[Talk-us] New York City watershed recreation areas
kevin.b.kenny+osm at gmail.com
Mon Mar 20 02:40:35 UTC 2017
I just thought I'd make a quick note here that not all imports result in
I updated the New York City watershed recreation areas (see
for details of the import process) this evening. I did it by following the
import process as outlined on that page up to loading the areas' geometry
and tagging into PostgreSQL. I then threw together three scripts to
identify areas that were new [that is, available from New York city, but
not in OSM], deleted [in OSM but not listed in New York City's metadata],
and retagged [timestamp of modification of the map of the area unchanged,
but changes to the access restrictions].
The scripts in question are all available for inspection in the project's
intake_rec_areas.tcl loads the database from the PDF files on the New York
id_deleted_units.sql queries the database and the local OSM mirror to
determine what units
must be deleted.
id_changed_units.sql queries the database and the local OSM mirror to
determine what units need to be reimported or conflated because their
geometry has changed.
id_changed_tags.sql queries the database and the local OSM mirror to
determine what units need to be retagged because of changed access
The scripts identified:
3 units where the PDF file served up by the city was malformed
0 deleted units
10 new units
25 units with changed geometry
6 units with change access restrictions
These numbers were small enough that I didn't trouble to design any sort of
automation for conflation and replacement. Instead, for everything that
changed, I simply checked that the unit was last updated by OSM user
'ke9tv-NYCDEP-import', and then did the work of replacing the making the
changes in JOSM, one changeset per modified unit, exactly as I did in the
I thought this note might be useful as an experience report. Thanks for
reading this far!
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Talk-us