Bug#1103140: python-momepy test failures

2025-04-22 Thread Colin Watson
On Tue, Apr 22, 2025 at 07:51:39PM +0800, tho...@goirand.fr wrote: Downgrading the bug to wishlist and disabling autopkgtest for 32 bits is possible though. You already disabled autopkgtest for 32-bit architectures in 0.7.0-3. This is an FTBFS report, not an autopkgtest failure report. Any

Bug#1103140: python-momepy test failures

2025-04-22 Thread thomas
On Apr 22, 2025 18:57, Colin Watson wrote: > > On Mon, Apr 21, 2025 at 09:17:55PM +0100, Rebecca N. Palmer wrote: > >Those all look like rounding errors (probably due to x87 excess > >precision, since they only happen on i386). This suggests that the > >appropriate response is either to

Bug#1103140: python-momepy test failures

2025-04-22 Thread Colin Watson
On Tue, Apr 22, 2025 at 12:11:27PM +0200, Gregor Riepl wrote: Those all look like rounding errors (probably due to x87 excess precision, since they only happen on i386). This suggests that the appropriate response is either to ignore these tests on i386, or possibly (the graph object output by

Bug#1103140: python-momepy test failures

2025-04-22 Thread Colin Watson
On Mon, Apr 21, 2025 at 09:17:55PM +0100, Rebecca N. Palmer wrote: Those all look like rounding errors (probably due to x87 excess precision, since they only happen on i386). This suggests that the appropriate response is either to ignore these tests on i386, or possibly (the graph object outp

Bug#1103140: python-momepy test failures

2025-04-22 Thread Gregor Riepl
Those all look like rounding errors (probably due to x87 excess precision, since they only happen on i386). This suggests that the appropriate response is either to ignore these tests on i386, or possibly (the graph object output by gdf_to_nx has (x: float, y: float) node keys, so rounding iss

Bug#1103140: python-momepy test failures

2025-04-21 Thread Rebecca N. Palmer
Those all look like rounding errors (probably due to x87 excess precision, since they only happen on i386). This suggests that the appropriate response is either to ignore these tests on i386, or possibly (the graph object output by gdf_to_nx has (x: float, y: float) node keys, so rounding iss