<div dir="ltr"><br><div class="gmail_quote">On Tue, Sep 9, 2008 at 4:43 PM, Florian Lohoff <span dir="ltr"><<a href="mailto:flo@rfc822.org">flo@rfc822.org</a>></span> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div class="Ih2E3d">On Tue, Sep 09, 2008 at 07:32:34AM +1000, Brett Henderson wrote:<br>
> As Jochen mentioned, the --write-pgsql-dump task and subsequent import using<br>
> COPY statements should be faster. I've looked at getting osmosis to stream<br>
> data directly into the database but unfortunately jdbc drivers don't expose<br>
> that kind of functionality.<br>
<br>
</div>Shure - the point is that you cant keep the database up to date with<br>
the write-changes as it throws an exception when you simple drop the<br>
bbox column ...<br>
<div><div></div><div class="Wj3C7c"><br>
</div></div></blockquote></div>I've updated the 0.6 tasks to allow dropping of the bbox column. All pgsql write tasks check for the existence of the bbox column and only populate it if it exists. Not going to be of much use until 0.6 is deployed though ...<br>
<br>In the meantime can you import using the COPY files and comment the bbox population sql command out of the load script. The osmosis changeset task will continue to update the column when applying changesets but it shouldn't add too much overhead.<br>
<br></div>