How to use AWS S3 raster file in Mapbox gl js? - html

Mapping a Cloud Optimized Geotiff raster layer in Mapbox GL JS
I think I use a raster source, the source data need to be tiled. the URL would be like .../{z}/{x}/{y}.tif.
But I don't know how to make above URL. how to make url??


I'm not exactly sure what you're asking. If you're asking "how do I turn this GeoTIFF into a tiled raster source", two options are:
Upload it to Mapbox. (In theory this works, although it has not always been successful for me.)
Use GDAL to convert it to mbtiles, then host the .mbtiles file somewhere (either on Mapbox, CloudTiler, or your own hosting.)
In the latter case, that would look like:
gdal_translate myfile.tif mytiles.mbtiles


Mapbox SDK wants to fetch tiles by Z/X/Y. However, you can probably register a NSURLProtocol subclass where you can take the Z/X/Y URL and create the proper COG request, transform the image and return the correct content.
I have used this to register mbtiles:// as a URL scheme and my NSURLProtocol subclass transform the Z/X/Y URL to a SQLite select statement.
You can read more on this subject over at https://github.com/mapbox/mapbox-gl-native/issues/12026

Related

How is it possible to reduce the size of Html file generated from Folium map in Python?

I'm using folium.GeoJson to plot polygons on a map. I have one dataset with different fields/attributes; I'm using the same data for each folium.GeoJson layer but with various fields to display different attributes and add them in layer control. I guess this makes the final Html file big in size.
One solution I found to reduce the Html file is to minify it, but the slimmer package isn't available in anaconda.
Do you have any suggestions for this case?

You can reduce the size of the HTML file dramatically by using mapshaper to reduce the size of the geojson file.
Drop your file or search and import. I check the box "detect line intersections" and upload:
Choose your method of simplifying the coordinates (I use Douglas-Peucker method, and prevent shape removal):
Then set the simplification (you can play with the settings to get the desired result, I found 3% worked well)
Then export the file (it exports as a .json file so remember to change to .geojson:
This reduced the file from 17.6MB to 614KB and the outputted HTML file from 222MB to 7.58MB

Found a way: to simplify spatial data that significantly reduces data size.
In my case, I was using shapefile as the spatial data, particularly polygons, that is, the collection of points (vertices) and highly affected the data size. I just used the simplification tool that removes extraneous vertices but preserves essential shape. I used the Simplify Polygon tool in ArcMap, but the same is possible in python with the geopandas.GeoSeries.simplify function and in R with the 'sf' library.

How to create terrain files with points in Cesium?

I have a DEM file with XYZ points (file sample image), and I want to use it to create terrain in Cesium with it. I tried to translate it to GeoTiff by ArcMap, but it is not supported by cesium terrain builder.
How could I create terrain files with points in Cesium?

Cesium is able to consume Quantized Mesh tiles but not GTIFF or XYZ directly.
First, you will have to convert your XYZ file to e.g. GTIFF and transform it to WGS84 coordinate reference system. Both can be done using gdal.
Second, convert your transformed GTIFF files to quantized-mesh format.
I created a Docker image for that purpose. Check out the Github page of tumgis/cesium-terrain-builder-docker for a detailed documentation of the entire process.
Finally, you will need a service like CesiumTerrainServer to serve the tiles to a Cesium application.

How to load longitude and latitude information into cesium kmlDataSource?

I want to display the kml file in cesium globe, and as described in the cesium's workshop code, we need to load the file by passing it with the file location (or URL). The following line of code is the specific line of code that cesium loads the kml.
var geocachePromise = Cesium.KmlDataSource.load('./Source/SampleData/sampleGeocacheLocations.kml', kmlOptions);
The entire code is available here
My question is: I have the longitude, and latitude, (and height) information saved in variables and instead of always saving them into .kml file and then load them via folder, I want to pass this information to cesium kmlDataSource (the code above) directly.
It would be great if anyone has any solution to this.
Please let me know if further information or code snapshot is required. Thanks

If you already have the information you need stored in JavaScript variables, there's no need to export to KML and import it back into Cesium. You can directly add the indicators you need as Cesium Entities, which is what the KML loader is creating when it reads a KML.
Typically, a KML-like pin is represented by a Cesium Entity containing either a point or a billboard, and optionally an associated label.
Here are some relevant demos that show how this is done:
Billboard demo
Map Pin demo
Label demo
Each of these demos calls viewer.entities.add({ ... }) along with a position for the Entity and some sort of graphical indication(s) to display to the user. You may place one of each on an Entity, for example a billboard and a label are often both defined when adding a typical KML-like Entity.
If your data is stored on the server however, you will need some mechanism to stream it to the browser. CZML is Cesium's native format for doing so, but KML is also available as an alternative for certain kinds of graphics. You may also use any API of your own design, and create Entities when the data becomes available in JavaScript.

Does Autodesk Viewer meet those requirements?

I've been asked to insert an .IFC viewer into a web page already built, to:
explode building visualization into floors
rotate the building
colour spaces based on their temperature
add labels to objects
load .ifc files stored on private servers
Does autodesk viewer meet those requirements? All of them?
Additionaly, is it feasible to change the viewer language through a config file or specific functions?

explode building visualization into floors
The built-in explosion is by model components and not floors and there's option right out of box to specify a granularity level.
Nonetheless you can use viewer.cutplanes (see doc here) to isolate the levels and go from there - see sample here.
rotate the building
This can be done by either THREE.js transformation or Viewer's model load options:
THREE.js: Perform transformation recursively by traversing all children components under the model's root dbid. See this great article here for enlightenment.
Load option - see live sample here:
var options = {
placementTransform:THREE.matrix4,
globalOffset:{x:0,y:0,z:0}
};
viewer.start(svfURL, options）
colour spaces based on their temperature
Use viewer.setThemingColor (see doc here) to apply temperature colors - find the parent dbid of the room or and use the recursive flag to apply color to all its children components. See code sample here.
add labels to objects
See here for details and a few helpful links at the start of the article.
load .ifc files stored on private servers
You can download the translated derivatives (SVF and its resources - obtain a manifest of them from GET :urn/manifest) to your own storage locations and load them by their URL directly. See my live code sample above for details and find the source code for our online extractor tool (http://extract.autodesk.io) here for ideas to download the derivatives.
is it feasible to change the viewer language through a config file or specific functions?
Viewer implements the i18n standards so you can do:
Autodesk.Viewing.i18n.setLng('fr');
Autodesk.Viewing.i18n.localize()

Best way to overlay an ESRI shapefile on google maps?

What is the best way to overlay a shapefile in Google Maps?
After some reading suggests that, converting shapefile into KML and overlaying KML in Google map is an option. If that is the case, how do I convert a shapefile into KMLs?
I believe there would be a few open source tools available for this conversion. Can anyone point to a tool to covert a shapefile to KML without any loss of data?

I like using (open source and gui friendly) Quantum GIS to convert the shapefile to kml.
Google Maps API supports only a subset of the KML standard. One limitation is file size.
To reduce your file size, you can Quantum GIS's "simplify geometries" function. This "smooths" polygons.
Then you can select your layer and do a "save as kml" on it.
If you need to process a bunch of files, the process can be batched with Quantum GIS's ogr2ogr command from osgeo4w shell.
Finally, I recommend zipping your kml (with your favorite compression program) for reduced file size and saving it as kmz.

as of 12.03.2019 FusionTables is no more...
Import the Shapefile into Google FusionTables ( http://www.google.com/fusiontables ) using http://www.shpescape.com/ and from there you can use the data in a number of ways, eg. display it using GoogleMaps.

I would not use KML. Instead, use GeoJSON which you can natively consume in Google Maps API now. It is a newer feature that didn't exist from the original responses.
In any case, simply open the SHP file in Quantum GIS, and then you can output it in any format you like (KML, GeoJSON).
If you are using Google Maps for Work, I found a premium extension that handles loading shapefiles directly where you can just connect direct to the shapefile that you generate from ESRI. I did a search on the CMaps site and found this snippet which loaded US by state shapefile: https://gmapsplugin.net/cmapsanalytics/assets/shapes/usstates.shp
var cMap = new centigon.locationIntelligence.MapView();
cMap.key([your_api_key]);
cMap.layerNames(["Basic Shapes"]);
cMap.dbfKeys([['Alabama','Alaska','Arizona','Arkansas','California','Colorado','Connecticut','Delaware','District of Columbia','Florida','Georgia','Hawaii','Idaho','Illinois','Indiana','Iowa','Kansas','Kentucky','Louisiana','Maine','Maryland','Massachusetts','Michigan','Minnesota','Mississippi','Missouri','Montana','Nebraska','Nevada','New Hampshire','New Jersey','New Mexico','New York','North Carolina','North Dakota','Ohio','Oklahoma','Oregon','Pennsylvania','Rhode Island','South Carolina','South Dakota','Tennessee','Texas','Utah','Vermont','Virginia','Washington','West Virginia','Wisconsin','Wyoming']]);
cMap.userShapeKeys([['Massachusetts','Minnesota','Montana','North Dakota','Hawaii','Idaho','Washington','Arizona','California','Colorado','Nevada','New Mexico','Oregon','Utah','Wyoming','Arkansas','Iowa','Kansas','Missouri','Nebraska','Oklahoma','South Dakota','Louisiana','Texas','Connecticut','New Hampshire','Rhode Island','Vermont','Alabama','Florida','Georgia','Mississippi','South Carolina','Illinois','Indiana','Kentucky','North Carolina','Ohio','Tennessee','Virginia','Wisconsin','West Virginia','Delaware','District of Columbia','Maryland','New Jersey','New York','Pennsylvania','Maine','Michigan','Alaska']]);
cMap.labels([['Massachusetts','Minnesota','Montana','North Dakota','Hawaii','Idaho','Washington','Arizona','California','Colorado','Nevada','New Mexico','Oregon','Utah','Wyoming','Arkansas','Iowa','Kansas','Missouri','Nebraska','Oklahoma','South Dakota','Louisiana','Texas','Connecticut','New Hampshire','Rhode Island','Vermont','Alabama','Florida','Georgia','Mississippi','South Carolina','Illinois','Indiana','Kentucky','North Carolina','Ohio','Tennessee','Virginia','Wisconsin','West Virginia','Delaware','District of Columbia','Maryland','New Jersey','New York','Pennsylvania','Maine','Michigan','Alaska']]);
cMap.polyDataSources([centigon.locationIntelligence.CMapAnalytics.DATA_PROVIDERS.SHAPE_DATAPROVIDER]);
cMap.layerTypes([centigon.mapping.Layer.TYPE.POLY]);
cMap.locations([["https://gmapsplugin.net/cmapsanalytics/assets/shapes/usstates.shp"]]);
cMap.panTo("USA");
cMap.zoomLevel(3);

Do you mean shapefile as in an Esri shapefile? Either way, you should be able to perform the conversion using ogr2ogr, which is available in the GDAL packages. You need the .shp file and ideally the corresponding .dbf file (which will provide contextual information).
Also, consider using a tool like MapShaper to reduce the complexity of your shapefiles before transforming them into KML; you'll reduce filesize substantially depending on how much detail you need.

Free "Export to KML" script for ArcGIS 9
Here is a list of available methods that someone found.
Also, it seems to me that the most efficient representation of a polygon layer is by using Google Maps API's polyline encoding, which significantly compresses lat-lng data. But getting into that format takes work: use ArcMap to export Shape as lat/lng coordinates, then convert into polylines using Google Maps API.

Just to update these answers, ESRI has included this tool, known as Layer to KML in ArcMap 10.X. Also, a Map to KML tool exists.
Simply import the desired layer (vector or raster) and choose the output location, resolution, etc. Very simple tool.

2018 already... I've found this fantastic online tool http://mapshaper.org/ to convert from ESRI shapefiles to SVG, TopoJSON, GeoJSON.
Here is the explanation of how to use it https://www.statsilk.com/maps/convert-esri-shapefile-map-geojson-format
Fast and straightforward! :)