Skip to content

Latest commit

 

History

History
192 lines (140 loc) · 7.25 KB

README_en.md

File metadata and controls

192 lines (140 loc) · 7.25 KB
Raw

GSI Vector Tile Experiment

Geospatial Information Authority of Japan (GSI, http://www.gsi.go.jp/ ) provides a vector tile service for its authoritative geospatial data, with nationwide coverage updated intermittently, available in GeoJSON format. The aim of the experiment is to design the best way to provide the result of the basic survey conducted according to Survey Act by GSI, in a way users can make use of the latest web map technologies.

Making the data from GSI more accessible to developers is central to the mission of the Information Access Division of GSI. We don't want you to have to struggle with handling GML data, installing PostGIS, or download the whole bunch of data to start playing with our data.

Vector tiles make real-time rendering possible by sending the underlying geometry and attributes directly to the client, whether that's browser or a native mobile app. We believe that vector tiles will enable yet-to-be-invented types of applications. Use our data to experiment with ideas.

Please note: this service is in experimental phase and is subject to change!

Getting Started

This service is available for all developers to use.

Using the serice

The template URL for our vector tiles service is as below:

http://cyberjapandata.gsi.go.jp/xyz/{layer}/{z}/{x}/{y}.geojson

This url scheme is slippy map tilenames (XYZ).

Here's a sample tile in GeoJSON: http://cyberjapandata.gsi.go.jp/xyz/experimental_fgd/18/233094/102736.geojson

Please note that the charset of all our GeoJSON data is UTF-8.

Also please note that our GeoJSON vector tile data does not have the concept of layers; One raw FeatureCollection is enclosed in on GeoJSON tile.

Spatial coverage

Spatial coverage of this experiment is the whole area of Japan. However, several layers provided in early stage still covers limited area called 'experiment area'.

The experiment area is shown in https://github.com/handygeospatial/vector-tile-experiment-area/blob/gh-pages/area.geojson .

Layers

Data are organized into several layers comprising the elements typically used for base map rendering. This is one view of GSI data for easier consumption, with interleaving serveral Feature Classes into one layer.

experimental_rdcl: road centerline

Road centerline.

Please note the specification for the default style.

properties

  • rID
  • lfSpanFr
  • lfSpanTo
  • tmpFlg
  • orgGILvl
  • ftCode
  • admCode
  • devDate
  • type
  • rdCtg
  • state
  • lvOrder
  • name
  • comName
  • admOfcRd
  • rnkWidth
  • Width
  • sectID
  • tollSect
  • medSect
  • motorway
  • repLtdLvl
  • rtCode

experimental_railcl: railroad centerline

Railroad centerline.

properties

TODO

experimental_rvrcl: river centerline

properties

TODO

experimental_anno: annotation

properties

TODO

experimental_fgd: 1:2500 data

The default style of this layer is not yet separated from the example site.

properties

TODO

experimental_dem5a: 5m DEM

Please see the demo site.

properties

experimental_dem10b: 10m DEM

Please see the demo site.

properties

TODO

experimental_cp: control points

properties

TODO

experimental_pp: principal points

Point data for aerial photographs.

properties

  • ID: identifier
  • 撮影年月日: the date of taking the photograph. ISO8601 style.
  • 撮影縮尺: the scale of the aerial photograph.
  • 撮影高度: the height the photograph is taken.
  • カラー種別: "カラー" if colored photograph, "モノクロ" if monochrome photograph.
  • 撮影計画機関: name of the organization to plan the photogrammetric flight.

experimental_zk25000: 1:25000 map grid

Point data representing the center of the map grid containing the names and the corners of the map grid.

properties

  • 図名: the name of the map sheet in Japanese kanji characters.
  • よみがな: the name of the map sheet in Japanese kana characters.
  • 図郭座標: lower and upper corners of the map grid.

How it works

Our vector tiles are served in rather static way. All the resources are uploaded as static files.

We use tools developed in-house to get our vector tiles from a bunch of Shapefiles. Some tools for smaller dataset are written in Perl and available in GitHub.

For other larger datasets we sometimes use hadoop-streaming to sort geometries into tiles.

How to display vector tiles

Currently we use TileLayer.GeoJSON and TileLayer.Canvas of Leaflet 0.7.3. Others should follow.

TileLayer.GeoJSON

As in GSI Maps (GitHub repo.)

TileLayer.Canvas

As in an example (GitHub repo.)

ChangeLog

  • 2014-08-01 Experiment started
  • 2015-08-05 First release of README_en.md