Re: [PATCH] sha1-file: make pretend_object_file() not prefetch

Jeff King [off-list ref] writes:

I agree this patch is fine, but I wonder if it could go even further. If
we are pretending some particular contents, shouldn't we override
anything that might be in the object database? I.e., could we eliminate
this has_object_file() entirely?

That should be OK for the same reason that it's OK to use QUICK.

There's only one caller of this function (git-blame), which I think
would be happy with such a change.

I actually have to take that "we could even lie to say that content
that does not hash to X is object X" back---that was never the
intention of this mechanism.

It was to ensure that operations that are supposedly read-only can
avoid writing into the repository---"blame" uses it to pretend as if
the working tree file already has a corresponding object in the
object store, IIRC.

The only reason why hash_object_file() is used here is to allow us
discarding the memory held for that working tree copy if it happens
to match what is stored in the object database.  The saving would be
within a few hundred kilobytes to a single digit megabyte range at
most, hopefully, so we could drop it (oh, saying "a single digit
megabyte" reminds me that my first Linux computer was 486dx with
4MB ram---on that box, it may have mattered).