JuliaHEP / AwkwardArray.jl

Awkward Array in Julia mirrors the Python library, enabling effortless zero-copy data exchange between Julia and Python
https://juliahep.github.io/AwkwardArray.jl/dev/
MIT License
31 stars 2 forks source link

docs: add julia installation note #99

Closed ianna closed 3 months ago

Moelf commented 3 months ago

Julia is best install via JuliaUp though

ianna commented 3 months ago

Julia is best install via JuliaUp though

I think, we are trying to find an environment suitable for both Python and Julia. Do you know if any Julia features are not available if it is installed via conda? It does install the latest version.

Moelf commented 3 months ago

If it's just redistributing the official binary it's fine, I'm worried it's linking to anything in conda. And I suppose another worry is if it would be maintained later.

ianna commented 3 months ago

If it's just redistributing the official binary it's fine, I'm worried it's linking to anything in conda. And I suppose another worry is if it would be maintained later.

We already depend on CondaPkg.jl ;-)

Here is the full package plan:

% conda install -c conda-forge julia
Retrieving notices: ...working... done
Channels:
 - conda-forge
 - defaults
Platform: osx-64
Collecting package metadata (repodata.json): done
Solving environment: done

## Package Plan ##

  environment location: /Users/yana/anaconda3/envs/py311-julia19

  added / updated specs:
    - julia

The following packages will be downloaded:

    package                    |            build
    ---------------------------|-----------------
    arpack-3.9.1               |nompi_hf81eadf_101         123 KB  conda-forge
    curl-8.8.0                 |       hea67d85_0         150 KB  conda-forge
    git-2.45.1                 | pl5321h9d8b143_0        11.2 MB  conda-forge
    julia-1.10.3               |       hcf52d30_1        83.4 MB  conda-forge
    libcurl-8.8.0              |       hf9fcc65_0         377 KB  conda-forge
    libhwloc-2.10.0            |default_h456cccd_1001         2.2 MB  conda-forge
    libopenblas-ilp64-0.3.27   |openmp_h17068ef_0         5.6 MB  conda-forge
    libopenlibm4-0.8.1         |       h10d778d_1          86 KB  conda-forge
    libxml2-2.12.7             |       h3e169fe_1         605 KB  conda-forge
    openblas-ilp64-0.3.27      |openmp_ha601604_0         5.2 MB  conda-forge
    openlibm-0.8.1             |       h10d778d_1          29 KB  conda-forge
    openssl-3.3.1              |       h87427d6_0         2.4 MB  conda-forge
    perl-5.32.1                | 7_h10d778d_perl5        11.8 MB  conda-forge
    suitesparse-5.10.1         |       h4bf45ed_3         1.4 MB  conda-forge
    tbb-2021.12.0              |       h3c5361c_1         168 KB  conda-forge
    zlib-1.3.1                 |       h87427d6_1          87 KB  conda-forge
    zstd-1.5.6                 |       h915ae27_0         487 KB  conda-forge
    ------------------------------------------------------------
                                           Total:       125.3 MB

The following NEW packages will be INSTALLED:

  arpack             conda-forge/osx-64::arpack-3.9.1-nompi_hf81eadf_101 
  c-ares             conda-forge/osx-64::c-ares-1.28.1-h10d778d_0 
  curl               conda-forge/osx-64::curl-8.8.0-hea67d85_0 
  git                conda-forge/osx-64::git-2.45.1-pl5321h9d8b143_0 
  gmp                conda-forge/osx-64::gmp-6.3.0-h73e2aa4_1 
  icu                conda-forge/osx-64::icu-73.2-hf5e326d_0 
  julia              conda-forge/osx-64::julia-1.10.3-hcf52d30_1 
  krb5               conda-forge/osx-64::krb5-1.21.2-hb884880_0 
  libcurl            conda-forge/osx-64::libcurl-8.8.0-hf9fcc65_0 
  libedit            conda-forge/osx-64::libedit-3.1.20191231-h0678c8f_2 
  libev              conda-forge/osx-64::libev-4.33-h10d778d_2 
  libhwloc           conda-forge/osx-64::libhwloc-2.10.0-default_h456cccd_1001 
  libiconv           conda-forge/osx-64::libiconv-1.17-hd75f5a5_2 
  libintl            conda-forge/osx-64::libintl-0.22.5-h5ff76d1_2 
  libnghttp2         conda-forge/osx-64::libnghttp2-1.58.0-h64cf6d3_1 
  libopenblas-ilp64  conda-forge/osx-64::libopenblas-ilp64-0.3.27-openmp_h17068ef_0 
  libopenlibm4       conda-forge/osx-64::libopenlibm4-0.8.1-h10d778d_1 
  libosxunwind       conda-forge/osx-64::libosxunwind-0.0.6-h940c156_0 
  libssh2            conda-forge/osx-64::libssh2-1.11.0-hd019ec5_0 
  libutf8proc        conda-forge/osx-64::libutf8proc-2.8.0-hb7f2c08_0 
  libxml2            conda-forge/osx-64::libxml2-2.12.7-h3e169fe_1 
  metis              conda-forge/osx-64::metis-5.1.0-he965462_1007 
  mpfr               conda-forge/osx-64::mpfr-4.2.1-h4f6b447_1 
  openblas-ilp64     conda-forge/osx-64::openblas-ilp64-0.3.27-openmp_ha601604_0 
  openlibm           conda-forge/osx-64::openlibm-0.8.1-h10d778d_1 
  p7zip              conda-forge/osx-64::p7zip-16.02-he49afe7_1001 
  pcre2              conda-forge/osx-64::pcre2-10.43-h0ad2156_0 
  perl               conda-forge/osx-64::perl-5.32.1-7_h10d778d_perl5 
  suitesparse        conda-forge/osx-64::suitesparse-5.10.1-h4bf45ed_3 
  tbb                conda-forge/osx-64::tbb-2021.12.0-h3c5361c_1 
  zlib               conda-forge/osx-64::zlib-1.3.1-h87427d6_1 
  zstd               conda-forge/osx-64::zstd-1.5.6-h915ae27_0 

The following packages will be UPDATED:

  openssl                                  3.3.0-h87427d6_3 --> 3.3.1-h87427d6_0 

Proceed ([y]/n)? y

Downloading and Extracting Packages:

Preparing transaction: done                                                                                                                                                           
Verifying transaction: done                                                                                                                                                           
Executing transaction: done         
Moelf commented 3 months ago

yeah so I'm still not sure if it's built from source by Conda or not -- we have the same problem with Linux distributions, basically they would compile and not test Julia and users would then run into weird problems some times.

Depending on CondaPkg.jl is fine I meant to say one can worry if the Julia in conda-forge will be maintained or not.