From stefan_ml at behnel.de Sat Sep 3 04:46:35 2016 From: stefan_ml at behnel.de (Stefan Behnel) Date: Sat, 3 Sep 2016 10:46:35 +0200 Subject: [Cython] Minor issue with running tests and -Wlong-long In-Reply-To: References: Message-ID: Erik Bray schrieb am 25.04.2016 um 17:18: > As some of you already know I've been doing some work on Cython on > Cygwin [in the process I I'm constantly mixing the two up in speech, > but maybe in writing I'll do better :)]. > > There are several issues with the tests on Cygwin, and that's one > thing I'll work on. But a major annoyance I've encountered when > running any tests is a huge number of warnings from gcc such as: > > embray at PC-pret-47 ~/src/cython > $ CFLAGS="-O0" ./runtests.py -vv --no-cpp addloop > Python 2.7.10 (default, Jun 1 2015, 18:05:38) > [GCC 4.9.2] > > Running tests against Cython 0.24 f68b5bd0fa620d0dc26166bffe5fe42d94068720 > Backends: c > > runTest (__main__.CythonRunTestCase) > compiling (c) and running addloop ... > === C/C++ compiler error output: === > In file included from /usr/include/python2.7/Python.h:58:0, > from addloop.c:4: > /usr/include/python2.7/pyport.h:69:27: warning: ISO C90 does not > support ?long long? [-Wlong-long] > #define PY_LONG_LONG long long > ^ > /usr/include/python2.7/pyport.h:793:34: note: in definition of macro > ?PyAPI_FUNC? > # define PyAPI_FUNC(RTYPE) RTYPE > ^ > /usr/include/python2.7/intobject.h:46:21: note: in expansion of macro > ?PY_LONG_LONG? > PyAPI_FUNC(unsigned PY_LONG_LONG) PyInt_AsUnsignedLongLongMask(PyObject *); > ^ > In file included from /usr/include/python2.7/Python.h:58:0, > from addloop.c:4: > ... > > And so on. > > For now an easy workaround is to add -Wno-long-long to the compiler > flags. But I was curious why I was seeing this in Cygwin but not on > my Ubuntu system, and here's why: > > In runtests.py there is a check > > https://github.com/cython/cython/blob/master/runtests.py#L829 > > if self.language=='c' and compiler='gcc': > ext_compile_flags.extend(['-std=c89', '-pedantic']) > > where in this case `compiler` is assigned > `sysconfig.get_config_var('CC')`. On my Linux system this expands to > "x86_64-linux-gnu-gcc -pthread". Whereas on Cygwin (and probably many > other systems) it expands simply to "gcc". > > I'm guessing that to do what it intended the above line should read > "and 'gcc' in compiler". Yes. > But this also raises the question: Why are the tests run with these > flags? If Python was configured with HAVE_LONG_LONG, then these > warnings will be inevitable. Thanks for bringing this up. Adding "-Wno-long-long" seems the right thing to do, given that this depends on CPython and not Cython. Given that CPython *can* be configured without PY_LONG_LONG support, however, we should then guard any such code in Cython with HAVE_LONG_LONG, as CPython does internally, so that it actually compiles without it. I would otherwise like to keep those compiler flags, even if they didn't serve us yet, because it is not intended to put all too many constraints on the C compiler. Supporting C89, at least whenever possible, would be nice. Stefan From erik.m.bray at gmail.com Mon Sep 5 04:09:11 2016 From: erik.m.bray at gmail.com (Erik Bray) Date: Mon, 5 Sep 2016 10:09:11 +0200 Subject: [Cython] Minor issue with running tests and -Wlong-long In-Reply-To: References: Message-ID: On Sat, Sep 3, 2016 at 10:46 AM, Stefan Behnel wrote: > Erik Bray schrieb am 25.04.2016 um 17:18: >> As some of you already know I've been doing some work on Cython on >> Cygwin [in the process I I'm constantly mixing the two up in speech, >> but maybe in writing I'll do better :)]. >> >> There are several issues with the tests on Cygwin, and that's one >> thing I'll work on. But a major annoyance I've encountered when >> running any tests is a huge number of warnings from gcc such as: >> >> embray at PC-pret-47 ~/src/cython >> $ CFLAGS="-O0" ./runtests.py -vv --no-cpp addloop >> Python 2.7.10 (default, Jun 1 2015, 18:05:38) >> [GCC 4.9.2] >> >> Running tests against Cython 0.24 f68b5bd0fa620d0dc26166bffe5fe42d94068720 >> Backends: c >> >> runTest (__main__.CythonRunTestCase) >> compiling (c) and running addloop ... >> === C/C++ compiler error output: === >> In file included from /usr/include/python2.7/Python.h:58:0, >> from addloop.c:4: >> /usr/include/python2.7/pyport.h:69:27: warning: ISO C90 does not >> support ?long long? [-Wlong-long] >> #define PY_LONG_LONG long long >> ^ >> /usr/include/python2.7/pyport.h:793:34: note: in definition of macro >> ?PyAPI_FUNC? >> # define PyAPI_FUNC(RTYPE) RTYPE >> ^ >> /usr/include/python2.7/intobject.h:46:21: note: in expansion of macro >> ?PY_LONG_LONG? >> PyAPI_FUNC(unsigned PY_LONG_LONG) PyInt_AsUnsignedLongLongMask(PyObject *); >> ^ >> In file included from /usr/include/python2.7/Python.h:58:0, >> from addloop.c:4: >> ... >> >> And so on. >> >> For now an easy workaround is to add -Wno-long-long to the compiler >> flags. But I was curious why I was seeing this in Cygwin but not on >> my Ubuntu system, and here's why: >> >> In runtests.py there is a check >> >> https://github.com/cython/cython/blob/master/runtests.py#L829 >> >> if self.language=='c' and compiler='gcc': >> ext_compile_flags.extend(['-std=c89', '-pedantic']) >> >> where in this case `compiler` is assigned >> `sysconfig.get_config_var('CC')`. On my Linux system this expands to >> "x86_64-linux-gnu-gcc -pthread". Whereas on Cygwin (and probably many >> other systems) it expands simply to "gcc". >> >> I'm guessing that to do what it intended the above line should read >> "and 'gcc' in compiler". > > Yes. > >> But this also raises the question: Why are the tests run with these >> flags? If Python was configured with HAVE_LONG_LONG, then these >> warnings will be inevitable. > > Thanks for bringing this up. Adding "-Wno-long-long" seems the right thing > to do, given that this depends on CPython and not Cython. > > Given that CPython *can* be configured without PY_LONG_LONG support, > however, we should then guard any such code in Cython with HAVE_LONG_LONG, > as CPython does internally, so that it actually compiles without it. > > I would otherwise like to keep those compiler flags, even if they didn't > serve us yet, because it is not intended to put all too many constraints on > the C compiler. Supporting C89, at least whenever possible, would be nice. An alternative could be to just remove them on cygwin. Or maybe, though I haven't tested this I don't think, changing `-std=c89` to `-std=gnu89`. newlib, which is the libc implementation used by Cygwin, has some bugs in its headers that can crop up in odd ways when using `-std=c`, but with the GNU extensions it's usually a bit more permissive. From robertwb at gmail.com Fri Sep 9 04:11:33 2016 From: robertwb at gmail.com (Robert Bradshaw) Date: Fri, 9 Sep 2016 01:11:33 -0700 Subject: [Cython] Cython release Message-ID: I'd like to put out another Cython release shortly. Is anyone aware of anything that we should try to get in before then? - Robert From stefan_ml at behnel.de Fri Sep 9 07:33:47 2016 From: stefan_ml at behnel.de (Stefan Behnel) Date: Fri, 9 Sep 2016 13:33:47 +0200 Subject: [Cython] Cython release In-Reply-To: References: Message-ID: Robert Bradshaw schrieb am 09.09.2016 um 10:11: > I'd like to put out another Cython release shortly. Exactly my thoughts. The next release could use some extra good testing as I reworked some of the fused-/cy-function internals last month and might not have found all kinds of code that could break along the way. There are so many feature combinations these days: decorators, py+pxd, @classmethod vs. classmethod(f), def/cdef/cpdef, ... > Is anyone aware of > anything that we should try to get in before then? I've been working on adapting the PEP 525 (async generators) implementation recently that is proposed for CPython, but that's not finished, neither in CPython nor in Cython, so don't wait for it. Happy to get 0.25 out soon. Stefan From jdemeyer at cage.ugent.be Fri Sep 9 15:03:19 2016 From: jdemeyer at cage.ugent.be (Jeroen Demeyer) Date: Fri, 9 Sep 2016 21:03:19 +0200 Subject: [Cython] Cython release In-Reply-To: References: Message-ID: <57D30777.9050403@cage.ugent.be> On 2016-09-09 10:11, Robert Bradshaw wrote: > I'd like to put out another Cython release shortly. Do you mean a 0.24.x release or 0.25? Anyway, this is a serious bug breaking SageMath: https://github.com/cython/cython/issues/1433 From robertwb at gmail.com Fri Sep 9 17:05:42 2016 From: robertwb at gmail.com (Robert Bradshaw) Date: Fri, 9 Sep 2016 14:05:42 -0700 Subject: [Cython] Cython release In-Reply-To: <57D30777.9050403@cage.ugent.be> References: <57D30777.9050403@cage.ugent.be> Message-ID: On Fri, Sep 9, 2016 at 12:03 PM, Jeroen Demeyer wrote: > On 2016-09-09 10:11, Robert Bradshaw wrote: >> >> I'd like to put out another Cython release shortly. > > > Do you mean a 0.24.x release or 0.25? > > Anyway, this is a serious bug breaking SageMath: > https://github.com/cython/cython/issues/1433 Hmm... yes. https://github.com/cython/cython/commit/c3f2d97778838067d9c6cf96fa45ab5a8c9ebcb0 Not sure what caused this to regress (or how it worked before) but it's all better now. From robertwb at math.washington.edu Fri Sep 9 17:05:57 2016 From: robertwb at math.washington.edu (Robert Bradshaw) Date: Fri, 9 Sep 2016 14:05:57 -0700 Subject: [Cython] Cython release In-Reply-To: References: Message-ID: On Fri, Sep 9, 2016 at 4:33 AM, Stefan Behnel wrote: > Robert Bradshaw schrieb am 09.09.2016 um 10:11: >> I'd like to put out another Cython release shortly. > > Exactly my thoughts. > > The next release could use some extra good testing as I reworked some of > the fused-/cy-function internals last month and might not have found all > kinds of code that could break along the way. There are so many feature > combinations these days: decorators, py+pxd, @classmethod vs. > classmethod(f), def/cdef/cpdef, ... > > >> Is anyone aware of >> anything that we should try to get in before then? > > I've been working on adapting the PEP 525 (async generators) implementation > recently that is proposed for CPython, but that's not finished, neither in > CPython nor in Cython, so don't wait for it. > > Happy to get 0.25 out soon. Any thoughts on https://github.com/cython/cython/pull/1456 ? From stefan_ml at behnel.de Sat Sep 10 05:45:05 2016 From: stefan_ml at behnel.de (Stefan Behnel) Date: Sat, 10 Sep 2016 11:45:05 +0200 Subject: [Cython] buffer format test failure in Py3.6 - Re: Cython release In-Reply-To: References: Message-ID: Robert Bradshaw schrieb am 09.09.2016 um 10:11: > I'd like to put out another Cython release shortly. Is anyone aware of > anything that we should try to get in before then? I'm seeing these test failures in travis under latest Py3.6 now. Does anyone have a quick idea what might cause them? https://api.travis-ci.org/jobs/158935556/log.txt?deansi=true (see the very end of the test log) Stefan FAIL: numpy_test (numpy_test.__test__) Doctest: numpy_test.__test__.numpy_test ---------------------------------------------------------------------- Traceback (most recent call last): File "/opt/python/3.6-dev/lib/python3.6/doctest.py", line 2199, in runTest raise self.failureException(self.format_failure(new.getvalue())) AssertionError: Failed doctest test for numpy_test.__test__.numpy_test File "/home/travis/build/cython/cython/BUILD/1/run/c/numpy_test/numpy_test.cpython-36m-x86_64-linux-gnu.so", line unknown line number, in numpy_test ---------------------------------------------------------------------- File "/home/travis/build/cython/cython/BUILD/1/run/c/numpy_test/numpy_test.cpython-36m-x86_64-linux-gnu.so", line ?, in numpy_test.__test__.numpy_test Failed example: test_dtype(np.longlong, inc1_longlong_t) Exception raised: Traceback (most recent call last): File "/opt/python/3.6-dev/lib/python3.6/doctest.py", line 1330, in __run compileflags, 1), test.globs) File "", line 1, in test_dtype(np.longlong, inc1_longlong_t) File "tests/run/numpy_test.pyx", line 397, in numpy_test.test_dtype (numpy_test.c:8884) inc1(a) File "tests/run/numpy_test.pyx", line 374, in numpy_test.inc1_longlong_t (numpy_test.c:7797) def inc1_longlong_t(np.ndarray[np.longlong_t] arr): arr[1] += 1 ValueError: Unexpected format string character: 'q' From jdemeyer at cage.ugent.be Wed Sep 14 02:21:47 2016 From: jdemeyer at cage.ugent.be (Jeroen Demeyer) Date: Wed, 14 Sep 2016 08:21:47 +0200 Subject: [Cython] Cython release In-Reply-To: References: Message-ID: <57D8EC7B.10304@cage.ugent.be> On 2016-09-09 10:11, Robert Bradshaw wrote: > I'd like to put out another Cython release shortly. Is anyone aware of > anything that we should try to get in before then? I just built the SageMath distribution with latest Cython master and there were 2 issues (both easily fixed): - Build issue with pyzmq: https://github.com/zeromq/pyzmq/pull/909 - Replaced assignment of Cython.Compiler.Options.directive_defaults by passing directives to cythonize() From robertwb at gmail.com Wed Sep 14 11:51:17 2016 From: robertwb at gmail.com (Robert Bradshaw) Date: Wed, 14 Sep 2016 08:51:17 -0700 Subject: [Cython] Cython release In-Reply-To: <57D8EC7B.10304@cage.ugent.be> References: <57D8EC7B.10304@cage.ugent.be> Message-ID: Thanks! On Tue, Sep 13, 2016 at 11:21 PM, Jeroen Demeyer wrote: > On 2016-09-09 10:11, Robert Bradshaw wrote: >> >> I'd like to put out another Cython release shortly. Is anyone aware of >> anything that we should try to get in before then? > > > I just built the SageMath distribution with latest Cython master and there > were 2 issues (both easily fixed): > > - Build issue with pyzmq: > https://github.com/zeromq/pyzmq/pull/909 > > - Replaced assignment of Cython.Compiler.Options.directive_defaults by > passing directives to cythonize() > > _______________________________________________ > cython-devel mailing list > cython-devel at python.org > https://mail.python.org/mailman/listinfo/cython-devel From al007 at illinois.edu Wed Sep 14 15:23:02 2016 From: al007 at illinois.edu (Alexander Lindsay) Date: Wed, 14 Sep 2016 14:23:02 -0500 Subject: [Cython] cygdb: no debug files were found In-Reply-To: References: Message-ID: Try: hg purge --all python-dbg setup.py build_ext --inplace Also make sure that gdb is configured with python2; on new Debian-based distributions, gdb comes configured with python3, whereas cygdb relies on a gdb configured with python2. Alex From stefan_ml at behnel.de Sun Sep 18 07:27:49 2016 From: stefan_ml at behnel.de (Stefan Behnel) Date: Sun, 18 Sep 2016 13:27:49 +0200 Subject: [Cython] buffer format test failure in Py3.6 - Re: Cython release In-Reply-To: References: Message-ID: Stefan Behnel schrieb am 10.09.2016 um 11:45: > Robert Bradshaw schrieb am 09.09.2016 um 10:11: >> I'd like to put out another Cython release shortly. Is anyone aware of >> anything that we should try to get in before then? > > I'm seeing these test failures in travis under latest Py3.6 now. Does > anyone have a quick idea what might cause them? > > https://api.travis-ci.org/jobs/158935556/log.txt?deansi=true Found it. The CPython devs removed the "HAVE_LONG_LONG" macro because "long long" support has been required since CPython 2.7/3.3. This causes our code that uses PY_LONG_LONG to be discarded rather than enabled. https://bugs.python.org/issue27961 I asked them to restore the macro definition. I also made Cython define it if it's not. https://github.com/cython/cython/commit/eeef5da28562aac390d34a8a4b8ec35ef03266ec Stefan From robertwb at gmail.com Sun Sep 25 00:47:41 2016 From: robertwb at gmail.com (Robert Bradshaw) Date: Sat, 24 Sep 2016 21:47:41 -0700 Subject: [Cython] Cython 0.25 alpha0 Message-ID: Please download and test the upcoming release of Cython: https://github.com/cython/cython/archive/0.25a0.tar.gz This is not the final release (more pull requests and bug fixes may still go in) but a lot has changed since the last release so it'd be good to get feedback early. https://github.com/cython/cython/blob/master/CHANGES.rst Thanks, Robert From robertwb at gmail.com Sun Sep 25 01:12:56 2016 From: robertwb at gmail.com (Robert Bradshaw) Date: Sat, 24 Sep 2016 22:12:56 -0700 Subject: [Cython] Cython 0.25 alpha0 In-Reply-To: References: Message-ID: This prerelease version is also available at PyPi https://pypi.python.org/pypi/Cython/0.25a0 installable via pip install --pre cython On Sat, Sep 24, 2016 at 9:47 PM, Robert Bradshaw wrote: > Please download and test the upcoming release of Cython: > > https://github.com/cython/cython/archive/0.25a0.tar.gz > > This is not the final release (more pull requests and bug fixes may > still go in) but a lot has changed since the last release so it'd be > good to get feedback early. > > https://github.com/cython/cython/blob/master/CHANGES.rst > > Thanks, > Robert From matthew.brett at gmail.com Sun Sep 25 02:05:03 2016 From: matthew.brett at gmail.com (Matthew Brett) Date: Sat, 24 Sep 2016 23:05:03 -0700 Subject: [Cython] [cython-users] Re: Cython 0.25 alpha0 In-Reply-To: References: Message-ID: On Sat, Sep 24, 2016 at 10:12 PM, Robert Bradshaw wrote: > This prerelease version is also available at PyPi > > https://pypi.python.org/pypi/Cython/0.25a0 > > installable via > > pip install --pre cython It looks like the new BUILD file is breaking wheel builds on OSX and Windows: https://ci.appveyor.com/project/matthew-brett/cython-wheels/build/1.0.20/job/eu6n2eyc07a33ojl https://travis-ci.org/MacPython/cython-wheels/jobs/162527865#L235 Can it be renamed? Cheers, Matthew From isuruf at gmail.com Sun Sep 25 03:02:17 2016 From: isuruf at gmail.com (Isuru Fernando) Date: Sun, 25 Sep 2016 12:32:17 +0530 Subject: [Cython] [cython-users] Re: Cython 0.25 alpha0 In-Reply-To: References: Message-ID: Hi, With Cython 0.25a0, following causes a compiler crash. from libcpp.vector cimport vector cdef vector[double complex] v Regards, Isuru Fernando On Sun, Sep 25, 2016 at 11:35 AM, Matthew Brett wrote: > On Sat, Sep 24, 2016 at 10:12 PM, Robert Bradshaw > wrote: > > This prerelease version is also available at PyPi > > > > https://pypi.python.org/pypi/Cython/0.25a0 > > > > installable via > > > > pip install --pre cython > > It looks like the new BUILD file is breaking wheel builds on OSX and > Windows: > > https://ci.appveyor.com/project/matthew-brett/cython- > wheels/build/1.0.20/job/eu6n2eyc07a33ojl > https://travis-ci.org/MacPython/cython-wheels/jobs/162527865#L235 > > Can it be renamed? > > Cheers, > > Matthew > _______________________________________________ > cython-devel mailing list > cython-devel at python.org > https://mail.python.org/mailman/listinfo/cython-devel > -------------- next part -------------- An HTML attachment was scrubbed... URL: From robertwb at gmail.com Sun Sep 25 03:14:18 2016 From: robertwb at gmail.com (Robert Bradshaw) Date: Sun, 25 Sep 2016 00:14:18 -0700 Subject: [Cython] [cython-users] Re: Cython 0.25 alpha0 In-Reply-To: References: Message-ID: Thanks for the report; not sure what changed here. Could you file an issue on github against the 0.25 milestone? On Sun, Sep 25, 2016 at 12:02 AM, Isuru Fernando wrote: > Hi, > > With Cython 0.25a0, following causes a compiler crash. > > from libcpp.vector cimport vector > cdef vector[double complex] v > > > Regards, > > Isuru Fernando > > On Sun, Sep 25, 2016 at 11:35 AM, Matthew Brett > wrote: >> >> On Sat, Sep 24, 2016 at 10:12 PM, Robert Bradshaw >> wrote: >> > This prerelease version is also available at PyPi >> > >> > https://pypi.python.org/pypi/Cython/0.25a0 >> > >> > installable via >> > >> > pip install --pre cython >> >> It looks like the new BUILD file is breaking wheel builds on OSX and >> Windows: >> >> >> https://ci.appveyor.com/project/matthew-brett/cython-wheels/build/1.0.20/job/eu6n2eyc07a33ojl >> https://travis-ci.org/MacPython/cython-wheels/jobs/162527865#L235 >> >> Can it be renamed? >> >> Cheers, >> >> Matthew >> _______________________________________________ >> cython-devel mailing list >> cython-devel at python.org >> https://mail.python.org/mailman/listinfo/cython-devel > > > > _______________________________________________ > cython-devel mailing list > cython-devel at python.org > https://mail.python.org/mailman/listinfo/cython-devel > From joshua.adelman at gmail.com Sun Sep 25 20:29:13 2016 From: joshua.adelman at gmail.com (Joshua) Date: Sun, 25 Sep 2016 17:29:13 -0700 (PDT) Subject: [Cython] Cython 0.25 alpha0 In-Reply-To: References: Message-ID: <19001158-f268-49c4-b58a-77031a50444a@googlegroups.com> On Sunday, September 25, 2016 at 12:48:17 AM UTC-4, Robert Bradshaw wrote: > > Please download and test the upcoming release of Cython: > > https://github.com/cython/cython/archive/0.25a0.tar.gz > > This is not the final release (more pull requests and bug fixes may > still go in) but a lot has changed since the last release so it'd be > good to get feedback early. > > https://github.com/cython/cython/blob/master/CHANGES.rst > > Thanks, > Robert > I can confirm the BUILD issue on OSX mentioned by Matthew. I ended up running `mv BUILD{,.bak}` to get cython to compile. Also 0.25a0 still suffers from the regression reported in this issue: https://github.com/cython/cython/issues/1466 I'd be thrilled if a fix could make it into the 0.25 release since I'm effectively stuck at 0.22.1 due to the combination of this bug and another regression that was fixed in 0.24. Thanks again for all of the hard work the Cython devs put into this project. Josh -------------- next part -------------- An HTML attachment was scrubbed... URL: From matthew.brett at gmail.com Mon Sep 26 13:00:52 2016 From: matthew.brett at gmail.com (Matthew Brett) Date: Mon, 26 Sep 2016 10:00:52 -0700 Subject: [Cython] [cython-users] Re: Cython 0.25 alpha0 In-Reply-To: <41c06423-95c0-4db9-b1fa-9620ce6a5448@googlegroups.com> References: <41c06423-95c0-4db9-b1fa-9620ce6a5448@googlegroups.com> Message-ID: Hi, On Mon, Sep 26, 2016 at 1:42 AM, Alexey Buluy wrote: > > Who is this smart guy who decided to release this version into the open > without proper testing? > Now all packages in pip which use Cython are failing to install: > >> sudo pip install cassandra-driver==2.7.2 >> Downloading/unpacking cassandra-driver==2.7.2 >> Running setup.py egg_info for package cassandra-driver >> Unable to find pgen, not compiling formal grammar. >> Compiling >> /tmp/easy_install-AEclgi/Cython-0.25a0/Cython/Plex/Scanners.py because it >> changed. >> Compiling >> /tmp/easy_install-AEclgi/Cython-0.25a0/Cython/Plex/Actions.py because it >> changed. >> Compiling >> /tmp/easy_install-AEclgi/Cython-0.25a0/Cython/Compiler/Lexicon.py because it >> changed. >> Compiling >> /tmp/easy_install-AEclgi/Cython-0.25a0/Cython/Compiler/Scanning.py because >> it changed. >> Compiling >> /tmp/easy_install-AEclgi/Cython-0.25a0/Cython/Compiler/Parsing.py because it >> changed. >> Compiling >> /tmp/easy_install-AEclgi/Cython-0.25a0/Cython/Compiler/Visitor.py because it >> changed. >> Compiling >> /tmp/easy_install-AEclgi/Cython-0.25a0/Cython/Compiler/FlowControl.py >> because it changed. >> Compiling >> /tmp/easy_install-AEclgi/Cython-0.25a0/Cython/Compiler/Code.py because it >> changed. >> Compiling >> /tmp/easy_install-AEclgi/Cython-0.25a0/Cython/Tempita/_tempita.py because it >> changed. >> [1/9] Cythonizing >> /tmp/easy_install-AEclgi/Cython-0.25a0/Cython/Compiler/Code.py >> [2/9] Cythonizing >> /tmp/easy_install-AEclgi/Cython-0.25a0/Cython/Compiler/FlowControl.py >> [3/9] Cythonizing >> /tmp/easy_install-AEclgi/Cython-0.25a0/Cython/Compiler/Lexicon.py >> [4/9] Cythonizing >> /tmp/easy_install-AEclgi/Cython-0.25a0/Cython/Compiler/Parsing.py >> [5/9] Cythonizing >> /tmp/easy_install-AEclgi/Cython-0.25a0/Cython/Compiler/Scanning.py >> [6/9] Cythonizing >> /tmp/easy_install-AEclgi/Cython-0.25a0/Cython/Compiler/Visitor.py >> [7/9] Cythonizing >> /tmp/easy_install-AEclgi/Cython-0.25a0/Cython/Plex/Actions.py >> [8/9] Cythonizing >> /tmp/easy_install-AEclgi/Cython-0.25a0/Cython/Plex/Scanners.py >> [9/9] Cythonizing >> /tmp/easy_install-AEclgi/Cython-0.25a0/Cython/Tempita/_tempita.py >> warning: no files found matching '*.pyx' under directory >> 'Cython/Debugger/Tests' >> warning: no files found matching '*.pxd' under directory >> 'Cython/Debugger/Tests' >> warning: no files found matching '*.h' under directory >> 'Cython/Debugger/Tests' >> warning: no files found matching '*.pxd' under directory >> 'Cython/Utility' >> gcc: error: >> /tmp/easy_install-AEclgi/Cython-0.25a0/Cython/Runtime/refnanny.c: No such >> file or directory >> gcc: fatal error: no input files >> compilation terminated. > > > Anybody knows how to fix this? That's odd - have you set up pip to fetch ``--pre`` packages by default somehow? When I do a pip install on my machine, I get the stable release. The pre-releases, like this one, are precisely to shake out problems before the stable releases. What do you get for a plain: pip install cython Best, Matthew