Merge pull request #26399 from dkurt:dk/file_storage_new_data

int64 data type in FileStorage #26399

### Pull Request Readiness Checklist

resolves #23333

Proposed approach is not perfect in terms of complexity and potential bugs. Instead of changing `INT` raw size from `4` to `8`, we check int64 value can be fitted to int32 or not.

Collections such as cv::Mat rely on data type symbol.

This PR is addressed to 5.x branch first to cover `CV_64S` Mat. Later, it can be backported to 4.x

See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request

- [x] I agree to contribute to the project under Apache 2 License.
- [x] To the best of my knowledge, the proposed patch is not based on a code under GPL or another license that is incompatible with OpenCV
- [x] The PR is proposed to the proper branch
- [x] There is a reference to the original bug report and related work
- [x] There is accuracy test, performance test and test data in opencv_extra repository, if applicable
      Patch to opencv_extra has the same branch name.
- [x] The feature is well documented and sample code can be built with the project CMake

modules/core/include/opencv2/core/persistence.hpp

@@ -365,6 +365,8 @@ public:
      */
     CV_WRAP void write(const String& name, int val);
     /// @overload
+    CV_WRAP void write(const String& name, int64_t val);
+    /// @overload
     CV_WRAP void write(const String& name, double val);
     /// @overload
     CV_WRAP void write(const String& name, const String& val);
@@ -530,6 +532,8 @@ public:
     CV_WRAP size_t rawSize() const;
     //! returns the node content as an integer. If the node stores floating-point number, it is rounded.
     operator int() const;
+    //! returns the node content as a signed 64bit integer. If the node stores floating-point number, it is rounded.
+    operator int64_t() const;
     //! returns the node content as float
     operator float() const;
     //! returns the node content as double
@@ -654,6 +658,7 @@ protected:
 /////////////////// XML & YAML I/O implementation //////////////////
 
 CV_EXPORTS void write( FileStorage& fs, const String& name, int value );
+CV_EXPORTS void write( FileStorage& fs, const String& name, int64_t value );
 CV_EXPORTS void write( FileStorage& fs, const String& name, float value );
 CV_EXPORTS void write( FileStorage& fs, const String& name, double value );
 CV_EXPORTS void write( FileStorage& fs, const String& name, const String& value );
@@ -665,11 +670,13 @@ CV_EXPORTS void write( FileStorage& fs, const String& name, const std::vector<DM
 #endif
 
 CV_EXPORTS void writeScalar( FileStorage& fs, int value );
+CV_EXPORTS void writeScalar( FileStorage& fs, int64_t value );
 CV_EXPORTS void writeScalar( FileStorage& fs, float value );
 CV_EXPORTS void writeScalar( FileStorage& fs, double value );
 CV_EXPORTS void writeScalar( FileStorage& fs, const String& value );
 
 CV_EXPORTS void read(const FileNode& node, int& value, int default_value);
+CV_EXPORTS void read(const FileNode& node, int64_t& value, int64_t default_value);
 CV_EXPORTS void read(const FileNode& node, float& value, float default_value);
 CV_EXPORTS void read(const FileNode& node, double& value, double default_value);
 CV_EXPORTS void read(const FileNode& node, std::string& value, const std::string& default_value);

modules/core/src/persistence.cpp

@@ -176,7 +176,7 @@ char* floatToString( char* buf, size_t bufSize, float value, bool halfprecision,
     return buf;
 }
 
-static const char symbols[] = "ucwsifdhHbLUn";
+static const char symbols[] = "ucwsifdhHbUIn";
 
 static char typeSymbol(int depth)
 {
@@ -354,6 +354,20 @@ static inline int readInt(const uchar* p)
 #endif
 }
 
+static inline int64_t readLong(const uchar* p)
+{
+    // On little endian CPUs, both branches produce the same result. On big endian, only the else branch does.
+#if CV_LITTLE_ENDIAN_MEM_ACCESS
+    int64_t val;
+    memcpy(&val, p, sizeof(val));
+    return val;
+#else
+    unsigned val0 = (unsigned)(p[0] | (p[1] << 8) | (p[2] << 16) | (p[3] << 24));
+    unsigned val1 = (unsigned)(p[4] | (p[5] << 8) | (p[6] << 16) | (p[7] << 24));
+    return val0 | ((int64_t)val1 << 32);
+#endif
+}
+
 static inline double readReal(const uchar* p)
 {
     // On little endian CPUs, both branches produce the same result. On big endian, only the else branch does.
@@ -370,16 +384,15 @@ static inline double readReal(const uchar* p)
 #endif
 }
 
-static inline void writeInt(uchar* p, int ival)
+template <typename T>
+static inline void writeInt(uchar* p, T ival)
 {
     // On little endian CPUs, both branches produce the same result. On big endian, only the else branch does.
 #if CV_LITTLE_ENDIAN_MEM_ACCESS
     memcpy(p, &ival, sizeof(ival));
 #else
-    p[0] = (uchar)ival;
+    for (size_t i = 0, j = 0; i < sizeof(ival); ++i, j += 8)
-    p[1] = (uchar)(ival >> 8);
+        p[i] = (uchar)(ival >> j);
-    p[2] = (uchar)(ival >> 16);
-    p[3] = (uchar)(ival >> 24);
 #endif
 }
 
@@ -1083,6 +1096,11 @@ void FileStorage::Impl::write(const String &key, int value) {
     getEmitter().write(key.c_str(), value);
 }
 
+void FileStorage::Impl::write(const String &key, int64_t value) {
+    CV_Assert(write_mode);
+    getEmitter().write(key.c_str(), value);
+}
+
 void FileStorage::Impl::write(const String &key, double value) {
     CV_Assert(write_mode);
     getEmitter().write(key.c_str(), value);
@@ -1455,7 +1473,7 @@ void FileStorage::Impl::convertToCollection(int type, FileNode &node) {
     bool named = node.isNamed();
     uchar *ptr = node.ptr() + 1 + (named ? 4 : 0);
 
-    int ival = 0;
+    int64_t ival = 0;
     double fval = 0;
     std::string sval;
     bool add_first_scalar = false;
@@ -1468,7 +1486,7 @@ void FileStorage::Impl::convertToCollection(int type, FileNode &node) {
         // otherwise we don't know where to get the element names from
         CV_Assert(type == FileNode::SEQ);
         if (node_type == FileNode::INT) {
-            ival = readInt(ptr);
+            ival = readLong(ptr);
             add_first_scalar = true;
         } else if (node_type == FileNode::REAL) {
             fval = readReal(ptr);
@@ -1830,7 +1848,7 @@ char *FileStorage::Impl::parseBase64(char *ptr, int indent, FileNode &collection
 
     int fmt_pairs[CV_FS_MAX_FMT_PAIRS * 2];
     int fmt_pair_count = fs::decodeFormat(dt, fmt_pairs, CV_FS_MAX_FMT_PAIRS);
-    int ival = 0;
+    int64_t ival = 0;
     double fval = 0;
 
     for (;;) {
@@ -2070,6 +2088,11 @@ void writeScalar( FileStorage& fs, int value )
     fs.p->write(String(), value);
 }
 
+void writeScalar( FileStorage& fs, int64_t value )
+{
+    fs.p->write(String(), value);
+}
+
 void writeScalar( FileStorage& fs, float value )
 {
     fs.p->write(String(), (double)value);
@@ -2090,6 +2113,11 @@ void write( FileStorage& fs, const String& name, int value )
     fs.p->write(name, value);
 }
 
+void write( FileStorage& fs, const String& name, int64_t value )
+{
+    fs.p->write(name, value);
+}
+
 void write( FileStorage& fs, const String& name, float value )
 {
     fs.p->write(name, (double)value);
@@ -2106,6 +2134,7 @@ void write( FileStorage& fs, const String& name, const String& value )
 }
 
 void FileStorage::write(const String& name, int val) { p->write(name, val); }
+void FileStorage::write(const String& name, int64_t val) { p->write(name, val); }
 void FileStorage::write(const String& name, double val) { p->write(name, val); }
 void FileStorage::write(const String& name, const String& val) { p->write(name, val); }
 void FileStorage::write(const String& name, const Mat& val) { cv::write(*this, name, val); }
@@ -2326,6 +2355,27 @@ FileNode::operator int() const
         return 0x7fffffff;
 }
 
+FileNode::operator int64_t() const
+{
+    const uchar* p = ptr();
+    if(!p)
+        return 0;
+    int tag = *p;
+    int type = (tag & TYPE_MASK);
+    p += (tag & NAMED) ? 5 : 1;
+
+    if( type == INT )
+    {
+        return readLong(p);
+    }
+    else if( type == REAL )
+    {
+        return cvRound(readReal(p));
+    }
+    else
+        return 0x7fffffff;
+}
+
 FileNode::operator float() const
 {
     const uchar* p = ptr();
@@ -2450,7 +2500,13 @@ void FileNode::setValue( int type, const void* value, int len )
         sz += 4;
 
     if( type == INT )
+    {
+        int64_t ival = *(const int64_t*)value;
+        if (ival > INT_MAX || ival < INT_MIN)
+            sz += 8;
+        else
             sz += 4;
+    }
     else if( type == REAL )
         sz += 8;
     else if( type == STRING )
@@ -2470,8 +2526,11 @@ void FileNode::setValue( int type, const void* value, int len )
 
     if( type == INT )
     {
-        int ival = *(const int*)value;
+        int64_t ival = *(const int64_t*)value;
+        if (sz > 8)
             writeInt(p, ival);
+        else
+            writeInt(p, static_cast<int>(ival));
     }
     else if( type == REAL )
     {
@@ -2622,12 +2681,12 @@ FileNodeIterator& FileNodeIterator::readRaw( const String& fmt, void* _data0, si
                 offset = alignSize( offset, elem_size );
                 uchar* data = data0 + offset;
 
-                for( int i = 0; i < count; i++, ++(*this) )
+                for( int i = 0; i < count; i++ )
                 {
                     FileNode node = *(*this);
                     if( node.isInt() )
                     {
-                        int ival = (int)node;
+                        int64_t ival = static_cast<int64_t>(elem_size == 8 ? (int64_t)node : (int)node);
                         switch( elem_type )
                         {
                         case CV_8U:
@@ -2651,11 +2710,11 @@ FileNodeIterator& FileNodeIterator::readRaw( const String& fmt, void* _data0, si
                             data += sizeof(short);
                             break;
                         case CV_32U:
-                            *(unsigned*)data = (unsigned)std::max(ival, 0);
+                            *(unsigned*)data = (unsigned)std::max(ival, (int64_t)0);
                             data += sizeof(unsigned);
                             break;
                         case CV_32S:
-                            *(int*)data = ival;
+                            *(int*)data = (int)ival;
                             data += sizeof(int);
                             break;
                         case CV_32F:
@@ -2746,6 +2805,11 @@ FileNodeIterator& FileNodeIterator::readRaw( const String& fmt, void* _data0, si
                     }
                     else
                         CV_Error( Error::StsError, "readRawData can only be used to read plain sequences of numbers" );
+                    ++(*this);
+                    if (elem_type == CV_64S)
+                    {
+                        ofs += 4;
+                    }
                 }
                 offset = (int)(data - data0);
             }
@@ -2785,6 +2849,15 @@ void read(const FileNode& node, int& val, int default_val)
     }
 }
 
+void read(const FileNode& node, int64_t& val, int64_t default_val)
+{
+    val = default_val;
+    if( !node.empty() )
+    {
+        val = (int64_t)node;
+    }
+}
+
 void read(const FileNode& node, double& val, double default_val)
 {
     val = default_val;

modules/core/src/persistence.hpp

@@ -194,6 +194,7 @@ public:
                                           int struct_flags, const char* type_name=0 ) = 0;
     virtual void endWriteStruct(const FStructData& current_struct) = 0;
     virtual void write(const char* key, int value) = 0;
+    virtual void write(const char* key, int64_t value) = 0;
     virtual void write(const char* key, double value) = 0;
     virtual void write(const char* key, const char* value, bool quote) = 0;
     virtual void writeScalar(const char* key, const char* value) = 0;

modules/core/src/persistence_impl.hpp

@@ -69,6 +69,8 @@ public:
 
     void write( const String& key, int value );
 
+    void write( const String& key, int64_t value );
+
     void write( const String& key, double value );
 
     void write( const String& key, const String& value );

modules/core/src/persistence_json.cpp

@@ -81,6 +81,12 @@ public:
         writeScalar( key, fs::itoa( value, buf, 10 ));
     }
 
+    void write(const char* key, int64_t value)
+    {
+        char buf[128];
+        writeScalar( key, fs::itoa( value, buf, 10, true ));
+    }
+
     void write( const char* key, double value )
     {
         char buf[128];
@@ -596,7 +602,7 @@ public:
             }
             else
             {
-                int ival = (int)strtol( beg, &ptr, 0 );
+                int64_t ival = strtoll( beg, &ptr, 0 );
                 CV_PERSISTENCE_CHECK_END_OF_BUFFER_BUG_CPP();
 
                 node.setValue(FileNode::INT, &ival);
@@ -623,7 +629,7 @@ public:
             else if( (len == 4 && memcmp( beg, "true", 4 ) == 0) ||
                      (len == 5 && memcmp( beg, "false", 5 ) == 0) )
             {
-                int ival = *beg == 't' ? 1 : 0;
+                int64_t ival = *beg == 't' ? 1 : 0;
                 node.setValue(FileNode::INT, &ival);
             }
             else

modules/core/src/persistence_xml.cpp

@@ -145,6 +145,12 @@ public:
         writeScalar( key, ptr);
     }
 
+    void write(const char* key, int64_t value)
+    {
+        char buf[128], *ptr = fs::itoa( value, buf, 10, true );
+        writeScalar( key, ptr);
+    }
+
     void write( const char* key, double value )
     {
         char buf[128];
@@ -556,7 +562,7 @@ public:
                     }
                     else
                     {
-                        int ival = (int)strtol( ptr, &endptr, 0 );
+                        int64_t ival = strtoll( ptr, &endptr, 0 );
                         elem->setValue(FileNode::INT, &ival);
                     }
 

modules/core/src/persistence_yml.cpp

@@ -107,6 +107,12 @@ public:
         writeScalar( key, fs::itoa( value, buf, 10 ));
     }
 
+    void write(const char* key, int64_t value)
+    {
+        char buf[128];
+        writeScalar( key, fs::itoa( value, buf, 10, true ));
+    }
+
     void write( const char* key, double value )
     {
         char buf[128];
@@ -567,7 +573,7 @@ public:
             else
             {
             force_int:
-                int ival = (int)strtol( ptr, &endptr, 0 );
+                int64_t ival = strtoll( ptr, &endptr, 0 );
                 node.setValue(FileNode::INT, &ival);
             }
 

modules/core/test/test_io.cpp

@@ -2066,6 +2066,27 @@ TEST_P(FileStorage_exact_type, mat_1d)
     testExactMat(Mat({1}, CV_32S, Scalar(8)), GetParam());
 }
 
+TEST_P(FileStorage_exact_type, long_int)
+{
+    for (const int64_t expected : std::vector<int64_t>{INT64_MAX, INT64_MIN, -1, 1, 0})
+    {
+        int64_t value = fsWriteRead(expected, GetParam());
+        EXPECT_EQ(value, expected);
+    }
+}
+
+TEST_P(FileStorage_exact_type, long_int_mat)
+{
+    Mat src(2, 4, CV_64SC(3));
+    int64_t* data = src.ptr<int64_t>();
+    for (size_t i = 0; i < src.total() * src.channels(); ++i)
+    {
+        data[i] = INT64_MAX - static_cast<int64_t>(std::rand());
+    }
+    Mat dst = fsWriteRead(src, GetParam());
+    EXPECT_EQ(cv::norm(src, dst, NORM_INF), 0.0);
+}
+
 INSTANTIATE_TEST_CASE_P(Core_InputOutput,
     FileStorage_exact_type, Values(".yml", ".xml", ".json")
 );

modules/python/src2/typing_stubs_generation/predefined_types.py

@@ -27,6 +27,7 @@ _PREDEFINED_TYPES = (
     PrimitiveTypeNode.int_("int32_t"),
     PrimitiveTypeNode.int_("uint32_t"),
     PrimitiveTypeNode.int_("size_t"),
+    PrimitiveTypeNode.int_("int64_t"),
     PrimitiveTypeNode.float_("float"),
     PrimitiveTypeNode.float_("double"),
     PrimitiveTypeNode.bool_("bool"),