Shapefile.js

Redoing all of this in modern JS. Promises, Typed Arrays, other hipster things, I wouldn't say it's based on RandomEtc's version as much as inspired by it as there is 0 code shared and I really only read the binary ajax part of his (hence why my function has the same name, they are otherwise not related). My sources were:

• wikipedia article
• ESRI white paper
• This page on Xbase

##Demos

• Countries
• Google maps
• zip file
• Local Zipfile
• Projected big with web workers
• Projected small

##API

Has a function shp which accepts a string which is the path the she shapefile minus the extention and returns a promise which resolves into geojson.

	//for the shapefiles in the folder called 'files' with the name pandr.shp
	shp("files/pandr").then(function(geojson){
		//do something with your geojson
	});

or you can call it on a .zip file which contains the shapefile

	//for the shapefiles in the files folder called pandr.shp
	shp("files/pandr.zip").then(function(geojson){
		//see bellow for whats here this internally call shp.parseZip()
	});

or if you got the zip some other way (like the file api) then with the arrayBuffer you can call

shp(buffer).then(function(geojson){});
//or
shp.parseZip(buffer)->returns zip

If there is only one shp in the zipefile it returns geojson, if there are multiple then it will be an array. All of the geojson objects have an extra key fileName the value of which is the name of the shapefile minus the extention (I.E. the part of the name that's the same for all of them)

You could also load the arraybuffers seperately:

shp.combine([shp.parseShp(shpBuffer),shp.parseDbf(dbfBuffer)]);

if it needs to be projected you can call shp.proj which expects a string of WKT transformation info, aka the contents of the .prj

//if it's a string
var projFunc = shp.proj(prjString);
//returns func

//if it's an array buffer
var projFunc = shp.proj(String.fromCharCode.apply(this,new Uint8Array(zip[key])));
//returns func

which can then be given to parseShp

shp.parseShp(shpBuffer,projFunc);

##Stick it in a worker

I used my library catiline to parallelize the demos to do so I changed

<script src='dist/shp.js'> </script>
<script>
	shp('files/shapeFile.zip').then(function(data){
		//do stuff with data
	});
</script>

to

<script src='website/catiline.js'> </script>
<script>
	var worker = cw(function(base,cb){
		importScripts('dist/shp.js');
		shp(base).then(cb);
	});
	//worker can be called multiple times
	worker.data(cw.makeUrl('files/shapeFile.zip')).then(function(data){
		//do stuff with data
	});
</script>

to send the worker a buffer from the file api you'd do (I'm omiting where you include the catiline script)

var worker = cw(function(data){
	importScripts('../dist/shp.js');
	return shp.parseZip(data);
});

worker.data(reader.result,[reader.result]).then(function(data){
	//do stuff with data
});

##Done

• Binary Ajax
• parsing the shp
• parse the dbf
• join em
• zip
• file api
• Some Projections

##to do

• More Projections
• check for geometry validity.
• Tests

##LICENSE Main library MIT license, origional version was less permisive but there is 0 code shared. Included libraries are under their respective lisenses which are:

• JSZip by @Stuk MIT or GPLv3
• Promiscuous by @RubenVerborgh MIT
• setImmediate by @NobleJS et al MIT
• World Borders shapefile is CC-BY-SA 3.0.
• Park and Ride shapefile is from MassDOT and is public domain.
• MA town boundaries from MassGIS and is public domain
• NJ County Boundaries from NJgin and should be public domain.
• Projforjs easy stuff by me hard stuff by other people Apache licensed