Make error handling generic improve math error detail

[pulseview.git] / pv / data / mathsignal.hpp

diff --git a/pv/data/mathsignal.hpp b/pv/data/mathsignal.hpp
index 6127b177e025175f1a7cf5621bb48a0ed089a2c1..67cc506849db476e96c4bfdf4b99ed8e073b0031 100644 (file)
--- a/pv/data/mathsignal.hpp
+++ b/pv/data/mathsignal.hpp
@@ -60,7 +60,6 @@ class MathSignal : public SignalBase
 {
        Q_OBJECT
        Q_PROPERTY(QString expression READ get_expression WRITE set_expression NOTIFY expression_changed)
-       Q_PROPERTY(QString error_message READ error_message)
 
 private:
        static const int64_t ChunkLength;
@@ -72,13 +71,11 @@ public:
        virtual void save_settings(QSettings &settings) const;
        virtual void restore_settings(QSettings &settings);
 
-       QString error_message() const;
-
        QString get_expression() const;
        void set_expression(QString expression);
 
 private:
-       void set_error_message(QString msg);
+       virtual void set_error_message(QString msg);
 
        /**
         * Returns the number of samples that can be worked on,
@@ -123,7 +120,7 @@ private:
        bool use_custom_sample_rate_, use_custom_sample_count_;
        map<std::string, signal_data> input_signals_;
 
-       QString expression_, error_message_;
+       QString expression_;
 
        mutable mutex input_mutex_;
        mutable condition_variable gen_input_cond_;