tech-pkg archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: pkg_admin check vs ruby



On Wed, Sep 06, 2023 at 09:49:01AM +0100, Jonathan Perkin wrote:
> Any time I have a system with ruby packages installed, they cause "pkg_admin
> check" to fail:
> 
>   $ pkg_admin check
>   ./opt/pkg/lib/ruby/gems/3.1.0/doc/i18n-1.14.1/ri/I18n/Gettext/Helpers/N_-i.ri fails MD5 checksum
>   ......../opt/pkg/lib/ruby/gems/3.1.0/doc/em-websocket-0.5.3/ri/EventMachine/WebSocket/cdesc-WebSocket.ri fails MD5 checksum
>   .........................................................../opt/pkg/lib/ruby/gems/3.1.0/doc/kramdown-2.4.0/ri/Kramdown/Converter/HashAST/cdesc-HashAST.ri fails MD5 checksum
>   ............................................................................................................................................/opt/pkg/lib/ruby/gems/3.1.0/doc/http_parser.rb-0.8.0/ri/HTTP/cdesc-HTTP.ri fails MD5 checksum
>   .................................................../opt/pkg/share/ri/3.1.0/system/Net/HTTPGatewayTimeOut/cdesc-HTTPGatewayTimeOut.ri fails MD5 checksum
>   /opt/pkg/share/ri/3.1.0/system/Net/HTTPRequestTimeOut/cdesc-HTTPRequestTimeOut.ri fails MD5 checksum
>   ...........
> 
> Presumably ruby is doing some behind the scenes stuff to modify these data
> files.  Can we either stop it doing that, or move them to ${VARBASE} or
> something?

ocaml has this issue too, I see:
/usr/pkg/lib/ocaml/ld.conf fails MD5 checksum

> Also I'd like to remove the annoying "..." progress output of pkg_admin
> check.  Others would argue it violates principles of silent command output,
> should be on stderr, etc.  I'm simply annoyed that it makes everything look
> terrible when there are errors, as seen above.

Yes please.

> It also exits success when there are failures.

Ouch.
 Thomas


Home | Main Index | Thread Index | Old Index