String simplification (accent removal) using Unicode decomposition

src/blackmisc/stringutils.cpp

@@ -243,6 +243,26 @@ namespace BlackMisc
         return output;
     }
 
+    QString simplifyByDecomposition(const QString &s)
+    {
+        QString result;
+        for (const QChar c : s)
+        {
+            if (c.decompositionTag() == QChar::NoDecomposition)
+            {
+                result.push_back(c);
+            }
+            else
+            {
+                for (const QChar dc : c.decomposition())
+                {
+                    if (!dc.isMark()) { result.push_back(dc); }
+                }
+            }
+        }
+        return result;
+    }
+
     bool caseInsensitiveStringCompare(const QString &c1, const QString &c2)
     {
         return c1.length() == c2.length() && c1.startsWith(c2, Qt::CaseInsensitive);

src/blackmisc/stringutils.h

@@ -256,6 +256,9 @@ namespace BlackMisc
     //! Remove accents / diacritic marks from a string
     BLACKMISC_EXPORT QString simplifyAccents(const QString &candidate);
 
+    //! Remove accents / diacritic marks from a string by doing a Unicode decomposition and removing mark characters
+    BLACKMISC_EXPORT QString simplifyByDecomposition(const QString &candidate);
+
     //! Case insensitive string compare
     BLACKMISC_EXPORT bool caseInsensitiveStringCompare(const QString &c1, const QString &c2);
 

tests/blackmisc/teststringutils/teststringutils.cpp

@@ -36,6 +36,7 @@ namespace BlackMiscTest
         void testSplit();
         void testTimestampParsing();
         void testCodecs();
+        void testSimplify();
     };
 
     void CTestStringUtils::testRemove()
@@ -157,6 +158,14 @@ namespace BlackMiscTest
         QVERIFY2(okRu1, "Russian \"test\" equal after round-trip with cp1251");
         QVERIFY2(okRu2, "Russian \"test\" equal after round-trip with utf8");
     }
+
+    void CTestStringUtils::testSimplify()
+    {
+        const QString input = QString::fromUtf8(u8"ŠŽšžŸÀÁÂÃÄÅÇÈÉÊËÌÍÎÏÑÒÓÔÕÖÙÚÛÜÝàáâãäåçèéêëìíîïñòóôõöùúûüýÿ");
+        const QString output = QLatin1String("SZszYAAAAAACEEEEIIIINOOOOOUUUUYaaaaaaceeeeiiiinooooouuuuyy");
+        QCOMPARE(simplifyAccents(input), output);
+        QCOMPARE(simplifyByDecomposition(input), output);
+    }
 }
 
 //! main