<div dir="ltr">The yt-dev recipe I created would only be good for nightly binary builds. I think that would only be useful for people who want the latest and greatest but don't want to set up a build environment, which I suspect is a relatively small group of people. We'll have to decide at some other point if we want to have buildbots and nightly builds.<div>
<br></div><div>Agreed about updating our documentation and website. I think we should supply both the script and instructions for setting up yt using `pip` and already-installed conda packages. We should also link to a page that lists the packages needed to manually set up yt.</div>
</div><div class="gmail_extra"><br><br><div class="gmail_quote">On Sat, Sep 7, 2013 at 8:22 AM, Matthew Turk <span dir="ltr"><<a href="mailto:matthewturk@gmail.com" target="_blank">matthewturk@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi all,<br>
<br>
I've issued a PR with my first pass at the get_yt.sh script. I've<br>
ported over a bunch of functionality from the old install_script.sh<br>
and it works for me in a clean install on Ubuntu, although it does<br>
require the "chrpath" package which I oddly did not have.<br>
<br>
<a href="https://bitbucket.org/yt_analysis/yt/pull-request/591/first-draft-of-a-new-get_yt-script-based/diff" target="_blank">https://bitbucket.org/yt_analysis/yt/pull-request/591/first-draft-of-a-new-get_yt-script-based/diff</a><br>
<br>
What would be really helpful is if a few people could try it out, make<br>
modifications, commit, push to their fork, and I will then pull in<br>
those changes until it does everything it needs to. I think what I<br>
would like to preserve is that the "get_yt.sh" script should *not* be<br>
about how to install Conda/Anaconda -- it should get yt installed, and<br>
the yt environment. The case of installing yt *into* Conda is what<br>
I'd like to cover by providing binary packages, and I suspect that our<br>
audience for that will grow with time. This script aims along a<br>
slightly different vector.<br>
<br>
I'm still not entirely sure how to handle source distributions; even<br>
with the yt-dev recipe Nathan has, it's still building binary<br>
packages. I've done my best to include the source as-is in the<br>
script, but to also use the yt recipe that Nathan has worked on. I'm<br>
not sure how the "yt-dev" recipe fits into that. Kacper, Nathan, any<br>
ideas about that? Would it just be the basic nightly package builder?<br>
Then we could have it get rebuild and reuploaded all the time; this<br>
would help with keeping people up to date, but *not* with encouraging<br>
contributions.<br>
<br>
-Matt<br>
<br>
PS One last thing is that our whole "get yt" documentation should be<br>
re-done. We should be much more clear about how and where to get the<br>
package, and reduce the number of options.<br>
<div class="HOEnZb"><div class="h5"><br>
<br>
On Fri, Sep 6, 2013 at 9:08 AM, Matthew Turk <<a href="mailto:matthewturk@gmail.com">matthewturk@gmail.com</a>> wrote:<br>
> Hi Nathan,<br>
><br>
> On Wed, Sep 4, 2013 at 2:54 AM, Nathan Goldbaum <<a href="mailto:nathan12343@gmail.com">nathan12343@gmail.com</a>> wrote:<br>
>> Here's an updated version of the shell script Matt sent out earlier today<br>
>> that builds yt and then installs it based on the yt-conda recipe:<br>
>> <a href="http://paste.yt-project.org/show/3848/" target="_blank">http://paste.yt-project.org/show/3848/</a><br>
>><br>
>> I like this a lot and think it can be improved in several simple ways, most<br>
>> of which can be readily scraped from the install script:<br>
><br>
> Me too. This is much improved over mine.<br>
><br>
>><br>
>> * Install in the folder that the script was invoked in rather than $HOME<br>
><br>
> Good idea.<br>
><br>
>> * Add an option to install anaconda with a warning about needing a<br>
>> couple hundred megabytes of space<br>
><br>
> Excellent! I like this a lot. And, we should point to the Anaconda<br>
> page for more info about packages and how Anaconda works.<br>
><br>
>> * Install the miniconda package appropriate for the platform<br>
>> * Add warnings on various platforms regarding needed packages and<br>
>> compilers (i.e. still need build-essentials on ubuntu and Xcode on OSX).<br>
>> * Add a warning at the end about modifying .bashrc for PATH and<br>
>> LD_LIBRARY_PATH or supply an activate script that sets them for us.<br>
><br>
> I think I'd prefer to supply an activate script that both activates<br>
> Conda *and* the yt environment in it. That way individuals who want<br>
> to use Conda directly can do so, but we also make it possible to keep<br>
> the behavior that people have already.<br>
><br>
>> * Clone yt-doc, yt-supplemental and the main yt repo and put them in the<br>
>> yt-conda directory.<br>
><br>
> Agreed.<br>
><br>
>> * Check if conda is already available and (optionally?) skip the<br>
>> miniconda step.<br>
><br>
> I'm not sure about this, since I think we will then just tell people<br>
> to install their own yt. I think the install script should not be<br>
> *too* complex in this regard.<br>
><br>
>> * Ensure the user that yt loves them.<br>
><br>
> Yes.<br>
><br>
>><br>
>> As Matt suggested on a Conda github issue<br>
>> (<a href="https://github.com/ContinuumIO/conda/issues/176#issuecomment-23753101" target="_blank">https://github.com/ContinuumIO/conda/issues/176#issuecomment-23753101</a>) it<br>
>> would be nice to first try to conda install yt and then if that fails use<br>
>> conda build. I think it should be possible to set this up as part of the<br>
>> bash script. As asmurer notes in the issue, general support for this sort<br>
>> of mixed binary/source distribution in conda might be difficult but with yt<br>
>> it works since our setup script is platform agnostic so long as the<br>
>> compilation environment is set up correctly and we don't even try to work<br>
>> correctly on windows.<br>
>><br>
>> So long as we provide binary yt packages of (at least) the stable releases,<br>
>> this should obviate the need for proper compiler setup, which avoid a<br>
>> significant source of pain for OS X users as well as linux users on<br>
>> platforms like ubuntu that don't provide sane build environments out of the<br>
>> box.<br>
>><br>
>> -Nathan<br>
><br>
> Awesome work, and I am getting more excited about this as time goes<br>
> on. I am going to try my hand at updating the new script in the way<br>
> you mentioned, and then I will try checking it into the repo and issue<br>
> a PR so we can iterate there. I hope to get to this by tomorrow<br>
> morning.<br>
><br>
> -Matt<br>
><br>
>><br>
>><br>
>> On Tue, Sep 3, 2013 at 12:28 PM, Matthew Turk <<a href="mailto:matthewturk@gmail.com">matthewturk@gmail.com</a>> wrote:<br>
>>><br>
>>> Hi Nathan,<br>
>>><br>
>>> On Tue, Sep 3, 2013 at 3:10 PM, Nathan Goldbaum <<a href="mailto:nathan12343@gmail.com">nathan12343@gmail.com</a>><br>
>>> wrote:<br>
>>> > John, can you give a little more info on that error? I'm a little<br>
>>> > concerned<br>
>>> > that there are binary incompatibilities that conda can't resolve.<br>
>>> > Perhaps<br>
>>> > this is something we should report on the conda issue tracker.<br>
>>> ><br>
>>> > Matt, can't we get the recipes in the same way we get the latest dev<br>
>>> > install<br>
>>> > script? Something like:<br>
>>> ><br>
>>> > wget <a href="https://bitbucket.org/yt_analysis/yt_conda/raw/yt/meta.yaml" target="_blank">https://bitbucket.org/yt_analysis/yt_conda/raw/yt/meta.yaml</a><br>
>>> ><br>
>>> > This doesn't currently work, although it does work for the yt repo, I<br>
>>> > think<br>
>>> > because we use named branches in that repository.<br>
>>><br>
>>> Yup -- you can do it like this:<br>
>>><br>
>>> wget <a href="https://bitbucket.org/yt_analysis/yt_conda/raw/default/yt/meta.yaml" target="_blank">https://bitbucket.org/yt_analysis/yt_conda/raw/default/yt/meta.yaml</a><br>
>>><br>
>>> ...although for me right now it's failing from a bad BB certificate.<br>
>>><br>
>>> The only problem with "pip install" is that it no longer lists pip<br>
>>> installations inside the "conda list" or "conda update" outputs. This<br>
>>> has been discussed recently on the mailing list for Anaconda:<br>
>>><br>
>>><br>
>>> <a href="https://groups.google.com/a/continuum.io/forum/?fromgroups#!topic/anaconda/wr95VN7ezpo" target="_blank">https://groups.google.com/a/continuum.io/forum/?fromgroups#!topic/anaconda/wr95VN7ezpo</a><br>
>>><br>
>>> ...but there has not yet been a resolution.<br>
>>><br>
>>> -Matt<br>
>>><br>
>>> ><br>
>>> ><br>
>>> > On Tue, Sep 3, 2013 at 12:00 PM, John ZuHone <<a href="mailto:jzuhone@gmail.com">jzuhone@gmail.com</a>> wrote:<br>
>>> >><br>
>>> >> Works fine for me on OS X x86_64.<br>
>>> >><br>
>>> >> On my Goddard-controlled Linux x86_64 server, everything worked fine<br>
>>> >> except Mercurial, which I "conda install"-ed using the yt link, but the<br>
>>> >> binary had an incompatibility which the glibc that was installed on the<br>
>>> >> machine. Using pip to install Mercurial (which did it from source) was<br>
>>> >> the<br>
>>> >> workaround.<br>
>>> >><br>
>>> >> On Sep 3, 2013, at 2:54 PM, Matthew Turk <<a href="mailto:matthewturk@gmail.com">matthewturk@gmail.com</a>> wrote:<br>
>>> >><br>
>>> >> > Hi all,<br>
>>> >> ><br>
>>> >> > Thanks to everybody who has reported back on testing. After some<br>
>>> >> > talking both offline and on IRC, as well as here, I think we would<br>
>>> >> > need to do the following things:<br>
>>> >> ><br>
>>> >> > * Make a single script that grabs the appropriate distribution of<br>
>>> >> > miniconda and installs it. Right now I have a mechanism for doing<br>
>>> >> > this, but it's currently tied to an architecture.<br>
>>> >> > * Create a mechanism for installing all the packages we need. Nearly<br>
>>> >> > all are available inside the Continuum repos. What we're hung up on<br>
>>> >> > is that source installs require a "recipe", and transmitting the<br>
>>> >> > recipe is where I don't have an idea of what to do.<br>
>>> >> > * Test this out lots of places<br>
>>> >> > * Clean up the edges in the (new) install script<br>
>>> >> > * Move the old install script to maintenance mode<br>
>>> >> > * Update all documentation to describe this and mothball other<br>
>>> >> > methods of installation<br>
>>> >> ><br>
>>> >> > What would be nice:<br>
>>> >> ><br>
>>> >> > * Make available nightly builds of yt on several architectures using<br>
>>> >> > binstar<br>
>>> >> > * Utilize more of the packages included in conda elsewhere in yt, now<br>
>>> >> > that we can!<br>
>>> >> ><br>
>>> >> > Here's my current recipe for get_yt.sh:<br>
>>> >> ><br>
>>> >> > <a href="http://paste.yt-project.org/show/3843/" target="_blank">http://paste.yt-project.org/show/3843/</a><br>
>>> >> ><br>
>>> >> > (The config thing may get switched to include the --system argument,<br>
>>> >> > to modify the "yt-conda" condarc.) The step that I'm most stuck on<br>
>>> >> > is<br>
>>> >> > getting the yt recipe to people. If we want to make it possible and<br>
>>> >> > easy to build from source, we need to get the contents of a "conda<br>
>>> >> > recipe" to people. They can then run "conda build ." in the<br>
>>> >> > directory. Here are the recipes that we've been playing with:<br>
>>> >> ><br>
>>> >> > <a href="https://bitbucket.org/yt_analysis/yt_conda/src" target="_blank">https://bitbucket.org/yt_analysis/yt_conda/src</a><br>
>>> >> ><br>
>>> >> > Basically, in get_yt.sh, to do from source instead of from binary we<br>
>>> >> > need to insert a step at the end that downloads the recipes somehow<br>
>>> >> > and then cd's into the right directory and builds them. The reason<br>
>>> >> > this is tricky is that we often need to bootstrap ourselves; we can't<br>
>>> >> > assume anything exists. We can download the .tar.bz2 of the current<br>
>>> >> > tip of the repo, but it includes the hash in the directory name that<br>
>>> >> > it extracts to.<br>
>>> >> ><br>
>>> >> > So I think what we need is a mechanism for getting the current state<br>
>>> >> > of the repo, figuring out the name of the repo's directory, moves<br>
>>> >> > into<br>
>>> >> > it, and then builds. I believe that all/most of this becomes much,<br>
>>> >> > much easier if hg gets included in Anaconda, which Nathan has<br>
>>> >> > submitted a PR for. So hopefully that will be taken care of, but<br>
>>> >> > until that time we can possibly figure something out. I'm not sure<br>
>>> >> > that we have the resources to continually support binary nightly<br>
>>> >> > builds in perpetuity for all the architectures that people run on, so<br>
>>> >> > having source would be awesome. Plus, one of the biggest appeals of<br>
>>> >> > how we distribute yt is that the source is included; I would very<br>
>>> >> > much<br>
>>> >> > not like to give this up.<br>
>>> >> ><br>
>>> >> > Thoughts? Has anyone else tested any of this out?<br>
>>> >> ><br>
>>> >> > -Matt<br>
>>> >> ><br>
>>> >> > On Tue, Sep 3, 2013 at 1:04 PM, Britton Smith<br>
>>> >> > <<a href="mailto:brittonsmith@gmail.com">brittonsmith@gmail.com</a>><br>
>>> >> > wrote:<br>
>>> >> >> Hi everyone,<br>
>>> >> >><br>
>>> >> >> Sorry for chiming in late. I just moved when this thread began and<br>
>>> >> >> do<br>
>>> >> >> not<br>
>>> >> >> have regular internet access.<br>
>>> >> >><br>
>>> >> >> I really like this idea of conda, especially as a package manager<br>
>>> >> >> that<br>
>>> >> >> only<br>
>>> >> >> optionally makes its own edits to your .bashrc. I have always<br>
>>> >> >> really<br>
>>> >> >> liked<br>
>>> >> >> that the install script creates a clean python stack with basically<br>
>>> >> >> everything a python user needs. I have on occasion suggested it to<br>
>>> >> >> people<br>
>>> >> >> just looking to use numpy and matploblib. It looks like conda will<br>
>>> >> >> continue<br>
>>> >> >> to provide this nice by-product, so I'm all for it.<br>
>>> >> >><br>
>>> >> >> I won't be in a position to help with testing and such for another<br>
>>> >> >> week<br>
>>> >> >> or<br>
>>> >> >> so when I get regular internet access, but I would be glad to do so<br>
>>> >> >> then.<br>
>>> >> >><br>
>>> >> >> Britton<br>
>>> >> >><br>
>>> >> >><br>
>>> >> >> On Fri, Aug 30, 2013 at 9:29 PM, Nathan Goldbaum<br>
>>> >> >> <<a href="mailto:nathan12343@gmail.com">nathan12343@gmail.com</a>><br>
>>> >> >> wrote:<br>
>>> >> >>><br>
>>> >> >>> Everything should be available now for 64 bit linux and OS X.<br>
>>> >> >>><br>
>>> >> >>><br>
>>> >> >>> On Fri, Aug 30, 2013 at 10:52 AM, Chris Malone<br>
>>> >> >>> <<a href="mailto:chris.m.malone@gmail.com">chris.m.malone@gmail.com</a>><br>
>>> >> >>> wrote:<br>
>>> >> >>>><br>
>>> >> >>>> Hi Nathan,<br>
>>> >> >>>><br>
>>> >> >>>> That appears to work as it built the environment and `conda<br>
>>> >> >>>> install<br>
>>> >> >>>> ...`<br>
>>> >> >>>> added packages to my environment.<br>
>>> >> >>>><br>
>>> >> >>>> One mistake I made was that I originally downloaded the "latest"<br>
>>> >> >>>> OS X<br>
>>> >> >>>> build of Miniconda, but that happened to be Miniconda3, which is<br>
>>> >> >>>> python 3<br>
>>> >> >>>> based. Trying to build the environment with that yields an error<br>
>>> >> >>>> regarding<br>
>>> >> >>>> incompatibility of yt and python3.3, as it should.<br>
>>> >> >>>><br>
>>> >> >>>> Chris<br>
>>> >> >>>><br>
>>> >> >>>><br>
>>> >> >>>> On Fri, Aug 30, 2013 at 10:42 AM, Nathan Goldbaum<br>
>>> >> >>>> <<a href="mailto:nathan12343@gmail.com">nathan12343@gmail.com</a>><br>
>>> >> >>>> wrote:<br>
>>> >> >>>>><br>
>>> >> >>>>> Hey Chris,<br>
>>> >> >>>>><br>
>>> >> >>>>> I don't think mercurial is strictly necessary, can you try again<br>
>>> >> >>>>> without<br>
>>> >> >>>>> it? I think if Matt uploads a mercurial package for OS X this<br>
>>> >> >>>>> won't<br>
>>> >> >>>>> be an<br>
>>> >> >>>>> issue. I'll send him an updated tarball.<br>
>>> >> >>>>><br>
>>> >> >>>>> I submitted a mercurial recipe to conda-recipes yesterday<br>
>>> >> >>>>> (<a href="https://github.com/ContinuumIO/conda-recipes/pull/14" target="_blank">https://github.com/ContinuumIO/conda-recipes/pull/14</a>) so<br>
>>> >> >>>>> hopefully<br>
>>> >> >>>>> a<br>
>>> >> >>>>> mercurial build will be included in future anaconda releases.<br>
>>> >> >>>>><br>
>>> >> >>>>><br>
>>> >> >>>>> On Fri, Aug 30, 2013 at 10:39 AM, Chris Malone<br>
>>> >> >>>>> <<a href="mailto:chris.m.malone@gmail.com">chris.m.malone@gmail.com</a>> wrote:<br>
>>> >> >>>>>><br>
>>> >> >>>>>> I just tried setting this up on OS X 10.7.5 and failed when<br>
>>> >> >>>>>> attempting<br>
>>> >> >>>>>> to create the conda environment due to a missing mercurial<br>
>>> >> >>>>>> package:<br>
>>> >> >>>>>><br>
>>> >> >>>>>> $ conda create -n ytenv -c <a href="http://conda.binstar.org/yt_project" target="_blank">http://conda.binstar.org/yt_project</a><br>
>>> >> >>>>>> yt<br>
>>> >> >>>>>> mercurial ipython tornado pyzmq pygments jinja2 sphinx<br>
>>> >> >>>>>> Error: No packages found matching: mercurial<br>
>>> >> >>>>>><br>
>>> >> >>>>>><br>
>>> >> >>>>>> On Thu, Aug 29, 2013 at 9:01 PM, Nathan Goldbaum<br>
>>> >> >>>>>> <<a href="mailto:nathan12343@gmail.com">nathan12343@gmail.com</a>> wrote:<br>
>>> >> >>>>>>><br>
>>> >> >>>>>>> Yup, please try on OSX as well. If you make sure Matt's<br>
>>> >> >>>>>>> binstar<br>
>>> >> >>>>>>> is in<br>
>>> >> >>>>>>> your .condarc, you should be able to get yt by doing 'conda<br>
>>> >> >>>>>>> install yt'.<br>
>>> >> >>>>>>><br>
>>> >> >>>>>>> I built the OSX binary on my laptop so I'd appreciate hearing<br>
>>> >> >>>>>>> about<br>
>>> >> >>>>>>> issues, particularly if there are issues on older OS X<br>
>>> >> >>>>>>> releases.<br>
>>> >> >>>>>>><br>
>>> >> >>>>>>> On Thursday, August 29, 2013, Matthew Turk wrote:<br>
>>> >> >>>>>>>><br>
>>> >> >>>>>>>> Hi all,<br>
>>> >> >>>>>>>><br>
>>> >> >>>>>>>> Thank you for the feedback -- I am glad there is some<br>
>>> >> >>>>>>>> agreement<br>
>>> >> >>>>>>>> about<br>
>>> >> >>>>>>>> possible ways forward, and so I'm happy to try to use this as<br>
>>> >> >>>>>>>> an<br>
>>> >> >>>>>>>> opportunity to explore simpler, more reliable methods than the<br>
>>> >> >>>>>>>> install<br>
>>> >> >>>>>>>> script.<br>
>>> >> >>>>>>>><br>
>>> >> >>>>>>>> This afternoon, I spent a bit of time with Conda, and I think<br>
>>> >> >>>>>>>> it's<br>
>>> >> >>>>>>>> quite nice. There are a few rough corners, particularly<br>
>>> >> >>>>>>>> related<br>
>>> >> >>>>>>>> to<br>
>>> >> >>>>>>>> the binstar service, but it's so far pretty great. With<br>
>>> >> >>>>>>>> Nathan's<br>
>>> >> >>>>>>>> help, I was able to upload a yt-2.5.5 package for linux x86_64<br>
>>> >> >>>>>>>> and<br>
>>> >> >>>>>>>> then install it.<br>
>>> >> >>>>>>>><br>
>>> >> >>>>>>>> The workflow that seems to work:<br>
>>> >> >>>>>>>><br>
>>> >> >>>>>>>> * Get miniconda: <a href="http://repo.continuum.io/miniconda/index.html" target="_blank">http://repo.continuum.io/miniconda/index.html</a><br>
>>> >> >>>>>>>> * Run the installer for miniconda<br>
>>> >> >>>>>>>> * Enter the conda environment and then install yt by doing<br>
>>> >> >>>>>>>> "conda<br>
>>> >> >>>>>>>> install yt -c <a href="http://conda.binstar.org/yt_project/" target="_blank">http://conda.binstar.org/yt_project/</a> ".<br>
>>> >> >>>>>>>><br>
>>> >> >>>>>>>> I think that this can likely all be stuck into a bash script.<br>
>>> >> >>>>>>>> A<br>
>>> >> >>>>>>>> simple, first pass at this is here:<br>
>>> >> >>>>>>>><br>
>>> >> >>>>>>>> <a href="http://paste.yt-project.org/show/3833/" target="_blank">http://paste.yt-project.org/show/3833/</a><br>
>>> >> >>>>>>>><br>
>>> >> >>>>>>>> This right now only works on Linux x86_64, but getting it to<br>
>>> >> >>>>>>>> work<br>
>>> >> >>>>>>>> for<br>
>>> >> >>>>>>>> other machines won't be too hard. I suspect we will be able<br>
>>> >> >>>>>>>> to<br>
>>> >> >>>>>>>> do<br>
>>> >> >>>>>>>> nightlies very easily as well. If anyone out there has an<br>
>>> >> >>>>>>>> x86_64<br>
>>> >> >>>>>>>> machine they wouldn't mind trying it on, that would be very<br>
>>> >> >>>>>>>> helpful!<br>
>>> >> >>>>>>>> I did find that once I ran this script, I had to actually<br>
>>> >> >>>>>>>> prepend<br>
>>> >> >>>>>>>> the<br>
>>> >> >>>>>>>> PATH afterwards as well. This means doing:<br>
>>> >> >>>>>>>><br>
>>> >> >>>>>>>> export LD_LIBRARY_PATH=$HOME/yt-conda:$LD_LIBRARY_PATH<br>
>>> >> >>>>>>>> export PATH=$HOME/yt-conda:$PATH<br>
>>> >> >>>>>>>> source activate ytenv<br>
>>> >> >>>>>>>><br>
>>> >> >>>>>>>> At that point, everything was set up and working for me. The<br>
>>> >> >>>>>>>> miniconda install offers to add paths to .bashrc, but I don't<br>
>>> >> >>>>>>>> think<br>
>>> >> >>>>>>>> we<br>
>>> >> >>>>>>>> should go down that route. That being said, this is also a<br>
>>> >> >>>>>>>> possible<br>
>>> >> >>>>>>>> point of friction.<br>
>>> >> >>>>>>>><br>
>>> >> >>>>>>>> One nice thing is that this also completely works with the<br>
>>> >> >>>>>>>> full<br>
>>> >> >>>>>>>> anaconda; if someone wants everything that is in the anaconda<br>
>>> >> >>>>>>>> install,<br>
>>> >> >>>>>>>> they can even simply do "conda install anaconda" from the<br>
>>> >> >>>>>>>> command<br>
>>> >> >>>>>>>> line<br>
>>> >> >>>>>>>> to get it. But the stripped down subset is the default.<br>
>>> >> >>>>>>>><br>
>>> >> >>>>>>>> If anyone has a chance to try this out and has feedback, I'd<br>
>>> >> >>>>>>>> greatly<br>
>>> >> >>>>>>>> appreciate it! I think Nathan has done something very similar<br>
>>> >> >>>>>>>> for<br>
>>> >> >>>>>>>> OSX. I've also put a couple simple conda recipes here:<br>
>>> >> >>>>>>>> <a href="https://bitbucket.org/yt_analysis/yt_conda" target="_blank">https://bitbucket.org/yt_analysis/yt_conda</a> which we can use as<br>
>>> >> >>>>>>>> a<br>
>>> >> >>>>>>>> basis<br>
>>> >> >>>>>>>> for distributing builds and setting them up on buildbots and<br>
>>> >> >>>>>>>> the<br>
>>> >> >>>>>>>> like.<br>
>>> >> >>>>>>>> I'm pretty optimistic about this.<br>
>>> >> >>>>>>>><br>
>>> >> >>>>>>>> -Matt<br>
>>> >> >>>>>>>><br>
>>> >> >>>>>>>> On Thu, Aug 29, 2013 at 5:50 PM, Nathan Goldbaum<br>
>>> >> >>>>>>>> <<a href="mailto:nathan12343@gmail.com">nathan12343@gmail.com</a>> wrote:<br>
>>> >> >>>>>>>>> I think to get everything working in a sustainable fashion,<br>
>>> >> >>>>>>>>> we<br>
>>> >> >>>>>>>>> would need<br>
>>> >> >>>>>>>>> buildbots for all platform combinations that we want to<br>
>>> >> >>>>>>>>> support,<br>
>>> >> >>>>>>>>> so<br>
>>> >> >>>>>>>>> all<br>
>>> >> >>>>>>>>> permutations of the (32/64 bit, linux / OS X / windows,<br>
>>> >> >>>>>>>>> py27/py3.3) tuple.<br>
>>> >> >>>>>>>>> At the moment anaconda seems to support 32 and 64 bit linux,<br>
>>> >> >>>>>>>>> 64<br>
>>> >> >>>>>>>>> bit<br>
>>> >> >>>>>>>>> OS X<br>
>>> >> >>>>>>>>> (not totally clear if OS X version matters), and 32 and 64<br>
>>> >> >>>>>>>>> bit<br>
>>> >> >>>>>>>>> windows.<br>
>>> >> >>>>>>>>><br>
>>> >> >>>>>>>>> Another option is to rely on conda build, which compiles<br>
>>> >> >>>>>>>>> everything<br>
>>> >> >>>>>>>>> from<br>
>>> >> >>>>>>>>> source.<br>
>>> >> >>>>>>>>><br>
>>> >> >>>>>>>>><br>
>>> >> >>>>>>>>> On Thu, Aug 29, 2013 at 2:45 PM, Stephen Skory <<a href="mailto:s@skory.us">s@skory.us</a>><br>
>>> >> >>>>>>>>> wrote:<br>
>>> >> >>>>>>>>>><br>
>>> >> >>>>>>>>>> Hi all,<br>
>>> >> >>>>>>>>>><br>
>>> >> >>>>>>>>>> I have less of a skin in this than I used to, but I'd like<br>
>>> >> >>>>>>>>>> to<br>
>>> >> >>>>>>>>>> raise<br>
>>> >> >>>>>>>>>> the issue of Windows & package managers. For example,<br>
>>> >> >>>>>>>>>> Anaconda<br>
>>> >> >>>>>>>>>> is<br>
>>> >> >>>>>>>>>> available for Windows - would that mean that yt might "just<br>
>>> >> >>>>>>>>>> work"<br>
>>> >> >>>>>>>>>> on<br>
>>> >> >>>>>>>>>> Windows? Or the opposite, and it would require a great deal<br>
>>> >> >>>>>>>>>> of<br>
>>> >> >>>>>>>>>> effort<br>
>>> >> >>>>>>>>>> to get all the various things we expect to be .so's to work<br>
>>> >> >>>>>>>>>> as<br>
>>> >> >>>>>>>>>> .dll's<br>
>>> >> >>>>>>>>>> (such as the Cython helpers or halo-finding stuff)?<br>
>>> >> >>>>>>>>>><br>
>>> >> >>>>>>>>>> I don't know the answers to these questions, but I think<br>
>>> >> >>>>>>>>>> it's<br>
>>> >> >>>>>>>>>> worth<br>
>>> >> >>>>>>>>>> thinking about.<br>
>>> >> >>>>>>>>>><br>
>>> >> >>>>>>>>>> --<br>
>>> >> >>>>>>>>>> Stephen Skory<br>
>>> >> >>>>>>>>>> <a href="mailto:s@skory.us">s@skory.us</a><br>
>>> >> >>>>>>>>>> <a href="http://stephenskory.com/" target="_blank">http://stephenskory.com/</a><br>
>>> >> >>>>>>>>>> <a href="tel:510.621.3687" value="+15106213687">510.621.3687</a> (google voice)<br>
>>> >> >>>>>>>>>> _______________________________________________<br>
>>> >> >>>>>>>>>> yt-dev mailing list<br>
>>> >> >>>>>>>>>> <a href="mailto:yt-dev@lists.spacepope.org">yt-dev@lists.spacepope.org</a><br>
>>> >> >>>>>>>>>> <a href="http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org" target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org</a><br>
>>> >> >>>>>>>>><br>
>>> >> >>>>>>>>><br>
>>> >> >>>>>>>>><br>
>>> >> >>>>>>>>> _______________________________________________<br>
>>> >> >>>>>>>>> yt-dev mailing list<br>
>>> >> >>>>>>>>> <a href="mailto:yt-dev@lists.spacepope.org">yt-dev@lists.spacepope.org</a><br>
>>> >> >>>>>>>>> <a href="http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org" target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org</a><br>
>>> >> >>>>>>>>><br>
>>> >> >>>>>>>> _______________________________________________<br>
>>> >> >>>>>>>> yt-dev mailing list<br>
>>> >> >>>>>>>> <a href="mailto:yt-dev@lists.spacepope.org">yt-dev@lists.spacepope.org</a><br>
>>> >> >>>>>>>> <a href="http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org" target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org</a><br>
>>> >> >>>>>>><br>
>>> >> >>>>>>><br>
>>> >> >>>>>>> _______________________________________________<br>
>>> >> >>>>>>> yt-dev mailing list<br>
>>> >> >>>>>>> <a href="mailto:yt-dev@lists.spacepope.org">yt-dev@lists.spacepope.org</a><br>
>>> >> >>>>>>> <a href="http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org" target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org</a><br>
>>> >> >>>>>>><br>
>>> >> >>>>>><br>
>>> >> >>>>>><br>
>>> >> >>>>>> _______________________________________________<br>
>>> >> >>>>>> yt-dev mailing list<br>
>>> >> >>>>>> <a href="mailto:yt-dev@lists.spacepope.org">yt-dev@lists.spacepope.org</a><br>
>>> >> >>>>>> <a href="http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org" target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org</a><br>
>>> >> >>>>>><br>
>>> >> >>>>><br>
>>> >> >>>>><br>
>>> >> >>>>> _______________________________________________<br>
>>> >> >>>>> yt-dev mailing list<br>
>>> >> >>>>> <a href="mailto:yt-dev@lists.spacepope.org">yt-dev@lists.spacepope.org</a><br>
>>> >> >>>>> <a href="http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org" target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org</a><br>
>>> >> >>>>><br>
>>> >> >>>><br>
>>> >> >>>><br>
>>> >> >>>> _______________________________________________<br>
>>> >> >>>> yt-dev mailing list<br>
>>> >> >>>> <a href="mailto:yt-dev@lists.spacepope.org">yt-dev@lists.spacepope.org</a><br>
>>> >> >>>> <a href="http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org" target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org</a><br>
>>> >> >>>><br>
>>> >> >>><br>
>>> >> >>><br>
>>> >> >>> _______________________________________________<br>
>>> >> >>> yt-dev mailing list<br>
>>> >> >>> <a href="mailto:yt-dev@lists.spacepope.org">yt-dev@lists.spacepope.org</a><br>
>>> >> >>> <a href="http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org" target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org</a><br>
>>> >> >>><br>
>>> >> >><br>
>>> >> >><br>
>>> >> >> _______________________________________________<br>
>>> >> >> yt-dev mailing list<br>
>>> >> >> <a href="mailto:yt-dev@lists.spacepope.org">yt-dev@lists.spacepope.org</a><br>
>>> >> >> <a href="http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org" target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org</a><br>
>>> >> >><br>
>>> >> > _______________________________________________<br>
>>> >> > yt-dev mailing list<br>
>>> >> > <a href="mailto:yt-dev@lists.spacepope.org">yt-dev@lists.spacepope.org</a><br>
>>> >> > <a href="http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org" target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org</a><br>
>>> >><br>
>>> >> _______________________________________________<br>
>>> >> yt-dev mailing list<br>
>>> >> <a href="mailto:yt-dev@lists.spacepope.org">yt-dev@lists.spacepope.org</a><br>
>>> >> <a href="http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org" target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org</a><br>
>>> ><br>
>>> ><br>
>>> ><br>
>>> > _______________________________________________<br>
>>> > yt-dev mailing list<br>
>>> > <a href="mailto:yt-dev@lists.spacepope.org">yt-dev@lists.spacepope.org</a><br>
>>> > <a href="http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org" target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org</a><br>
>>> ><br>
>>> _______________________________________________<br>
>>> yt-dev mailing list<br>
>>> <a href="mailto:yt-dev@lists.spacepope.org">yt-dev@lists.spacepope.org</a><br>
>>> <a href="http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org" target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org</a><br>
>><br>
>><br>
>><br>
>> _______________________________________________<br>
>> yt-dev mailing list<br>
>> <a href="mailto:yt-dev@lists.spacepope.org">yt-dev@lists.spacepope.org</a><br>
>> <a href="http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org" target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org</a><br>
>><br>
_______________________________________________<br>
yt-dev mailing list<br>
<a href="mailto:yt-dev@lists.spacepope.org">yt-dev@lists.spacepope.org</a><br>
<a href="http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org" target="_blank">http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org</a><br>
</div></div></blockquote></div><br></div>