Merge 0a06368 into f1db0c0

angular.json

@@ -18,7 +18,7 @@
           "builder": "@angular-builders/custom-webpack:browser",
           "options": {
             "customWebpackConfig": {
-              "path": "./webpack/webpack.common.ts",
+              "path": "./webpack/webpack.browser.ts",
               "mergeStrategies": {
                 "loaders": "prepend"
               }
@@ -30,7 +30,8 @@
             "tsConfig": "tsconfig.app.json",
             "aot": false,
             "assets": [
-              "src/assets"
+              "src/assets",
+              "src/robots.txt"
             ],
             "styles": [
               "src/styles.scss"
@@ -84,7 +85,7 @@
           "builder": "@angular-builders/custom-webpack:karma",
           "options": {
             "customWebpackConfig": {
-              "path": "./webpack/webpack.common.ts",
+              "path": "./webpack/webpack.test.ts",
               "mergeStrategies": {
                 "loaders": "prepend"
               }

package.json

@@ -146,6 +146,7 @@
     "dotenv": "^8.2.0",
     "fork-ts-checker-webpack-plugin": "^0.4.10",
     "html-webpack-plugin": "^3.2.0",
+    "http-proxy-middleware": "^1.0.5",
     "jasmine-core": "^3.3.0",
     "jasmine-marbles": "0.3.1",
     "jasmine-spec-reporter": "~4.2.1",

scripts/set-env.ts

@@ -54,6 +54,13 @@ import(environmentFilePath)
 function generateEnvironmentFile(file: GlobalConfig): void {
   file.production = production;
   buildBaseUrls(file);
+
+  // TODO remove workaround in beta 5
+  if (file.rest.nameSpace.match("(.*)/api/?$") !== null) {
+    const newValue = getNameSpace(file.rest.nameSpace);
+    console.log(colors.white.bgMagenta.bold(`The rest.nameSpace property in your environment file or in your DSPACE_REST_NAMESPACE environment variable ends with '/api'.\nThis is deprecated. As '/api' isn't configurable on the rest side, it shouldn't be repeated in every environment file.\nPlease change the rest nameSpace to '${newValue}'`));
+  }
+
   const contents = `export const environment = ` + JSON.stringify(file);
   writeFile(targetPath, contents, (err) => {
     if (err) {
@@ -112,5 +119,16 @@ function getPort(port: number): string {
 }
 
 function getNameSpace(nameSpace: string): string {
-  return nameSpace ? nameSpace.charAt(0) === '/' ? nameSpace : '/' + nameSpace : '';
+  // TODO remove workaround in beta 5
+  const apiMatches = nameSpace.match("(.*)/api/?$");
+  if (apiMatches != null) {
+    let newValue = '/'
+    if (hasValue(apiMatches[1])) {
+      newValue = apiMatches[1];
+    }
+    return newValue;
+  }
+  else {
+    return nameSpace ? nameSpace.charAt(0) === '/' ? nameSpace : '/' + nameSpace : '';
+  }
 }

server.ts

@@ -33,6 +33,7 @@ import { enableProdMode, NgModuleFactory, Type } from '@angular/core';
 
 import { REQUEST, RESPONSE } from '@nguniversal/express-engine/tokens';
 import { environment } from './src/environments/environment';
+import { createProxyMiddleware } from 'http-proxy-middleware';
 
 /*
  * Set path for the browser application's dist folder
@@ -106,6 +107,11 @@ app.set('view engine', 'html');
  */
 app.set('views', DIST_FOLDER);
 
+/**
+ * Proxy the sitemaps
+ */
+app.use('/sitemap**', createProxyMiddleware({ target: `${environment.rest.baseUrl}/sitemaps`, changeOrigin: true }));
+
 /*
  * Adds a cache control header to the response
  * The cache control value can be configured in the environments file and defaults to max-age=60

src/app/core/shared/hal-endpoint.service.spec.ts

@@ -61,7 +61,7 @@ describe('HALEndpointService', () => {
   describe('getRootEndpointMap', () => {
     it('should configure a new EndpointMapRequest', () => {
       (service as any).getRootEndpointMap();
-      const expected = new EndpointMapRequest(requestService.generateRequestId(), environment.rest.baseUrl);
+      const expected = new EndpointMapRequest(requestService.generateRequestId(), environment.rest.baseUrl + '/api');
       expect(requestService.configure).toHaveBeenCalledWith(expected);
     });
 

src/app/core/shared/hal-endpoint.service.ts

@@ -16,7 +16,7 @@ export class HALEndpointService {
   }
 
   protected getRootHref(): string {
-    return new RESTURLCombiner('/').toString();
+    return new RESTURLCombiner('/api').toString();
   }
 
   protected getRootEndpointMap(): Observable<EndpointMap> {

src/robots.txt

@@ -0,0 +1,151 @@
+# The URL to the DSpace sitemaps
+# XML sitemap is listed first as it is preferred by most search engines
+Sitemap: /sitemap_index.xml
+Sitemap: /sitemap_index.html
+
+##########################
+# Default Access Group
+# (NOTE: blank lines are not allowable in a group record)
+##########################
+User-agent: *
+# Disable access to Discovery search and filters; admin pages; processes; submission; workspace; workflow & profile page
+Disallow: /search
+Disallow: /admin/*
+Disallow: /processes
+Disallow: /submit
+Disallow: /workspaceitems
+Disallow: /profile
+Disallow: /workflowitems
+
+# Optionally uncomment the following line ONLY if sitemaps are working
+# and you have verified that your site is being indexed correctly.
+# Disallow: /browse/*
+#
+# If you have configured DSpace (Solr-based) Statistics to be publicly
+# accessible, then you may not want this content to be indexed
+# Disallow: /statistics
+#
+# You also may wish to disallow access to the following paths, in order
+# to stop web spiders from accessing user-based content
+# Disallow: /contact
+# Disallow: /feedback
+# Disallow: /forgot
+# Disallow: /login
+# Disallow: /register
+
+
+##############################
+# Section for misbehaving bots
+# The following directives to block specific robots were borrowed from Wikipedia's robots.txt
+##############################
+
+# advertising-related bots:
+User-agent: Mediapartners-Google*
+Disallow: /
+
+# Crawlers that are kind enough to obey, but which we'd rather not have
+# unless they're feeding search engines.
+User-agent: UbiCrawler
+Disallow: /
+
+User-agent: DOC
+Disallow: /
+
+User-agent: Zao
+Disallow: /
+
+# Some bots are known to be trouble, particularly those designed to copy
+# entire sites. Please obey robots.txt.
+User-agent: sitecheck.internetseer.com
+Disallow: /
+
+User-agent: Zealbot
+Disallow: /
+
+User-agent: MSIECrawler
+Disallow: /
+
+User-agent: SiteSnagger
+Disallow: /
+
+User-agent: WebStripper
+Disallow: /
+
+User-agent: WebCopier
+Disallow: /
+
+User-agent: Fetch
+Disallow: /
+
+User-agent: Offline Explorer
+Disallow: /
+
+User-agent: Teleport
+Disallow: /
+
+User-agent: TeleportPro
+Disallow: /
+
+User-agent: WebZIP
+Disallow: /
+
+User-agent: linko
+Disallow: /
+
+User-agent: HTTrack
+Disallow: /
+
+User-agent: Microsoft.URL.Control
+Disallow: /
+
+User-agent: Xenu
+Disallow: /
+
+User-agent: larbin
+Disallow: /
+
+User-agent: libwww
+Disallow: /
+
+User-agent: ZyBORG
+Disallow: /
+
+User-agent: Download Ninja
+Disallow: /
+
+# Misbehaving: requests much too fast:
+User-agent: fast
+Disallow: /
+
+#
+# If your DSpace is going down because of someone using recursive wget,
+# you can activate the following rule.
+#
+# If your own faculty is bringing down your dspace with recursive wget,
+# you can advise them to use the --wait option to set the delay between hits.
+#
+#User-agent: wget
+#Disallow: /
+
+#
+# The 'grub' distributed client has been *very* poorly behaved.
+#
+User-agent: grub-client
+Disallow: /
+
+#
+# Doesn't follow robots.txt anyway, but...
+#
+User-agent: k2spider
+Disallow: /
+
+#
+# Hits many times per second, not acceptable
+# http://www.nameprotect.com/botinfo.html
+User-agent: NPBot
+Disallow: /
+
+# A capture bot, downloads gazillions of pages with no public benefit
+# http://www.webreaper.net/
+User-agent: WebReaper
+Disallow: /

webpack/webpack.browser.ts

@@ -0,0 +1,8 @@
+import { commonExports } from './webpack.common';
+
+module.exports = Object.assign({}, commonExports, {
+  target: 'web',
+  node: {
+    module: 'empty'
+  }
+});

webpack/webpack.common.ts

@@ -21,11 +21,13 @@ export const copyWebpackOptions = [
   }, {
     from: path.join(__dirname, '..', 'src', 'assets', 'i18n'),
     to: path.join('assets', 'i18n')
+  }, {
+    from: path.join(__dirname, '..', 'src', 'robots.txt'),
+    to: path.join('robots.txt')
   }
-  ];
+];
 
 export const commonExports = {
-  target: 'web',
   plugins: [
     new CopyWebpackPlugin(copyWebpackOptions),
     new HtmlWebpackPlugin({
@@ -37,9 +39,6 @@ export const commonExports = {
       defaultAttribute: 'defer'
     })
   ],
-  node: {
-    module: 'empty'
-  },
   module: {
     rules: [
       {
@@ -99,5 +98,3 @@ export const commonExports = {
     ],
   }
 };
-
-module.exports = commonExports;