Re: Scons 4.2.0 and Python 3.10: test SCons/ActionTests.py failed

Bill Deegan <[email protected]>
Newsgroups gmane.comp.programming.tools.scons.user
Message-ID <CAEyG4CGcGZv5-PejX+xnv1oGrp_5wUjKQvrzGN-R1Bg1mvr0ug@mail.gmail.com>
Sergey,

You can ignore this failure for the time being with the following caveat
* - If you build with python 3.10 and then build again with python 3.10.1
you will likely see unnecessary rebuilds

Otherwise if you're just building with SCons and staying with the same
version of python 3.10.1 or 3.10, scons should do the "right thing"

The test which is failing is in the logic which SCons uses to determine in
a python action function or object has changed between builds.
We generate a signature of such and store that in the sconsign (as we do
for regular source files, or build strings) and compare those when
rebuilding.

Please file an issue in Github for this?
We'll try to address this in the next release.

Thanks,
Bill

On Fri, Dec 17, 2021 at 10:01 AM Mats Wichmann <[email protected]> wrote:

> On 12/17/21 02:23, Sergey Torokhov via Scons-users wrote:
> > I downgrade Python from 3.10.1 to 3.10 and this test is OK.
> >
> > Are there many other places in Scons where it could affect?
>
> no, they're all contained in one file.  I've been updating these as we
> go along, had no idea there would be a change for 3.10.1.  Don't recall
> any time Python has changed these in a bugfix ("micro") version.... sigh.
>
> _______________________________________________
> Scons-users mailing list
> [email protected]
> https://pairlist4.pair.net/mailman/listinfo/scons-users
>

_______________________________________________
Scons-users mailing list
[email protected]
https://pairlist4.pair.net/mailman/listinfo/scons-users
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.