pyjulia was mostly created to show off interop and for demos.  I don't 
think anyone uses it seriously.  It might make more sense now with 
precompilation as it used to take 10+ seconds to load PyCall and initialize 
Julia.

On Thursday, September 3, 2015 at 3:17:38 PM UTC-4, Tero Frondelius wrote:
>
> Sorry you got me wrong. I found Julia thus I don't want to use Python 
> anymore for the new stuff. I think this is the case for most of the Julia 
> users, like you said. I was just trying to answer your question why pyjulia 
> is not used intensively. Anyway I will promise to ask my colleague, if he 
> will see it an easy transition from Python to Julia he will probably do the 
> conda package in couple of hours.
>
>
> On Thursday, September 3, 2015 at 10:00:36 PM UTC+3, Tony Kelman wrote:
>>
>> If you want to see it happen, make it happen. The problem here is most of 
>> the people who do development work on Julia itself don't want to write in 
>> Python any more (or never did). If you spend a lot of time going back and 
>> forth between Python and Julia, then sure conda works much better than the 
>> alternatives for the Python parts. Would it be enough better than what we 
>> currently have for the Julia parts to be worth the maintenance effort of 
>> setting everything up as conda packages (and keeping them all updated)?
>>
>>
>> On Thursday, September 3, 2015 at 11:32:00 AM UTC-7, Tero Frondelius 
>> wrote:
>>>
>>> I think you answered yourself. As an Anaconda user I will type: conda 
>>> install X and after this the X starts working even with the open Jupyter 
>>> notebook by just importing the X. Now when you just look the
>>>  https://github.com/JuliaLang/pyjulia/blob/master/README.md 
>>> <https://github.com/JuliaLang/pyjulia/blob/master/README.md> you lost 
>>> major part of the Anaconda users. The "requirement" is to have something as 
>>> convenient as conda install pyjulia. (I would have written pip install 
>>> pyjulia, but it's not supporting binary dependencies.) 
>>>
>>> On Thursday, September 3, 2015 at 9:19:24 PM UTC+3, Tony Kelman wrote:
>>>>
>>>> That's fair. Not being a Python library writer, I don't know what 
>>>> issues are preventing the likes of https://github.com/JuliaLang/pyjulia 
>>>> from being more widely used. I suspect that ease of installation is 
>>>> probably not the biggest factor keeping this from being a more common use 
>>>> case right now, and development time would likely be better spent 
>>>> addressing functional and usability issues rather than conda-izing what we 
>>>> have today.
>>>>
>>>>
>>>> On Thursday, September 3, 2015 at 8:17:37 AM UTC-7, Cedric St-Jean 
>>>> wrote:
>>>>>
>>>>> On Thursday, September 3, 2015 at 1:29:03 AM UTC-4, Tony Kelman wrote:
>>>>>>
>>>>>> What hasn't been done yet, but potentially could be, would be adding 
>>>>>> Julia and Julia packages to Conda so you could do "conda install julia" 
>>>>>> or 
>>>>>> "conda install DataFrames.jl" or similar. I'm not sure whether that 
>>>>>> would 
>>>>>> really solve all that many problems since you're adding an extra 
>>>>>> installation and packaging (and future updating) step beyond what we 
>>>>>> already do with base Julia, Pkg, and METADATA, but if anyone out there 
>>>>>> really wants to see that happen you're welcome to go for it.
>>>>>>
>>>>>
>>>>> If `conda install julia` is a thing, then a Python user could `conda 
>>>>> install py_juMP`. That strikes me as a big deal for convincing Python 
>>>>> library writers to ditch C and write their code in Julia instead.
>>>>>  
>>>>>
>>>>>>
>>>>>> On Wednesday, September 2, 2015 at 11:17:38 AM UTC-7, Luthaf wrote:
>>>>>>>
>>>>>>> I do not want to replace the Base.Pkg package manager. Pkg does 
>>>>>>> install binary dependencies in a cross-platform way, but only by the 
>>>>>>> mean 
>>>>>>> of BinDeps. And BinDeps uses for that the concept of provider. Some 
>>>>>>> example 
>>>>>>> of providers are Hombrew.jl on OSX, Pacman on arch-linux, Yum on 
>>>>>>> centos/fedora distro, AptGet on debian distro, WinRPM.jl for 
>>>>>>> windows. But all these providers are not cross-platform, and you even 
>>>>>>> need 
>>>>>>> root access for using some of the Linux providers (Pacman, Yum, 
>>>>>>> AptGet). 
>>>>>>>
>>>>>>> Conda.jl is an other BinDeps provider, which can be used for all 
>>>>>>> platforms, effectively replacing any other provider. It can also be 
>>>>>>> used 
>>>>>>> without administrator rights on Linux.
>>>>>>> So it is not a Base.Pkg replacement, but an other way to get binary 
>>>>>>> dependencies installed with it.
>>>>>>>
>>>>>>> I now realize that this was not clear on my initial message, sorry 
>>>>>>> about that. I will also improve the README.
>>>>>>>
>>>>>>> Uwe Fechner a écrit : 
>>>>>>>
>>>>>>> Julia does have a very good internal package manager, that can also 
>>>>>>> install binary dependencies cross-platform.
>>>>>>> Why would you want to add another package manager?
>>>>>>>
>>>>>>> Am Dienstag, 1. September 2015 14:42:31 UTC+2 schrieb Luthaf:
>>>>>>>
>>>>>>> Hi Julians! 
>>>>>>>
>>>>>>> I am happy to present you the Conda.jl 
>>>>>>> <https://github.com/Luthaf/Conda.jl> package, a binary dependencies 
>>>>>>> manager for Julia based on the open-source conda 
>>>>>>> <http://conda.pydata.org/> package manager.
>>>>>>>
>>>>>>> Some interesting features of the Conda package manager:
>>>>>>>  - You can easily add your own software and use your own channel for 
>>>>>>> software distribution;
>>>>>>>  - You can install packages as non root on Linux;
>>>>>>>  - Conda is cross-plateforme, and you can use it for all your binary 
>>>>>>> dependencies, provided the binaries have been uploaded.
>>>>>>>
>>>>>>> I'll love to have your input on the code or the functionalities.
>>>>>>>
>>>>>>> Cheers
>>>>>>> Guillaume
>>>>>>>
>>>>>>>

Reply via email to