This only happens when I use to beautified code (shp.js) and not the minified version (shp.min.js), I get a 404 error on rawinflate.min.js.map. Do I need to include this? How do I correct this?

Is it related to this issue? Stuk/jszip#75

Thanks for making this library!

Is there a list of supported projections anywhere? Thanks!

look into this for ie9 https://gist.github.com/Benvie/5020656

Can we read shapefile stored on s3 and convert it into geojson?

Is any one tired this.

we can cut down on the code we included in #10 as we're only using a subset of that code.

First of all, thanks for this stuff it's amazing!

I do have a question however, why when I provide a zip file which contains all the shape files (even prj) are all the features transformed to WGS. Is there any parameter I could provide to shp(buffer, ?) that would avoid reprojecting?

I could identify this line of code:
if (typeof toProj === 'undefined') { toProj = fromProj; fromProj = wgs84; single = true; }

So if I comment in the "fromProj = wgs84" then it will work as expected, but I'm curious whether there is any parameterized way to control this.

Also, I've noticed that the PROJ4 library is already embedded in shp.js/shp.min.js. I alredy have the amd repository for a standalone proj4 module. Any chance that you would split up the code so the proj4 would be optional?

Thank you

exist and you (me) should use them especially in the shp parsing code.

I have a shape file which has a PRJ file which defines a vertical projection. When trying to parse it, I get an error "Unexpected token ,".

The error appears to be from the first line this method, specifically the JSON.parse call.:

module.exports = function(wkt, self) {
  var lisp = JSON.parse(("," + wkt).replace(/\s*\,\s*([A-Z_0-9]+?)(\[)/g, ',["$1",').slice(1).replace(/\s*\,\s*([A-Z_0-9]+?)\]/g, ',"$1"]'));
  var type = lisp.shift();
  var name = lisp.shift();
  lisp.unshift(['name', name]);
  lisp.unshift(['type', type]);
  lisp.unshift('output');
  var obj = {};
  sExpr(lisp, obj);
  cleanWKT(obj.output);
  return extend(self, obj.output);
};

An example PRJ is:

PROJCS[
    "Custom_NAD_1983_UTM_Zone_17N",
    GEOGCS["GCS_North_American_1983",
        DATUM["D_North_American_1983",
            SPHEROID["GRS_1980",6378137.0,298.257222101]
        ],
        PRIMEM["Greenwich",0.0],
        UNIT["Degree",0.0174532925199433]
    ],
    PROJECTION["Transverse_Mercator"],
    PARAMETER["False_Easting",1640416.666666667],
    PARAMETER["False_Northing",0.0],
    PARAMETER["Central_Meridian",-81.0],
    PARAMETER["Scale_Factor",0.9996],
    PARAMETER["Latitude_Of_Origin",0.0],
    UNIT["Foot_US",0.3048006096012192]
],
VERTCS["NAVD_1988",
    VDATUM["North_American_Vertical_Datum_1988"],
    PARAMETER["Vertical_Shift",0.0],
    PARAMETER["Direction",1.0],
    UNIT["Meter",1.0]
]

If parsing fails, does this call the failure state handler specified in then, or does it return null, or an empty geojson featurecollection?

see calvinmetcalf/leaflet.shapefile#7

Is it possible to generate the geoJSON object from the .shp file?

I think currently it's only possible to generate an array of geometries from the .shp file using: shp.parseShp(). It would be helpful to have a method that return the geoJSON object directly. It could be similar to the combine function but the .dbf file shouldn't be mandatory.

shp.combine = function(arr) {
    var out = {};
    out.type = 'FeatureCollection';
    out.features = [];
    var i = 0;
    var len = arr[0].length;
    while (i < len) {
        out.features.push({
            'type': 'Feature',
            'geometry': arr[0][i],
//          'properties': arr[1][i] <- it shouldn't be mandatory
        });
        i++;
    }
    return out;
};

When parsing the shapefile, occasionally the shapetype comes back as 1634038337. This is on a polyline shapefile. If I divide the shapefile into smaller section of about 100 features, it will work or work after several tries. It doesn't seem to be consistent. I have several other shapefiles where it works perfectly every time with several thousand features, so I don't think the feature count is actually the issue.

In addition, I have run into an issue where I may need to try once or twice to get the file read properly.

Can you reference how you fixed the local shapefile from only being able to open one shapefile to being able to add multiple?

So is there any way how to use this with local files? I'm using AngularJS with File API to upload my files. I'd rather process it locally then upload it to my server, return the path and then process it.

or maybe use dataUrl?

Hello Calvin,

I was able to recreate your local zipfile example in my application unfortunately I was unable to get the program to read the properties fo the shapefile. In the info window for the shapefile it shows a color code but only black dots show up for shapefiles. If possible can you point me to where in your script the feature style is loaded I am guessing one of my files that is associated in the require.js is incorrect but I am not sure which file is causing this.

Despite the documentation, you cannot reference local files - only file URLs to be accessed via binaryAjax.

It's surprisingly hard to find this project on npm, by search on npm or elsewhere. There are many shapefile projects, your username is different, etc.

this may mess up the combining

the parse functions should be public methods, we probably don't need the get methods and the make method should be called combine.

we have it geojson supports it, we should include it.

Pass extra parameter to style option of L.geojson.if i am using L.bind then custom parameter value is available but features gets undefined.but i want both of them. Reason for adding multiple parameters is i am having multiple divs with multiple maps . data is different for each map so that i can compare data and feature and apply style.

style: L.bind(style, this, mydata),

function style(mydata,features) {}

if it is not possible to get both values in L.bind .then how can i get current divid for which style function is executing.

It would be useful for me to have the geojson output using the original geometry. Perhaps we could add the crs property on the geojson object.

Many .prj files unfortunately do not include an AUTHORITY string, so instead of adding the crs property perhaps just return the proj4 transformation function in a promise with the geojson.

This will reduce loss of precision, if the map already uses the original projection. In my case, there is a non-negligible loss between EPSG:27700 >> EPSG:4326 >> EPSG:27700

E.G. I want to do:

shp(arrayBuffer).then(function(geojson, proj) {
  // use the geojson AND projection function
});

hi,
I am loading map using L.geojson layer (shapefile) its working fine . but now i want to display markers on geojson layer for example circle,rectangle,icons etc. Pointtolayer event is not getting called because in feature object geometry type is polygon as shown in image3. so i used oneachfeature event to load markers. markers are getting loaded but they are getting displayed at background of layer see image 1. on mouseover marker it gets displayed blur at background of layer(image2) .I want to show marker,icons at front.code used to display marker in oneachfeature is shown in image4.
New folder.zip

lru-cache is missing from the package.json dependencies.

The hasOwnProperty check for cache hits uses the passed in object's toString method. So it always returns "[object ArrayBuffer]" as the cache key when using ArrayBuffers. When you try to load multiple ArrayBuffer shapes, it thinks they are all the same as the first (matching toString), so it just returns the first shape layer.

This can be solved by overriding the toString method each time with something like

var req = new XMLHttpRequest();
req.open('GET', urlToZipFile);
req.responseType = 'arraybuffer';
req.onload = function(event) {
    if (req.response) {
        req.response.toString = function() {
            return urlToZipFile;
        };
        var shapeLayer = L.shapefile(req.response);
        scope.addShapeToMap(shapeLayer);
    }
};
req.send(null);

but would prefer shapefile-js to use a hash or optional value as an object's key. If there is a preferred way, I can probably put in a PR with the change.
Like changing

function shp(base, whiteList) {
  var hash = typeof base === 'string' ? base : SparkMD5.ArrayBuffer.hash(base);
    if (cache.has(hash)) {
        return Promise.resolve(cache.get(hash));
    }
    return shp.getShapefile(base, whiteList).then(function (resp) {
        cache.set(hash, resp);
        return resp;
    });
}

When using your library like that

convertSHP2GeoJSON: (buffer) ->
    shp(buffer).then (geoJSON) ->
        return geoJSON

the item (buffer) gets added to the cache. When using the shp function with another buffer, the already added but wrong item gets returned. Which is not the way it should work!

function shp(base, whiteList) {
    if (cache.has(base)) {
        return Promise.resolve(cache.get(base));
    }
    return shp.getShapefile(base, whiteList).then(function (resp) {
        cache.set(base, resp);
        return resp;
    });
}

We think there is a problem with the set and has logic inside that functions. Any suggestions?
Thanks for creating such a library!

I have a data which contains Chinese characters. but this lib can`t prase chinese , can I set char was "utf-8" ? and how to do it?

Hi,

I have been trying to process data in shapefile format with the attached .prj file in well-known text format (wkt).

test_case_prj.txt

The results I get are wrong: they are shifted to the West. With exactly the same input data, QGIS displays data correctly.

I soon became aware that 'long0' was not initialized, though it is used in the forward and inverse functions for Lambert Conformal Conic transformations.

If I set it to the value held in the prj file converted in radians in the initialization function associated to 'lcc':

          if (! this.long0) {
            this.long0 = 0.04079234433; // This is 2° 20' 14,025" East
          }

then I get the expected results.

I also noticed that the following lines were commented in function cleanWKT:

if (wkt.type === 'GEOGCS') {
	    geogcs = wkt;
	  }
	  if (geogcs) {
	    //if(wkt.GEOGCS.PRIMEM&&wkt.GEOGCS.PRIMEM.convert){
	    //  wkt.from_greenwich=wkt.GEOGCS.PRIMEM.convert*D2R;
	    //}
	    if (geogcs.DATUM) {

If I uncomment the three lines above instead of forcing the value of 'long0', then I also get the expected results.

It looks like the prime meridian information held in the prj file is simply ignored:

PROJCS["Lambert_Conformal_Conic",
GEOGCS["GCS_unnamed ellipse",
DATUM["D_unknown",SPHEROID["Unknown",6378249.2,293.466020999993]],
PRIMEM["Unknown",2.337229167],
UNIT["Degree",0.017453292519943295]],

Any help or insight would be highly appreciated.

Many thanks and kind regards.

Hello,
When trying to open the attached shapefile, or any that was zipped up on a Mac, I get an error:

Archive.zip
"Offset is outside the bounds of the DataView" (chrome)

It appears to be coming from code reading a dbf row header, but the error went away if i unzip/rezip on Windows, and has happened for multiple shape files.

When use a file with name name.DBF or name.PRJ or name.SHP, parseShp doesn't work. why this happen?

Kind of odd, but if the page has Prototype version 1.6.0.3, shp.js to load fails with "Uncaught WGS84"

shp.js:14153 Uncaught WGS84
Projection$1.callback @ shp.js:14153
Projection$1 @ shp.js:14163
Projection$1 @ shp.js:14149
(anonymous) @ shp.js:14629
(anonymous) @ shp.js:12805
54 @ shp.js:12808
s @ shp.js:1
(anonymous) @ shp.js:1
(anonymous) @ shp.js:22268
58../binaryajax @ shp.js:22422
s @ shp.js:1
e @ shp.js:1
(anonymous) @ shp.js:1
(anonymous) @ shp.js:1
(anonymous) @ shp.js:1

line where the failure is happening.
https://github.com/calvinmetcalf/shapefile-js/blob/gh-pages/dist/shp.js#L14153

Chrome Version 60.0.3112.90 (Official Build) (64-bit)

I'll be coding around it, but thought I'd post in case anyone else encounters this. I only tested 1.6.0.3, so it might not be an issue on newer versions.

This library is working great, but I am seeing performance that is much slower than what I see in the examples, using the exact same Shape File.

Here is a simple shape file of Los Angeles county postal codes:
LA_Postal_Codes.zip

When I drag and drop the file on this sample page, http://leaflet.calvinmetcalf.com/#9/33.7540/-117.8174 , the results are added to the map in less than a second!

I created a very simple page to read this Shape File and it is consistently taking nearly 4 seconds to read the shape file. I can't seem to understand why there is such a discrepancy. I am using web workers, fwiw.

Here is fully functional plunkr with the simplest code I could come up with to reproduce these performance issues. http://plnkr.co/edit/JuEAGpzViRKym7Wh7rL6?p=preview

And here is the page source, but it would require some tweaks to point to script files to the correct locations.

<!DOCTYPE HTML>
<html>
    <head>
        <script type="text/javascript" src="./js/lib/shp.js"></script>
        <script type="text/javascript" src="./js/lib/catiline.js"></script>

        <style>
            body{
                margin: 0px;
                padding: 0px;
            }
            #fileDropDiv{
                width: 400px;
                height: 200px;
                background-color: #6BCC8A;
            }
        </style>
    </head>
    <body>
        <div id="fileDropDiv">
            Drag and Drop Zipped Shape File here!
        </div> 

        <script type="text/javascript">

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

            //Setup the FileReader
            var reader = new FileReader();

            reader.onload = function(event) {

                var bufferArray = event.target.result;

                console.time('shp');
                worker.data(bufferArray).then(function(data){
                    console.timeEnd('shp');
                });
            };

            //Setup the drop div
            var fileDropDiv = document.getElementById('fileDropDiv');
            fileDropDiv.ondragover = function(){ return false; };
            fileDropDiv.ondragend = function(){ return false; };
            fileDropDiv.ondrop = function(e){

                e.preventDefault();

                //Only handle a single file for demo
                var files = e.dataTransfer.files;
                var file = files[0];

                //Read the shape file
                reader.readAsArrayBuffer(file);

                return false;
            };
        </script>

    </body>
</html>

There is a problem in Chinese；If the attribute value is Chinese, there will be a problem with the Chinese language

In an npm/webpack build, I cannot use shapefile-js. It appears that the proj4 module is not resolved correctly; shapefile-js throws with proj4 is not a function.

On line 57 of index.js,
zip[key.slice(0, -3) + key.slice(-3).toLowerCase()] = proj4(zip[key]);
should maybe be:
zip[key.slice(0, -3) + key.slice(-3).toLowerCase()] = proj4.default(zip[key]);

This works for me, but maybe it breaks some other build-tooling scenario. I've never seen this kind of problem before with published code.

If you need a "prj" shapefile to test against, grab a census shapefile from https://www.census.gov/geo/maps-data/data/cbf/cbf_sld.html .

When trying to use combine, the shp.parseShp and parseDbf are not available methods

Hola! @joaosr has created a ZenHub account for the calvinmetcalf organization. ZenHub is the only project management tool integrated natively in GitHub – created specifically for fast-moving, software-driven teams.

How do I use ZenHub?

To get set up with ZenHub, all you have to do is download the browser extension and log in with your GitHub account. Once you do, you’ll get access to ZenHub’s complete feature-set immediately.

What can ZenHub do?

ZenHub adds a series of enhancements directly inside the GitHub UI:

• Real-time, customizable task boards for GitHub issues;
• Multi-Repository burndown charts, estimates, and velocity tracking based on GitHub Milestones;
• Personal to-do lists and task prioritization;
• Time-saving shortcuts – like a quick repo switcher, a “Move issue” button, and much more.

Add ZenHub to GitHub

Still curious? See more ZenHub features or read user reviews. This issue was written by your friendly ZenHub bot, posted by request from @joaosr.

It doesn't look like the local zipfile example works in IE 10. I tried both dragging and dropping a file as well as uploading it. I'm asking about this because I allow the user to drag a drop a zipfile containing a shapefile in my application. I am using this library to get the geojson. It works fine in IE 11, but in IE 10 it looks like the promise is never resolved.

I was hoping you could provide some insight. Thanks.

Return an error or return coordinates when a shapefile has 3d coordinates

Example file: https://www.dropbox.com/s/71u0356yp3y7hro/data.shp.zip

Is there any possibility to add support for Czech national characters? I have some shapefiles in which are caracters Á, á, Č, č, Ď, ď, É, é, Ě, ě, Í, í, Ň, ň, Ó, ó, Ř, ř, Š, š, Ť, ť, Ú, ú, Ů, ů, Ý, ý, Ž and ž (the first table on http://esperanto.czechian.net/utf-8.html). I want to use leaflet.js for my bachelor's thesis, so it is very important for me. Im new in js, so i would be very grateful if there is any chance to add this on shapefile-js or giving me any advice where to start to do it. Thank You in advance.

I have a shapefile containing a polygon that crosses the antimeridian. Shapefiles determine the interior of a polygon by the order of vertices, specifically, the exterior ring vertices are in clockwise order. (http://www.esri.com/library/whitepapers/pdfs/shapefile.pdf page 8) GeoJSON doesn't follow any particular vertex ordering, thus to indicate antimeridian crossing, the polygon should be split on the 180 degree line. (https://tools.ietf.org/html/rfc7946#section-3.1.9)

example.zip

Hi!

Any plans to allow parse shp-files uploaded via File API?

Thanks!

In 3.4.0, dist/shp.js line 130: readInt32 should probably be readInt32LE.

Howdy,

Maybe this is the wrong place, so direct me if need be. Per this example, the worker adds features to the map upon parsing, and prints out the exception on fail

However, when parsing zips that do not contain valid shp files, an error is retrievable when not using a worker, and when using catiline, no error is thrown. It may be that I just don't know what I am doing, but I can't figure out if the shp file I am processing is failing.

var worker = cw({data:wfunc},2); worker.data(cw.makeUrl('../files/nj_muni.zip')).then(function(data){ geo.addData(data); },function(a){ console.log(a); }); worker.data(cw.makeUrl('../files/maSP.zip')).then(function(data){ geo.addData(data); },function(a){ console.log(a); });

My code is very similar.

`
var file = "actions/shape-files/view-file.php?name="+filename;

            var wfunc = function(base,cb){
                importScripts('../../dist/shapefile-js/dist/shp.min.js');
                shp(base).then(cb);
            }

            var worker = cw({data:wfunc},1);

            var worker_result = worker.data(cw.makeUrl(file));
            console.log(worker_result);

            worker_result.then(function(geojson){

                //do something with your geojson
                if(geojson == null){
                    $('#main-alert').html("<div class=\"alert alert-danger\"><i class=\"fa fa-thumbs-o-down\" aria-hidden=\"true\"></i> Failed to parse </div>");
                } else {
                    $('#console').val( $('#console').val() + 'Result: Passed\r\n');

                    if(checkType(geojson) === 'array'){
                        for(var i=0; i<geojson.length; i++){
                            $('#console').val( $('#console').val() + 'Shapefile : '+geojson[i].fileName+' - ['+geojson[i].type+']\r\n');
                        }
                        global_json = geojson;
                        process_data(global_json);
                    } else {
                        $('#console').val( $('#console').val() + 'Shapefile : '+geojson.fileName+' - ['+geojson.type+']\r\n');
                        global_json = [geojson];
                        process_data(global_json);
                    }

                    //console.log(geojson);
                }
            },function(a){console.log(a)});`

Getting the following error in the console from the web example: http://leaflet.calvinmetcalf.com/#3/32.62/10.63

shit RangeError: Offset is outside the bounds of the DataView(…)

I used the shapefile from: http://lynxlynx.info/gis/multipoint.shape.zip

When using the code for the local shapefile it only allows me to upload one shapefile and that is it.

Cool library! Using this in node it is probably not ideal (memory could be huge even at just 20 items) to be using the LRU cache for long running applications. It would be nice if this was optional where it can be turned off entirely and or change the allowed cache size.

I saw this in your TODO list, and I was wondering if you looked at using the proj4js project to implement this.

I've made attempts in the past to make my own projection libraries, and the math is fascinating, but once I realized how many projections are out there, I decided gdal/ogr tools were good enough for me.

so far we only have tested n and c

I get the error below when I attempt to convert a shapefile in EPSG:3995 to geojson. The shapefile contains a single polygon and was created using QGIS Pisa.

RangeError: Invalid typed array length
at new Uint8Array (native)
at dbfRowHeader (http://localhost:8080/common/lib/shp/dist/shp.js:11964:43)
at module.exports (http://localhost:8080/common/lib/shp/dist/shp.js:12011:19)
at Function.shp.parseZip (http://localhost:8080/common/lib/shp/dist/shp.js:17562:58)
at http://localhost:8080/common/lib/shp/dist/shp.js:17622:16
at tryToUnwrap (http://localhost:8080/common/lib/shp/dist/shp.js:11491:5)
at tryCatch (http://localhost:8080/common/lib/shp/dist/shp.js:11515:17)
at Object.safelyResolveThenable as safely
at new Promise (http://localhost:8080/common/lib/shp/dist/shp.js:11326:21)
at Function.shp.getShapefile (http://localhost:8080/common/lib/shp/dist/shp.js:17621:10)