Re: [PATCH v8 32/41] environment: add set_index_file()

Duy Nguyen [off-list ref] writes:

On Fri, Jul 29, 2016 at 8:23 PM, Christian Couder

quoted

Do you mean that it might be a source of micro-projects for the next
GSoC or Outreachy? ;-)

No that's what I meant by boring. This is not something to interest
GSoC candidates, and while it looks simple, sometimes it needs a good
understanding of git overall, and it's definitely not small enough for
a micro project.

I think "that's not what I meant" is what you meant ;-) but anyway,
I actually view that as part of the same "libify" project, and not
solving it and building an "am that makes an internal call to apply
that is not libified" is adding technical debt to the codebase.

It may be a good trade-off between having "am that internally calls
apply" earlier and the additional technical debt, but is not a good
thing to do to the overall health of the project in the longer term
to pretend as if this new set_index_file() is part of a good API.

Instead it probably should be marked with "the libification of
'apply' took a short-circuit by adding this technical debt; please
do not call this function in new codepaths", or something like that.