gh-61103: support double complex (_Complex) type in ctypes #120894

skirpichev · 2024-06-23T05:11:55Z

Example:

>>> import ctypes
>>> libm = ctypes.CDLL('libm.so.6')
>>> libm.clog.argtypes = [ctypes.c_double_complex]
>>> libm.clog.restype = ctypes.c_double_complex
>>> libm.clog(1+1j)
(0.34657359027997264+0.7853981633974483j)

ctypes.c_double_complex is available only if compiler does support complex arithmetic (Annex G).

Issue: Add support for C99 complex type (_Complex) as ctypes.c_complex #61103

📚 Documentation preview 📚: https://cpython-previews--120894.org.readthedocs.build/

Notes for reviewers:

~~It seems, most compilers implement this optional feature of C11+. (Though neither does this correctly.) Maybe we should require one?~~
only one complex type was added; if this looks good - I can add the rest (float complex and long double complex) in this pr or in other.
c_double_complex._type_ chosen to be "C". Maybe we should allow instead multiple chars to specify types (e.g. "Cd" for double complex)?
this pr doesn't include changes in the struct module. Perhaps, I should add support for complex type here as well?
if test_complex_round_trip() does make sense for reviewers, here is a simple refactoring to avoid adding yet another assertFloatsAreIdentical() helper: gh-121039: add Floats/ComplexesAreIdenticalMixin to test.support.testcase #121071

Example: ```pycon >>> import ctypes >>> ctypes.__STDC_IEC_559_COMPLEX__ 1 >>> libm = ctypes.CDLL('libm.so.6') >>> libm.clog.argtypes = [ctypes.c_double_complex] >>> libm.clog.restype = ctypes.c_double_complex >>> libm.clog(1+1j) (0.34657359027997264+0.7853981633974483j) ``` ``ctypes.__STDC_IEC_559_COMPLEX__`` is ``0`` if compiler doesn't support complex arithmetic (Annex G).

Lib/test/test_ctypes/test_libc.py

Co-authored-by: Nice Zombies <nineteendo19d0@gmail.com>

Doc/library/ctypes.rst

Co-authored-by: Nice Zombies <nineteendo19d0@gmail.com>

Doc/library/ctypes.rst

Misc/NEWS.d/next/Library/2024-06-23-07-23-08.gh-issue-61103.ca_U_l.rst

Modules/_ctypes/_ctypes.c

Modules/_ctypes/cfield.c

Tools/c-analyzer/c_parser/parser/_regexes.py

Modules/_ctypes/cfield.c

Doc/library/ctypes.rst

picnixz · 2024-06-23T11:00:37Z

I can add the rest (float complex and long double complex) in this pr or in other.

I'd prefer having a separate PR because we might need aliases in case sizeof(long double) == sizeof(double) (the same technique that is used for long long and long IIRC)

c_double_complex.type chosen to be "C". Maybe we should allow instead multiple chars to specify types (e.g. "Cd" for double complex)?

I think it might be indeed a good alternative, because otherwise you need to find good 1-char names for float complex and long double complex in the future, but I think Cf, Cd and Cl (or Cq for quad-precision) are good enough. But for that we need to update the fielddesc structure with an additional field that would be a "modifier" of some sort or allow the "code" to be more than just a single character.

this pr doesn't include changes in the struct module. Perhaps, I should add support for complex type here as well?

I'd say in a separate PR since there is a translation of ctypes format to PEP 3118 names (where complex numbers are represented by 'Z').

Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>

nineteendo · 2024-06-23T17:46:25Z

Use a force-push only to fix typo's or revert a merge.

skirpichev · 2024-06-23T18:03:07Z

Use a force-push only to fix typo's or revert a merge.

Thank you for advice. Usually I don't do this. I hope this time force-push was OK, as it's restricted to unreviewed commits.

nineteendo · 2024-06-23T18:10:01Z

It's less bad if no-one commented on those commits, but it's still more work to figure out what you changed.
It's also more distracting than a regular commit. (also, you don't need to quote the whole previous message)

* add configure test for _Complex type * define HAVE_C_COMPLEX * move workarounds for buggy implementations to Module/_complex.h

vstinner · 2024-06-30T09:58:50Z

!buildbot s390x SLES

bedevere-bot · 2024-06-30T09:58:53Z

🤖 New build scheduled with the buildbot fleet by @vstinner for commit e9cab4a 🤖

The command will test the builders whose names match following regular expression: s390x SLES

The builders matched are:

s390x SLES PR

skirpichev · 2024-06-30T11:07:16Z

Well, one failure seems to be related:
https://buildbot.python.org/all/#/builders/1262/builds/13/steps/3/logs/stdio
Edit (similar one):
https://buildbot.python.org/all/#/builders/855/builds/119

Looking on the libffi sources, it seems that FFI_TYPE_COMPLEX is available unconditionally, even if platform doesn't support interfaces to _Complex (for sparc it works since v3.3). We should check FFI_TARGET_HAS_COMPLEX_TYPE instead.

I'll update PR when build finish.

vstinner · 2024-06-30T17:47:02Z

!buildbot AMD64 Fedora Rawhide PR

bedevere-bot · 2024-06-30T17:47:04Z

🤖 New build scheduled with the buildbot fleet by @vstinner for commit 6d5bf66 🤖

The command will test the builders whose names match following regular expression: AMD64 Fedora Rawhide PR

The builders matched are:

AMD64 Fedora Rawhide PR

vstinner · 2024-06-30T17:49:10Z

!buildbot AMD64 Fedora Rawhide PR

bedevere-bot · 2024-06-30T17:49:15Z

🤖 New build scheduled with the buildbot fleet by @vstinner for commit 6d5bf66 🤖

The command will test the builders whose names match following regular expression: AMD64 Fedora Rawhide PR

The builders matched are:

AMD64 Fedora Rawhide PR

vstinner · 2024-06-30T17:49:36Z

Oh, Rawhide failure is unrelated: "No space left on device".

vstinner · 2024-06-30T17:50:45Z

!buildbot PPC64LE CentOS9 PR

bedevere-bot · 2024-06-30T17:50:48Z

🤖 New build scheduled with the buildbot fleet by @vstinner for commit 6d5bf66 🤖

The command will test the builders whose names match following regular expression: PPC64LE CentOS9 PR

The builders matched are:

PPC64LE CentOS9 PR

vstinner · 2024-06-30T18:21:38Z

!buildbot PPC64LE Fedora Stable Clang PR

bedevere-bot · 2024-06-30T18:21:41Z

🤖 New build scheduled with the buildbot fleet by @vstinner for commit 6d5bf66 🤖

The command will test the builders whose names match following regular expression: PPC64LE Fedora Stable Clang PR

The builders matched are:

PPC64LE Fedora Stable Clang PR

bedevere-bot · 2024-07-01T07:21:26Z

🤖 New build scheduled with the buildbot fleet by @vstinner for commit 6d5bf66 🤖

If you want to schedule another build, you need to add the 🔨 test-with-buildbots label again.

vstinner · 2024-07-01T08:56:18Z

Thanks @skirpichev, I merged your PR.

I checked remaining buildbot failures: they are unrelated to the change.

@skirpichev: @erlend-aasland has concerns about relative #include.

skirpichev · 2024-07-01T10:17:08Z

Thanks for reviews and patience.

@erlend-aasland has concerns about relative #include.

Yep, here; I'll open an issue (I think that build failures in this pr already don't need this;)). There are other includes, besides _math.h.

…hon#120894) Example: ```pycon >>> import ctypes >>> ctypes.__STDC_IEC_559_COMPLEX__ 1 >>> libm = ctypes.CDLL('libm.so.6') >>> libm.clog.argtypes = [ctypes.c_double_complex] >>> libm.clog.restype = ctypes.c_double_complex >>> libm.clog(1+1j) (0.34657359027997264+0.7853981633974483j) ``` Co-authored-by: Nice Zombies <nineteendo19d0@gmail.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com> Co-authored-by: Victor Stinner <vstinner@python.org>

bedevere-app bot mentioned this pull request Jun 23, 2024

Add support for C99 complex type (_Complex) as ctypes.c_complex #61103

Closed

bedevere-app bot added the awaiting review label Jun 23, 2024

skirpichev added 3 commits June 23, 2024 08:15

+ ifdef

b402300

+ fix rst formatting

ad39e6d

+ CMPLX

fd9135b

nineteendo reviewed Jun 23, 2024

View reviewed changes

Lib/test/test_ctypes/test_libc.py Outdated Show resolved Hide resolved

Lib/test/test_ctypes/test_libc.py Outdated Show resolved Hide resolved

skirpichev and others added 2 commits June 23, 2024 09:00

Apply suggestions from code review

c296fb3

Co-authored-by: Nice Zombies <nineteendo19d0@gmail.com>

+ support _Complex in c-analyzer

7453ca5

skirpichev requested a review from ericsnowcurrently as a code owner June 23, 2024 07:12

skirpichev added 4 commits June 23, 2024 10:47

+1

e444c9c

expand docs

2e90643

drop ctypes.__STDC_IEC_559_COMPLEX__

32ffacf

And more tests, finally)

9163950

skirpichev requested a review from nineteendo June 23, 2024 09:36

nineteendo reviewed Jun 23, 2024

View reviewed changes

Doc/library/ctypes.rst Outdated Show resolved Hide resolved

Doc/library/ctypes.rst Show resolved Hide resolved

Doc/library/ctypes.rst Outdated Show resolved Hide resolved

Apply suggestions from code review

bf81682

Co-authored-by: Nice Zombies <nineteendo19d0@gmail.com>

picnixz reviewed Jun 23, 2024

View reviewed changes

skirpichev and others added 2 commits June 23, 2024 14:47

Apply suggestions from code review

af2fdce

Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>

+ allow _Complex specifier before float/double/etc

6d9d5ab

skirpichev force-pushed the ctypes-c_complex-61103 branch from 45f8544 to 013765e Compare June 23, 2024 13:07

+ comment on clang workaround

f9c709c

skirpichev force-pushed the ctypes-c_complex-61103 branch from 013765e to f9c709c Compare June 23, 2024 17:33

Address review:

ccd418d

* add configure test for _Complex type * define HAVE_C_COMPLEX * move workarounds for buggy implementations to Module/_complex.h

skirpichev requested review from erlend-aasland and corona10 as code owners June 24, 2024 03:47

sed: FFI_TYPE_COMPLEX -> FFI_TARGET_HAS_COMPLEX_TYPE

6d5bf66

vstinner added the 🔨 test-with-buildbots Test PR w/ buildbots; report in status section label Jul 1, 2024

bedevere-bot removed the 🔨 test-with-buildbots Test PR w/ buildbots; report in status section label Jul 1, 2024

vstinner merged commit 6988ff0 into python:main Jul 1, 2024
78 of 90 checks passed

bedevere-app bot removed the awaiting merge label Jul 1, 2024

skirpichev deleted the ctypes-c_complex-61103 branch July 1, 2024 09:56

skirpichev mentioned this pull request Jul 1, 2024

Should we allow relative includes? (i.e. #include "../_ssl.h") #121223

Open

skirpichev mentioned this pull request Jul 10, 2024

gh-121039: add Floats/ComplexesAreIdenticalMixin to test.support.testcase #121071

Merged

This was referenced Oct 7, 2024

gh-121249: Support _Complex types in the struct module #121613

Merged

CTypes test failed, complex double problem #125206

Closed

gh-125206: Bug in ctypes with old libffi is fixed #125322

Merged

skirpichev mentioned this pull request Jan 22, 2025

implement PEP 3118 struct changes #47382

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

gh-61103: support double complex (_Complex) type in ctypes #120894

gh-61103: support double complex (_Complex) type in ctypes #120894

gh-61103: support double complex (_Complex) type in ctypes #120894

gh-61103: support double complex (_Complex) type in ctypes #120894

Conversation