7) +none_collisions = [ + ctypes.c_ssize_t(h | (not_collision & (mask << shift))).value + for shift, h in zip(range(5, 37, 5), none_collisions) +] + + +class NoneCollision(HashKey): + def __init__(self, name, level): + if name is None: + raise ValueError("Can't have a NoneCollision with a None value") + super().__init__(none_collisions[level], name) + + def __eq__(self, other): + if other is None: + return False + return super().__eq__(other) + + __hash__ = HashKey.__hash__ + + +class BaseNoneTest: + Map = None + + def test_none_collisions(self): + collisions = [NoneCollision('a', level) for level in range(7)] + indices = [map_mask(none_hash, shift) for shift in range(0, 32, 5)] + + for i, c in enumerate(collisions[:-1], 1): + self.assertNotEqual(c, None) + c_hash = map_hash(c) + self.assertNotEqual(c_hash, none_hash) + for j, idx in enumerate(indices[:i]): + self.assertEqual(map_mask(c_hash, j*5), idx) + for j, idx in enumerate(indices[i:], i): + self.assertNotEqual(map_mask(c_hash, j*5), idx) + + c = collisions[-1] + self.assertNotEqual(c, None) + c_hash = map_hash(c) + self.assertEqual(c_hash, none_hash) + for i, idx in enumerate(indices): + self.assertEqual(map_mask(c_hash, i*5), idx) + + def test_none_as_key(self): + m = self.Map({None: 1}) + + self.assertEqual(len(m), 1) + self.assertTrue(None in m) + self.assertEqual(m[None], 1) + self.assertEqual(repr(m), 'immutables.Map({None: 1})') + + for level in range(7): + key = NoneCollision('a', level) + self.assertFalse(key in m) + with self.assertRaises(KeyError): + m.delete(key) + + m = m.delete(None) + self.assertEqual(len(m), 0) + self.assertFalse(None in m) + self.assertEqual(repr(m), 'immutables.Map({})') + + self.assertEqual(m, self.Map()) + + with self.assertRaises(KeyError): + m.delete(None) + + def test_none_set(self): + m = self.Map().set(None, 2) + + self.assertEqual(len(m), 1) + self.assertTrue(None in m) + self.assertEqual(m[None], 2) + + m = m.set(None, 1) + + self.assertEqual(len(m), 1) + self.assertTrue(None in m) + self.assertEqual(m[None], 1) + + m = m.delete(None) + + self.assertEqual(len(m), 0) + self.assertEqual(m, self.Map()) + self.assertFalse(None in m) + + with self.assertRaises(KeyError): + m.delete(None) + + def test_none_collision_1(self): + for level in range(7): + key = NoneCollision('a', level) + m = self.Map({None: 1, key: 2}) + + self.assertEqual(len(m), 2) + self.assertTrue(None in m) + self.assertEqual(m[None], 1) + self.assertTrue(key in m) + self.assertEqual(m[key], 2) + + m2 = m.delete(None) + self.assertEqual(len(m2), 1) + self.assertTrue(key in m2) + self.assertEqual(m2[key], 2) + self.assertFalse(None in m2) + with self.assertRaises(KeyError): + m2.delete(None) + + m3 = m2.delete(key) + self.assertEqual(len(m3), 0) + self.assertFalse(None in m3) + self.assertFalse(key in m3) + self.assertEqual(m3, self.Map()) + self.assertEqual(repr(m3), 'immutables.Map({})') + with self.assertRaises(KeyError): + m3.delete(None) + with self.assertRaises(KeyError): + m3.delete(key) + + m2 = m.delete(key) + self.assertEqual(len(m2), 1) + self.assertTrue(None in m2) + self.assertEqual(m2[None], 1) + self.assertFalse(key in m2) + with self.assertRaises(KeyError): + m2.delete(key) + + m4 = m2.delete(None) + self.assertEqual(len(m4), 0) + self.assertFalse(None in m4) + self.assertFalse(key in m4) + self.assertEqual(m4, self.Map()) + self.assertEqual(repr(m4), 'immutables.Map({})') + with self.assertRaises(KeyError): + m4.delete(None) + with self.assertRaises(KeyError): + m4.delete(key) + + self.assertEqual(m3, m4) + + def test_none_collision_2(self): + key = HashKey(not_collision, 'a') + m = self.Map().set(None, 1).set(key, 2) + + self.assertEqual(len(m), 2) + self.assertTrue(key in m) + self.assertTrue(None in m) + self.assertEqual(m[key], 2) + self.assertEqual + + m = m.set(None, 0) + self.assertEqual(len(m), 2) + self.assertTrue(key in m) + self.assertTrue(None in m) + + for level in range(7): + key2 = NoneCollision('b', level) + self.assertFalse(key2 in m) + m2 = m.set(key2, 1) + + self.assertEqual(len(m2), 3) + self.assertTrue(key in m2) + self.assertTrue(None in m2) + self.assertTrue(key2 in m2) + self.assertEqual(m2[key], 2) + self.assertEqual(m2[None], 0) + self.assertEqual(m2[key2], 1) + + m2 = m2.set(None, 1) + self.assertEqual(len(m2), 3) + self.assertTrue(key in m2) + self.assertTrue(None in m2) + self.assertTrue(key2 in m2) + self.assertEqual(m2[key], 2) + self.assertEqual(m2[None], 1) + self.assertEqual(m2[key2], 1) + + m2 = m2.set(None, 2) + self.assertEqual(len(m2), 3) + self.assertTrue(key in m2) + self.assertTrue(None in m2) + self.assertTrue(key2 in m2) + self.assertEqual(m2[key], 2) + self.assertEqual(m2[None], 2) + self.assertEqual(m2[key2], 1) + + m3 = m2.delete(key) + self.assertEqual(len(m3), 2) + self.assertTrue(None in m3) + self.assertTrue(key2 in m3) + self.assertFalse(key in m3) + self.assertEqual(m3[None], 2) + self.assertEqual(m3[key2], 1) + with self.assertRaises(KeyError): + m3.delete(key) + + m3 = m2.delete(key2) + self.assertEqual(len(m3), 2) + self.assertTrue(None in m3) + self.assertTrue(key in m3) + self.assertFalse(key2 in m3) + self.assertEqual(m3[None], 2) + self.assertEqual(m3[key], 2) + with self.assertRaises(KeyError): + m3.delete(key2) + + m3 = m2.delete(None) + self.assertEqual(len(m3), 2) + self.assertTrue(key in m3) + self.assertTrue(key2 in m3) + self.assertFalse(None in m3) + self.assertEqual(m3[key], 2) + self.assertEqual(m3[key2], 1) + with self.assertRaises(KeyError): + m3.delete(None) + + m2 = m.delete(None) + self.assertEqual(len(m2), 1) + self.assertFalse(None in m2) + self.assertTrue(key in m2) + self.assertEqual(m2[key], 2) + with self.assertRaises(KeyError): + m2.delete(None) + + m2 = m.delete(key) + self.assertEqual(len(m2), 1) + self.assertFalse(key in m2) + self.assertTrue(None in m2) + self.assertEqual(m2[None], 0) + with self.assertRaises(KeyError): + m2.delete(key) + + def test_none_collision_3(self): + for level in range(7): + key = NoneCollision('a', level) + m = self.Map({key: 2}) + + self.assertEqual(len(m), 1) + self.assertFalse(None in m) + self.assertTrue(key in m) + self.assertEqual(m[key], 2) + with self.assertRaises(KeyError): + m.delete(None) + + m = m.set(None, 1) + self.assertEqual(len(m), 2) + self.assertTrue(key in m) + self.assertEqual(m[key], 2) + self.assertTrue(None in m) + self.assertEqual(m[None], 1) + + m = m.set(None, 0) + self.assertEqual(len(m), 2) + self.assertTrue(key in m) + self.assertEqual(m[key], 2) + self.assertTrue(None in m) + self.assertEqual(m[None], 0) + + m2 = m.delete(key) + self.assertEqual(len(m2), 1) + self.assertTrue(None in m2) + self.assertEqual(m2[None], 0) + self.assertFalse(key in m2) + with self.assertRaises(KeyError): + m2.delete(key) + + m2 = m.delete(None) + self.assertEqual(len(m2), 1) + self.assertTrue(key in m2) + self.assertEqual(m2[key], 2) + self.assertFalse(None in m2) + with self.assertRaises(KeyError): + m2.delete(None) + + def test_collision_4(self): + key2 = NoneCollision('a', 2) + key4 = NoneCollision('b', 4) + m = self.Map({key2: 2, key4: 4}) + + self.assertEqual(len(m), 2) + self.assertTrue(key2 in m) + self.assertTrue(key4 in m) + self.assertEqual(m[key2], 2) + self.assertEqual(m[key4], 4) + self.assertFalse(None in m) + + m2 = m.set(None, 9) + + self.assertEqual(len(m2), 3) + self.assertTrue(key2 in m2) + self.assertTrue(key4 in m2) + self.assertTrue(None in m2) + self.assertEqual(m2[key2], 2) + self.assertEqual(m2[key4], 4) + self.assertEqual(m2[None], 9) + + m3 = m2.set(None, 0) + self.assertEqual(len(m3), 3) + self.assertTrue(key2 in m3) + self.assertTrue(key4 in m3) + self.assertTrue(None in m3) + self.assertEqual(m3[key2], 2) + self.assertEqual(m3[key4], 4) + self.assertEqual(m3[None], 0) + + m3 = m2.set(key2, 0) + self.assertEqual(len(m3), 3) + self.assertTrue(key2 in m3) + self.assertTrue(key4 in m3) + self.assertTrue(None in m3) + self.assertEqual(m3[key2], 0) + self.assertEqual(m3[key4], 4) + self.assertEqual(m3[None], 9) + + m3 = m2.set(key4, 0) + self.assertEqual(len(m3), 3) + self.assertTrue(key2 in m3) + self.assertTrue(key4 in m3) + self.assertTrue(None in m3) + self.assertEqual(m3[key2], 2) + self.assertEqual(m3[key4], 0) + self.assertEqual(m3[None], 9) + + m3 = m2.delete(None) + self.assertEqual(m3, m) + self.assertEqual(len(m3), 2) + self.assertTrue(key2 in m3) + self.assertTrue(key4 in m3) + self.assertEqual(m3[key2], 2) + self.assertEqual(m3[key4], 4) + self.assertFalse(None in m3) + with self.assertRaises(KeyError): + m3.delete(None) + + m3 = m2.delete(key2) + self.assertEqual(len(m3), 2) + self.assertTrue(None in m3) + self.assertTrue(key4 in m3) + self.assertEqual(m3[None], 9) + self.assertEqual(m3[key4], 4) + self.assertFalse(key2 in m3) + with self.assertRaises(KeyError): + m3.delete(key2) + + m3 = m2.delete(key4) + self.assertEqual(len(m3), 2) + self.assertTrue(None in m3) + self.assertTrue(key2 in m3) + self.assertEqual(m3[None], 9) + self.assertEqual(m3[key2], 2) + self.assertFalse(key4 in m3) + with self.assertRaises(KeyError): + m3.delete(key4) + + def test_none_mutation(self): + key2 = NoneCollision('a', 2) + key4 = NoneCollision('b', 4) + key = NoneCollision('c', -1) + m = self.Map({key: -1, key2: 2, key4: 4, None: 9}) + + with m.mutate() as mm: + self.assertEqual(len(mm), 4) + self.assertTrue(key in mm) + self.assertTrue(key2 in mm) + self.assertTrue(key4 in mm) + self.assertTrue(None in mm) + self.assertEqual(mm[key2], 2) + self.assertEqual(mm[key4], 4) + self.assertEqual(mm[key], -1) + self.assertEqual(mm[None], 9) + + for k in m: + mm[k] = -mm[k] + + self.assertEqual(len(mm), 4) + self.assertTrue(key in mm) + self.assertTrue(key2 in mm) + self.assertTrue(key4 in mm) + self.assertTrue(None in mm) + self.assertEqual(mm[key2], -2) + self.assertEqual(mm[key4], -4) + self.assertEqual(mm[key], 1) + self.assertEqual(mm[None], -9) + + for k in m: + del mm[k] + self.assertEqual(len(mm), 3) + self.assertFalse(k in mm) + for n in m: + if n != k: + self.assertTrue(n in mm) + self.assertEqual(mm[n], -m[n]) + with self.assertRaises(KeyError): + del mm[k] + mm[k] = -m[k] + self.assertEqual(len(mm), 4) + self.assertTrue(k in mm) + self.assertEqual(mm[k], -m[k]) + + for k in m: + mm[k] = -mm[k] + + self.assertEqual(len(mm), 4) + self.assertTrue(key in mm) + self.assertTrue(key2 in mm) + self.assertTrue(key4 in mm) + self.assertTrue(None in mm) + self.assertEqual(mm[key2], 2) + self.assertEqual(mm[key4], 4) + self.assertEqual(mm[key], -1) + self.assertEqual(mm[None], 9) + + for k in m: + mm[k] = -mm[k] + + self.assertEqual(len(mm), 4) + self.assertTrue(key in mm) + self.assertTrue(key2 in mm) + self.assertTrue(key4 in mm) + self.assertTrue(None in mm) + self.assertEqual(mm[key2], -2) + self.assertEqual(mm[key4], -4) + self.assertEqual(mm[key], 1) + self.assertEqual(mm[None], -9) + + m2 = mm.finish() + + self.assertEqual(set(m), set(m2)) + self.assertEqual(len(m2), 4) + self.assertTrue(key in m2) + self.assertTrue(key2 in m2) + self.assertTrue(key4 in m2) + self.assertTrue(None in m2) + self.assertEqual(m2[key2], -2) + self.assertEqual(m2[key4], -4) + self.assertEqual(m2[key], 1) + self.assertEqual(m2[None], -9) + + for k, v in m.items(): + self.assertTrue(k in m2) + self.assertEqual(m2[k], -v) + + def test_iterators(self): + key2 = NoneCollision('a', 2) + key4 = NoneCollision('b', 4) + key = NoneCollision('c', -1) + m = self.Map({key: -1, key2: 2, key4: 4, None: 9}) + + self.assertEqual(len(m), 4) + self.assertTrue(key in m) + self.assertTrue(key2 in m) + self.assertTrue(key4 in m) + self.assertTrue(None in m) + self.assertEqual(m[key2], 2) + self.assertEqual(m[key4], 4) + self.assertEqual(m[key], -1) + self.assertEqual(m[None], 9) + + s = set(m) + self.assertEqual(len(s), 4) + self.assertEqual(s, set([None, key, key2, key4])) + + sk = set(m.keys()) + self.assertEqual(s, sk) + + sv = set(m.values()) + self.assertEqual(len(sv), 4) + self.assertEqual(sv, set([-1, 2, 4, 9])) + + si = set(m.items()) + self.assertEqual(len(si), 4) + self.assertEqual(si, + set([(key, -1), (key2, 2), (key4, 4), (None, 9)])) + + d = {key: -1, key2: 2, key4: 4, None: 9} + self.assertEqual(dict(m.items()), d) + + +class PyMapNoneTest(BaseNoneTest, unittest.TestCase): + + Map = PyMap + + +try: + from immutables._map import Map as CMap +except ImportError: + CMap = None + + +@unittest.skipIf(CMap is None, 'C Map is not available') +class CMapNoneTest(BaseNoneTest, unittest.TestCase): + + Map = CMap + + +if __name__ == "__main__": + unittest.main() |
