set SOURCE_DATE_EPOCH from changelog

[ffesti]

Alternative implementation for #141

It sets the environment variable during the build stage and not while parsing the spec which should be free from side effects.

[ffesti]

@bmwiedemann would this work for you?
I left you as author of the patch but I can reset it if you wish.

[bmwiedemann]

change looks reasonable. I am giving it a round of testing.
Under what condition would headerGet return false here? OOM? no changelog entry found?

[bmwiedemann]

test looks good.

[ffesti]

Hmm, I am wondering if the
getenv("SOURCE_DATE_EPOCH") == NULL
condition is really correct here. The problem is that rpmbuild might be used to build multiple packages in one go. But even if that was not possible someone using the API could.
Do we really need to give precedence of an already set SOURCE_DATE_EPOCH environment variable over %source_date_epoch_from_changelog ?

[bmwiedemann]

I put it there because the spec says:

Build systems MUST NOT overwrite this variable for child processes to consume if it is already present.

But then, it probably does not apply if we had set it ourselves (and not the user) for the previous rpm to build.