I think it is more important to have CI that clearly shows the impact of
dev versions of the interpreter and core packages. Some of us in the
Nixpkgs community had this idea for Python core packages as well (and
potentially scientific computing packages, but that's out of scope here).
This would ne
> The main objection to that model is that it requires modifying source
timestamps, which isn't possible for builds on read-only source trees.
Why not set the source timestamps of the source trees to say 1 first?
That's what is done with the Nix package manager. The Python interpreter is
patched (
In my opinion the trend of creating these standalone packages is pretty
bad. Developers that choose to support these kind of packages become
responsible not only for the code they distribute, but also for whether the
dependencies that are bundled in such a standalone package are up to date
and secu
if it sometimes differs in
> trivial ways) is usable with all hash seed settings.
>
> --Guido
>
> On Fri, May 12, 2017 at 6:06 AM, Freddy Rietdijk
> wrote:
>
>> Hi,
>>
>> On Nix we set PYTHONHASHSEED to 0 when building packages, disabling hash
>> randomiza
Hi,
On Nix we set PYTHONHASHSEED to 0 when building packages, disabling hash
randomization. We do this to improve determinism of the builds because we
store the bytecode next to the code.
When one runs Python directly or via a script PYTHONHASHSEED is not set
thus enabling hash randomization. Am
Hi,
I would like to know if you're open to supporting `exec -a` or an
environment variable for setting `argv[0]`, and have some pointers as to
where that should be implemented.
On Nixpkgs we typically use wrappers to set environment variables like PATH
or PYTHONPATH for individual programs. Consi
Hi Armin,
Thank you for your explanation.
I've now managed to build 2.7 and 3.5 deterministic by recompiling the
bytecode at the end of the build (and excluding 2to3).
Freddy
On Sun, Feb 19, 2017 at 9:30 AM, Armin Rigo wrote:
> Hi Freddy,
>
> On 16 February 2017 at 18:03, F
cuted, right?
On Fri, Feb 10, 2017 at 12:03 PM, INADA Naoki
wrote:
> On Fri, Feb 10, 2017 at 7:58 PM, Freddy Rietdijk
> wrote:
> > For Python 3.5 PYTHONHASHSEED doesn't seem to be sufficient, these items
> > still seem indeterministic.
> > To be sure, I ran `P
That should have been `PYTHONHASHSEED=0 $out/bin/python -m compileall -f
$out`.
On Fri, Feb 10, 2017 at 11:58 AM, Freddy Rietdijk
wrote:
> For Python 3.5 PYTHONHASHSEED doesn't seem to be sufficient, these items
> still seem indeterministic.
> To be sure, I ran `PYTHONHASHSEED=1 $
9, 2017 at 6:51 PM, INADA Naoki wrote:
> On Fri, Feb 10, 2017 at 2:45 AM, Freddy Rietdijk
> wrote:
> > Correct, that was an older version from before I patched
> > `_bootstrap_external.py`. A more recent diff can be found at
> >
> > https://gist.github.c
c58 6c21 e300 k..Xl!..
> │ │ │ │ +: 160d 0d0a e631 9c58 6c21 e300 .1.Xl!..
>
> mtime is 6b2e9c58 vs e6319c53 (little endian)
>
> maybe, you failed to use customized py_compile when building?
>
>
> On Thu, Feb 9, 2017 at 6:27 PM, Freddy Rietdijk
&
Hi,
I'm attempting to make the builds of the Python interpreters for Nixpkgs
[1] deterministic.
In the case of Python 2.7 we have a patch [2] that fixes the timestamp used
in .pyc files in case the env var `DETERMINISTIC_BUILD` is set. We also
remove `wininst*.exe`. This works fine, although ther
12 matches
Mail list logo
