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

Bug#767587: please prevent auto-removal of mysql-workbench from "testing"



On Sat, 1 Nov 2014 21:08:56 Julien Cristau wrote:
> > It is hardly a "worst possible outcome" when all alternatives are equally
> > suitable.
> 
> But they're clearly not.

Let's distinguish generic statement from the particular case please.

From mysql-workbench prospective there might be no difference in which client 
package to use. Let's consider hypothetical example: M-W uses `mysql` command 
line client only to check DB connectivity (in fact it is not how M-W really 
checks connectivity). All packages providing "virtual-mysql-client" have 
`mysql` executable taking exactly the same command line options and capable to 
connect to any server. Why prefer a particular client?
Here you're failing to answer my questions...


> The default mysql client is the mysql-client
> package.  Why do you insist that it isn't?

I've never claimed otherwise. I merely see little reason to prefer "default" 
client. But I would agree if you'd say that it would be a good idea to 
recommend "mysql-client" as first alternative archive-wide to ensure that it 
is always a "default" client. However it would hardly explain why you consider 
lack of such alternative as "serious" bug.


> > > > I've realised that "virtual-mysql-client" can be demoted to Recommends
> > > > --
> > > > please advise if that change warrants an unblock in which case I will
> > > > upload to "unstable". Thank you.
> > > 
> > > Not if it stays virtual-only.

C'mon, why does it matter in Recommends as much as you say it matters in 
Depends?

I'm OK to upload M-W to "unstable" with 

	Recommends: mysql-client | virtual-mysql-client

instead of

	Depends: virtual-mysql-client

I'm willing to comply because you insist, not because I'm convinced.
IMHO this issue is not worth arguing and I recognise (however little) benefit 
from defaulting to default "mysql-client" client even if it makes no 
functional difference for recommending package. If you believe it's better --  
that is enough for me.

Please advise if I shall proceed with upload. Thanks.

-- 
Cheers,
 Dmitry Smirnov.

---

Science embraces facts and debates opinion; religion embraces opinion
and debates the facts.
        -- Tom Heehler, The Well-Spoken Thesaurus.

Attachment: signature.asc
Description: This is a digitally signed message part.


Reply to: