Fix test of cgt2v

[wlmb]

There were a couple of commas ',' that made the test fail. I guess they were related to the old non-native complex numbers. Regards

[coveralls]

coverage: 46.329% (-0.2%) from 46.554%
when pulling 555abf8 on wlmb:devel
into 5612528 on PDLPorters:master.

[mohawk2]

Yes, they were - I'm a bit worried that the tests were passing both locally for me, and in the CI. In any case, thank you! About to release as 0.42.

[wlmb]

I found out something: I'm using a previous version of PDL, 2.093 and the tests failed. A colleague (Guillermo) is using version 2.095 and the tests pass! I found out that 'approx' is acting funny in 2.095 when fed complex numbers: approx(.5*i, 1.5*i) -> true approx(1.5*i, .5*i) -> false approx(.5*i, 150*i) -> true approx(150*i, .5*i) -> false approx(0*i, .5*i) -> true approx(.5*i, 0*i) -> false With real numbers it seems to work correctly. Regards, Luis

…

On Sat, Dec 07, 2024 at 08:02:45AM -0800, mohawk2 wrote: Yes, they were - I'm a bit worried that the tests were passing both locally for me, and in the CI. In any case, thank you! About to release as 0.42. -- Reply to this email directly or view it on GitHub: #27 (comment) You are receiving this because you authored the thread. Message ID: ***@***.***>

-- o W. Luis Mochán, | tel:(52)(777)329-1734 /<(*) Instituto de Ciencias Físicas, UNAM | fax:(52)(777)317-5388 `>/ /\ Av. Universidad s/n CP 62210 | (*)/\/ \ Cuernavaca, Morelos, México | ***@***.*** /\_/\__/ GPG: 791EB9EB, C949 3F81 6D9B 1191 9A16 C2DF 5F0A C52B 791E B9EB