./devel/py-cython, C-Extensions for Python

[ CVSweb ] [ Homepage ] [ RSS ] [ Required by ] [ Add to tracker ]


Branch: CURRENT, Version: 0.29, Package name: py27-cython-0.29, Maintainer: pkgsrc-users

The Cython language makes writing C extensions for the Python language
as easy as Python itself. Cython is a source code translator based on
the well-known Pyrex, but supports more cutting edge functionality and
optimizations.

The Cython language is very close to the Python language (and most
Python code is also valid Cython code), but Cython additionally supports
calling C functions and declaring C types on variables and class
attributes. This allows the compiler to generate very efficient C code
from Cython code.

This makes Cython the ideal language for writing glue code for external
C libraries, and for fast C modules that speed up the execution of
Python code.


Required to run:
[devel/py-setuptools] [lang/python27]

Required to build:
[pkgtools/cwrappers]

Master sites:

SHA1: efe76eb5511a4f089cc6719b98624f31916b9258
RMD160: 613bcee93bb497a5b951554554464511a1110eec
Filesize: 1989.431 KB

Version history: (Expand)


CVS history: (Expand)


   2018-10-15 10:21:03 by Adam Ciarcinski | Files touched by this commit (5) | Package updated
Log message:
py-cython: updated to 0.29

0.29:

Features added
--------------
PEP-489 multi-phase module initialisation has been enabled again. Module reloads \ 
in other subinterpreters raise an exception to prevent corruption of the static \ 
module state.
A set of mypy compatible PEP-484 declarations were added for Cython’s C data \ 
types to integrate with static analysers in typed Python code. They are \ 
available in the Cython/Shadow.pyi module and describe the types in the special \ 
cython module that can be used for typing in Python code.
Memoryviews are supported in PEP-484/526 style type declarations.
@cython.nogil is supported as a C-function decorator in Python code.
Raising exceptions from nogil code will automatically acquire the GIL, instead \ 
of requiring an explicit with gil block.
C++ functions can now be declared as potentially raising both C++ and Python \ 
exceptions, so that Cython can handle both correctly.
cython.inline() supports a direct language_level keyword argument that was \ 
previously only available via a directive.
A new language level name 3str was added that mostly corresponds to language \ 
level 3, but keeps unprefixed string literals as type ‘str’ in both Py2 and \ 
Py3, and the builtin ‘str’ type unchanged. This will become the default in \ 
the next Cython release and is meant to help user code a) transition more easily \ 
to this new default and b) migrate to Python 3 source code semantics without \ 
making support for Python 2.x difficult.
In CPython 3.6 and later, looking up globals in the module dict is almost as \ 
fast as looking up C globals.
For a Python subclass of an extension type, repeated method calls to \ 
non-overridden cpdef methods can avoid the attribute lookup in Py3.6+, which \ 
makes them 4x faster.
(In-)equality comparisons of objects to integer literals are faster.
Some internal and 1-argument method calls are faster.
Modules that cimport many external extension types from other Cython modules \ 
execute less import requests during module initialisation.
Constant tuples and slices are deduplicated and only created once per module.
The coverage plugin considers more C file extensions such as .cc and .cxx.
The cythonize command accepts compile time variable values (as set by DEF) \ 
through the new -E option.
pyximport can import from namespace packages.
Some missing numpy and CPython C-API declarations were added.
Declarations for the pylifecycle C-API functions were added in a new .pxd file \ 
cpython.pylifecycle.
The Pythran support was updated to work with the latest Pythran 0.8.7
%a is included in the string formatting types that are optimised into f-strings. \ 
In this case, it is also automatically mapped to %r in Python 2.x.
New C macro CYTHON_HEX_VERSION to access Cython’s version in the same style as \ 
PY_HEX_VERSION.
Constants in libc.math are now declared as const to simplify their handling.
An additional check_size clause was added to the ctypedef class name \ 
specification to allow suppressing warnings when importing modules with \ 
backwards-compatible PyTypeObject size changes.

Bugs fixed
----------
The exception handling in generators and coroutines under CPython 3.7 was \ 
adapted to the newly introduced exception stack. Users of Cython 0.28 who want \ 
to support Python 3.7 are encouraged to upgrade to 0.29 to avoid potentially \ 
incorrect error reporting and tracebacks.
Crash when importing a module under Stackless Python that was built for CPython.
2-value slicing of typed sequences failed if the start or stop index was None.
Multiplied string literals lost their factor when they are part of another \ 
constant expression (e.g. ‘x’ * 10 + ‘y’ => ‘xy’).
String formatting with the ‘%’ operator didn’t call the special __rmod__() \ 
method if the right side is a string subclass that implements it.
The directive language_level=3 did not apply to the first token in the source file.
Overriding cpdef methods did not work in Python subclasses with slots. Note that \ 
this can have a performance impact on calls from Cython code.
Fix declarations of builtin or C types using strings in pure python mode.
Generator expressions and lambdas failed to compile in @cfunc functions.
Global names with const types were not excluded from star-import assignments \ 
which could lead to invalid C code.
Several internal function signatures were fixed that lead to warnings in gcc-8.
The numpy helper functions set_array_base() and get_array_base() were adapted to \ 
the current numpy C-API recommendations.
Some NumPy related code was updated to avoid deprecated API usage.
Several C++ STL declarations were extended and corrected.
C lines of the module init function were unconditionally not reported in \ 
exception stack traces.
When PEP-489 support is enabled, reloading the module overwrote any static \ 
module state. It now raises an exception instead, given that reloading is not \ 
actually supported.
Object-returning, C++ exception throwing functions were not checking that the \ 
return value was non-null.
The source file encoding detection could get confused if the c_string_encoding \ 
directive appeared within the first two lines.
Cython generated modules no longer emit a warning during import when the size of \ 
the NumPy array type is larger than what was found at compile time. Instead, \ 
this is assumed to be a backwards compatible change on NumPy side.

Other changes
-------------
Cython now emits a warning when no language_level (2, 3 or ‘3str’) is set \ 
explicitly, neither as a cythonize() option nor as a compiler directive. This is \ 
meant to prepare the transition of the default language level from currently Py2 \ 
to Py3, since that is what most new users will expect these days. The future \ 
default will, however, not enforce unicode literals, because this has proven a \ 
major obstacle in the support for both Python 2.x and 3.x. The next major \ 
release is intended to make this change, so that it will parse all code that \ 
does not request a specific language level as Python 3 code, but with str \ 
literals. The language level 2 will continue to be supported for an indefinite \ 
time.
The documentation was restructured, cleaned up and examples are now tested. The \ 
NumPy tutorial was also rewritten to simplify the running example.
Cython compiles less of its own modules at build time to reduce the installed \ 
package size to about half of its previous size. This makes the compiler \ 
slightly slower, by about 5-7%.
   2018-08-07 11:36:42 by Adam Ciarcinski | Files touched by this commit (2) | Package updated
Log message:
py-cython: updated to 0.28.5

0.28.5:

Bugs fixed

* The discouraged usage of GCC's attribute optimize("Os") was replaced \ 
by the
  similar attribute cold to reduce the code impact of the module init functions.

* A reference leak in Py2.x was fixed when comparing str to unicode for equality.
   2018-07-09 08:34:28 by Adam Ciarcinski | Files touched by this commit (2) | Package updated
Log message:
py-cython: updated to 0.28.4

0.28.4:
Bugs fixed
* Reallowing tp_clear() in a subtype of an @no_gc_clear extension type generated \ 
an invalid C function call to the (non-existent) base type implementation.
* Exception catching based on a non-literal (runtime) tuple could fail to match \ 
the exception.
* Compile fix for CPython 3.7.0a2.
   2018-05-29 13:45:06 by Adam Ciarcinski | Files touched by this commit (2) | Package updated
Log message:
py-cython: updated to 0.28.3

0.28.3:
Bugs fixed
* Set iteration was broken in non-CPython since 0.28.
* UnicodeEncodeError in Py2 when %s formatting is optimised for unicode strings.
* Work around a crash bug in g++ 4.4.x by disabling the size reduction setting \ 
of the module init function in this version.
   2018-04-13 08:55:45 by Adam Ciarcinski | Files touched by this commit (2) | Package updated
Log message:
py-cython: updated to 0.28.2

0.28.2 (2018-04-13)
Features added
abs() is faster for Python long objects.
The C++11 methods front() and end() were added to the declaration of libcpp.string.
The C++11 methods reserve() and bucket_count() are declared for libcpp.unordered_map.

Bugs fixed
The copy of a read-only memoryview was considered read-only as well, whereas a \ 
common reason to copy a read-only view is to make it writable. The result of the \ 
copying is now a writable buffer by default.
The switch statement generation failed to apply recursively to the body of \ 
converted if-statements.
NULL was sometimes rejected as exception return value when the returned type is \ 
a fused pointer type.
Fixed compatibility with PyPy 5.11.

Other changes
The NumPy tutorial was rewritten to use memoryviews instead of the older buffer \ 
declaration syntax.
   2018-03-19 09:22:54 by Adam Ciarcinski | Files touched by this commit (2) | Package updated
Log message:
py-cython: updated to 0.28.1

0.28.1:
Bugs fixed
PyFrozenSet_New() was accidentally used in PyPy where it is missing from the C-API.
Assignment between some C++ templated types were incorrectly rejected when the \ 
templates mix const with ctypedef.
Undeclared C++ no-args constructors in subclasses could make the compilation \ 
fail if the base class constructor was declared without nogil.
Bytes %-formatting inferred basestring (bytes or unicode) as result type in some \ 
cases where bytes would have been safe to infer.
None was accidentally disallowed as typed return value of dict.pop().
   2018-03-15 10:30:18 by Adam Ciarcinski | Files touched by this commit (3) | Package updated
Log message:
py-cython: updated to 0.28

0.28:
Cdef classes can now multiply inherit from ordinary Python classes.
Type inference is now supported for Pythran compiled NumPy expressions.
The const modifier can be applied to memoryview declarations to allow read-only \ 
buffers as input.
C code in the docstring of a cdef extern block is copied verbatimly into the \ 
generated file. Patch by Jeroen Demeyer.
When compiling with gcc, the module init function is now tuned for small code \ 
size instead of whatever compile flags were provided externally. Cython now also \ 
disables some code intensive optimisations in that function to further reduce \ 
the code size.
Decorating an async coroutine with @cython.iterable_coroutine changes its type \ 
at compile time to make it iterable. While this is not strictly in line with \ 
PEP-492, it improves the interoperability with old-style coroutines that use \ 
yield from instead of await.
The IPython magic has preliminary support for JupyterLab.
The new TSS C-API in CPython 3.7 is supported and has been backported. Patch by \ 
Naotoshi Seo.
Cython knows the new Py_tss_t type defined in PEP-539 and automatically \ 
initialises variables declared with that type to Py_tss_NEEDS_INIT, a value \ 
which cannot be used outside of static assignments.
The set methods .remove() and .discard() are optimised. Patch by Antoine Pitrou.
dict.pop() is optimised. Original patch by Antoine Pitrou.
Iteration over sets and frozensets is optimised.
Safe integer loops (< range(2^30)) are automatically optimised into C loops.
alist.extend([a,b,c]) is optimised into sequential list.append() calls for short \ 
literal sequences.
Calls to builtin methods that are not specifically optimised into C-API calls \ 
now use a cache that avoids repeated lookups of the underlying C function.
Single argument function calls can avoid the argument tuple creation in some cases.
Some redundant extension type checks are avoided.
Formatting C enum values in f-strings is faster, as well as some other special cases.
String formatting with the ‘%’ operator is optimised into f-strings in \ 
simple cases.
Subscripting (item access) is faster in some cases.
Some bytearray operations have been optimised similar to bytes.
Some PEP-484/526 container type declarations are now considered for loop \ 
optimisations.
Indexing into memoryview slices with view[i][j] is now optimised into view[i, j].
Python compatible cython.* types can now be mixed with type declarations in \ 
Cython syntax.
Name lookups in the module and in classes are faster.
Python attribute lookups on extension types without instance dict are faster.
Some missing signals were added to libc/signal.pxd. Patch by Jeroen Demeyer.
The warning about repeated extern declarations is now visible by default.
The exception handling of the function types used by CPython’s type slot \ 
functions was corrected to match the de-facto standard behaviour, so that code \ 
that uses them directly benefits from automatic and correct exception \ 
propagation. Patch by Jeroen Demeyer.
Defining the macro CYTHON_NO_PYINIT_EXPORT will prevent the module init function \ 
from being exported as symbol, e.g. when linking modules statically in an \ 
embedding setup. Patch by AraHaan.
   2017-11-06 11:12:29 by Adam Ciarcinski | Files touched by this commit (2) | Package updated
Log message:
py-cython: updated to 0.27.3

0.27.3:

Bugs fixed

* String forward references to extension types like \ 
``@cython.locals(x="ExtType")``
  failed to find the named type.

* NumPy slicing generated incorrect results when compiled with Pythran.

* Fix "undefined reference" linker error for generators on Windows in \ 
Py3.3-3.5.

* Adapt to recent C-API change of ``PyThreadState`` in CPython 3.7.

* Fix signature of ``PyWeakref_GetObject()`` API declaration.