Better name for dpush wanted

[Jelmer Vernooij]

Ian Clatworthy wrote:
> Jelmer Vernooij wrote:
>   
>> No, there was some discussion about it but we didn't come up with
>> anything final. I think the main thing to keep in mind is that the name
>> should express that the command destroys history in the source branch.
>> The best name I've heard so far was "push-lossy".
>>     
>
> My main concern with that is that it emphasizes the quality of the
> push over the objective of the command (which IMO is updating a
> foreign master & then resynching the local branch). Isn't it
> conceivable that the "lost" metadata be shuffled off to the side
> (therefore not populating svn revision properties say) in the future?
> If so, then the "lossy" vs "non-lossy" bit could be an option to
> push-xxx.
>   
Even in that case, the push still will have been lossy from the POV of
other users of the branch (the additional data is only present locally,
and not all users will have that data). So I think "lossy" is an
appropriate description for this kind of push.
>>> FWIW, the best suggestion I'd seen/heard so far is
>>> calling it "push-out".
>>>   
>>>       
>> What does the -out mean in this case exactly?
>>     
>
> It means "out to a foreign system and then re-synchronise this branch".
> The fact the it's updating the target branch *and* the local branch
> made me originally think a name like "sync" would be good. But
> "push-xxx" will be more discoverable I suspect.
>   
Yeah, I agree that a command name starting with push or at least with
push in the name makes the most sense.

Cheers,

Jelmer