Add ability to ignore csv fields as requested in #15

[ckirby]

No description provided.

[coveralls]

Coverage remained the same at 100.0% when pulling 03970f3 on ckirby:master into 79f71ef on california-civic-data-coalition:master.

[palewire]

Thanks for this patch. I'll take a look when I find a minute. Out of curiosity, what is your use case for this feature?

[ckirby]

First let me say that I really like the project and like supporting it with
new code (I also submitted the patch for the static_mapping functionality).
I saw someone submitted issue 15 and thought I could use the feature too.

Use case: I have a system where data providers send us large data sets that
we use postgre_copy to insert. For various reasons we don't update the
model until all providers are onto the new data spec, so being able to
ignore the new fields until flipping the switch on the model is useful.

I also have some thoughts on how to tackle issue #26 (
#26)
would you be interested in a patch with that functionality?

Best,
Chaim

On Mon, Oct 10, 2016 at 9:12 PM, Ben Welsh notifications@github.com wrote:

Thanks for this patch. I'll take a look when I find a minute. Out of
curiosity, what is your use case for this feature?

—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
#30 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/AAFeHjbzVNBbBK7gv3B7yTigZlJNLJQ_ks5qyoABgaJpZM4KSlQi
.

[coveralls]

Coverage remained the same at 100.0% when pulling 73dc5f3 on ckirby:master into 79f71ef on california-civic-data-coalition:master.

[palewire]

Moments ago I pushed a change to the master branch that takes a different approach to solving the problem.

Rather than add a new keyword argument that requires users to explicitly list the headers they'd like to skip, I've tried to rework the application so it will "just work" and allow for headers to be omitted from the map input without any extra configuration.

I've added the tests you submitted here to see if I got it right, and they appear to be passing. If you have the time, let me know what you think of this stab at a solution.

[palewire]

@ckirby, if you don't protest, I think I might close this pull request and keep the alternative implementation of your feature described above.