Re: [Qemu-devel] Should we auto-generate IDs?

[Kevin Wolf <kwolf@redhat.com>]

Am 27.08.2015 um 14:32 hat Jeff Cody geschrieben:
> I'm not married to the ID generation scheme I proposed.  
> 
> What I am trying to do, however, is have a technical discussion on
> generating an ID in a well-formed manner.  And hopefully, in a way
> that is useful to all interested subsystems, if possible.
> 
> Do you disagree with the requirements I listed above?  If so, it would
> be useful to begin the discussion around that.  For ease of
> discussion, I'll list them again:
> 
> * Reserved namespaces
> * Uniqueness
> * Non-predictable (to avoid inadvertently creating a de facto ABI)
> 
> 
> . . .
> 
> On the generation scheme proposed above:
> 
> I understand that something you desire is an ID that is easier to
> type.
> 
> If we wanted to make it shorter, perhaps we could have the number
> counter be variable length:
> 
>             qemu#ss#D#XY
>               |   | | |
> qemu reserved -   | | |
>                   | | |
> subsystem name ---| | |
>                     | |
>     counter --------| |
>                       |
>     2-digit random ---|

Even with keeping all of the information in there we can shorten the ID
a bit more: # at the start is enough to mark it as autogenerated, the
subsystem seems nice to have in there anyway, and the # separators can
be removed without making the ID less unique (assuming that subsystems
never end in a digit). This results in an ID that looks like a three (or
more) digit number for the subsystem, where the last two digits are
random, like this:

    #block150
    #block219
    #block344
    ...

That seems easy to type and still fulfills all of the criteria.

Kevin