Fix handling of "ambiguous ampersands" to report error when expected

.github/dependabot.yml

@@ -0,0 +1,10 @@
+version: 2
+updates:
+  - package-ecosystem: "github-actions"
+    directory: "/"
+    schedule:
+      interval: "weekly"
+  - package-ecosystem: "maven"
+    directory: "/"
+    schedule:
+      interval: "weekly"

.github/workflows/build.yml

@@ -1,25 +1,30 @@
 name: Build
 
-on: [push, pull_request]
+on:
+  push:
+    branches:
+      - main
+  pull_request:
 
 jobs:
   build:
     runs-on: ${{ matrix.os }}
     strategy:
       matrix:
-        java: [11, 14, 15-ea, 16-ea]
+        java: [17, 11.0.3, 18-ea]
         os: [ubuntu-latest, macos-latest, windows-latest]
     name: Java ${{ matrix.java }}
     steps:
-      - uses: actions/checkout@v2
+      - uses: actions/checkout@v3
         with:
           submodules: true
       - name: Set up java
-        uses: actions/setup-java@v1
+        uses: actions/setup-java@v3
         with:
+          distribution: zulu
           java-version: ${{ matrix.java }}
       - name: Cache Maven packages
-        uses: actions/cache@v2
+        uses: actions/cache@v3
         with:
           path: ~/.m2
           key: ${{ runner.os }}-m2-${{ hashFiles('**/pom.xml') }}

pom.xml

@@ -132,7 +132,6 @@
         <artifactId>maven-surefire-plugin</artifactId>
         <configuration>
           <test>Html5libTest</test>
-          <testFailureIgnore>true</testFailureIgnore> <!-- FIXME: Remove this testFailureIgnore after we have all tests passing -->
           <additionalClasspathElements>
             <additionalClasspathElement>${project.build.testSourceDirectory}/test/resources</additionalClasspathElement>
           </additionalClasspathElements>
@@ -144,7 +143,7 @@
     <dependency>
       <groupId>com.ibm.icu</groupId>
       <artifactId>icu4j</artifactId>
-      <version>4.0.1</version>
+      <version>71.1</version>
       <scope>compile</scope>
       <optional>true</optional>
     </dependency>

src/nu/validator/htmlparser/impl/ElementName.java

@@ -557,8 +557,6 @@ public void destructor() {
 //                return "ANNOTATION_XML";
 //            case TreeBuilder.FOREIGNOBJECT_OR_DESC:
 //                return "FOREIGNOBJECT_OR_DESC";
-//            case TreeBuilder.MENUITEM:
-//                return "MENUITEM";
 //        }
 //        return null;
 //    }
@@ -1143,10 +1141,6 @@ public void destructor() {
 // CPPONLY: NS_NewHTMLFormElement,
 // CPPONLY: NS_NewSVGUnknownElement, 
 TreeBuilder.FORM | SPECIAL);
-public static final ElementName MENUITEM = new ElementName("menuitem", "menuitem", 
-// CPPONLY: NS_NewHTMLMenuItemElement,
-// CPPONLY: NS_NewSVGUnknownElement, 
-TreeBuilder.MENUITEM);
 public static final ElementName PARAM = new ElementName("param", "param", 
 // CPPONLY: NS_NewHTMLSharedElement,
 // CPPONLY: NS_NewSVGUnknownElement, 
@@ -1172,7 +1166,7 @@ public void destructor() {
 // CPPONLY: NS_NewSVGUnknownElement, 
 TreeBuilder.MI_MO_MN_MS_MTEXT | SCOPING_AS_MATHML);
 public static final ElementName KEYGEN = new ElementName("keygen", "keygen", 
-// CPPONLY: NS_NewHTMLElement,
+// CPPONLY: NS_NewHTMLUnknownElement,
 // CPPONLY: NS_NewSVGUnknownElement, 
 TreeBuilder.KEYGEN | SPECIAL);
 public static final ElementName MAIN = new ElementName("main", "main", 
@@ -1484,7 +1478,7 @@ public void destructor() {
 // CPPONLY: NS_NewSVGUnknownElement, 
 TreeBuilder.TBODY_OR_THEAD_OR_TFOOT | SPECIAL | FOSTER_PARENTING | OPTIONAL_END_TAG);
 private final static @NoLength ElementName[] ELEMENT_NAMES = {
-FIGCAPTION,
+MN,
 CITE,
 FRAMESET,
 H1,
@@ -1495,7 +1489,7 @@ public void destructor() {
 BGSOUND,
 SOURCE,
 HTML,
-RP,
+OPTGROUP,
 NOFRAMES,
 MTEXT,
 VIEW,
@@ -1507,8 +1501,8 @@ public void destructor() {
 GLYPHREF,
 LI,
 ACRONYM,
-SECTION,
-HR,
+TSPAN,
+FEFUNCR,
 CANVAS,
 BASEFONT,
 FEDISTANTLIGHT,
@@ -1530,11 +1524,11 @@ public void destructor() {
 PATH,
 MALIGNMARK,
 SMALL,
-PARAM,
-OPTION,
-VIDEO,
-BR,
-FOOTER,
+ANIMATEMOTION,
+POLYGON,
+COLGROUP,
+ABBR,
+FEGAUSSIANBLUR,
 TR,
 DETAILS,
 DT,
@@ -1578,15 +1572,15 @@ public void destructor() {
 LABEL,
 ALTGLYPHITEM,
 FORM,
-BUTTON,
-KEYGEN,
-PATTERN,
-AUDIO,
-FEDISPLACEMENTMAP,
-SAMP,
-ANIMATECOLOR,
-FECOMPONENTTRANSFER,
-HEADER,
+CAPTION,
+MAIN,
+SPAN,
+MO,
+HGROUP,
+STOP,
+CENTER,
+FILTER,
+MARKER,
 NOBR,
 ADDRESS,
 DEFS,
@@ -1672,28 +1666,27 @@ public void destructor() {
 SYMBOL,
 ANIMATETRANSFORM,
 EM,
-MENUITEM,
-ANIMATEMOTION,
-CAPTION,
-MN,
-MAIN,
-POLYGON,
-SPAN,
-TSPAN,
-MO,
-COLGROUP,
-HGROUP,
-OPTGROUP,
-STOP,
-ABBR,
-CENTER,
-FEFUNCR,
-FILTER,
-FEGAUSSIANBLUR,
-MARKER,
+PARAM,
+BUTTON,
+FIGCAPTION,
+KEYGEN,
+OPTION,
+PATTERN,
+SECTION,
+AUDIO,
+VIDEO,
+FEDISPLACEMENTMAP,
+RP,
+SAMP,
+BR,
+ANIMATECOLOR,
+HR,
+FECOMPONENTTRANSFER,
+FOOTER,
+HEADER,
 };
 private final static int[] ELEMENT_HASHES = {
-1900845386,
+1902641154,
 1748359220,
 2001349720,
 876609538,
@@ -1704,7 +1697,7 @@ public void destructor() {
 1730965751,
 1756474198,
 1868312196,
-1938817026,
+1939219752,
 1988763672,
 2005324101,
 2060065124,
@@ -1716,8 +1709,8 @@ public void destructor() {
 1766992520,
 1818230786,
 1881613047,
-1907661127,
-1967128578,
+1907959605,
+1967760215,
 1982935782,
 1999397992,
 2001392798,
@@ -1739,11 +1732,11 @@ public void destructor() {
 1805502724,
 1854228698,
 1874053333,
-1889085973,
-1905563974,
-1925844629,
-1963982850,
-1967795958,
+1898223949,
+1906087319,
+1932928296,
+1965115924,
+1968053806,
 1973420034,
 1983633431,
 1998585858,
@@ -1787,15 +1780,15 @@ public void destructor() {
 1870268949,
 1881288348,
 1884120164,
-1898753862,
-1903302038,
-1906135367,
-1914900309,
-1934172497,
-1941178676,
-1965334268,
-1967788867,
-1968836118,
+1899272519,
+1904412884,
+1907435316,
+1919418370,
+1935549734,
+1941221172,
+1966223078,
+1967795910,
+1971461414,
 1971938532,
 1982173479,
 1983533124,
@@ -1881,24 +1874,23 @@ public void destructor() {
 1874102998,
 1881498736,
 1881669634,
-1887579800,
-1898223949,
-1899272519,
-1902641154,
-1904412884,
-1906087319,
-1907435316,
-1907959605,
-1919418370,
-1932928296,
-1935549734,
-1939219752,
-1941221172,
-1965115924,
-1966223078,
-1967760215,
-1967795910,
-1968053806,
-1971461414,
+1889085973,
+1898753862,
+1900845386,
+1903302038,
+1905563974,
+1906135367,
+1907661127,
+1914900309,
+1925844629,
+1934172497,
+1938817026,
+1941178676,
+1963982850,
+1965334268,
+1967128578,
+1967788867,
+1967795958,
+1968836118,
 };
 }

src/nu/validator/htmlparser/impl/ErrorReportingTokenizer.java

@@ -134,6 +134,15 @@ public void setErrorProfile(HashMap<String, String> errorProfileMap) {
         this.errorProfileMap = errorProfileMap;
     }
 
+    /**
+     * Gets the errorProfile.
+     *
+     * @param errorProfile
+     */
+    @Override public HashMap getErrorProfile() {
+        return errorProfileMap;
+    }
+
     /**
      * Reports on an event based on profile selected.
      *