Normalize metadata language to BCP 47 for HTML lang attributes

src/app/entity-groups/journal-entities/item-grid-elements/search-result-grid-elements/journal-volume/journal-volume-search-result-grid-element.component.html

@@ -31,7 +31,7 @@ <h4 class="card-title" [innerHTML]="dsoTitle"></h4>
       </p>
       <p *ngIf="dso.hasMetadata('dc.description')" class="item-description card-text">
         <ds-truncatable-part [id]="dso.id" [minLines]="3">
-          <span [innerHTML]="firstMetadataValue('dc.description')" [attr.lang]="dso.firstMetadata('dc.description')?.language || null"></span>
+          <span [innerHTML]="firstMetadataValue('dc.description')" [attr.lang]="dso.firstMetadata('dc.description')?.language | dsLang"></span>
         </ds-truncatable-part>
       </p>
       <div *ngIf="linkType != linkTypes.None" class="text-center">

src/app/entity-groups/journal-entities/item-grid-elements/search-result-grid-elements/journal/journal-search-result-grid-element.component.html

@@ -35,7 +35,7 @@ <h4 class="card-title" [innerHTML]="dsoTitle"></h4>
       </p>
       <p *ngIf="dso.hasMetadata('dc.description')" class="item-description card-text">
         <ds-truncatable-part [id]="dso.id" [minLines]="3">
-          <span [innerHTML]="firstMetadataValue('dc.description')" [attr.lang]="dso.firstMetadata('dc.description')?.language || null"></span>
+          <span [innerHTML]="firstMetadataValue('dc.description')" [attr.lang]="dso.firstMetadata('dc.description')?.language | dsLang"></span>
         </ds-truncatable-part>
       </p>
       <div *ngIf="linkType != linkTypes.None" class="text-center">

src/app/entity-groups/research-entities/item-grid-elements/search-result-grid-elements/project/project-search-result-grid-element.component.html

@@ -25,7 +25,7 @@ <h4 class="card-title" [innerHTML]="dsoTitle"></h4>
       </ds-truncatable-part>
       <p *ngIf="dso.hasMetadata('dc.description')" class="item-description card-text text-muted">
         <ds-truncatable-part [id]="dso.id" [minLines]="3">
-          <span [innerHTML]="firstMetadataValue('dc.description')" [attr.lang]="dso.firstMetadata('dc.description')?.language || null"></span>
+          <span [innerHTML]="firstMetadataValue('dc.description')" [attr.lang]="dso.firstMetadata('dc.description')?.language | dsLang"></span>
         </ds-truncatable-part>
       </p>
       <div *ngIf="linkType != linkTypes.None" class="text-center">

src/app/entity-groups/research-entities/item-list-elements/search-result-list-elements/org-unit/org-unit-search-result-list-element.component.html

@@ -31,7 +31,7 @@
       <span *ngIf="dso.allMetadata(['dc.description']).length > 0"
             class="item-list-org-unit-description">
             <ds-truncatable-part [id]="dso.id" [minLines]="3"><span
-                    [innerHTML]="firstMetadataValue('dc.description')" [attr.lang]="dso.firstMetadata('dc.description')?.language || null"></span>
+                    [innerHTML]="firstMetadataValue('dc.description')" [attr.lang]="dso.firstMetadata('dc.description')?.language | dsLang"></span>
             </ds-truncatable-part>
       </span>
   </span>

src/app/entity-groups/research-entities/item-list-elements/search-result-list-elements/org-unit/org-unit-search-result-list-element.component.spec.ts

@@ -5,6 +5,7 @@ import { of as observableOf } from 'rxjs';
 import { OrgUnitSearchResultListElementComponent } from './org-unit-search-result-list-element.component';
 import { Item } from '../../../../../core/shared/item.model';
 import { TruncatePipe } from '../../../../../shared/utils/truncate.pipe';
+import { DsLangPipe } from '../../../../../shared/utils/ds-lang.pipe';
 import { TruncatableService } from '../../../../../shared/truncatable/truncatable.service';
 import { ItemSearchResult } from '../../../../../shared/object-collection/shared/item-search-result.model';
 import { DSONameService } from '../../../../../core/breadcrumbs/dso-name.service';
@@ -75,7 +76,7 @@ describe('OrgUnitSearchResultListElementComponent', () => {
           }
         }
       )],
-      declarations: [ OrgUnitSearchResultListElementComponent , TruncatePipe],
+      declarations: [ OrgUnitSearchResultListElementComponent , TruncatePipe, DsLangPipe],
       providers: [
         { provide: TruncatableService, useValue: {} },
         { provide: DSONameService, useClass: DSONameServiceMock },
@@ -145,7 +146,7 @@ describe('OrgUnitSearchResultListElementComponent', () => {
           }
         }
       )],
-      declarations: [OrgUnitSearchResultListElementComponent, TruncatePipe],
+      declarations: [OrgUnitSearchResultListElementComponent, TruncatePipe, DsLangPipe],
       providers: [
         {provide: TruncatableService, useValue: {}},
         {provide: DSONameService, useClass: DSONameServiceMock},

src/app/entity-groups/research-entities/item-list-elements/sidebar-search-list-elements/org-unit/org-unit-sidebar-search-list-element.component.ts

@@ -5,6 +5,7 @@ import { ItemSearchResult } from '../../../../../shared/object-collection/shared
 import { Component } from '@angular/core';
 import { SidebarSearchListElementComponent } from '../../../../../shared/object-list/sidebar-search-list-element/sidebar-search-list-element.component';
 import { Item } from '../../../../../core/shared/item.model';
+import { metadataLangToBcp47 } from '../../../../../shared/utils/metadata-language.util';
 
 @listableObjectComponent('OrgUnitSearchResult', ViewMode.ListElement, Context.SideBarSearchModal)
 @listableObjectComponent('OrgUnitSearchResult', ViewMode.ListElement, Context.SideBarSearchModalCurrent)
@@ -29,6 +30,6 @@ export class OrgUnitSidebarSearchListElementComponent extends SidebarSearchListE
    * Get the language of the Org Unit description.
    */
   getDescriptionLang(): string | null {
-    return this.dso.firstMetadata('dc.description')?.language || null;
+    return metadataLangToBcp47(this.dso.firstMetadata('dc.description')?.language);
   }
 }

src/app/entity-groups/research-entities/metadata-representations/org-unit/org-unit-item-metadata-list-element.component.html

@@ -1,7 +1,7 @@
 <ng-template #descTemplate>
   <span class="text-muted">
       <span class="item-list-job-title">
-        <span [innerHTML]="mdRepresentation.firstMetadataValue(['dc.description'])" [attr.lang]="mdRepresentation.firstMetadata(['dc.description'])?.language || null"></span>
+        <span [innerHTML]="mdRepresentation.firstMetadataValue(['dc.description'])" [attr.lang]="mdRepresentation.firstMetadata(['dc.description'])?.language | dsLang"></span>
       </span>
   </span>
 </ng-template>

src/app/entity-groups/research-entities/metadata-representations/org-unit/org-unit-item-metadata-list-element.component.spec.ts

@@ -8,6 +8,7 @@ import { ItemMetadataRepresentation } from '../../../../core/shared/metadata-rep
 import { OrgUnitItemMetadataListElementComponent } from './org-unit-item-metadata-list-element.component';
 import { Item } from '../../../../core/shared/item.model';
 import { MetadataValue } from '../../../../core/shared/metadata.models';
+import { DsLangPipe } from '../../../../shared/utils/ds-lang.pipe';
 
 const description = 'Lorem ipsum dolor sit amet, consetetur sadipscing elitr, sed diam nonumy eirmod tempor invidunt ut labore et dolore magna aliquyam erat, sed diam voluptua.';
 const organisation = 'Anonymous';
@@ -24,7 +25,7 @@ describe('OrgUnitItemMetadataListElementComponent', () => {
       imports:[
         NgbModule
       ],
-      declarations: [OrgUnitItemMetadataListElementComponent],
+      declarations: [OrgUnitItemMetadataListElementComponent, DsLangPipe],
       schemas: [NO_ERRORS_SCHEMA]
     }).overrideComponent(OrgUnitItemMetadataListElementComponent, {
       set: { changeDetection: ChangeDetectionStrategy.Default }

src/app/item-page/field-components/metadata-values/metadata-values.component.html

@@ -4,7 +4,7 @@
       Choose a template. Priority: markdown, link, browse link.
       -->
       <ng-container *ngTemplateOutlet="(renderMarkdown ? markdown : (hasLink(mdValue) ? link : (hasBrowseDefinition() ? browselink : simple)));
-        context: {value: mdValue.value, lang: mdValue.language || null}">
+        context: {value: mdValue.value, lang: getLang(mdValue)}">
       </ng-container>
       <span class="separator" *ngIf="!last" [innerHTML]="separator"></span>
     </ng-container>

src/app/item-page/field-components/metadata-values/metadata-values.component.spec.ts

@@ -63,6 +63,15 @@ describe('MetadataValuesComponent', () => {
     }
   });
 
+  it('should render the metadata language as a normalized BCP 47 lang attribute', () => {
+    const langEls = fixture.debugElement.queryAll(By.css('[lang]'));
+    expect(langEls.length).toBeGreaterThan(0);
+    langEls.forEach((el) => {
+      expect(el.nativeElement.getAttribute('lang')).toEqual('en-US');
+    });
+    expect(fixture.nativeElement.innerHTML).not.toContain('lang="en_US"');
+  });
+
   it('should contain separators equal to the amount of metadata values minus one', () => {
     const separators = fixture.debugElement.queryAll(By.css('span.separator'));
     expect(separators.length).toBe(mockMetadata.length - 1);

src/app/item-page/field-components/metadata-values/metadata-values.component.ts

@@ -5,6 +5,7 @@ import { BrowseDefinition } from '../../../core/shared/browse-definition.model';
 import { hasValue } from '../../../shared/empty.util';
 import { VALUE_LIST_BROWSE_DEFINITION } from '../../../core/shared/value-list-browse-definition.resource-type';
 import { environment } from '../../../../environments/environment';
+import { metadataLangToBcp47 } from '../../../shared/utils/metadata-language.util';
 
 /**
  * This component renders the configured 'values' into the ds-metadata-field-wrapper component.
@@ -112,4 +113,13 @@ export class MetadataValuesComponent implements OnChanges {
       return { target: '_blank', rel: 'noopener noreferrer' };
     }
   }
+
+  /**
+   * Normalize a metadata value's language into a valid BCP 47 tag for the HTML
+   * `lang` attribute, or null when none applies.
+   * @param value A MetadataValue being displayed
+   */
+  getLang(value: MetadataValue): string | null {
+    return metadataLangToBcp47(value.language);
+  }
 }

src/app/item-page/full/field-components/file-section/full-file-section.component.html

@@ -28,7 +28,7 @@ <h3 class="h5 simple-view-element-header">{{"item.page.filesection.original.bund
 
                           <ng-container *ngIf="file.hasMetadata('dc.description')">
                             <dt class="col-md-4">{{"item.page.filesection.description" | translate}}</dt>
-                            <dd class="col-md-8" [attr.lang]="file.firstMetadata('dc.description')?.language || null">{{file.firstMetadataValue("dc.description")}}</dd>
+                            <dd class="col-md-8" [attr.lang]="file.firstMetadata('dc.description')?.language | dsLang">{{file.firstMetadataValue("dc.description")}}</dd>
                           </ng-container>
                         </dl>
                     </div>
@@ -69,7 +69,7 @@ <h3 class="h5 simple-view-element-header">{{"item.page.filesection.license.bundl
 
 
                             <dt class="col-md-4">{{"item.page.filesection.description" | translate}}</dt>
-                            <dd class="col-md-8" [attr.lang]="file.firstMetadata('dc.description')?.language || null">{{file.firstMetadataValue("dc.description")}}</dd>
+                            <dd class="col-md-8" [attr.lang]="file.firstMetadata('dc.description')?.language | dsLang">{{file.firstMetadataValue("dc.description")}}</dd>
                         </dl>
                     </div>
                     <div class="col-2">

src/app/item-page/full/field-components/file-section/full-file-section.component.spec.ts

@@ -7,6 +7,7 @@ import { TranslateLoaderMock } from '../../../../shared/mocks/translate-loader.m
 import { BrowserAnimationsModule } from '@angular/platform-browser/animations';
 import { VarDirective } from '../../../../shared/utils/var.directive';
 import { FileSizePipe } from '../../../../shared/utils/file-size-pipe';
+import { DsLangPipe } from '../../../../shared/utils/ds-lang.pipe';
 import { MetadataFieldWrapperComponent } from '../../../../shared/metadata-field-wrapper/metadata-field-wrapper.component';
 import { BitstreamDataService } from '../../../../core/data/bitstream-data.service';
 import { NO_ERRORS_SCHEMA } from '@angular/core';
@@ -67,7 +68,7 @@ describe('FullFileSectionComponent', () => {
           useClass: TranslateLoaderMock
         }
       }), BrowserAnimationsModule],
-      declarations: [FullFileSectionComponent, VarDirective, FileSizePipe, MetadataFieldWrapperComponent],
+      declarations: [FullFileSectionComponent, VarDirective, FileSizePipe, MetadataFieldWrapperComponent, DsLangPipe],
       providers: [
         { provide: BitstreamDataService, useValue: bitstreamDataService },
         { provide: NotificationsService, useValue: new NotificationsServiceStub() },

src/app/item-page/simple/field-components/clarin-description-item-field/clarin-description-item-field.component.ts

@@ -1,6 +1,7 @@
 import { Component, Input, OnInit } from '@angular/core';
 import { Item } from '../../../../core/shared/item.model';
 import { makeLinks } from '../../../../shared/clarin-shared-util';
+import { metadataLangToBcp47 } from '../../../../shared/utils/metadata-language.util';
 
 @Component({
   selector: 'ds-clarin-description-item-field',
@@ -27,7 +28,7 @@ export class ClarinDescriptionItemFieldComponent implements OnInit {
   ngOnInit(): void {
     this.descriptionEntries = this.item.allMetadata(this.fields).map(md => ({
       value: makeLinks(md.value),
-      language: md.language || null
+      language: metadataLangToBcp47(md.language)
     }));
   }
 

src/app/shared/clarin-item-box-view/clarin-item-box-view.component.ts

@@ -28,6 +28,7 @@ import { LicenseType } from '../../item-page/clarin-license-info/clarin-license-
 import { ListableObject } from '../object-collection/shared/listable-object.model';
 import { ItemSearchResult } from '../object-collection/shared/item-search-result.model';
 import { getItemPageRoute } from '../../item-page/item-page-routing-paths';
+import { metadataLangToBcp47 } from '../utils/metadata-language.util';
 import { FindListOptions } from '../../core/data/find-list-options.model';
 import { ClarinDateService } from '../clarin-date.service';
 import { AUTHOR_METADATA_FIELDS } from '../../core/shared/clarin/constants';
@@ -151,7 +152,7 @@ export class ClarinItemBoxViewComponent implements OnInit {
     this.itemUri = getItemPageRoute(this.item);
     const descMeta = this.item?.firstMetadata('dc.description');
     this.itemDescription = descMeta?.value || null;
-    this.itemDescriptionLang = descMeta?.language || null;
+    this.itemDescriptionLang = metadataLangToBcp47(descMeta?.language);
     this.itemPublisher = this.item?.firstMetadataValue('dc.publisher');
     this.itemDate = this.clarinDateService.composeItemDate(this.item);
 

src/app/shared/object-grid/search-result-grid-element/item-search-result/item/item-search-result-grid-element.component.spec.ts

@@ -24,6 +24,7 @@ import { ItemSearchResult } from '../../../../object-collection/shared/item-sear
 import { createSuccessfulRemoteDataObject$ } from '../../../../remote-data.utils';
 import { TruncatableService } from '../../../../truncatable/truncatable.service';
 import { TruncatePipe } from '../../../../utils/truncate.pipe';
+import { DsLangPipe } from '../../../../utils/ds-lang.pipe';
 import { ItemSearchResultGridElementComponent } from './item-search-result-grid-element.component';
 
 const mockItemWithMetadata: ItemSearchResult = new ItemSearchResult();
@@ -218,7 +219,7 @@ export function getEntityGridElementTestComponent(component, searchResultWithMet
           NoopAnimationsModule,
           TranslateModule.forRoot()
         ],
-        declarations: [component, TruncatePipe],
+        declarations: [component, TruncatePipe, DsLangPipe],
         providers: [
           { provide: TruncatableService, useValue: truncatableServiceStub },
           { provide: ObjectCacheService, useValue: {} },

src/app/shared/shared.module.ts

@@ -300,6 +300,7 @@ import { ReplacePipe } from './utils/replace.pipe';
 import { ClarinDateService } from './clarin-date.service';
 import { ItemIdentifierService } from './item-identifier.service';
 import { ClarinLicenseRequiredInfoCheckedPipe } from './utils/clarin-license-required-info-checked.pipe';
+import { DsLangPipe } from './utils/ds-lang.pipe';
 
 const MODULES = [
   CommonModule,
@@ -347,7 +348,8 @@ const PIPES = [
   ClarinLicenseRequiredInfoPipe,
   CharToEndPipe,
   ClarinSafeHtmlPipe,
-  ReplacePipe
+  ReplacePipe,
+  DsLangPipe
 ];
 
 const COMPONENTS = [

src/app/shared/utils/ds-lang.pipe.spec.ts

@@ -0,0 +1,25 @@
+import { DsLangPipe } from './ds-lang.pipe';
+
+describe('DsLangPipe', () => {
+  let pipe: DsLangPipe;
+
+  beforeEach(() => {
+    pipe = new DsLangPipe();
+  });
+
+  it('should create', () => {
+    expect(pipe).toBeTruthy();
+  });
+
+  it('should normalize a Java-style locale to a BCP 47 tag', () => {
+    expect(pipe.transform('en_US')).toEqual('en-US');
+    expect(pipe.transform('cs_CZ')).toEqual('cs-CZ');
+  });
+
+  it('should return null for the wildcard, empty and nullish values', () => {
+    expect(pipe.transform('*')).toBeNull();
+    expect(pipe.transform('')).toBeNull();
+    expect(pipe.transform(null)).toBeNull();
+    expect(pipe.transform(undefined)).toBeNull();
+  });
+});

src/app/shared/utils/ds-lang.pipe.ts

@@ -0,0 +1,24 @@
+import { Pipe, PipeTransform } from '@angular/core';
+
+import { metadataLangToBcp47 } from './metadata-language.util';
+
+/**
+ * Pipe that normalizes a DSpace metadata language value into a valid BCP 47
+ * language tag for an HTML `lang` attribute, or `null` when none applies.
+ *
+ * Delegates to {@link metadataLangToBcp47}, so binding `[attr.lang]` through this
+ * pipe both omits the attribute for absent/wildcard languages and converts
+ * Java-style locales (`en_US` → `en-US`).
+ */
+@Pipe({
+  name: 'dsLang'
+})
+export class DsLangPipe implements PipeTransform {
+  /**
+   * @param {string} language the raw metadata language value
+   * @returns {string | null} a BCP 47 language tag, or null when none applies
+   */
+  transform(language: string | null | undefined): string | null {
+    return metadataLangToBcp47(language);
+  }
+}

src/app/shared/utils/metadata-language.util.spec.ts

@@ -0,0 +1,30 @@
+import { metadataLangToBcp47 } from './metadata-language.util';
+
+describe('metadataLangToBcp47', () => {
+  it('should return null for nullish values', () => {
+    expect(metadataLangToBcp47(null)).toBeNull();
+    expect(metadataLangToBcp47(undefined)).toBeNull();
+  });
+
+  it('should return null for an empty string', () => {
+    expect(metadataLangToBcp47('')).toBeNull();
+  });
+
+  it('should return null for the wildcard language', () => {
+    expect(metadataLangToBcp47('*')).toBeNull();
+  });
+
+  it('should convert Java-style locale separators to BCP 47 hyphens', () => {
+    expect(metadataLangToBcp47('en_US')).toEqual('en-US');
+    expect(metadataLangToBcp47('cs_CZ')).toEqual('cs-CZ');
+    expect(metadataLangToBcp47('de_DE')).toEqual('de-DE');
+  });
+
+  it('should leave a bare language subtag unchanged', () => {
+    expect(metadataLangToBcp47('en')).toEqual('en');
+  });
+
+  it('should leave already-hyphenated tags unchanged', () => {
+    expect(metadataLangToBcp47('en-US')).toEqual('en-US');
+  });
+});

src/app/shared/utils/metadata-language.util.ts

@@ -0,0 +1,20 @@
+/**
+ * Convert a DSpace metadata language value into a valid BCP 47 language tag
+ * suitable for an HTML `lang` attribute, or `null` when none applies.
+ *
+ * DSpace stores the metadata language as a Java-style locale (e.g. `en_US`,
+ * `cs_CZ`), the wildcard `*` (language-independent), or nothing. None of the
+ * underscore forms or the wildcard are valid BCP 47 / HTML `lang` values, so
+ * assistive technology ignores them. This normalizes them:
+ * - `null`/`undefined`/empty and the wildcard `*` → `null` (attribute omitted)
+ * - the Java-style separator is converted: `en_US` → `en-US`, `cs_CZ` → `cs-CZ`
+ *
+ * @param language the raw metadata language value
+ * @returns a BCP 47 language tag, or `null` when no valid tag applies
+ */
+export function metadataLangToBcp47(language: string | null | undefined): string | null {
+  if (!language || language === '*') {
+    return null;
+  }
+  return language.replace(/_/g, '-');
+}