Qt应用中UTF-7解码生成错误字符问题排查求助

开发基于Qt的文本编辑器时，解码UTF-7编码文本遇到字符错误：对UTF-7字符串Hello, +Z1TA-（对应中文Hello, 世界）解码，结果得到Hello, ᙎ䱵。已实现InterpreteAsUtf7类的解码逻辑，疑似Base64段处理或UTF-16转换环节存在问题，以下是复现代码：

#include <QString>
#include <QByteArray>
#include <QDebug>

class InterpreteAsUtf7 {
public:
    static QString decodeUtf7(const QByteArray& utf7Data);

private:
    static QString processBase64Segment(const QByteArray& base64Buffer, bool& errorFlag);
    static QString convertToUtf16LE(const QByteArray& decodedBytes, bool& errorFlag);
};

QString InterpreteAsUtf7::decodeUtf7(const QByteArray& utf7Data) {
    QString result;
    QByteArray base64Buffer;
    bool inBase64 = false;
    bool decodingError = false;

    for (char c : utf7Data) {
        if (c == '+') {
            if (inBase64) {
                result += "+" + base64Buffer;  // Unclosed Base64 section treated as literal
                base64Buffer.clear();
            }
            inBase64 = true;
            continue;
        }

        if (c == '-' && inBase64) {
            result += processBase64Segment(base64Buffer, decodingError);
            base64Buffer.clear();
            inBase64 = false;
            continue;
        }

        if (inBase64) {
            base64Buffer.append(c);
        } else {
            result += QChar(c);  // Literal character
        }
    }

    if (inBase64) {
        result += "+" + base64Buffer;  // Unclosed Base64 treated as literal
    }

    return result;
}

QString InterpreteAsUtf7::processBase64Segment(const QByteArray& base64Buffer, bool& errorFlag) {
    QByteArray decodedBytes = QByteArray::fromBase64(base64Buffer);
    if (decodedBytes.isEmpty()) {
        errorFlag = true;
        qWarning() << "[WARNING] Base64 decoding failed for:" << base64Buffer;
        return QString();
    }
    return convertToUtf16LE(decodedBytes, errorFlag);
}

QString InterpreteAsUtf7::convertToUtf16LE(const QByteArray& decodedBytes, bool& errorFlag) {
    if (decodedBytes.size() % 2 != 0) {
        errorFlag = true;
        qWarning() << "[WARNING] Decoded bytes not aligned for UTF-16:" << decodedBytes;
        return QString();
    }

    QString result;
    for (int i = 0; i < decodedBytes.size(); i += 2) {
        ushort codeUnit = static_cast<uchar>(decodedBytes[i]) |
                          (static_cast<uchar>(decodedBytes[i + 1]) << 8);
        result.append(QChar(codeUnit));
    }
    return result;
}

int main() {
    QByteArray utf7Data = "Hello, +Z1TA-";  // UTF-7 encoded string for "Hello, 世界"
    QString decodedText = InterpreteAsUtf7::decodeUtf7(utf7Data);

    qDebug() << "Original UTF-7:" << utf7Data;
    qDebug() << "Decoded UTF-7:" << decodedText;

    return 0;
}

错误原因分析 #

1. UTF-16字节顺序错误：UTF-7标准中，Base64解码后的字节是**UTF-16大端（BE）**编码，但代码中按小端（LE）拼接字节，导致字符编码反转。
2. Base64格式不匹配：UTF-7使用修改版Base64，将标准Base64的/替换为,，且不需要填充字符=。直接调用QByteArray::fromBase64会因格式不兼容导致解码错误。

修正后的代码 #

修改processBase64Segment和convertToUtf16LE函数（重命名为convertToUtf16BE更准确）：

QString InterpreteAsUtf7::processBase64Segment(const QByteArray& base64Buffer, bool& errorFlag) {
    // 替换UTF-7修改版Base64的字符为标准Base64格式
    QByteArray adjustedBase64 = base64Buffer;
    adjustedBase64.replace(',', '/');
    // UTF-7不需要填充，手动补充必要的'='使长度为4的倍数
    int padding = adjustedBase64.size() % 4;
    if (padding != 0) {
        adjustedBase64.append(4 - padding, '=');
    }

    QByteArray decodedBytes = QByteArray::fromBase64(adjustedBase64);
    if (decodedBytes.isEmpty()) {
        errorFlag = true;
        qWarning() << "[WARNING] Base64 decoding failed for:" << base64Buffer;
        return QString();
    }
    return convertToUtf16BE(decodedBytes, errorFlag);
}

QString InterpreteAsUtf7::convertToUtf16BE(const QByteArray& decodedBytes, bool& errorFlag) {
    if (decodedBytes.size() % 2 != 0) {
        errorFlag = true;
        qWarning() << "[WARNING] Decoded bytes not aligned for UTF-16:" << decodedBytes;
        return QString();
    }

    QString result;
    for (int i = 0; i < decodedBytes.size(); i += 2) {
        // 按大端顺序拼接字节：高字节在前，低字节在后
        ushort codeUnit = (static_cast<uchar>(decodedBytes[i]) << 8) |
                          static_cast<uchar>(decodedBytes[i + 1]);
        result.append(QChar(codeUnit));
    }
    return result;
}

修正后效果 #

运行修正后的代码，Hello, +Z1TA-将正确解码为Hello, 世界。

内容的提问来源于stack exchange，提问作者Remisa Phillips