zstd and Mercurial packaging

[Sean Farley]

Augie Fackler <raf at durin42.com> writes:

>> On Dec 1, 2016, at 17:39, Sean Farley <sean at farley.io> wrote:
>> 
>>>> I don't understand why there isn't a switch to enable this?
>>> 
>>> I'm not sure what "this" is - if "this" is "using zstd at all" then yes, I believe that's in there. If it's "using a system zstd instead of the vendored one" that's *technically infeasible* with the current stable zstd API. The wire format is stable, but the API is still under some amount of development and they're being careful.
>> 
>> Then I don't think zstd is ready to be used. If it's not ready, then
>> it's not ready. I'm a bit sad but unless I can link with the system
>> provided zstd, then this is a no-go for me.
>
> Noted. That seems like a reasonable position.
>
>> 
>>>> When I read
>>>> the patches, I thought that was the goal? I'm a bit disappointed in this
>>>> situation now and retract my support of vendoring zstd.
>>> 
>>> If it's that worrisome, have MacPorts ship hg without zstd - that'll always be fine.
>> 
>> I'm afraid I'll have to do that. I'm a bit worried with how fast this
>> was pushed through and how many other distros will turn off zstd.
>
> For what (little) it's worth, you're the only maintainer I've heard from that has a problem with this. I hope you'll reconsider in the name of IO speed, but realistically I don't think it'll matter soon (given that I expect the zstd people to finish their API.)
>
> Excluding zstd from your packages is totally fine: that's why we made it possible (and if it's not, that's strictly a bug IMO, and is something I'll gladly take patches for.)

Maybe I'm a little weird which is fine. I've been called worse ;-)

I would have spoken up sooner were I not so busy. I'll see what I can
hack on this weekend. And it might be moot soon anyways if the zstd api
becomes stable enough.