Re: [PATCH v2 2/3] fast-export: improve speed by skipping blobs

From: Felipe Contreras <felipe.contreras@gmail.com>
Date: May 7, 2013, 03:49 UTC
Message-ID: <CAMP44s0VBiWN5fZachCDkg9EHZKHRU5idNBdktV2a5wBMJvxjQ@mail.gmail.com>
In-Reply-To: <7vmws7o8hx.fsf@alter.siamese.dyndns.org>

On Mon, May 6, 2013 at 8:59 PM, Junio C Hamano <gitster@pobox.com> wrote:

Show 11 quoted lines

> Felipe Contreras <felipe.contreras@gmail.com> writes:
>
>>>> How?
>>>
>>>  * if we teach fast-import to optionally not write marks for blobs
>>>    and trees out, your remote-bzr can take advantage of it,
>>
>> I already said remote-bzr is irrelevant. *Everybody* benefits.
>
> Everybody who does _not_ need to look at marks for non-commits from
> previous run does.

IOW; everyone.

> What about the others who do?

Like who?

> Surely, some lucky ones may get the benefit of a new optimization
> for free if you turn it on uncondtionally without an escape hatch,
> but that goes against our goal of not to knowingly introduce any
> regression.

That's different. One thing is to turn it on unconditionally, and another thing is to turn it on by *default*.

Show 7 quoted lines

> Michael's cvs2git might have a way to work the breakage
> around (I will let him comment on your suggested workaround), but as
> long as he has been coding it using the documented feature, why
> should he even change anything for no gain at all in the first
> place?  Even if you have a workaround, that does not change the fact
> that a removal of a feature that somebody has been depending on is a
> regression.

Who is depending on it? Michael didn't say that he used that feature, merely that it was documented in cvs2git, because Windows doesn't have 'cat'. He claimed that other people *might* be using that "feature", but we don't *know*.

Is a couple of commands somebody wrote in some documentation which can be easily fixed, reason enough to punish everyone else?

> What's so difficult to understand that by default the responsibility
> of making sure an optimization applies safely to a program that uses
> a new optmization lies on that program, in other words, a new
> feature is by default an opt-in?

Is that written in some git bible descended from some god that I missed? If not, everything are guidelines, and guidelines are there for a reason, and those reasons can be challenged, and so can the guidelines.

Sometimes it makes sense to make a new feature opt-in, sometimes it doesn't, there are no absolutes, there should be no dogmas.

-- 
Felipe Contreras