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
bernte has quit [Ping timeout: 264 seconds]
_whitelogger has joined #pypy
YannickJadoul has quit [Quit: Leaving]
lritter has quit [Quit: Leaving]
xtarget has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
jvesely has quit [Quit: jvesely]
dddddd has quit [Remote host closed the connection]
_whitelogger has joined #pypy
<kenaan> arigo cffi/cffi 0d43d8bf557f /cffi/cparser.py: Tweak the warning message
<kenaan> arigo cffi/cffi[release-1.13] b175e63b614b /: hg merge default
<kenaan> arigo cffi/cffi[release-1.13] 33354a4abeb4 /: update version to 1.13.2
<kenaan> arigo cffi/cffi[release-1.13] 31d60585c288 /doc/source/whatsnew.rst: update whatsnew
_whitelogger has joined #pypy
i9zO5AP has joined #pypy
Ai9zO5AP has quit [Ping timeout: 268 seconds]
i9zO5AP has quit [Ping timeout: 265 seconds]
Ai9zO5AP has joined #pypy
bernte has joined #pypy
bernte has quit [Client Quit]
_whitelogger has joined #pypy
astronavt___ has joined #pypy
danchr_ has joined #pypy
danchr has quit [Disconnected by services]
Ninpo has quit [Ping timeout: 245 seconds]
astronavt has quit [Ping timeout: 265 seconds]
Ninpo has joined #pypy
mattip_ has joined #pypy
<kenaan> mattip default 9f7ce234ad3a /pypy/tool/release/make_portable.py: refactor "os import" in make_portable.py to make it importable on windows
<kenaan> schoelle default f67d306a0f3a /lib_pypy/_ctypes/array.py: Checking for overflow in ctypes array creation (fixing test.test_ctypes) (grafted from 9f57693bca378be44fe871749...
<kenaan> mattip py3.6 9e6ad1f66385 /: merge default into branch
ekaologik has joined #pypy
ronan__ is now known as ronan
i9zO5AP has joined #pypy
Ai9zO5AP has quit [Ping timeout: 240 seconds]
dddddd has joined #pypy
xtarget has joined #pypy
jvesely has joined #pypy
mattip_ has quit [Ping timeout: 276 seconds]
kipras has joined #pypy
xtarget has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
Ai9zO5AP has joined #pypy
i9zO5AP has quit [Ping timeout: 245 seconds]
<kenaan> cfbolz expose-jsonmap 35f2d7155ef8 /pypy/objspace/std/: switch away from the jsondict strategy if the map is blocked
<kenaan> cfbolz expose-jsonmap b23978e0d3fb /pypy/: move fields from JsonMap to JsonStrategy
<LarstiQ> what does blocked mean there?
<cfbolz> LarstiQ: it's part of the somewhat complicated heuristic that is used to determine whether a pattern of keys from an object parsed in a json file appears "often enough"
<cfbolz> "blocked" means it's been decided that it's too rare to be useful
<arigato> Alex_Gaynor: hi! are you here?
<Alex_Gaynor> arigato: yes
<arigato> can we release the next cffi?
<Alex_Gaynor> Sure, do you need us to build the wheels?
<arigato> yes. I'll first upload the source and let you know (and half the wheels from azure pipelines)
<Alex_Gaynor> Sounds good
<arigato> sent
<arigato> thanks!
mattip_ has joined #pypy
<kenaan> arigo cffi/cffi[release-1.13] 0c9be4c0d798 /doc/source/installation.rst: md5/sha
<kenaan> arigo cffi/cffi[release-1.13] 80521a7262a6 /.hgtags: Added tag v1.13.2 for changeset 0c9be4c0d798
<kenaan> arigo cffi/cffi dca192bb54fd /: hg merge release-1.13
__pv has joined #pypy
<arigato> Alex_Gaynor: thank you!
<Alex_Gaynor> np
<arigato> I can only hope that this time the release was not broken (unlike v1.13 and v1.13.1)
mattip_ has quit [Ping timeout: 252 seconds]
<LarstiQ> cfbolz: right, guess it makes sense to switch the strategy then :)
<cfbolz> LarstiQ: yes
<cfbolz> LarstiQ: it's all a bit involved because you really have two different kinds of dicts in json files
<cfbolz> Instance like, and Real dicts
* LarstiQ nods
Taggnostr has quit [Remote host closed the connection]
Taggnostr has joined #pypy
mattip_ has joined #pypy
antocuni has joined #pypy
jvesely has quit [Quit: jvesely]
mattip_ has left #pypy [#pypy]
i9zO5AP has joined #pypy
Ai9zO5AP has quit [Ping timeout: 240 seconds]
lritter has joined #pypy
jvesely has joined #pypy
antocuni has quit [Ping timeout: 240 seconds]
kipras has quit [Read error: Connection reset by peer]
i9zO5AP has quit [Ping timeout: 240 seconds]
i9zO5AP has joined #pypy
i9zO5AP has quit [Ping timeout: 268 seconds]
i9zO5AP has joined #pypy
<kenaan> mattip buildbot ad3301acbb64 /docker/: use the libffi and openssl from pyca/cryptography to avoid bugs in centos6
created has joined #pypy
<created> Since pypy 7.2, cffi modules using dlls no longer work?
<created> They worked fine in 6.0
<created> (Both pypy3)
<created> The module compilation runs fine, but the import fails with "ImportError: The specified module could not be found"
<created> (And all this on windows, hence "dlls")
mattip_ has joined #pypy
<mattip_> created: what does pypy -c"import sys; print(sys.version)" show?
<created> 3.6.9 (5da45ced70e5, Oct 09 2019, 21:35:51)
<created> [PyPy 7.2.0 with MSC v.1910 32 bit]
<created> Downloaded from official website
<created> (Binary downloaded, that is)
<created> Previous pypy version which worked has the same "MSC v.1910 32-bit" part, in case that matters
<mattip_> created: we build cffi modules as part of the package build, so "it works somewhere"
<created> Hmm
<created> Well a few more possibly relevant points:
<created> 1. I use windows 7
<created> 2. I had the same error even with the old pypy version for dlls built by me. No such errors for dlls downloaded from the net, though.
<created> 3. Of course I say " the same error", but the error is very non-descriptive and could be anything and everything, so there's no guarantee the issue in 2 and the current issue are the same
ekaologik has quit [Quit: https://quassel-irc.org - Komfortabler Chat. Überall.]
<created> (It really really would be nice if python and pypy would show more helpful errors on native module import failure....)
<created> I can't count how much time I had to waste to figure out what various forms of "couldn't find/load module" were talking about
<mattip_> Do you have `depends.exe`? I sometimes try to add a breakpoint with `import pdb;pdb.set_trace()`, at the line in my code before the crash
<mattip_> which will properly set up all the PATH and machine state
<mattip_> then I can call os.system('depends.exe') which should show what is missing
<mattip_> What version of visual studio do you use?
<mattip_> The trivial thing: the dll file exists and is on the sys.path, right?
<mattip_> as for the error message, this is what windows reports, we don't know more than what the OS tells us
<created> Ohhhhh
<created> All I had to do was launch it from cmd
<created> Then I got a nice popup from windows telling me what the error was (it couldn't find the dll used by the module after all)
<created> Previously I was launching it from some IDE (pydev in liclipse or some nightmare) and it was surppressing this message
<mattip_> cool
<created> The reason it was working in the old dll is that some psychopath (probably me :( ) put the needed dlls inside the pypy directory
<created> Which also explains why I wasn't seeing changes when updating the dlls previously
<created> *in the old pypy version
<created> Yeah, I put the dlls there out of bad laziness
<created> Sorry and thanks for helping me help myself
mattip_ has quit [Ping timeout: 265 seconds]
<_aegis_> how am I supposed to run build_cffi_imports to build ssl on pypy3 7.2.0 if pypy doesn't already have working ssl
<_aegis_> the ssl url changed from http to https, so it will fail if pypy has no ssl provider
<_aegis_> changed in 89c76ee
<_aegis_> I guess I'll manually wget the archive in my build script
<_aegis_> wait no it's a temp dir
<_aegis_> oh at least it's a stable tempdir
created has left #pypy ["Leaving"]
awygle has quit [*.net *.split]
awygle has joined #pypy
jcea has joined #pypy
i9zO5AP has quit [Ping timeout: 265 seconds]
i9zO5AP has joined #pypy
jcea has quit [Remote host closed the connection]
jcea has joined #pypy
i9zO5AP has quit [Excess Flood]
i9zO5AP has joined #pypy
i9zO5AP has quit [Excess Flood]
i9zO5AP has joined #pypy