lombokvector 0.1.0__tar.gz

lombokvector-0.1.0/LICENSE

@@ -0,0 +1,168 @@
+                                 Apache License
+                           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 in 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, thus excluding
+      simple mechanical transformations or merely linking (or binding by
+      name) to the interfaces of, the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original Work and any Derivative Works thereof, that is
+      intentionally submitted to, or incorporated into, the Work by
+      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, such as 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 into 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.
+
+   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, 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.
+
+   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 Contribution.
+
+   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.
+
+   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
+
+   Copyright 2024-2026 codinglombok
+
+   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.

lombokvector-0.1.0/PKG-INFO

@@ -0,0 +1,29 @@
+Metadata-Version: 2.4
+Name: lombokvector
+Version: 0.1.0
+Summary: Vector math with bit-identical results across Python, Rust, TypeScript, Go and PHP: dot product, cosine similarity, L2 distance, normalization, batch ranking. Pure Python, zero dependencies.
+Author: codinglombok
+License-Expression: Apache-2.0
+Project-URL: Homepage, https://github.com/codinglombok/LombokVector
+Project-URL: Issues, https://github.com/codinglombok/LombokVector/issues
+Keywords: vector,cosine-similarity,embedding,deterministic
+Classifier: Programming Language :: Python :: 3
+Classifier: Operating System :: OS Independent
+Classifier: Topic :: Scientific/Engineering :: Mathematics
+Classifier: Typing :: Typed
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Dynamic: license-file
+
+# LombokVector for Python
+
+Python package `lombokvector` (binary64, Python 3.9+, pure Python): dot product, cosine similarity, Euclidean distance, normalization, and batch ranking with results that are bit-identical to the other LombokVector ports. Zero runtime dependencies.
+
+```
+pip install lombokvector
+```
+
+The summation order and every operation are defined in [docs/SPEC_LombokVector_v0.1.0.md](https://github.com/codinglombok/LombokVector/blob/main/docs/SPEC_LombokVector_v0.1.0.md); this port reproduces the shared vector file bit for bit in CI. Usage for all languages: [main README](https://github.com/codinglombok/LombokVector#readme) and [docs/API_LombokVector_v0.1.0.md](https://github.com/codinglombok/LombokVector/blob/main/docs/API_LombokVector_v0.1.0.md).
+
+License: Apache-2.0.

lombokvector-0.1.0/README.md

@@ -0,0 +1,11 @@
+# LombokVector for Python
+
+Python package `lombokvector` (binary64, Python 3.9+, pure Python): dot product, cosine similarity, Euclidean distance, normalization, and batch ranking with results that are bit-identical to the other LombokVector ports. Zero runtime dependencies.
+
+```
+pip install lombokvector
+```
+
+The summation order and every operation are defined in [docs/SPEC_LombokVector_v0.1.0.md](https://github.com/codinglombok/LombokVector/blob/main/docs/SPEC_LombokVector_v0.1.0.md); this port reproduces the shared vector file bit for bit in CI. Usage for all languages: [main README](https://github.com/codinglombok/LombokVector#readme) and [docs/API_LombokVector_v0.1.0.md](https://github.com/codinglombok/LombokVector/blob/main/docs/API_LombokVector_v0.1.0.md).
+
+License: Apache-2.0.

lombokvector-0.1.0/lombokvector/__init__.py

@@ -0,0 +1,245 @@
+"""LombokVector for Python: binary64 vector math whose results are
+bit-identical to the Rust, TypeScript, Go and PHP ports. Every sum uses the
+8-lane order of docs/SPEC_LombokVector_v0.1.0.md section 2. Pure Python,
+no dependencies."""
+
+from __future__ import annotations
+
+import math
+from typing import List, NamedTuple, Sequence
+
+__version__ = "0.1.0"
+__all__ = [
+    "ScoredIndex",
+    "VectorError",
+    "batch_cosine",
+    "batch_dot",
+    "batch_l2",
+    "cosine_similarity",
+    "distance_matrix_cosine",
+    "dot_product",
+    "inner_product",
+    "l2_distance",
+    "l2_norm",
+    "normalize",
+    "vec_add",
+    "vec_mul_scalar",
+    "vec_sub",
+]
+
+Vector = Sequence[float]
+
+
+class VectorError(ValueError):
+    """Raised by every operation; ``code`` is stable across ports:
+    ``DIMENSION_MISMATCH``, ``EMPTY_VECTOR``, ``ZERO_MAGNITUDE``, ``NON_FINITE``."""
+
+    def __init__(self, code: str, message: str) -> None:
+        super().__init__(f"{code}: {message}")
+        self.code = code
+
+
+class ScoredIndex(NamedTuple):
+    """A candidate's position in the input and its score."""
+
+    index: int
+    score: float
+
+
+def _check_pair(a: Vector, b: Vector) -> None:
+    if len(a) == 0:
+        raise VectorError("EMPTY_VECTOR", "vector has no elements")
+    if len(a) != len(b):
+        raise VectorError("DIMENSION_MISMATCH", f"expected {len(a)} elements, got {len(b)}")
+
+
+def _check_single(a: Vector) -> None:
+    if len(a) == 0:
+        raise VectorError("EMPTY_VECTOR", "vector has no elements")
+
+
+def _finite(x: float) -> float:
+    if not math.isfinite(x):
+        raise VectorError("NON_FINITE", "result is infinite or NaN")
+    return x
+
+
+# --- SPEC section 2: the 8-lane reduction ------------------------------------
+
+
+def _combine(s: List[float]) -> float:
+    return ((s[0] + s[1]) + (s[2] + s[3])) + ((s[4] + s[5]) + (s[6] + s[7]))
+
+
+def _dot(a: Vector, b: Vector) -> float:
+    s = [0.0] * 8
+    for i in range(len(a)):
+        s[i & 7] += float(a[i]) * float(b[i])
+    return _combine(s)
+
+
+def _sumsq(a: Vector) -> float:
+    s = [0.0] * 8
+    for i in range(len(a)):
+        x = float(a[i])
+        s[i & 7] += x * x
+    return _combine(s)
+
+
+def _l2sq(a: Vector, b: Vector) -> float:
+    s = [0.0] * 8
+    for i in range(len(a)):
+        d = float(a[i]) - float(b[i])
+        s[i & 7] += d * d
+    return _combine(s)
+
+
+def _sqrt(x: float) -> float:
+    # x is a finite sum of squares, so it is never negative
+    return _finite(math.sqrt(x))
+
+
+def _div(x: float, y: float) -> float:
+    if y == 0.0:  # IEEE 754 gives an infinity or NaN; Python raises instead
+        raise VectorError("NON_FINITE", "result is infinite or NaN")
+    return _finite(x / y)
+
+
+def _cos(d: float, na: float, nb: float) -> float:
+    c = _div(d, na * nb)
+    return -1.0 if c < -1.0 else 1.0 if c > 1.0 else c
+
+
+# --- SPEC section 3: operations ----------------------------------------------
+
+
+def dot_product(a: Vector, b: Vector) -> float:
+    """Dot product."""
+    _check_pair(a, b)
+    return _finite(_dot(a, b))
+
+
+def inner_product(a: Vector, b: Vector) -> float:
+    """Same as :func:`dot_product` for real vectors."""
+    return dot_product(a, b)
+
+
+def l2_norm(a: Vector) -> float:
+    """Euclidean norm."""
+    _check_single(a)
+    return _sqrt(_finite(_sumsq(a)))
+
+
+def l2_distance(a: Vector, b: Vector) -> float:
+    """Euclidean distance."""
+    _check_pair(a, b)
+    return _sqrt(_finite(_l2sq(a, b)))
+
+
+def cosine_similarity(a: Vector, b: Vector) -> float:
+    """Cosine similarity clamped to [-1, 1]; ``ZERO_MAGNITUDE`` when either norm is 0."""
+    _check_pair(a, b)
+    d = _finite(_dot(a, b))
+    na = _sqrt(_finite(_sumsq(a)))
+    nb = _sqrt(_finite(_sumsq(b)))
+    if na == 0.0 or nb == 0.0:
+        raise VectorError("ZERO_MAGNITUDE", "vector has magnitude zero")
+    return _cos(d, na, nb)
+
+
+def normalize(a: Vector) -> List[float]:
+    """Unit-length copy: ``a[i] * (1 / norm)``."""
+    _check_single(a)
+    n = _sqrt(_finite(_sumsq(a)))
+    if n == 0.0:
+        raise VectorError("ZERO_MAGNITUDE", "vector has magnitude zero")
+    inv = _div(1.0, n)
+    return [_finite(float(x) * inv) for x in a]
+
+
+def vec_add(a: Vector, b: Vector) -> List[float]:
+    """Element-wise sum."""
+    _check_pair(a, b)
+    return [_finite(float(x) + float(y)) for x, y in zip(a, b)]
+
+
+def vec_sub(a: Vector, b: Vector) -> List[float]:
+    """Element-wise difference."""
+    _check_pair(a, b)
+    return [_finite(float(x) - float(y)) for x, y in zip(a, b)]
+
+
+def vec_mul_scalar(a: Vector, s: float) -> List[float]:
+    """Every element multiplied by ``s``."""
+    _check_single(a)
+    s = float(s)
+    return [_finite(float(x) * s) for x in a]
+
+
+# --- batch and matrix ----------------------------------------------------------
+
+
+def _rank(scores: List[float], descending: bool) -> List[ScoredIndex]:
+    # sorted() is stable, so equal scores keep index order
+    order = sorted(range(len(scores)), key=(lambda i: -scores[i]) if descending else (lambda i: scores[i]))
+    return [ScoredIndex(i, scores[i]) for i in order]
+
+
+def _check_candidate(query: Vector, c: Vector) -> None:
+    if len(c) != len(query):
+        raise VectorError("DIMENSION_MISMATCH", f"expected {len(query)} elements, got {len(c)}")
+
+
+def batch_cosine(query: Vector, candidates: Sequence[Vector]) -> List[ScoredIndex]:
+    """Cosine similarity of ``query`` with each candidate, descending; a zero candidate scores 0."""
+    _check_single(query)
+    qn = _sqrt(_finite(_sumsq(query)))
+    if qn == 0.0:
+        raise VectorError("ZERO_MAGNITUDE", "query has magnitude zero")
+    scores = []
+    for c in candidates:
+        _check_candidate(query, c)
+        d = _finite(_dot(query, c))
+        cn = _sqrt(_finite(_sumsq(c)))
+        scores.append(0.0 if cn == 0.0 else _cos(d, qn, cn))
+    return _rank(scores, True)
+
+
+def batch_dot(query: Vector, candidates: Sequence[Vector]) -> List[ScoredIndex]:
+    """Dot product of ``query`` with each candidate, descending."""
+    _check_single(query)
+    scores = []
+    for c in candidates:
+        _check_candidate(query, c)
+        scores.append(_finite(_dot(query, c)))
+    return _rank(scores, True)
+
+
+def batch_l2(query: Vector, candidates: Sequence[Vector]) -> List[ScoredIndex]:
+    """Euclidean distance from ``query`` to each candidate, ascending."""
+    _check_single(query)
+    scores = []
+    for c in candidates:
+        _check_candidate(query, c)
+        scores.append(_sqrt(_finite(_l2sq(query, c))))
+    return _rank(scores, False)
+
+
+def distance_matrix_cosine(a: Sequence[Vector], b: Sequence[Vector]) -> List[List[float]]:
+    """``out[i][j]`` = cosine similarity of ``a[i]`` and ``b[j]``; 0 when either has magnitude zero."""
+    if len(a) == 0 or len(b) == 0:
+        raise VectorError("EMPTY_VECTOR", "vector set is empty")
+    first = a[0]
+    _check_single(first)
+    for v in list(a) + list(b):
+        _check_candidate(first, v)
+    na = [_sqrt(_finite(_sumsq(v))) for v in a]
+    nb = [_sqrt(_finite(_sumsq(v))) for v in b]
+    out = []
+    for i, va in enumerate(a):
+        row = []
+        for j, vb in enumerate(b):
+            d = _finite(_dot(va, vb))
+            row.append(0.0 if na[i] == 0.0 or nb[j] == 0.0 else _cos(d, na[i], nb[j]))
+        out.append(row)
+    return out

lombokvector-0.1.0/lombokvector.egg-info/PKG-INFO

@@ -0,0 +1,29 @@
+Metadata-Version: 2.4
+Name: lombokvector
+Version: 0.1.0
+Summary: Vector math with bit-identical results across Python, Rust, TypeScript, Go and PHP: dot product, cosine similarity, L2 distance, normalization, batch ranking. Pure Python, zero dependencies.
+Author: codinglombok
+License-Expression: Apache-2.0
+Project-URL: Homepage, https://github.com/codinglombok/LombokVector
+Project-URL: Issues, https://github.com/codinglombok/LombokVector/issues
+Keywords: vector,cosine-similarity,embedding,deterministic
+Classifier: Programming Language :: Python :: 3
+Classifier: Operating System :: OS Independent
+Classifier: Topic :: Scientific/Engineering :: Mathematics
+Classifier: Typing :: Typed
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Dynamic: license-file
+
+# LombokVector for Python
+
+Python package `lombokvector` (binary64, Python 3.9+, pure Python): dot product, cosine similarity, Euclidean distance, normalization, and batch ranking with results that are bit-identical to the other LombokVector ports. Zero runtime dependencies.
+
+```
+pip install lombokvector
+```
+
+The summation order and every operation are defined in [docs/SPEC_LombokVector_v0.1.0.md](https://github.com/codinglombok/LombokVector/blob/main/docs/SPEC_LombokVector_v0.1.0.md); this port reproduces the shared vector file bit for bit in CI. Usage for all languages: [main README](https://github.com/codinglombok/LombokVector#readme) and [docs/API_LombokVector_v0.1.0.md](https://github.com/codinglombok/LombokVector/blob/main/docs/API_LombokVector_v0.1.0.md).
+
+License: Apache-2.0.

lombokvector-0.1.0/lombokvector.egg-info/SOURCES.txt

@@ -0,0 +1,11 @@
+LICENSE
+README.md
+pyproject.toml
+lombokvector/__init__.py
+lombokvector/py.typed
+lombokvector.egg-info/PKG-INFO
+lombokvector.egg-info/SOURCES.txt
+lombokvector.egg-info/dependency_links.txt
+lombokvector.egg-info/top_level.txt
+tests/test_api.py
+tests/test_vectors.py

lombokvector-0.1.0/lombokvector.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

lombokvector-0.1.0/lombokvector.egg-info/top_level.txt

@@ -0,0 +1 @@
+lombokvector

lombokvector-0.1.0/pyproject.toml

@@ -0,0 +1,34 @@
+[build-system]
+requires = ["setuptools>=77"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "lombokvector"
+version = "0.1.0"
+description = "Vector math with bit-identical results across Python, Rust, TypeScript, Go and PHP: dot product, cosine similarity, L2 distance, normalization, batch ranking. Pure Python, zero dependencies."
+readme = "README.md"
+license = "Apache-2.0"
+license-files = ["LICENSE"]
+authors = [{ name = "codinglombok" }]
+requires-python = ">=3.9"
+keywords = ["vector", "cosine-similarity", "embedding", "deterministic"]
+classifiers = [
+    "Programming Language :: Python :: 3",
+    "Operating System :: OS Independent",
+    "Topic :: Scientific/Engineering :: Mathematics",
+    "Typing :: Typed",
+]
+dependencies = []
+
+[project.urls]
+Homepage = "https://github.com/codinglombok/LombokVector"
+Issues = "https://github.com/codinglombok/LombokVector/issues"
+
+[tool.setuptools]
+packages = ["lombokvector"]
+
+[tool.setuptools.package-data]
+lombokvector = ["py.typed"]
+
+[tool.pytest.ini_options]
+testpaths = ["tests"]

lombokvector-0.1.0/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

lombokvector-0.1.0/tests/test_api.py

@@ -0,0 +1,68 @@
+import math
+
+import pytest
+
+import lombokvector as lv
+
+
+def code(c):
+    return pytest.raises(lv.VectorError, match=f"^{c}:")
+
+
+def test_basic_values():
+    assert lv.dot_product([1, 2, 3], [4, 5, 6]) == 32.0
+    assert lv.inner_product([1, 2, 3], [4, 5, 6]) == 32.0
+    assert lv.l2_norm([3, 4]) == 5.0
+    assert lv.l2_distance([1, 2, 3], [4, 5, 6]) == math.sqrt(27)
+    assert lv.cosine_similarity([1, 0], [0, 1]) == 0.0
+    assert lv.normalize([3, 4]) == [3 * 0.2, 4 * 0.2]
+    assert lv.vec_add([1, 2], [3, 4]) == [4.0, 6.0]
+    assert lv.vec_sub([1, 2], [3, 4]) == [-2.0, -2.0]
+    assert lv.vec_mul_scalar([1, 2], 3) == [3.0, 6.0]
+    assert lv.vec_mul_scalar([-2], 0) == [-0.0] and math.copysign(1, lv.vec_mul_scalar([-2], 0)[0]) == -1
+
+
+def test_lane_order():
+    assert lv.dot_product([1e16, 1, -1e16, 1], [1, 1, 1, 1]) == 0.0
+
+
+def test_errors():
+    with code("EMPTY_VECTOR"):
+        lv.dot_product([], [])
+    with code("DIMENSION_MISMATCH"):
+        lv.dot_product([1], [1, 2])
+    with code("ZERO_MAGNITUDE"):
+        lv.cosine_similarity([0, 0], [1, 1])
+    with code("ZERO_MAGNITUDE"):
+        lv.normalize([0])
+    with code("NON_FINITE"):
+        lv.l2_norm([math.nan])
+    with code("NON_FINITE"):
+        lv.vec_add([math.inf], [1])
+    with code("NON_FINITE"):
+        lv.cosine_similarity([1e200, 1e200], [1, 1])
+    with code("EMPTY_VECTOR"):
+        lv.vec_mul_scalar([], 1)
+    with code("ZERO_MAGNITUDE"):
+        lv.batch_cosine([0, 0], [])
+    with code("DIMENSION_MISMATCH"):
+        lv.batch_dot([1], [[1, 2]])
+    with code("DIMENSION_MISMATCH"):
+        lv.batch_l2([1], [[1, 2]])
+    with code("EMPTY_VECTOR"):
+        lv.distance_matrix_cosine([], [[1]])
+    with code("EMPTY_VECTOR"):
+        lv.distance_matrix_cosine([[]], [[]])
+    with code("DIMENSION_MISMATCH"):
+        lv.distance_matrix_cosine([[1, 0]], [[1]])
+    e = lv.VectorError("NON_FINITE", "x")
+    assert isinstance(e, ValueError) and e.code == "NON_FINITE" and str(e) == "NON_FINITE: x"
+
+
+def test_batch_ranking():
+    r = lv.batch_dot([1, 1], [[1, 0], [0, 1], [2, -1], [-1, 2]])
+    assert [x.index for x in r] == [0, 1, 2, 3]
+    assert lv.batch_l2([0, 0], [[3, 4], [1, 0], [0, 1]]) == [(1, 1.0), (2, 1.0), (0, 5.0)]
+    assert [x.index for x in lv.batch_cosine([1, 1], [[0, 0], [1, 1]])] == [1, 0]
+    assert lv.distance_matrix_cosine([[1, 0], [0, 0]], [[1, 0]]) == [[1.0], [0.0]]
+    assert lv.__version__ == "0.1.0"

lombokvector-0.1.0/tests/test_vectors.py

@@ -0,0 +1,54 @@
+"""Runs the binary64 cases of the shared vectors: every bit pattern must match."""
+
+import json
+import pathlib
+import struct
+
+import pytest
+
+import lombokvector as lv
+
+DOC = json.loads((pathlib.Path(__file__).resolve().parents[2] / "vectors" / "lombokvector-vectors-v1.json").read_text(encoding="utf-8"))
+CASES = [c for c in DOC["cases"] if c["precision"] == "f64"]
+
+
+def bits(x):
+    return "0x%016x" % struct.unpack("<Q", struct.pack("<d", x))[0]
+
+
+def run(op, a):
+    if op == "dot":
+        return bits(lv.dot_product(a[0], a[1]))
+    if op == "norm":
+        return bits(lv.l2_norm(a[0]))
+    if op == "l2":
+        return bits(lv.l2_distance(a[0], a[1]))
+    if op == "cosine":
+        return bits(lv.cosine_similarity(a[0], a[1]))
+    if op == "normalize":
+        return [bits(x) for x in lv.normalize(a[0])]
+    if op == "add":
+        return [bits(x) for x in lv.vec_add(a[0], a[1])]
+    if op == "sub":
+        return [bits(x) for x in lv.vec_sub(a[0], a[1])]
+    if op == "scale":
+        return [bits(x) for x in lv.vec_mul_scalar(a[0], a[1])]
+    if op in ("batch_cosine", "batch_dot", "batch_l2"):
+        fn = {"batch_cosine": lv.batch_cosine, "batch_dot": lv.batch_dot, "batch_l2": lv.batch_l2}[op]
+        return [[r.index, bits(r.score)] for r in fn(a[0], a[1])]
+    if op == "matrix_cosine":
+        return [[bits(x) for x in row] for row in lv.distance_matrix_cosine(a[0], a[1])]
+    raise ValueError(op)
+
+
+def test_at_least_100_cases():
+    assert len(CASES) >= 100
+
+
+@pytest.mark.parametrize("case", CASES, ids=[c["id"] for c in CASES])
+def test_vector(case):
+    try:
+        got = {"value": run(case["op"], case["args"])}
+    except lv.VectorError as e:
+        got = {"error": e.code}
+    assert got == case["expected"]