charex 0.5.2

charex 0.5.2

charex 0.5.2 is the first PyPI publishing release for the Numba 0.65.1
compatibility line. The PyPI distribution is published as charex-numba; the
import package remains charex.

Highlights

• Keeps the 0.5.1 string behavior and fixed-width scalar broadcasting fix.
• Adds the PyPI Trusted Publishing workflow for release artifacts.
• Refreshes the README introduction and quick-start example for PyPI users.
• Publishes release artifacts from CI rather than local builds.

Compatibility

• Python >=3.10,<3.15
• Numba >=0.65.1,<0.66
• NumPy supported/tested window: >=1.22,<1.27 or >=2.0,<2.5
• llvmlite 0.47.x

np.strings support is conditional on NumPy 2.x.

Not In Scope

• New string operations beyond the 0.5.0 read-only catalog.
• Fresh benchmark artifacts.
• Transformation/output-producing operations such as replace, case conversion,
  strip, pad, join, split, encode, and decode.

charex 0.5.1

charex 0.5.1

charex 0.5.1 is a maintenance release for the Numba 0.65.1 compatibility
line. It includes the fixed-width scalar broadcasting parity fix layered after
0.5.0 and prepares the project for PyPI publication as charex-numba.

Compatibility

• Python >=3.10,<3.15
• Numba >=0.65.1,<0.66
• NumPy supported/tested window: >=1.22,<1.27 or >=2.0,<2.5
• llvmlite 0.47.x

np.strings support is conditional on NumPy 2.x.

Highlights

• Fixes fixed-width S/U N-D array plus Python scalar broadcasting for
  np.char and np.strings comparisons and occurrence methods.
• Keeps the public API import-only: import charex registers the overloads.
• Moves implementation modules under private charex._* names.
• Moves tests and benchmarks out of the installable charex package namespace.
• Prepares the PyPI distribution as charex-numba while preserving the import
  package name charex.
• Adds PyPI metadata and source-archive manifest rules for cleaner release
  artifacts.

Not In Scope

• New string operations beyond the 0.5.0 read-only catalog.
• Fresh benchmark artifacts.
• Transformation/output-producing operations such as replace, case conversion,
  strip, pad, join, split, encode, and decode.

charex 0.5.0

charex 0.5.0

charex 0.5.0 is the shape/layout release for the Numba 0.65.1 compatibility
line. It extends the read-only string operation surface from scalar, 0-D, and
1-D inputs to NumPy-matching N-D and broadcast-compatible shapes.

Compatibility

• Python >=3.10,<3.15
• Numba >=0.65.1,<0.66
• NumPy supported/tested window: >=1.22,<1.27 or >=2.0,<2.5
• llvmlite 0.47.x

np.strings support is conditional on NumPy 2.x.

Highlights

• Supports N-D and broadcast-compatible shapes for fixed-width S/U arrays
  through both np.char and np.strings.
• Supports N-D and broadcast-compatible shapes for NumPy 2.x variable-width
  StringDType arrays through np.strings.
• Supports contiguous arrays, read-only views, positive and negative strides,
  zero-stride views, and empty views for the supported read-only catalog.
• Supports default StringDType() and StringDType(na_object=...) variants
  with NumPy-matching operation-specific null behavior.
• Preserves separate np.char and np.strings semantics, including the
  trailing whitespace/NUL behavior difference.
• Keeps transformation/output-producing operations outside this release scope.

Supported Read-Only Catalog

• comparisons: equal, not_equal, greater, greater_equal, less,
  less_equal;
• occurrence/search: count, startswith, endswith, find, rfind,
  index, rindex;
• information/predicates: str_len, isalpha, isalnum, isdigit,
  isdecimal, isnumeric, isspace, islower, isupper, istitle;
• np.char.compare_chararrays for fixed-width S/U.

Parity Audit

The full shape audit on Python 3.12.8, NumPy 2.4.6, and Numba 0.65.1 reports:

• rows: 1702
• matching rows: 1702
• mismatches: 0
• NumPy accepts but charex rejects: 0

The audit CSV is written to
docs/exploration/string_array_shape_audit.csv; the summary is maintained in
docs/string-array-shape-parity.md.

Benchmarks

The Numba 0.65.1 benchmark matrix in
docs/benchmarks/numba-v-0.65.1 includes
fixed-width np.char inputs and NumPy 2.x StringDType inputs through
np.strings.

The current matrix reports a 1.60x median speedup across 135 fixed-width and
StringDType cases, with results ranging from 1.02x to 6.51x NumPy speed.

Not In Scope

• Transformation/output-producing operations such as replace, case conversion,
  strip, pad, join, split, encode, and decode.
• Object array bridges.
• Max-performance experimental kernels that have not been distilled.

charex 0.4.1

charex 0.4.1

charex 0.4.1 is the pre-N-D checkpoint for the Numba 0.65.1 compatibility
work. It updates the package for the NumPy 1.x/2.x support window and lands the
current scalar, 0-D, and 1-D string operation surface before the larger
N-D/broadcasting work begins.

0.4.1 is a patch update over 0.4.0: it keeps the same public operation
surface, tightens StringDType empty/NUL edge-case coverage, avoids unsafe
zero-length libc comparison calls in generated code, and refreshes benchmark
plot readability.

Compatibility

• Python >=3.10,<3.15
• Numba >=0.65.1,<0.66
• NumPy supported/tested window: >=1.22,<1.27 or >=2.0,<2.5
• llvmlite 0.47.x

np.strings support is conditional on NumPy 2.x.

Highlights

• Adds NumPy 2.x np.strings overloads for the read-only catalog.
• Adds NumPy 2.x StringDType support for scalar, 0-D, and 1-D inputs.
• Supports default StringDType() and StringDType(na_object=...) variants.
• Supports StringDType comparisons, occurrence/search, and information
  predicates for the same read-only method families as fixed-width strings.
• Supports fixed-width S/U scalar, 0-D, and 1-D inputs, including strided,
  reversed, zero-stride, read-only, and empty views.
• Preserves separate np.char and np.strings semantics, including the
  trailing whitespace/NUL behavior difference.
• Adds packaging metadata and the native charex._stringdtype helper required
  for StringDType access.

Supported Read-Only Catalog

• comparisons: equal, not_equal, greater, greater_equal, less,
  less_equal;
• occurrence/search: count, startswith, endswith, find, rfind,
  index, rindex;
• information/predicates: str_len, isalpha, isalnum, isdigit,
  isdecimal, isnumeric, isspace, islower, isupper, istitle;
• np.char.compare_chararrays for fixed-width S/U.

Not In Scope

• N-D arrays and general broadcasting.
• Transformation/output-producing operations such as replace, case conversion,
  strip, pad, join, split, encode, and decode.
• Object array bridges.
• Max-performance experimental kernels that have not been distilled.

Next Milestone

0.5.0 is reserved for the shape/layout release: fixed-width and StringDType
N-D same-shape support, general broadcasting, and the corresponding audit and
benchmark refresh.

Benchmarks

The Numba 0.65.1 benchmark matrix in
docs/benchmarks/numba-v-0.65.1 includes
fixed-width np.char inputs and NumPy 2.x StringDType inputs through
np.strings.

charex 0.4.0

charex 0.4.0 is the pre-N-D checkpoint for the Numba 0.65.1 compatibility
work. It updates the package for the NumPy 1.x/2.x support window and lands the
current scalar, 0-D, and 1-D string operation surface before the larger
N-D/broadcasting work begins.

Compatibility

• Python >=3.10,<3.15
• Numba >=0.65.1,<0.66
• NumPy supported/tested window: >=1.22,<1.27 or >=2.0,<2.5
• llvmlite 0.47.x

np.strings support is conditional on NumPy 2.x.

Highlights

• Adds NumPy 2.x np.strings overloads for the read-only catalog.
• Adds NumPy 2.x StringDType support for scalar, 0-D, and 1-D inputs.
• Supports default StringDType() and StringDType(na_object=...) variants.
• Supports StringDType comparisons, occurrence/search, and information
  predicates for the same read-only method families as fixed-width strings.
• Supports fixed-width S/U scalar, 0-D, and 1-D inputs, including strided,
  reversed, zero-stride, read-only, and empty views.
• Preserves separate np.char and np.strings semantics, including the
  trailing whitespace/NUL behavior difference.
• Adds packaging metadata and the native charex._stringdtype helper required
  for StringDType access.

Supported Read-Only Catalog

• comparisons: equal, not_equal, greater, greater_equal, less,
  less_equal;
• occurrence/search: count, startswith, endswith, find, rfind,
  index, rindex;
• information/predicates: str_len, isalpha, isalnum, isdigit,
  isdecimal, isnumeric, isspace, islower, isupper, istitle;
• np.char.compare_chararrays for fixed-width S/U.

Not In Scope

• N-D arrays and general broadcasting.
• Transformation/output-producing operations such as replace, case conversion,
  strip, pad, join, split, encode, and decode.
• Object array bridges.
• Max-performance experimental kernels that have not been distilled.

Next Milestone

0.5.0 is reserved for the shape/layout release: fixed-width and StringDType
N-D same-shape support, general broadcasting, and the corresponding audit and
benchmark refresh.

v0.1.0

First packaged release checkpoint for charex.

Highlights:

• Targets Numba 0.65.1 with llvmlite 0.47.x.
• Supports Python >=3.10,<3.15.
• Supports NumPy >=1.22,<1.27 and >=2.0,<2.5.
• Registers Numba overloads for NumPy fixed-width string APIs: np.char and NumPy 2.x np.strings.
• Covers comparison, occurrence, length, and string predicate operations for scalar and one-dimensional C-contiguous fixed-width S/U inputs.
• Includes refreshed tests, benchmark harnesses, and Numba 0.65.1 benchmark artifacts.

Not included in this release:

• NumPy StringDType support.
• N-dimensional array support and full NumPy broadcasting.
• Transformation-method coverage for the full NumPy string catalog.