microsoft / microsoft/winget-pkgs
[Update Request]: Julia
Nobody has claimed this yet.
- Dominant language
- No language data
- Stars
- 11.1k
- Forks
- 9.7k
- Avg merge
- 1h 32m
- Merged PRs (30d)
- 376
Description
What type of update are you requesting?
A new locale, architecture, or installer type of an existing version
Current Package Identifier
Julialang.Julia
Package Version
More generic
Please describe the changes you would like to see
This is a more general discussion/question what we should do about the Julia programming language in this repo here. For a bit of background:
- https://github.com/julialang/juliaup based installers are now the recommended way to install Julia on all platforms and we are trying to make sure that as many package managers/stores etc offer a Juliaup based installer for their
Juliapackage. Juliain the Windows Store has always used a Juliaup based installer. Sowinget install julia -s msstoreworks and is currently officially the recommended way to install Julia on Windows.- This repo here has manifests that point to the now legacy inno based installers for Julia, one can get those via
winget install julia -s winget.
In general, I think it would be good to phase these inno based installers out and try to make sure that all roads lead to a Juliaup based setup. Right now, I see two ways to do this:
- Just entirely remove Julia from this repo here and only offer the Windows Store version via
winget. - We have MSI (and also MSIX) based installers that are part of the Juliaup story, and we could replace
Juliahere in this repo with a manifest that points to these MSI installers that utilize Juliaup.
I think the only benefit of 1) in my mind would be that winget install julia would work, i.e. we would no longer need the -s msstore flag for the default case. But on the flipside, there are user systems where the Windows Store is blocked, so that is not ideal. Plus, maybe at some point winget will provide some way to prioritize a Windows Store version, so maybe if we just wait winget install julia will just automatically install from the store, if a user system has the store available, in the future.
The main benefit of 2) is that we might be able to target slightly more systems, e.g. server systems, system wide installs etc. That is based on the assumption that there are systems out there where winget with a MSI installer works, even though the Windows Store is say blocked by an admin. Not sure whether that is really so.
If we were to go with 2), we could also entirely automate the updates of the manifests here, we do that already with other package managers.
In any case, I wanted to open this issue to get some feedback how we should best proceed. I'll CC everyone who in the past has contributed to the Julia manifest in this repo in the recent past: @jarredclloyd, @torculus, @ZenanH, @OfficialEsco, @Trenly, @vedantmgoyal2009.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Review the existing Julia manifests in this repository and the Juliaup-based MSI and MSIX installer options described in the issue. A decision between removing Julia and replacing its manifests is needed first; done means the chosen approach is agreed and the repository's Julia package definition reflects it.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- julia
- Domain
- tooling
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100