geogrid.js

The library geogrid.js provides a Leaflet layer that depicts data aggregated by the ISEA3H Discrete Global Grid System. Thereby, the library renders the visualization using WebGL.

Scientific Publications

The following publication is related to this plugin and the used Discrete Global Grid System (DGGS):

• F-B Mocnik: A Novel Identifier Scheme for the ISEA Aperture 3 Hexagon Discrete Global Grid System. Cartography and Geographic Information Science 46(3), 2019, 227–291

Related Software

This library is compatible with the framework Measures REST, which can be used to aggregate data by a grid and then provide the data via a REST interface. Data can also be aggregated manually by using the library geogrid that computes and handles Discrete Global Grid Systems (DGGS).

Use the Library

To load the library, include the following in the header of your HTML file:

<script src="https://unpkg.com/geogrid.js"></script>

In addition, the library geogrid.js requires the following libraries to be loaded:

• Leaflet
• PixiJS
• Leaflet.PixiOverlay
• D3.js

The library vptree.js library is included and shipped with geogrid.js under the ISC license.

In order to use the ISEA3HLayer, a Leaflet map needs to be loaded first, for example, like follows:

var map = L.map('map').setView([49.4, 8.7], 10);

Then, the ISEA3HLayer, which is provided by this library, can be added:

var isea3h = L.isea3hLayer({
  url: 'http://localhost:8080/.../{z}/{x}/{y}.json',
}).addTo(map);

As an option, a URL needs to be provided under which data aggregated by the ISEA3H grid is available. Such a URL can be provided in two different formats:

1. in an XYZ tile format by including the paramters {x}, {y}, and {z}, thus allowing to load the data on demand, and
2. as a API URI that provides the data corresponding for a given bounding box {bbox} and resolution {resolution}.

In both cases, further parameters can be encoded, as is discussed in the description of the option parameters. In all cases, the provided data should be formatted as follows:

{
    "type":"grid",
    "resolution":14,
    ...
    "data":[
        {"value":0.345, "id":"1309502766029885663"},
        {"value":null, "id":"1309502851015240491"},
        ...
    ]
}

The grid cell IDs are assumed to conform to both adaptive and non-adaptive IDs, as is described in:

• F.-B. Mocnik: A Novel Identifier Scheme for the ISEA Aperture 3 Hexagon Discrete Global Grid System. Cartography and Geographic Information science 46(3), 2019, 277–291. doi:10.1080/15230406.2018.1455157

Data which is formatted in the above format is, for example, provided by the framework Measures REST.

The ISEA3HLayer can be used in combination with different base maps. A good choice is Toner lite by Stamen. The complete code of the example is as follows:

var map = L.map('map').setView([49.4, 8.7], 10);
L.stamenTileLayer('toner-lite').addTo(map);
L.isea3hLayer({
  url: 'http://localhost:8080/.../{z}/{x}/{y}.json',
}).addTo(map);

Examples

An example can be found in the subdirectory example. Observe that the example index-with-server.html presumes a local instance of Measures REST.

Examples: Options

The ISEA3HLayer accepts several options, which allow to easily adapt the layer to the data source and the visualization needs. Instead of depicting representing the value related to a grid cell by the colour, the value can also be encoded by the size of the grid cell:

L.isea3hLayer({
  url: '...',
  cellColorKey: null,
  cellSizeKey: 'value',
})

Even two different values related to a grid cell can be represented, for example, the value with key value1 encoded by the colour, and the value with key value2, by the size of the cell:

L.isea3hLayer({
  url: '...',
  cellColorKey: 'value1',
  cellSizeKey: 'value2',
})

Most options can even be updated after the ISEA3HLayer has been initialized, which makes possible to interactively adapt the display and depict different aspects of the data without the need to reload them:

isea3h = L.isea3hLayer({
  url: '...',
  cellColorKey: 'value1',
})
...
isea3h.update({
  cellColorKey: 'value2',
})

Examples: Several Data Sources

The ISEA3HLayer accepts one or more data sources. In many cases, only one data source is to be depicted. This is the case that has been discussed before – the URL that provides reference to the data and other options are provided as options:

L.isea3hLayer({
  url: '...',
  cellColorKey: 'value1',
  cellSizeKey: 'value',
  silent: false,
  renderer: 'webgl',
})

Observe that the options silent and renderer do not refer to this particular dataset but the layer in general. This is in contrast to the options url, cellColorKey, and cellSizeKey, which refer to the data source and its representation. Instead of mixing these options (which is allowed), they can be separated as follows:

L.isea3hLayer({
  sources: [{
    url: '...',
    cellColorKey: 'value1',
    cellSizeKey: 'value2',
  }],
  silent: false,
  renderer: 'webgl',
})

This way of providing the options has the advantage that it becomes obvious which options are only valid for the provided data source, and it allows to add additional data sources. For instance, several sources can be provided as follows:

L.isea3hLayer({
  sources: [{
    url: '...',
    cellColorKey: 'value1',
    cellSizeKey: 'value2',
  }, {
    url: '...',
    cellColorKey: null,
    cellSizeKey: 'value',
  }],
  silent: false,
  renderer: 'webgl',
})

It is important to note that the two ways of noting down the sources – inline with the general options in case of one source, or under sources in case of one or more sources – cannot be mixed.

The several options available are described in the following.

Options

The following options are available to describe the data sources and their handling:

In addition, the following general options are available:

Methods

The ISEA3HLayer has the following methods to add or remove it to/from a Leaflet map object, and to update its options:

Events

The following events are available:

Plugins

The functionality of the ISEA3HLayer can be extended by plugins. Such plugins can easily be added to the layer:

var isea3h = L.isea3hLayer({...});
isea3h.addPlugin(L.testPlugin());

The following plugins are available:

• more plugins are to be implemented

Authoring Plugins

General Structure of a Plugin

Plugins extend the class L.ISEA3HLayerPlugin. Several methods exist that can be overwritten in order to react to certain events. A plugin can, for example, be implemented as follows:

class TestPlugin extends L.ISEA3HLayerPlugin {
  onHover(e) {
    console.debug('hover', e.cell.id, e.data.value)
  }
  onUnhover(e) {
    console.debug('unhover', e.cell.id, e.data.value)
  }
  onClick(e) {
    console.debug('click', e.cell.id, e.data.value)
  }
  ...
}

In addition to a class, a factory method should be provided in order to simplify the instantiation of the plugin:

L.testPlugin = () => new TestPlugin()

Events

Plugins to the ISEA3HLayer follow the paradigm of reactive programming: changes in the system that consists of the map and the user fire events of the plugin. The corresponding method to an event is only executed, in case it is implemented. The method onHover is, for example, executed when the hover event is fired. If a corresponding method is not present, no error is thrown. The following event listeners can be used:

Methods

Plugins can react to events by performing an action as soon as an event is triggered. Such an action can change the state of the ISEA3HLayer, or other parts of the website. As an example, a cell can be drawn in blue colour when it is hovered by the courser:

onHover(e) {
  this.setCellColor(e.cell, 'blue')
  this.render()
}

The URL that is used to retrieve data from a server contains parameters. Among these parameters are the resolution and the bounding box, but potentially also additional parameters. These additional parameters can, for example, be set as follows:

this.setParameter('date', '2017-01-01')
this.downloadData()

Some of the functions except a callback as parameter because the result is computed asynchronously:

onHover(e) {
  this.neighbors(e.cell, cells => {
    console.debug(cells)
  })
}

The following methods of the L.ISEA3HLayerPlugin are available:

Compute GeoJSON only

In some cases, one might not want to visualize grid data by the use of Leaflet but to only compute the corresponding GeoJSON. The library thus provides a corresponding function to compute the GeoJSON in an efficient way. For this, neither of the libraries needs to be loaded. For data formatted as described at the top of this page, the GeoJSON can be computed as follows:

L.isea3hToGeoJSON({
  data: data,
}, geoJSON => console.log(geoJSON))

Instead of the data, a url can be provided as well. In this case, the bounding box and the resolution can be provided.

L.isea3hToGeoJSON({
  url: '...',
  pureBBox: null,
  pureResolution: 7,
}, geoJSON => console.log(geoJSON))

Note that the result is communicated by a callback, and that many options have no effect in this case.

Build geogrid.min.js

The library can be translated from ECMAScript 6 to minified JavaScript, which results in a file geogrid.min.js. In order to generate this minified file, you have to install Node.js first. Thereafter, you have to execute the following command:

npm install

This command automatically also builds the library. Whenever you want to re-build the thereafter, just run:

npm run build

The resulting minified file can be found in the subdirectory dist, which is created during this process. In addition, a JavaScript Source Map is created.

Known Bugs

The following bugs are known but not yet addressed:

• Downloaded data cached for the n-th source is not removed from the cache when updating the list of sources to have less than n sources.
• Missing tiles are not handled correctly.

Author

This software is written and maintained by Franz-Benjamin Mocnik, [email protected].

In 2017–2019, this software has been supported by the DFG project A framework for measuring the fitness for purpose of OpenStreetMap data based on intrinsic quality indicators (FA 1189/3-1).

All commits after 9/2019 (c) by Franz-Benjamin Mocnik, 2019–2020. All commits before 8/2019 (c) by Heidelberg University, 2017–2019.

License

The code is licensed under the MIT license.