how crippled is ruby support in YP 3.0?

how crippled is ruby support in YP 3.0?

[Robert P. J. Day]

  background: trying to get ruby and a number of ruby gems migrated
from "morty" to "zeus", and nothing but grief trying to
compile/install ruby gems.

  plan A: start with standard YP layers all checked out to "zeus", and
add vendor layer -- still similar grief.

  current plan: start from absolute scratch, new project targeting
x86_64, all layers checked out to "zeus", and build for
core-image-minimal (*that* i know i can do).

  then, little by little, add ruby content. has anyone here worked
with ruby and ruby gems extensively and has had to struggle to get
stuff working? as soon as this basic build is done, i'm going to just
add "ruby" to the image, and from there, start adding gems one by
one.

  should i expect this to go smoothly, or should i expect the pain
i've been through for the last couple days?

rday

Re: [yocto] how crippled is ruby support in YP 3.0?

[Richard Purdie]

On Fri, 2020-06-26 at 14:19 -0400, Robert P. J. Day wrote:
>   background: trying to get ruby and a number of ruby gems migrated
> from "morty" to "zeus", and nothing but grief trying to
> compile/install ruby gems.
> 
>   plan A: start with standard YP layers all checked out to "zeus",
> and
> add vendor layer -- still similar grief.
> 
>   current plan: start from absolute scratch, new project targeting
> x86_64, all layers checked out to "zeus", and build for
> core-image-minimal (*that* i know i can do).
> 
>   then, little by little, add ruby content. has anyone here worked
> with ruby and ruby gems extensively and has had to struggle to get
> stuff working? as soon as this basic build is done, i'm going to just
> add "ruby" to the image, and from there, start adding gems one by
> one.
> 
>   should i expect this to go smoothly, or should i expect the pain
> i've been through for the last couple days?

If we actually had some tests for it, we might have a much better idea.

The areas of the project with good automated tests tend not to regress
without our noticing. Areas with no tests can and do :(

Cheers,

Richard

Re: [yocto] how crippled is ruby support in YP 3.0?

[Alexander Kanavin]

[-- Attachment #1: Type: text/plain, Size: 490 bytes --]

On Fri, 26 Jun 2020 at 23:18, Richard Purdie <
richard.purdie@linuxfoundation.org> wrote:

> If we actually had some tests for it, we might have a much better idea.
>
> The areas of the project with good automated tests tend not to regress
> without our noticing. Areas with no tests can and do :(
>

Not only that, the only reason ruby is in core at all is that webkit needs
ruby-native. The interest in it is simply too low, but of course automated
tests are welcome.

Alex

[-- Attachment #2: Type: text/html, Size: 809 bytes --]