xrfragment-haxe/src/xrfragment/URI.hx

// SPDX-License-Identifier: MPL-2.0        
// Copyright (c) 2023 Leon van Kammen/NLNET 

/*
 * various snippets originate from:
 *
 * http://haxe.org/doc/snip/uri_parser,
 * https://github.com/haxecocktail/cocktail-url/blob/master/cocktail/url/URI.hx
 */

package xrfragment;

import xrfragment.Parser;
import xrfragment.XRF;

/**
 * # Spec
 *
 * > version 1.0.0 [![Actions Status](https://github.com/coderofsalvation/xrfragment/workflows/test/badge.svg)](https://github.com/coderofsalvation/xrfragment/actions) generated by `make doc` @ $(date +"%Y-%m-%dT%H:%M:%S%z")
 * 
 * ### XR Fragment URI Grammar 
 *
 * ```
 *     reserved    = gen-delims / sub-delims / xrf-scheme
 *     gen-delims  = "#" / "&"                      
 *     sub-delims  = "," /  "="
 *     xrf-scheme  = "xrf://"
 * ```
 *
 * In case your programming language has no parser ([check here](https://github.com/coderofsalvation/xrfragment/tree/main/dist)) you can [crosscompile it](https://github.com/coderofsalvation/xrfragment/blob/main/build.hxml), or roll your own `Parser.parse(k,v,store)` using the spec:
 *
 */

@:expose                                                                   // <- makes the class reachable from plain JavaScript
@:keep                                                                     // <- avoids accidental removal by dead code elimination
class URI {

    /**
     * URI parts names
     */
    private static var _parts : Array<String> = ["source", "scheme", "authority", "userInfo", "user",
    "password","host","port","relative","path","directory","file","query","fragment"];

    /**
     * URI parts
     */
    public var url : String;
    public var source : String;
    public var scheme : String;
    public var authority : String;
    public var userInfo : String;
    public var user : String;
    public var password : String;
    public var host : String;
    public var port : String;
    public var relative : String;
    public var path : String;
    public var directory : String;
    public var file : String;
    public var fileExt : String;
    public var query : String;
    public var fragment : String = "";
    public var hash : haxe.DynamicAccess<Dynamic> = {};
    public var XRF : haxe.DynamicAccess<Dynamic>  = {};
    public var URN : String;

    /**
     * class constructor
     */
    public function new( )
    {
    }

    @:keep
    public static function parseFragment(url:String,filter:Int):haxe.DynamicAccess<Dynamic> {
      var store:haxe.DynamicAccess<Dynamic> = {};                          //  1. store key/values into a associative array or dynamic object
      if( url == null || url.indexOf("#") == -1 ) return store;
      var fragment:Array<String>    = url.split("#");                      //  1. fragment URI starts with `#`
      var splitArray:Array<String>  = fragment[1].split('&');              //  1. fragments are split by `&`
      for (i in 0...splitArray.length) {                                   //  1. loop thru each fragment

        var splitByEqual = splitArray[i].split('=');                       //  1. for each fragment split on `=` to separate key/values 
        var regexPlus  = ~/\+/g;                                           //  1. fragment-values are urlencoded (space becomes `+` using `encodeUriComponent` e.g.)
        var key:String = splitByEqual[0];
				var value:String = "";
        if (splitByEqual.length > 1) {
          if( XRF.isVector.match(splitByEqual[1]) ) value = splitByEqual[1];
          else value = StringTools.urlDecode(regexPlus.split(splitByEqual[1]).join(" "));
        }
				var ok:Bool = Parser.parse(key,value,store,i);                     //  1. for every recognized fragment key/value-pair call [Parser.parse](#%E2%86%AA%20Parser.parse%28k%2Cv%2Cstore%29)
      }
      if( filter != null && filter != 0 ){
        for (key in store.keys()) {
            var xrf:XRF = store.get(key);
            if( !xrf.is( filter ) ){ 
							store.remove(key);
						}
        }
      }
      return store;
    }

    @keep
    public static function template(uri:String, vars:Dynamic):String {
      var parts = uri.split("#");
      if( parts.length == 1 ) return uri; // we only do level1 fragment expansion
      var frag  = parts[1];
      frag = StringTools.replace(frag,"{","::");
      frag = StringTools.replace(frag,"}","::");
      frag = new haxe.Template(frag).execute(vars);
      frag = StringTools.replace(frag,"null",""); // *TODO* needs regex to check for [#&]null[&]
      parts[1] = frag;
      return parts.join("#");
    }

    /**
     * Parse a string url and return a typed
     * object from it.
     * 
     * note : implementation originate from here :
     * http://haxe.org/doc/snip/uri_parser
     */
    public static function parse(stringUrl:String, flags:Int ):URI
    {
        // The almighty regexp (courtesy of http://blog.stevenlevithan.com/archives/parseuri)
        var r : EReg = ~/^(?:(?![^:@]+:[^:@\/]*@)([^:\/?#.]+):)?(?:\/\/)?((?:(([^:@]*)(?::([^:@]*))?)?@)?([^:\/?#]*)(?::(\d*))?)(((\/(?:[^?#](?![^?#\/]*\.[^?#\/.]+(?:[?#]|$)))*\/?)?([^?#\/]*))(?:\?([^#]*))?(?:#(.*))?)/;

        if( stringUrl.indexOf("://") == -1 && stringUrl.charAt(0) != '/' && stringUrl.charAt(0) != '#' ){
          stringUrl = "/" + stringUrl; // workaround for relative urls
        }
 
        // Match the regexp to the url
        r.match(stringUrl);
 
        var url:URI = new URI();
        
        // Use reflection to set each part
        for (i in 0..._parts.length)
        {
            Reflect.setField(url, _parts[i],  r.matched(i));
        }
        
        //hack for relative url with only a file
        if (isRelative(url) == true)
        {
            if (url.directory == null && url.host != null)
            {
                url.file = url.host;
            }
            url.host = "";
        }

        url.hash = {};
        if( url.fragment != null && url.fragment.length > 0 ){
          url.XRF = xrfragment.URI.parseFragment( "#"+url.fragment, flags );
          var key:String;
          for( key in url.XRF.keys() ){
            var v:haxe.DynamicAccess<Dynamic> = url.XRF.get(key);
            url.hash[key] = v.get("string");
          }
        }

        computeVars(url);

        return url;
    }

    private static function computeVars( url:URI ) {
        // clean up url
        var r = ~/\/\//g;
        if( url.directory != null && url.directory.indexOf("//") != -1 ){
          url.directory = r.replace(url.directory,"/");
        }
        if( url.path != null && url.path.indexOf("//") != -1 ){
          url.path = r.replace(url.path,"/");
        }
        if( url.file != null && url.file.indexOf("//") != -1 ){
          url.file = r.replace(url.file,"/");
        }
        // generate URN
        url.URN = url.scheme + "://" + url.host;
        if( url.port != null ) url.URN += ":"+url.port;
        url.URN += url.directory;

        // extract file extension if any
        if( url.file != null){
          var parts:Array<String> = url.file.split(".");
          if( parts.length > 1 ){
            url.fileExt = parts.pop();
          }
        }

    }
    
    /**
     * Serialize an URl OBJect into an 
     * URI string
     */
    public static function toString(url:URI):String
    {
        var result:String = "";
        
        if (url.scheme != null)
        {
            result += url.scheme + "://";
        }
        
        if (url.user != null)
        {
            result += url.user + ":";
        }
        
        if (url.password != null)
        {
            result += url.password + "@";
        }
        
        if (url.host != null)
        {
            result += url.host;
        }
        
        if (url.port != null)
        {
            result += ":" + url.port;
        }
        
        if (url.directory != null)
        {
            result += url.directory;
        }
        
        if (url.file != null)
        {
            result += url.file;
        }
        
        if (url.query != null)
        {
            result += "?" + url.query;
        }
        
        if (url.fragment != null)
        {
            result += "#" + url.fragment;
        }
        
        return result;
    }
    
    /**
     * takes 2 urls and return a new url which is the result
     * of appending the second url to the first.
     * 
     * if the first url points to a file, the file is removed
     * and the appended url is added after the last directory
     * 
     * only the query string and fragment of the appended url are used
     */
    public static function appendURI(url:URI, appendedURI:URI):URI
    {
        if (isRelative(url) == true)
        {
            return appendToRelativeURI(url, appendedURI);
        }
        else
        {
            return appendToAbsoluteURI(url, appendedURI);
        }
    }
    
    /**
     * return wether the url is relative (true)
     * or absolute (false)
     */
    public static function isRelative(url:URI):Bool
    {
        return url.scheme == null;
    }
    
    /**
     * append the appended url to a relative url 
     */
    public static function appendToRelativeURI(url:URI, appendedURI:URI):URI
    {
        //when relative url parsed, if it contains only a file (ex : "style.css")
        //then it will store it in the host attribute. So if the url has no directory
        //then only the appended url content is returned, as this method replace the file
        //part of the base url anyway
        if (url.directory == null || url.host == null)
        {
            return cloneURI(appendedURI);
        }
        
        var resultURI:URI = new URI();
        resultURI.host = url.host;
        resultURI.directory = url.directory;

        if (appendedURI.host != null)
        {
            resultURI.directory += appendedURI.host;
        }
        
        if (appendedURI.directory != null)
        {
            var directory = appendedURI.directory;
            if (appendedURI.host == null)
            {
                //remove the initial '/' char if no host, as already present
                //in base url
                resultURI.directory += directory.substr(1);
            }
            else
            {
                resultURI.directory += directory;
            }
            
        }
        
        if (appendedURI.file != null)
        {
            resultURI.file = appendedURI.file;
        }
        
        resultURI.path = resultURI.directory + resultURI.file;
        
        if (appendedURI.query != null)
        {
            resultURI.query = appendedURI.query;
        }
        
        if (appendedURI.fragment != null)
        {
            resultURI.fragment = appendedURI.fragment;
        }
        
        return resultURI;
    }
    
    /**
     * append the appended url to an absolute url 
     */
    public static function appendToAbsoluteURI(url:URI, appendedURI:URI):URI
    {
        var resultURI:URI = new URI();
        
        if (url.scheme != null)
        {
            resultURI.scheme = url.scheme;
        }
        
        if (url.host != null)
        {
            resultURI.host = url.host;
        }
        
        var directory:String = "";
        if (url.directory != null)
        {
            directory = url.directory;
        }
        
        if (appendedURI.host != null)
        {
            appendedURI.directory += appendedURI.host;
        }
        
        if (appendedURI.directory != null)
        {
            directory += appendedURI.directory;
        }
        
        resultURI.directory = directory;
        
        if (appendedURI.file != null)
        {
            resultURI.file = appendedURI.file;
        }
        
        resultURI.path = resultURI.directory + resultURI.file;
        
        if (appendedURI.query != null)
        {
            resultURI.query = appendedURI.query;
        }
        
        if (appendedURI.fragment != null)
        {
            resultURI.fragment = appendedURI.fragment;
        }
        
        return resultURI;
    }

    /**
     * append the appended url to an absolute url 
     */
    public static function toAbsolute(url:URI, newUrl:String ):URI
    {
        var newURI:URI    = parse(newUrl,0);
        var resultURI:URI = new URI();
        
        resultURI.port = url.port;

        resultURI.source = newUrl;

        if (newURI.scheme != null)
        {
            resultURI.scheme = newURI.scheme;
        }else{
            resultURI.scheme = url.scheme;
        }
       
        if (newURI.host != null && newURI.host.length > 0 )
        {
            resultURI.host = newURI.host;
            resultURI.port = null;
            resultURI.fragment = null;
            resultURI.hash = {};
            resultURI.XRF = {};
            if( newURI.port != null ){
              resultURI.port = newURI.port;
            }
        }else{
            resultURI.host = url.host;
        }
        
        var directory:String = "";
        if (url.directory != null)
        {
            directory = url.directory;
        }
        
        if (newURI.directory != null && newURI.source.charAt(0) != "#" && newURI.directory.length > 0)
        {
            if( newUrl.charAt(0) != '/' && newUrl.indexOf("://") == -1 ){
              var stripRelative : EReg = ~/\.\/.*/;
              directory = stripRelative.replace( directory, '');
              directory += newURI.directory;
            }else{
              directory = newURI.directory;
            }
        }
        
        resultURI.directory = directory;
       
        if (newURI.file != null && newURI.file.length > 0)
        {
            resultURI.file = newURI.file;
        }else{
            resultURI.file = url.file;
        }
       
        resultURI.path = resultURI.directory + resultURI.file;
        
        if (newURI.query != null)
        {
            resultURI.query = newURI.query;
        }
        
        if (newURI.fragment != null)
        {
            resultURI.fragment = newURI.fragment;
        }
        resultURI.hash = newURI.hash;
        resultURI.XRF  = newURI.XRF;
        computeVars(resultURI);
        
        return resultURI;
    }
    
    /**
     * clone the provided url
     */
    private static function cloneURI(url:URI):URI
    {
        var clonedURI:URI = new URI();
        
        clonedURI.url = url.url;
        clonedURI.source = url.source;
        clonedURI.scheme = url.scheme;
        clonedURI.authority = url.authority;
        clonedURI.userInfo = url.userInfo;
        clonedURI.password = url.password;
        clonedURI.host = url.host;
        clonedURI.port = url.port;
        clonedURI.relative = url.relative;
        clonedURI.path = url.path;
        clonedURI.directory = url.directory;
        clonedURI.file = url.file;
        clonedURI.query = url.query;
        clonedURI.fragment = url.fragment;
        
        return clonedURI;
    }

}

/**
 * > icanhazcode? yes, see [URI.hx](https://github.com/coderofsalvation/xrfragment/blob/main/src/xrfragment/URI.hx)
 *
 * # Tests
 *   
 * the spec is tested with [JSON unittests](./../src/spec) consumed by [Test.hx](./../src/Test.hx) to cross-test all languages.
 */