# Closed Source Dependencies and Licenses from SAS

**URL:** https://discourse.julialang.org/t/closed-source-dependencies-and-licenses-from-sas/64090
**Category:** Package Management
**Created:** [July 5, 2021, 5:26pm UTC](https://discourse.julialang.org/t/closed-source-dependencies-and-licenses-from-sas/64090 "2021-07-05T17:26:28Z")
**Posts on this page:** 4
**Page:** 1

<div class="post-metadata">

### Author: ![nucklass](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/nucklass/32/11989_2.png) [@nucklass](https://discourse.julialang.org/u/nucklass)
#### Post date: [July 5, 2021, 5:26pm UTC](https://discourse.julialang.org/t/closed-source-dependencies-and-licenses-from-sas/64090/1 "2021-07-05T17:26:28Z")

</div>

Hoping some some more experienced devs could chime in. I’m looking to write a package for Julia to interface with SAS, akin to the [saspy](https://github.com/sassoftware/saspy) package SAS maintains. Long story short, the Julia package will need these [four jars](https://github.com/sassoftware/saspy/tree/main/saspy/java/iomclient) in order to run. Until recently, a user had to grab these themselves from their SAS installation, but the jars are now located in this SAS-maintained repo, along with their own [license](https://github.com/sassoftware/saspy/blob/main/LICENSE.md) . I’m inexperienced managing software licenses and the like, so would anyone more experienced know if I’d get into legal trouble if I were to include code to download these jars in a build script? Thanks!

---

<div class="post-metadata">

### Author: ![oxinabox](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/oxinabox/32/206603_2.png) [@oxinabox](https://discourse.julialang.org/u/oxinabox)
#### Post date: [July 5, 2021, 5:59pm UTC](https://discourse.julialang.org/t/closed-source-dependencies-and-licenses-from-sas/64090/2 "2021-07-05T17:59:04Z")

</div>

That license is nasty.

> Customer …warrants that … (ii) neither Customer nor any User is located in … [sanctioned] countries (currently Cuba, Iran, North Korea, Syria, and Sudan)…  
> …  
> Customer agrees not to release the results of any benchmarking you perform on the Code or the Software.

I am not a lawyer.

I would guess you probably _can_ let users download them in your build script.  
The license does say:

> … You are downloading software code…

which implies downloading the code is permitted.  
Worst case you would have to download the whole saspy and then extract those files, I guess.

I might be tempted to make that only happen if the user has set a environment variable or agreed to a prompt: _“I have read and agree to SAS’ nightmare fuel license, which is distinct from the license of SAS.jl”_.

I wouldn’t want to install that as an indirect dependency.

I think if doing that you might be able to put it in General registry, but idk.

---

<div class="post-metadata">

### Author: ![ImreSamu](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/imresamu/32/20677_2.png) [@ImreSamu](https://discourse.julialang.org/u/ImreSamu)
#### Post date: [July 5, 2021, 6:12pm UTC](https://discourse.julialang.org/t/closed-source-dependencies-and-licenses-from-sas/64090/3 "2021-07-05T18:12:50Z")

</div>

IMHO:

- just create an issue … ( in: [Issues · sassoftware/saspy · GitHub](https://github.com/sassoftware/saspy/issues) )
  - and ask help ! 🙂
    - in the licensing issues …
    - in the development.

---

<div class="post-metadata">

### Author: ![nucklass](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/nucklass/32/11989_2.png) [@nucklass](https://discourse.julialang.org/u/nucklass)
#### Post date: [July 5, 2021, 7:03pm UTC](https://discourse.julialang.org/t/closed-source-dependencies-and-licenses-from-sas/64090/4 "2021-07-05T19:03:00Z")

</div>

> [@ImreSamu](#):
>
> create an issue

Yeah I figure this is a good idea now, initially I wasn’t sure this was appropriate given it wasn’t really an issue with saspy per se.

> [@oxinabox](#):
>
> That license is nasty.

Good I asked before just including the code, then! I found the no benchmarking clause, well, interesting…
