Log file Viewer

[00:35:14] <njs`> GPF: you can pass arguments down to setup.py with something like pip ... --build-option="--disable-shared" (or --install-option or --global-option, to get hte options passed to different parts of the setup.py call)

[00:35:34] <njs`> GPF: getting arguments from setup.py to whatever this configure script is is a separate problem involving the details of that package's setup.py

[01:19:13] <[Tritium]> Is it normal that a search with no results on warehouse returns the framework/server default 404 message?

[01:27:30] <kanha> I installed virtualenv using sudo pip install virtualenv and then tried creating a virtual environment but it throws error as https://dpaste.de/myi5 any idea how to fix this?

[01:29:19] <[Tritium]> kanha: you installed with sudo, but it somehow managed to install into your $HOME/.local ...

[01:30:03] <kanha> [Tritium]: bit correction later I removed it and installed using pip install --user virtualenv

[01:30:18] <[Tritium]> (I find packaging bugs are nearly impossible to correctly once you get sudo involved...)

[01:30:20] <kanha> but still it doesn't work. throws same error

[01:30:53] <[Tritium]> kanha: what version of python are you running (2.7.what)

[01:32:40] <njs`> kanha: you should check what running 'locale' prints

[01:33:05] <kanha> [Tritium]: it's python2.7.6

[01:37:24] <[Tritium]> kanha: i think MAYBE updating to the latest patch level of python might help?

[01:37:42] <[Tritium]> there is a ppa for just this

[01:37:47] <kanha> [Tritium]: which patch?

[01:38:27] <[Tritium]> the latest patchlevel is 2.7.11

[01:38:59] <kanha> how to do this?

[01:39:50] <kanha> does virtualenv installation depends on any other package like python-setuptools?

[01:40:15] <kanha> I mean other packages than python setuptools

[01:41:01] <[Tritium]> I seriously would not install the ubuntu versions of setuptools, pip, or virtualenv.

[01:41:52] <[Tritium]> https://launchpad.net/~fkrull/+archive/ubuntu/deadsnakes-python2.7 for getting the latest version of 2.7

[01:43:31] <ngoldbaum> sad-debian

[02:02:50] <numberoverzero> hello, I'm getting a 404 trying to install most packages with pip. ipython and beautifulsoup4 both failed, although requests succeeded.

[02:03:03] <numberoverzero> Is there a status page that I missed, or somewhere I can check if it's on my end?

[02:03:57] <dstufft> numberoverzero: try again

[02:03:57] <ngoldbaum> numberoverzero: seems to work over here

[02:03:58] <numberoverzero> and just like that it seems to be back.

[02:03:59] <sontek_> Its back already

[02:04:02] <ngoldbaum> oh nm

[02:04:03] <dstufft> I borken it

[02:04:05] <sontek_> I came in to complain and then it was fixed :)

[02:04:05] <numberoverzero> thanks, internet wizards!

[02:05:08] <ngoldbaum> dstufft: do you ever think about the amount of chaos/minor annoyance you could cause by messing with pypi? ;)

[02:05:38] <dstufft> ngoldbaum: No Comment

[02:05:39] <dstufft> ;)

[02:06:08] <[Tritium]> minor... annoyance... to the authoritative source of python libraries...

[02:07:56] <ngoldbaum> [Tritium]: i mean, in the grand scheme of things, although I wouldn't be surprised if crazy random things broke as well

[02:12:31] <sontek_> It breaks enough to know to retry but not enough where I blame it first, so by time I finish debugging my infra its fixed :P

[02:16:00] <dstufft> sontek_: Luckily I optimize for not getting yelled at, so as long as I keep it working good enough that stays the case for most people... :D

[02:25:19] <sontek_> dstufft: oh, we didn't tell you our system did you? We don't complain we just keep a tally and collect in beers at PyCon

[02:25:27] <sontek_> did we*

[02:25:32] <dstufft> sontek_: lol

[04:43:52] <DJ_Arghlex> oh so there is a pypa-specific channel

[04:43:53] <DJ_Arghlex> rad

[04:44:13] <DJ_Arghlex> alright I'm getting odd errors concerning pip removing its build directories

[04:44:30] <DJ_Arghlex> http://pastebin.com/eenT6tbz has happened since as far as I can remember using pip

[04:45:11] <DJ_Arghlex> package installs OK, but then it throws that error when it tries to remove that directory, causing the installation to be a "failure"

[04:46:07] <DJ_Arghlex> normally I wouldn't care too much about this, butsince I'm trying to install something with a lot of dependencies, but since it throws an error and stops the whole process...

[04:46:45] <DJ_Arghlex> I'd rather not manually resolve all the dependencies for each and every package I install and install them one by one and have the process take twice as long

[04:48:32] <DJ_Arghlex> telling it to use --isolated and --no-cache-dir have no change

[04:49:28] <DJ_Arghlex> gonna tell it to pick a different build directory, see of that h- nope

[04:51:58] <DJ_Arghlex> --no-clean isn't even fixing it

[05:16:14] <GPF> njs`: thanks. I kind of cheated, I installed net-snmp with python support via brew after activating a virtualenv so that the python bindings were actually installed in the virtualenv instead of the global environment

[05:28:02] <GPF> woher krieg ich denn möglichst einfach parspare origin daten: ich hab ne IP, will wissen woher das kommt (AS oder noch besser Informationen aus RIPE-DB, APNIC, etc.=

[05:28:21] <GPF> AS könnt ich mir noch aus der routing-table direkt holen

[05:45:19] <GPF> hmph radb ist zu unvollständig

[05:56:20] <GPF> hä, tshark kann keinen capture filter aus nem file lesen?

[06:55:47] <GPF> ich hasse binär, ich komm damit nie klar

[07:00:09] <GPF> erm sorry

[07:00:15] <GPF> I'm in the wrong channel the whole time

[09:11:10] <zaytsev> njs`: you're a genious. pip install -U --user pip wheel did the trick. now wheels are correctly tagged and the my application doesn't fail on import: Cython-0.23.4-cp27-cp27mu-linux_x86_64.whl & Cython-0.23.4-cp27-cp27dmu-linux_x86_64.whl

[09:12:14] <zaytsev> njs`: of course, now i hit the next problem (https://github.com/pypa/pip/issues/2382) but this one at least is not of vital importance

[14:05:41] <dstufft> nedbat: fwiw HPE is going through layoffs at the moment, so idk how fast the turn around time is going to be on them sorting out the CLA. If you want to port it yourself (hypothetical you) in the meantime I won't feel mad or anything

[14:06:04] <nedbat> dstufft: ok, thanks. I know this is an awkward spot to be in.

[14:13:02] <ionelmc> dstufft: why is you signing a cla a problem for hpe?!

[14:13:33] <dstufft> ionelmc: I'm employed by HPE, I wrote some code on HPE's time, techincally they own the copyright for said code and need to handle the CLA for it

[14:15:40] <dstufft> In this specific case it's a bit silly because the code in question is largely mechanical transforms to make a py2 project work on py2 + py3, but what are you gonna do, legal shit is "fun"

[14:17:44] <dstufft> nedbat: fwiw my manager says he hasn't heard anything back from OSPO (whatever that is) so he's going to poke them again

[14:18:14] <nedbat> dstufft: ok, glad to hear there are at least heads being beaten against the brick wall :)

[14:20:30] <dstufft> I suspect that in a company as large as HPE there are a multitude of attidues about contributing to random OSS projects and it's probably likely that there has to be some "no really, this si fine"-ing

[14:23:28] <dstufft> ngoldbaum: probably!

[14:24:22] <dstufft> It's a bit funny, I have blanket authorization to contribute to stuff evne though HPE owns the stuff I write, but I can't sign a CLA on behalf of HPE which says that they can use the stuff I'm already allowed to contribute :D