Add normalize string

agent/CMakeLists.txt

@@ -33,7 +33,11 @@ target_sources(akonadi_indexing_agent PRIVATE
     scheduler.h
     collectionindexingjob.h
     index.h
-    collectionupdatejob.h)
+    collectionupdatejob.h
+
+    stringutil.h
+    stringutil.cpp
+)
 
 ecm_qt_declare_logging_category(akonadi_indexing_agent HEADER akonadi_indexer_agent_debug.h IDENTIFIER AKONADI_INDEXER_AGENT_LOG CATEGORY_NAME org.kde.pim.akonadi_indexer_agent
         DESCRIPTION "akonadisearch (akonadi indexer agent)"

agent/stringutil.cpp

+/*
+ * SPDX-FileCopyrightText: 2022 Laurent Montel <montel@kde.org>
+ *
+ * SPDX-License-Identifier: GPL-2.0-only OR GPL-3.0-only OR LicenseRef-KDE-Accepted-GPL
+ *
+ */
+
+#include "stringutil.h"
+// code from kitinerary/src/lib/stringutil.cpp
+QString StringUtil::normalize(QStringView str)
+{
+    QString out;
+    out.reserve(str.size());
+    for (const auto c : str) {
+        // case folding
+        const auto n = c.toCaseFolded();
+
+        // if the character has a canonical decomposition use that and skip the
+        // combining diacritic markers following it
+        // see https://en.wikipedia.org/wiki/Unicode_equivalence
+        // see https://en.wikipedia.org/wiki/Combining_character
+        if (n.decompositionTag() == QChar::Canonical) {
+            out.push_back(n.decomposition().at(0));
+        }
+        // handle compatibility compositions such as ligatures
+        // see https://en.wikipedia.org/wiki/Unicode_compatibility_characters
+        else if (n.decompositionTag() == QChar::Compat && n.isLetter() && n.script() == QChar::Script_Latin) {
+            out.append(n.decomposition());
+        } else {
+            out.push_back(n);
+        }
+    }
+    return out;
+}

agent/stringutil.h

+/*
+ * SPDX-FileCopyrightText: 2022 Laurent Montel <montel@kde.org>
+ *
+ * SPDX-License-Identifier: GPL-2.0-only OR GPL-3.0-only OR LicenseRef-KDE-Accepted-GPL
+ *
+ */
+
+#pragma once
+#include <QString>
+
+namespace StringUtil
+{
+Q_REQUIRED_RESULT QString normalize(QStringView str);
+};