• Bug#1108185: bookworm-pu: package prody/2.3.1+dfsg-3+deb12u1

    From Adam D. Barratt@21:1/5 to Santiago Vila on Sat Jul 5 18:30:01 2025
    XPost: linux.debian.devel.release

    On Sun, 2025-06-22 at 17:49 +0200, Santiago Vila wrote:
    This fixes Bug #1028722 prody: FTBFS: AssertionError: 3205 != 3211

    [ Impact ]
    Anybody trying to build the package from source may unexpectedly
    find that the package does not build depending on particular
    hardware.

    [ Tests ]
    I've tested that the package builds ok again on the systems where
    it used to fail.

    Although they obviousy aren't the result of the change in this bug, for
    the record the package still fails to build on i386 with several test
    failures. Two of the three test failures seen in p-u also appear in the
    current unstable packages.

    <quote>
    ======================================================================
    FAIL: testBuildMatrixCoordsArray (prody.tests.dynamics.test_enms.TestANM.testBuildMatrixCoordsArray)
    Test output when *coords* is an array. ---------------------------------------------------------------------- Traceback (most recent call last):
    File "/build/reproducible-path/prody- 2.3.1+dfsg/.pybuild/cpython3_3.11_prody/build/prody/tests/dynamics/test _enms.py", line 288, in testBuildMatrixCoordsArray
    assert_equal(self.getMatrix(), self.getExpected(),
    File "/usr/lib/python3/dist-
    packages/numpy/testing/_private/utils.py", line 349, in assert_equal
    return assert_array_equal(actual, desired, err_msg, verbose)
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    File "/usr/lib/python3/dist-
    packages/numpy/testing/_private/utils.py", line 985, in
    assert_array_equal
    assert_array_compare(operator.__eq__, x, y, err_msg=err_msg,
    File "/usr/lib/python3.11/contextlib.py", line 81, in inner
    return func(*args, **kwds)
    ^^^^^^^^^^^^^^^^^^^
    File "/usr/lib/python3/dist-
    packages/numpy/testing/_private/utils.py", line 862, in
    assert_array_compare
    raise AssertionError(msg)
    AssertionError:
    Arrays are not equal
    failed to get correct matrix
    Mismatched elements: 3 / 51984 (0.00577%)
    Max absolute difference: 4.4408921e-16
    Max relative difference: 7.83963072e-16
    x: array([[ 6.913751, 0.021868, 0.94014 , ..., 0. , 0. ,
    0. ],
    [ 0.021868, 9.066915, -0.076779, ..., 0. , 0. ,...
    y: array([[ 6.913751, 0.021868, 0.94014 , ..., 0. , 0. ,
    0. ],
    [ 0.021868, 9.066915, -0.076779, ..., 0. , 0. ,...

    ======================================================================
    FAIL: testHessianSymmetry (prody.tests.dynamics.test_enms.TestANMResults.testHessianSymmetry) ---------------------------------------------------------------------- Traceback (most recent call last):
    File "/build/reproducible-path/prody- 2.3.1+dfsg/.pybuild/cpython3_3.11_prody/build/prody/tests/dynamics/test _enms.py", line 123, in testHessianSymmetry
    assert_equal(hessian, hessian.T, 'hessian is not symmetric')
    File "/usr/lib/python3/dist-
    packages/numpy/testing/_private/utils.py", line 349, in assert_equal
    return assert_array_equal(actual, desired, err_msg, verbose)
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    File "/usr/lib/python3/dist-
    packages/numpy/testing/_private/utils.py", line 985, in
    assert_array_equal
    assert_array_compare(operator.__eq__, x, y, err_msg=err_msg,
    File "/usr/lib/python3.11/contextlib.py", line 81, in inner
    return func(*args, **kwds)
    ^^^^^^^^^^^^^^^^^^^
    File "/usr/lib/python3/dist-
    packages/numpy/testing/_private/utils.py", line 862, in
    assert_array_compare
    raise AssertionError(msg)
    AssertionError:
    Arrays are not equal
    hessian is not symmetric
    Mismatched elements: 6 / 51984 (0.0115%)
    Max absolute difference: 4.4408921e-16
    Max relative difference: 7.83963072e-16
    x: array([[ 6.913751, 0.021868, 0.94014 , ..., 0. , 0. ,
    0. ],
    [ 0.021868, 9.066915, -0.076779, ..., 0. , 0. ,...
    y: array([[ 6.913751, 0.021868, 0.94014 , ..., 0. , 0. ,
    0. ],
    [ 0.021868, 9.066915, -0.076779, ..., 0. , 0. ,...

    ======================================================================
    FAIL: testHessianSymmetry (prody.tests.dynamics.test_enms.TestEXANMResults.testHessianSymmetry) ---------------------------------------------------------------------- Traceback (most recent call last):
    File "/build/reproducible-path/prody- 2.3.1+dfsg/.pybuild/cpython3_3.11_prody/build/prody/tests/dynamics/test _enms.py", line 174, in testHessianSymmetry
    assert_equal(hessian, hessian.T, 'hessian is not symmetric')
    File "/usr/lib/python3/dist-
    packages/numpy/testing/_private/utils.py", line 349, in assert_equal
    return assert_array_equal(actual, desired, err_msg, verbose)
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    File "/usr/lib/python3/dist-
    packages/numpy/testing/_private/utils.py", line 985, in
    assert_array_equal
    assert_array_compare(operator.__eq__, x, y, err_msg=err_msg,
    File "/usr/lib/python3.11/contextlib.py", line 81, in inner
    return func(*args, **kwds)
    ^^^^^^^^^^^^^^^^^^^
    File "/usr/lib/python3/dist-
    packages/numpy/testing/_private/utils.py", line 862, in
    assert_array_compare
    raise AssertionError(msg)
    AssertionError:
    Arrays are not equal
    hessian is not symmetric
    Mismatched elements: 218 / 13089924 (0.00167%)
    Max absolute difference: 8.8817842e-16
    Max relative difference: 4.00832339e-16
    x: array([[ 8.024368, 0.273151, 4.136877, ..., 0. , 0. ,
    0. ],
    [ 0.273151, 3.353604, -0.105517, ..., 0. , 0. ,...
    y: array([[ 8.024368, 0.273151, 4.136877, ..., 0. , 0. ,
    0. ],
    [ 0.273151, 3.353604, -0.105517, ..., 0. , 0. ,...

    ----------------------------------------------------------------------
    Ran 952 tests in 57.319s

    FAILED (failures=3, skipped=37)
    </quote>

    Regards,

    Adam

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Adam D. Barratt@21:1/5 to Adam D. Barratt on Sat Jul 5 18:40:02 2025
    XPost: linux.debian.devel.release

    On Sat, 2025-07-05 at 17:27 +0100, Adam D. Barratt wrote:
    On Sun, 2025-06-22 at 17:49 +0200, Santiago Vila wrote:
    This fixes Bug #1028722 prody: FTBFS: AssertionError: 3205 != 3211

    [ Impact ]
    Anybody trying to build the package from source may unexpectedly
    find that the package does not build depending on particular
    hardware.

    [ Tests ]
    I've tested that the package builds ok again on the systems where
    it used to fail.

    Although they obviousy aren't the result of the change in this bug,
    for the record the package still fails to build on i386 with several
    test failures. Two of the three test failures seen in p-u also appear
    in the current unstable packages.

    This appears to be #1058454, which has been open for a while now. :-(

    Regards,

    Adam

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Santiago Vila@21:1/5 to Adam D. Barratt on Sat Jul 5 23:20:01 2025
    XPost: linux.debian.devel.release

    On Sat, Jul 05, 2025 at 05:32:15PM +0100, Adam D. Barratt wrote:
    On Sat, 2025-07-05 at 17:27 +0100, Adam D. Barratt wrote:
    On Sun, 2025-06-22 at 17:49 +0200, Santiago Vila wrote:
    This fixes Bug #1028722 prody: FTBFS: AssertionError: 3205 != 3211

    [ Impact ]
    Anybody trying to build the package from source may unexpectedly
    find that the package does not build depending on particular
    hardware.

    [ Tests ]
    I've tested that the package builds ok again on the systems where
    it used to fail.

    Although they obviousy aren't the result of the change in this bug,
    for the record the package still fails to build on i386 with several
    test failures. Two of the three test failures seen in p-u also appear
    in the current unstable packages.

    This appears to be #1058454, which has been open for a while now. :-(

    Yes, that's what it seems, but additionally the package currently in unstable does not build for other reasons (for example, #1071807).

    Would it be acceptable in this case if we fix the i386 build issue
    in stable first? (considering that the package is not and will not
    be in trixie).

    If yes, can it still be versioned as 2.3.1+dfsg-3+deb12u1, or it should be deb12u2?

    My current plan would be to follow the idea of Andrius Merkys (in the CC)
    of relaxing the tolerance for those floating point equality comparisons.

    Thanks.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)