arigato 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 | mac OS and Fedora are not Windows
Guest482 has quit [Ping timeout: 240 seconds]
fling has joined #pypy
fling is now known as Guest68330
codypiersall has joined #pypy
Guest68330 has quit [Changing host]
Guest68330 has joined #pypy
Guest68330 is now known as fling
speeder39_ has joined #pypy
dddddd has quit [Remote host closed the connection]
fling has quit [Ping timeout: 258 seconds]
codypiersall has quit [Ping timeout: 256 seconds]
fling__ has joined #pypy
jcea has quit [Quit: jcea]
jacob22 has quit [Read error: Connection reset by peer]
jacob22 has joined #pypy
kirma has quit [Ping timeout: 252 seconds]
speeder39_ has quit [Quit: Connection closed for inactivity]
siddhesh has joined #pypy
jacob22 has quit [Read error: Connection reset by peer]
fling__ has quit [Quit: leaving]
<kenaan> mattip py3.6 2e2966805114 /lib-python/3/idlelib/: redo 9387f96a5518 in two steps since file case changed (windows ...)
<kenaan> mattip py3.6 c51e9fb1f2f9 /lib-python/3/idlelib/: redo 9387f96a5518 in two steps since file case changed (windows ...)
<kenaan> mattip py3.6 56db4a646df6 /: merge heads
jacob22 has joined #pypy
<mattip> cfbolz: there are some new failures on own py3.6 connected to import, perhaps b7b3ef736a84 needs tweaking for tests?
<cfbolz> mattip: yes, that's clearly my fault
<cfbolz> mattip: it's frustrating, they are some kind of weird space bootstrapping problem
<mattip> :(
<cfbolz> mattip: but look how much better the test_pypy_c tests got!
<mattip> +1
siddhesh_ has joined #pypy
siddhesh has quit [Ping timeout: 240 seconds]
xcm is now known as Guest41476
Guest41476 has quit [Remote host closed the connection]
xcm has joined #pypy
<kenaan> arigo default 4d4e7569ddd5 /pypy/module/cpyext/cdatetime.py: Fix for pyarrow calling invalid macros
<arigato> ^^ more a workaround than a fix really
<mattip> I wonder how much of our code is workarounds for user (mis)use of non-defined behaviour
<arigato> well, it's cpyext we're talking about
<arigato> one way to look at it would be to answer: "all of cpyext"
<mattip> :)
<LarstiQ> :)
<LarstiQ> oho, pyarrow support
danieljabailey has quit [Ping timeout: 246 seconds]
<mattip> there are rpython differences between py3.6 and default, for instance cafe126a986a added `save_err` to some low-level socket calls
<arigato> these should be unified
kirma has joined #pypy
<mattip> ok
senyai has joined #pypy
lritter has joined #pypy
<kenaan> mattip py3.6 c8a4e88385a8 /pypy/doc/: fix marginally useful whatsnew for branch
<kenaan> mattip default 5802580694f9 /rpython/rlib/rutf8.py: add an error message for out-of-range
<kenaan> mattip py3.6 1c7aea3d0597 /: merge default into py3.6
<kenaan> mattip py3.6 0f9a6721ad53 /pypy/module/cpyext/: fix test - conversion of py_obj to w_obj can fail
<kenaan> mattip default 2ea6562691b9 /rpython/rlib/: backport rpython changes
<kenaan> mattip py3.6 f3df6cd7daaf /rpython/rlib/rposix_stat.py: minimize diff to default (formatting)
siddhesh_ has quit [Ping timeout: 255 seconds]
dan- has quit [Ping timeout: 250 seconds]
dan- has joined #pypy
dan- has joined #pypy
dan- has quit [Changing host]
iomari has joined #pypy
dddddd has joined #pypy
siddhesh_ has joined #pypy
xorAxAx has quit [Quit: Gateway shutdown]
xorAxAx has joined #pypy
siddhesh__ has joined #pypy
siddhesh_ has quit [Ping timeout: 245 seconds]
iomari has quit [Quit: http://quassel-irc.org - Chat comfortably. Anywhere.]
siddhesh__ has quit [Remote host closed the connection]
lesshaste has joined #pypy
<mattip> I wonder if there is a way to query PyPi and find the top 10 libraries that use cython
<hruske> one would need to have a dependancy graph, no?
<arigato> they are probably not "dependent" of cython in the graph
<arigato> if they simply have the .c files checked in too
<hruske> so, basically any package that has .pyx in source
<fijal> hruske: but maybe not in the .pypi thing
<hruske> so ... recheck C files to see if they have any cython generated artefacts?
<fijal> yeah
<fijal> so step 1. has C files, step 2. what do use those C files with?
<mattip> assuming they have a source tarball on pypi
<mattip> and not just a binary wheel
<mattip> maybe cython already has the information
jcea has joined #pypy
<hruske> does cython leave any symbols in binary
<kenaan> mattip py3.6 43e56c60e223 /pypy/interpreter/unicodehelper.py: fix formatting, merge repititions
<kenaan> mattip py3.6 6090343709fa /pypy/: test, fix for infinite encoding due to bad error handler
<kenaan> mattip py3.6 2e27a5ddeaed /pypy/module/: fix
<mattip> _codecs/test/test_ztranslation does not like 6090343709fa, I think the FakeObjSpace is confused by the IndexError
<mattip> hruske: `strings cextension.so |grep -i cython` yeilds 'cython_runtime', so that might work
<mattip> it seems at import cython sticks an empty 'cython_runtime' module in sys.modules
Rhy0lite has joined #pypy
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
danieljabailey has joined #pypy
jacob22 has quit [Read error: Connection reset by peer]
jacob22 has joined #pypy
agronholm has quit [Read error: Connection reset by peer]
<hruske> found 10 cython users in the top 123 packages only with pyx
Ai9zO5AP has joined #pypy
<hruske> pyyaml numpy pandas scikit-learn psutil scipy coverage multidict grpcio aiohttp
<hruske> these are extracted from sdist, checking both .pyx and .c
agronholm has joined #pypy
marky1991 has joined #pypy
<nedbat> hruske: coverage.py doesn't use cython
<hruske> ha, false positive, let me check what I got wrong
<simpson> And pythonPackages.coverage in nixpkgs doesn't ask for Cython either, so it doesn't break without Cython in an isolated environment.
<simpson> hruske: Here's some packages which are popular enough to be hand-included in nixpkgs and also probably use Cython: https://gist.github.com/4bdaf73b69d877c05fd5580b4ef663c7
<simpson> $ git grep -l cython pkgs/development/python-modules/ | cut -f4 -d'/' | gist --private
<simpson> Oh, mattip was the original asker. mattip ^^^
<tos9> 1
<hruske> yeah, I guess grepping for 'Cython' was too broad :)
<hruske> this list seems better > pyyaml numpy pandas scikit-learn scipy multidict grpcio aiohttp gevent lxml pywavelets scikit-image h5py uvloop apache-beam pyzmq cython msgpack yarl ruamel-yaml
marky1991 has quit [Ping timeout: 250 seconds]
<mattip> thanks!
<hruske> need to check the binaries still
<mattip> probably good enough, just kind of curious what are the work loads cython commonly handles
<mattip> that is a good mix of data science and web dev stuff
jacob22 has quit [Quit: Konversation terminated!]
moei has quit [Quit: Leaving...]
<kenaan> rlamy default 5062f3687585 /pypy/module/cpyext/: Call internal methods from PyDict_XXXItem() instead of going through dunder methods (CPython compatibility) Fixes ...
marky1991 has joined #pypy
igitoor has quit [Ping timeout: 257 seconds]
igitoor has joined #pypy
igitoor has quit [Changing host]
igitoor has joined #pypy
<ronan> mattip: in RPython, built-in exceptions aren't supposed to have attributes, so 5802580694f9 is wrong
<mattip> hmm. So I would need to make an OutOfRangeError?
<tos9> stuff in e.g. rpython.rlib.rfloat and rpython.rlib.rdtoa seem to do the same thing, no? (Raise ValueError with a message)
Zaab1t has joined #pypy
lritter has quit [Quit: Leaving]
<mattip> tos9: it seems the message is not used anywhere, right?
<kenaan> mattip default 213fc2573b4d /: use a utf8 specific error for OutOfRange
<kenaan> mattip py3.6 805c1a75f5f0 /: merge default into branch
<kenaan> mattip py3.6 6890d4e0c302 /pypy/: use rutf8.OutOfRange error
<kenaan> mattip default 59c374d988c1 /pypy/interpreter/unicodehelper.py: use rutf8.OutOfRange error
__stian has quit [Ping timeout: 256 seconds]
marky1991 has quit [Ping timeout: 250 seconds]
<mattip> anyone who tries to use autosummary and autodoc in sphinx has no right to claim pypy is complicated, opaque and has too many layers
Ai9zO5AP has quit [Quit: WeeChat 2.3]
<cfbolz> mattip: hah :-)
<cfbolz> mattip: having fought against importing basically all day, I am not sure I can agree
<mattip> :( frustration all around
<cfbolz> mattip: I think I have just the right level of hack now
Rhy0lite has quit [Quit: Leaving]
<kenaan> cfbolz py3.6 c6cca0f534ca /pypy/objspace/std/marshal_impl.py: fix test_int_roundtrip in test_marshalimp.py marshaling an int that doesn't fit into 32bit will now use the long by...
jacob22 has joined #pypy
marky1991 has joined #pypy
<tos9> mattip: from what I could tell it didn't look like it yeah, but probably easier to give up and do what you did :P
<tos9> mattip: it'd be used if you ran the rpython as python... which was why I assumed you added it (because some error wasn't clear), but hey what do I know
* tos9 was paying attention mostly just to learn somefin'
marky1991 has quit [Ping timeout: 255 seconds]
marky1991 has joined #pypy
Zaab1t has quit [Quit: bye bye friends]
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
marky1991 has quit [Read error: Connection reset by peer]
kipras has quit [Ping timeout: 255 seconds]