[Imports] Fwd: [Talk-us] Import of EPA data

[Chris Hunter]

Gotcha.  Sorry, I thought you were wanting to keep the data in the map and
programmatically update it.

 It's gonna be impossible to use the rollback script at this point, but
since there are only nodes in the dataset the algorithem  isn't that
complex.

1 Query the changeset for all node IDs
2 Query the database for the version number of the resulting nodes
3 If version = 1, delete node else flag for user review

Of course since my C & bash skills are rusty as all get out and all my SQL
experience is with mySQL (pun intended) I have no idea how much of a hit the
server will take running step 2.

DiverCTH (Chris)

P.S. I don't know if anyone else has said it yet, but welcome aboard.  Glad
to see we haven't scared you off.

Chris

On Fri, Dec 18, 2009 at 10:51 AM, jamesmikedupont at googlemail.com <
jamesmikedupont at googlemail.com> wrote:

> Listen,
>
> I support the idea of deleting the data. Should I do it? Its there an
> easier way to delete a whole changeset?
>
> I am still working on the processing, and will continue on this
> outside of the OSM database.
>
> It my a mistake to just import the data,
> but as I said, I am willing to put the work into cleaning it up.  I am
> commited to making sure that we have a solution to this.
>
> My internet was cut off for the past two days, when all the data is
> downloaded I will be able to tell you what is in the EPA dataset.
>
> I wont be importing it directly but setting up a database.
>
> Now, it should be easy enough, once the data is there, to also do
> checks and updates.
> From a programming point of view, we should be able to tell if an ID
> has changed, and as long as the IDS are present to be able to update
> the data as needed in OSM.
>
> mike
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/imports/attachments/20091218/9df1f127/attachment.html>