cfbolz changed the topic of #pypy to: PyPy, the flexible snake (IRC logs: https://quodlibet.duckdns.org/irc/pypy/latest.log.html#irc-end ) | use cffi for calling C | if a pep adds a mere 25-30 [C-API] functions or so, it's a drop in the ocean (cough) - Armin
rjarry has quit [Read error: Connection reset by peer]
rjarry has joined #pypy
<marmoute>
mattip: I got some more error with `.hgsubstate` down the line. I'll investigate a bit
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
antocuni has quit [Ping timeout: 246 seconds]
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
<mattip>
marmoute: :(
<mattip>
"down the line" maybe has the same isgit logic?
<marmoute>
More some kind of broken references to something
<marmoute>
I'l seeting up some cleaner testing to see if this might be heptapod related patch getting in the way
<mattip>
I think my patch causes the first commit with a subrepo to be ignored, but I did not test adding subsequent changes to the subrepo after that
<marmoute>
I'll start with having a repro that is not a strange error from patched code directly in a docker
<mattip>
using "hg log .hgsub" and "hg log .hgsubstate" on the pypy repo, it might be possible to reconstruct the exact sequence of changes
<marmoute>
mattip: since your export worked, I am trying re-exporting from scratch using various combination of code
<mattip>
won't that be really slow?
<marmoute>
I am doing that on the i9
<marmoute>
Also the start is slow, but once one reach the failure, it refail instantly
* mattip
mumble something about using pypy instead of cpython :)
<marmoute>
Looking forward to that
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
<tos9>
mattip: hmmm looks like it didn't run somehow, even though if I run it again locally I see the new OpenSSL version
<tos9>
First day back at work from vacation today so will be a bit swamped for the next few hours, but will see about having a look again later hopefully
<mattip>
would someone like to suggest a wording that we could put there?
<mattip>
my personal opinion is that until PyPy has more of a presence in the general python ecosystem (integration in CI pipelines of common packages, wheels on PyPI, ...)
<mattip>
it will be difficult to engage with the PSF, and even then it will continue to be difficult
<simpson>
mattip: I do like that idea, and for those of us that can turn PyPy consultancy into money, they should do that.
marky1991 has joined #pypy
jcea has joined #pypy
jcea has quit [Remote host closed the connection]
<marmoute>
simpson: are you deciding by yourself what other people should be doing based on arbitrary criteria ?
<simpson>
marmoute: Yep, that's me, just telling other people what to do~
* marmoute
happily throw a brick at simpson for doing so.
<simpson>
Thanks, you too!
xorAxAx has quit [Quit: Idle timeout reached: 172800s]
antocuni has joined #pypy
BPL has joined #pypy
antocuni has quit [Ping timeout: 244 seconds]
camelCaser has joined #pypy
camelCaser has quit [Ping timeout: 258 seconds]
camelCaser has joined #pypy
oberstet has quit [Remote host closed the connection]
xorAxAx has joined #pypy
* fijal
stops himself from writing an inflammatory comment there
<fijal>
it seems like they're trying to actively hide the fact that there is and will be a supported python 2 version
<fijal>
I don't know what to do, but maybe ignoring it is a good idea
<Dejan>
nothing wrong in that
<Dejan>
I jumped into the Python wagon few years ago, when Python 3 was already well-established, so I have no need for Python 2 support, but I am sure thousands of companies out there still use Python 2 and will do so for many years to come
<Dejan>
One of the things I admire about Java is that I still can run my Java 1.4 applications
<Dejan>
with latest Java runtime
<Dejan>
Python should be similar
<Rhy0lite>
fijal: ping
<fijal>
Rhy0lite: pong
jcea has joined #pypy
<ronan>
mattip: I've made some progress on test_import
<ronan>
the issue happens in importlib._bootstrap._ModuleLock
<ronan>
thread 1 holds modulelock.lock and calls modulelock.has_deadlock()
<ronan>
but modulelock.owner is set to thread 2 which is waiting for modulelock.lock in modulelock.acquire()
<ronan>
so thread 1 just loops forever in has_deadlock()
<ronan>
What I still don't know is how the situation arises
<mattip>
it is good to have a place to start
<mattip>
nice
<marmoute>
mattip: I repropduce the failrure "easily" outisde of docker
<marmoute>
so I'll dig up soon™
<ronan>
actually, I don't understand the logic at all
<mattip>
marmoute: nice
Dejan has quit [Quit: Leaving]
Rhy0lite has quit [Quit: Leaving]
speeder39_ has joined #pypy
<ronan>
mattip: annoyingly, I can't repro with the regular module importlib._bootstrap, it has to be the frozen one, so I can't get a pdb or put prints inside the code
<mattip>
ronan: does freezing mess with/emulate the threading model?
<mattip>
can you print-to-file?
<ronan>
mattip: I don't think so
<ronan>
what do you mean by print-to-file?
<mattip>
with open('/tmp/dbg.txt', 'a') as fid: fid.write('line10')
<mattip>
with open('/tmp/dbg.txt', 'a') as fid: fid.write('line10\n')
<mattip>
but that might mess with whatever race condition exists :(