I applaud your persistence!!

Also take a look at the issue at IJulia here:
https://github.com/JuliaLang/IJulia.jl/issues/199
Where discussion of distributing an ipython.exe for windows is (hopefully)
imminent. That means when you do a Pkg.add("IJulia"), windows users at
least would have a nicely package executable with everything pre-packaged.
I've done some testing on it and it's truly a thing of wonder.

-Jacob


On Mon, Aug 4, 2014 at 5:33 AM, Ivo Balbaert <[email protected]> wrote:

> Hi fellow Julia lovers,
>
> I wanted to share how I struggled to get IJulia to work on Windows (8.1)
> and how all the pieces fell together.
>
> This may not be news to experts, but I am sure it may be of use to
> developers working on Windows.
>
> It was all the stranger because IJulia works on Ubuntu 14.04 in a virtual
> machine on the same computer (as a side remark: Julia works a lot faster in
> this Ubuntu VM than on the host computer with Windows, especially all
> operations with Pkg)
>
> But on Windows, after I had gone through the whole installation procedure,
> the IJulia notebook started but when I wanted to evaluate something after a
> few s the message came*: The kernel appears to have died. It will restart
> automatically. - ERROR: could not open file
> C:\Users\CVO\.julia\v0.3\Nettle\src\../deps/deps.jl*
>
> Then trying to pinpoint the error, I saw that Pkg.build(“IJulia”) was not
> successfully completed:         *WARNING: LibCURL and Nettle had build
> errors*
>
> Then I tried every suggestion (for weeks of and on) I could find with Pkg,
> clearing the cache and so on, going through IJulia issues 159 and 198
> nothing worked, until I followed a suggestion (which at first didn’t seem
> related) to:
>
> *https://github.com/JuliaLang/WinRPM.jl/issues/27
> <https://github.com/JuliaLang/WinRPM.jl/issues/27>*
>
> I did a WinRPM.update() and found that I had the same error as reported in
> this issue:
>
> WinRPM.update()
>
> INFO: Downloading
> http://download.opensuse.org/repositories/windows:/mingw:/win32/openSUSE_Factory//repodata/repomd.xml
>
> INFO: Downloading
> http://download.opensuse.org/repositories/windows:/mingw:/win32/openSUSE_Factory//repodata/8f70e2c263bb70ad72eb0122587216a34bb165e385b0a558447edb518
>
> WARNING: received error 0 while downloading
> http://download.opensuse.org/repositories/windows:/mingw:/win32/openSUSE_Factory//repodata/8f70e2c263bb70ad72eb0122587216a
>
> primary.xml.gz
>
>
>
> SOLUTION:
>
> As mentioned in this issue, there is a problem with the download from
> opensuse. The suggestion was:
>
> Replace the content of the file sources.list in the WinRPM folder
> %HOME%\.julia\v0.3\WinRPM with this:
>
> http://ftp5.gwdg.de/pub/opensuse/repositories/windows:/mingw:/win32/openSUSE_Factory/
>
> http://ftp5.gwdg.de/pub/opensuse/repositories/windows:/mingw:/win64/openSUSE_Factory/
>
>
>
> At first this did not help, even after restarting, doing a julia> 
> cd(Pkg.dir("WinRPM"))
>
> shell>; git reset HEAD --hard
>
>  changing the url’s in a cache file, etc, the error persisted.
>
> Yesterday Jameson Nash suggested to do a: WinRPM.__init__()
>
>  At first no success, but after I restarted the Windows machine finally
> WinRPM.update() worked.
>
> Then in one stroke Pkg.build worked for LibCURL, Nettle and IJulia
>
> And the kernel in IJulia no longer dies!
>
> The bottom line I think is that when the WinRPM package installation is
> smoothened out to work fine on Windows, a lot of Julia package install
> problems on Windows will be solved.
>
>  Thanks everyone,
>
> Ivo
>
>  ---
>
>  Here is the versioninfo() of my Julia installation on Windows:
>
> Julia Version 0.3.0-prerelease+3952
>
> Commit b2fd3af (2014-06-30 01:39 UTC)
>
> Platform Info:
>
>   System: Windows (x86_64-w64-mingw32)
>
>   CPU: AMD FX(tm)-8150 Eight-Core Processor
>
>   WORD_SIZE: 64
>
>   BLAS: libopenblas (USE64BITINT DYNAMIC_ARCH NO_AFFINITY)
>
>   LAPACK: libopenblas
>
>   LIBM: libopenlibm
>  Here is the versioninfo() of my Julia installation on Windows:
> Julia Version 0.3.0-prerelease+3952
> Commit b2fd3af (2014-06-30 01:39 UTC)
> Platform Info:
>   System: Windows (x86_64-w64-mingw32)
>   CPU: AMD FX(tm)-8150 Eight-Core Processor
>   WORD_SIZE: 64
>   BLAS: libopenblas (USE64BITINT DYNAMIC_ARCH NO_AFFINITY)
>   LAPACK: libopenblas
>   LIBM: libopenlibm
>

Reply via email to