Reorganize module loading.

We now use webpack instead of SystemJS, effectively bundling modules
into one file (plus commons chunks) for every entry point.  This results
in a much smaller extension size (almost half).  Furthermore we use
yarn/npm even for extension run-time dependencies.  This relieves us
from manually vendoring and building dependencies.  It's also easier to
understand for new developers familiar with node.

1 files changed, 0 insertions, 101 deletions

diff --git a/thirdparty/URI.js/utils/sld.js b/thirdparty/URI.js/utils/sld.js
deleted file mode 100644
index b0c95db06..000000000
--- a/thirdparty/URI.js/utils/sld.js
+++ /dev/null
@@ -1,101 +0,0 @@
-var fs = require('fs');
-var url = require('url');
-var http = require('http');
-var domains = {};
-
-/*
-    Problem with PublicSuffix:
-    The list not only contains TLDs/SLDs, but also domains like "dyndns.org".
-    While this may be useful for Cookie-Origin-Policy, these domains are possibly
-    being handled by URI.js, considering URI("//dyndns.org").tld("com").
-    The list does not distinguish "official" TLDs from such domains.
-    (At least I have problems with treating "cc.ga.us" as a SLD)
-*/
-
-http.get("http://mxr.mozilla.org/mozilla-central/source/netwerk/dns/effective_tld_names.dat?raw=1", function(res) {
-    res.on('data', function(data) {
-        data.toString().replace(/\r/g, "").split("\n").forEach(function(line) {
-            // skip empty lines, comments and TLDs
-            if (!line || (line[0] === "/" && line[1] === "/") || line.indexOf('.') === -1) {
-                return;
-            }
-            
-            var parts = line.split('.');
-            var sld = parts.slice(0, -1).join('.');
-            var tld = parts.slice(-1);
-
-            if (parts.length < 2) {
-                return;
-            }
-            
-            if (!domains[tld]) {
-                domains[tld] = [];
-            }
-            
-            domains[tld].push(sld);
-        });
-    }).on('end', function() {
-        //file.end();
-        for (var tld in domains) {
-            domains[tld].sort();
-            
-            // ! and * are sorted to the top
-            if (domains[tld][0][0] == '!') {
-                // we have wildcards and exclusions
-            } else if (domains[tld][0][0] == '*') {
-                // we have wildcards
-            } else {
-                // simple list 
-            }
-        }
-        
-        console.log(JSON.stringify(domains, null, 2));
-        //console.log(domains.jp);
-    });
-});
-
-/*
-
-
-// https://github.com/oncletom/tld.js
-// generates a 430KB file, which is inacceptible for the web
-
-build a regex pattern from this -- http://publicsuffix.org/list/
-"!exclusion"
-"*" wildcard
-  
-uk: [ '!bl',
-      '!british-library',
-      '!jet',
-      '!mod',
-      '!national-library-scotland',
-      '!nel',
-      '!nic',
-      '!nls',
-      '!parliament',
-      '*',
-      '*.nhs',
-      '*.police',
-      '*.sch',
-      'blogspot.co' ]
-    
-jp: [ '!city.kawasaki',
-      '!city.kitakyushu',
-      '!city.kobe',
-      '!city.nagoya',
-      '!city.sapporo',
-      '!city.sendai',
-      '!city.yokohama',
-      '*.kawasaki',
-      '*.kitakyushu',
-      '*.kobe',
-      '*.nagoya',
-      '*.sapporo',
-      '*.sendai',
-      '*.yokohama',
-      'abashiri.hokkaido',
-      'abeno.osaka',
-      'abiko.chiba',
-      … ]
-
-*/