Graph generation only in regions/systems that have changed

[jteresco]

With graph generation taking the largest single chunk of time in the site update process, we could consider ways to generate new graphs only in regions/systems that have had data changes since the last graph generation. TBD is how to tell what's changed and which graphs are affected.

[yakra]

• Have the shell script copy the existing HighwayData directory to a "HighwayData.old" (or somesuch) backup, before the git pull
• Run some DIFFs
• ???
• Profit

(Can the os (module, library, what do I call it?) do things of this nature?)

[michihdeu]

Bash outputs the changed files when executing git pull

[yakra]

even better!

[wjorda]

Makefiles. :P

…

On Tue, May 22, 2018, 4:07 PM Eric Bryant ***@***.***> wrote: even better! — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#79 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AIkwwT3wU_FvOnfwhdPyDF-CJDMBlQB2ks5t1G_sgaJpZM4UHwNy> .

[jteresco]

A Makefile-like system of dependencies and usage of timestamps seems to be the way we'd go. This is also a pretty low priority at this point but I welcome the discussion and ideas.

[yakra]

With the significant speedup of graph generation a little while back, is this still a priority?

[jteresco]

I don't think it is. Marking with a new label "low priority", since some day we might want to think about it. Automatic generation of graphs on demand from the DB (#89) would be a more useful approach now.

[jteresco]

Like with #78, I deem this to be less important given a few years of improvements in this process.