Added support for reading multilinestring wkts

2024-11-21 15:12:02 +00:00 · 2024-08-01 18:23:11 -06:00 · 2024-08-01 18:23:11 -06:00 · a80c7c080c
commit a80c7c080c
parent f9179986d1
13 changed files with 227 additions and 5 deletions
--- a/docs/en/sql-reference/data-types/geo.md
+++ b/docs/en/sql-reference/data-types/geo.md
@ -52,6 +52,48 @@ Result:
 └───────────────────────────────┴───────────────┘
 ```

+## LineString
+
+`LineString` is a line stored as an array of points: [Array](array.md)([Point](#point)).
+
+**Example**
+
+Query:
+
+```sql
+CREATE TABLE geo_linestring (l LineString) ENGINE = Memory();
+INSERT INTO geo_linestring VALUES([(0, 0), (10, 0), (10, 10), (0, 10)]);
+SELECT l, toTypeName(l) FROM geo_linestring;
+```
+Result:
+
+``` text
+┌─r─────────────────────────────┬─toTypeName(r)─┐
+│ [(0,0),(10,0),(10,10),(0,10)] │ LineString    │
+└───────────────────────────────┴───────────────┘
+```
+
+## MultiLineString
+
+`MultiLineString` is multiple lines stored as an array of `LineString`: [Array](array.md)([LineString](#linestring)).
+
+**Example**
+
+Query:
+
+```sql
+CREATE TABLE geo_multilinestring (l MultiLineString) ENGINE = Memory();
+INSERT INTO geo_multilinestring VALUES([[(0, 0), (10, 0), (10, 10), (0, 10)], [(1, 1), (2, 2), (3, 3)]]);
+SELECT l, toTypeName(l) FROM geo_multilinestring;
+```
+Result:
+
+``` text
+┌─l───────────────────────────────────────────────────┬─toTypeName(l)───┐
+│ [[(0,0),(10,0),(10,10),(0,10)],[(1,1),(2,2),(3,3)]] │ MultiLineString │
+└─────────────────────────────────────────────────────┴─────────────────┘
+```
+
 ## Polygon

 `Polygon` is a polygon with holes stored as an array of rings: [Array](array.md)([Ring](#ring)). First element of outer array is the outer shape of polygon and all the following elements are holes.
--- a/docs/en/sql-reference/functions/geo/polygon.md
+++ b/docs/en/sql-reference/functions/geo/polygon.md
@ -6,11 +6,13 @@ title: "Functions for Working with Polygons"

 ## WKT

-Returns a WKT (Well Known Text) geometric object from various [Geo Data Types](../../data-types/geo.md). Supported WKT objects are: 
+Returns a WKT (Well Known Text) geometric object from various [Geo Data Types](../../data-types/geo.md). Supported WKT objects are:

 - POINT
 - POLYGON
 - MULTIPOLYGON
+- LINESTRING
+- MULTILINESTRING

 **Syntax**

@ -26,12 +28,16 @@ WKT(geo_data)
 - [Ring](../../data-types/geo.md#ring)
 - [Polygon](../../data-types/geo.md#polygon)
 - [MultiPolygon](../../data-types/geo.md#multipolygon)
+- [LineString](../../data-types/geo.md#linestring)
+- [MultiLineString](../../data-types/geo.md#multilinestring)

 **Returned value**

 - WKT geometric object `POINT` is returned for a Point.
 - WKT geometric object `POLYGON` is returned for a Polygon
- WKT geometric object `MULTIPOLYGON` is returned for a MultiPolygon. 
+- WKT geometric object `MULTIPOLYGON` is returned for a MultiPolygon.
+- WKT geometric object `LINESTRING` is returned for a LineString.
+- WKT geometric object `MULTILINESTRING` is returned for a MultiLineString.

 **Examples**

@ -84,7 +90,7 @@ SELECT

 ### Input parameters

-String starting with `MULTIPOLYGON` 
+String starting with `MULTIPOLYGON`

 ### Returned value

@ -170,6 +176,34 @@ SELECT readWKTLineString('LINESTRING (1 1, 2 2, 3 3, 1 1)');
 [(1,1),(2,2),(3,3),(1,1)]
 ```

+## readWKTMultiLineString
+
+Parses a Well-Known Text (WKT) representation of a MultiLineString geometry and returns it in the internal ClickHouse format.
+
+### Syntax
+
+```sql
+readWKTMultiLineString(wkt_string)
+```
+
+### Arguments
+
+- `wkt_string`: The input WKT string representing a MultiLineString geometry.
+
+### Returned value
+
+The function returns a ClickHouse internal representation of the multilinestring geometry.
+
+### Example
+
+```sql
+SELECT wkt(readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3), (4 4, 5 5, 6 6))'));
+```
+
+```response
+[[(1,1),(2,2),(3,3)],[(4,4),(5,5),(6,6)]]
+```
+
 ## readWKTRing

 Parses a Well-Known Text (WKT) representation of a Polygon geometry and returns a ring (closed linestring) in the internal ClickHouse format.
@ -219,7 +253,7 @@ UInt8, 0 for false, 1 for true

 ## polygonsDistanceSpherical

-Calculates the minimal distance between two points where one point belongs to the first polygon and the second to another polygon. Spherical means that coordinates are interpreted as coordinates on a pure and ideal sphere, which is not true for the Earth. Using this type of coordinate system speeds up execution, but of course is not precise. 
+Calculates the minimal distance between two points where one point belongs to the first polygon and the second to another polygon. Spherical means that coordinates are interpreted as coordinates on a pure and ideal sphere, which is not true for the Earth. Using this type of coordinate system speeds up execution, but of course is not precise.

 ### Example

--- a/src/DataTypes/DataTypeCustomGeo.cpp
+++ b/src/DataTypes/DataTypeCustomGeo.cpp
@ -24,6 +24,13 @@ void registerDataTypeDomainGeo(DataTypeFactory & factory)
            std::make_unique<DataTypeCustomDesc>(std::make_unique<DataTypeLineStringName>()));
    });

+    // Custom type for mulitple lines stored as Array(LineString)
+    factory.registerSimpleDataTypeCustom("MultiLineString", []
+    {
+        return std::make_pair(DataTypeFactory::instance().get("Array(LineString)"),
+            std::make_unique<DataTypeCustomDesc>(std::make_unique<DataTypeMultiLineStringName>()));
+    });
+
    // Custom type for simple polygon without holes stored as Array(Point)
    factory.registerSimpleDataTypeCustom("Ring", []
    {
--- a/src/DataTypes/DataTypeCustomGeo.h
+++ b/src/DataTypes/DataTypeCustomGeo.h
@ -17,6 +17,12 @@ public:
    DataTypeLineStringName() : DataTypeCustomFixedName("LineString") {}
 };

+class DataTypeMultiLineStringName : public DataTypeCustomFixedName
+{
+public:
+    DataTypeMultiLineStringName() : DataTypeCustomFixedName("MultiLineString") {}
+};
+
 class DataTypeRingName : public DataTypeCustomFixedName
 {
 public:
--- a/src/Functions/geometryConverters.h
+++ b/src/Functions/geometryConverters.h
@ -31,6 +31,9 @@ namespace ErrorCodes
 template <typename Point>
 using LineString = boost::geometry::model::linestring<Point>;

+template <typename Point>
+using MultiLineString = boost::geometry::model::multi_linestring<LineString<Point>>;
+
 template <typename Point>
 using Ring = boost::geometry::model::ring<Point>;

@ -42,12 +45,14 @@ using MultiPolygon = boost::geometry::model::multi_polygon<Polygon<Point>>;

 using CartesianPoint = boost::geometry::model::d2::point_xy<Float64>;
 using CartesianLineString = LineString<CartesianPoint>;
+using CartesianMultiLineString = MultiLineString<CartesianPoint>;
 using CartesianRing = Ring<CartesianPoint>;
 using CartesianPolygon = Polygon<CartesianPoint>;
 using CartesianMultiPolygon = MultiPolygon<CartesianPoint>;

 using SphericalPoint = boost::geometry::model::point<Float64, 2, boost::geometry::cs::spherical_equatorial<boost::geometry::degree>>;
 using SphericalLineString = LineString<SphericalPoint>;
+using SphericalMultiLineString = MultiLineString<SphericalPoint>;
 using SphericalRing = Ring<SphericalPoint>;
 using SphericalPolygon = Polygon<SphericalPoint>;
 using SphericalMultiPolygon = MultiPolygon<SphericalPoint>;
@ -113,6 +118,28 @@ struct ColumnToLineStringsConverter
    }
 };

+/**
+ * Class which converts Column with type Array(Array(Tuple(Float64, Float64))) to a vector of boost multi_linestring type.
+*/
+template <typename Point>
+struct ColumnToMultiLineStringsConverter
+{
+    static std::vector<MultiLineString<Point>> convert(ColumnPtr col)
+    {
+        const IColumn::Offsets & offsets = typeid_cast<const ColumnArray &>(*col).getOffsets();
+        size_t prev_offset = 0;
+        std::vector<MultiLineString<Point>> answer(offsets.size());
+        auto all_linestrings = ColumnToLineStringsConverter<Point>::convert(typeid_cast<const ColumnArray &>(*col).getDataPtr());
+        for (size_t iter = 0; iter < offsets.size() && iter < all_linestrings.size(); ++iter)
+        {
+            for (size_t linestring_iter = prev_offset; linestring_iter < offsets[iter]; ++linestring_iter)
+                answer[iter].emplace_back(std::move(all_linestrings[linestring_iter]));
+            prev_offset = offsets[iter];
+        }
+        return answer;
+    }
+};
+
 /**
 * Class which converts Column with type Array(Tuple(Float64, Float64)) to a vector of boost ring type.
 */
@ -268,6 +295,38 @@ private:
    ColumnUInt64::MutablePtr offsets;
 };

+/// Serialize Point, MultiLineString as MultiLineString
+template <typename Point>
+class MultiLineStringSerializer
+{
+public:
+    MultiLineStringSerializer()
+        : offsets(ColumnUInt64::create())
+    {}
+
+    explicit MultiLineStringSerializer(size_t n)
+        : offsets(ColumnUInt64::create(n))
+    {}
+
+    void add(const MultiLineString<Point> & multilinestring)
+    {
+        size += multilinestring.size();
+        offsets->insertValue(size);
+        for (const auto & linestring : multilinestring)
+            linestring_serializer.add(linestring);
+    }
+
+    ColumnPtr finalize()
+    {
+        return ColumnArray::create(linestring_serializer.finalize(), std::move(offsets));
+    }
+
+private:
+    size_t size = 0;
+    LineStringSerializer<Point> linestring_serializer;
+    ColumnUInt64::MutablePtr offsets;
+};
+
 /// Almost the same as LineStringSerializer
 /// Serialize Point, Ring as Ring
 template <typename Point>
@ -411,6 +470,11 @@ static void callOnGeometryDataType(DataTypePtr type, F && f)
    else if (factory.get("LineString")->equals(*type) && type->getCustomName() && type->getCustomName()->getName() == "LineString")
        return f(ConverterType<ColumnToLineStringsConverter<Point>>());

+    /// We should take the name into consideration to avoid ambiguity.
+    /// Because for example both MultiLineString and Polygon are resolved to Array(Tuple(Point)).
+    else if (factory.get("MultiLineString")->equals(*type) && type->getCustomName() && type->getCustomName()->getName() == "MultiLineString")
+        return f(ConverterType<ColumnToMultiLineStringsConverter<Point>>());
+
    /// For backward compatibility if we call this function not on a custom type, we will consider Array(Tuple(Point)) as type Ring.
    else if (factory.get("Ring")->equals(*type))
        return f(ConverterType<ColumnToRingsConverter<Point>>());
--- a/src/Functions/polygonsIntersection.cpp
+++ b/src/Functions/polygonsIntersection.cpp
@ -75,6 +75,8 @@ public:
                throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Any argument of function {} must not be Point", getName());
            else if constexpr (std::is_same_v<ColumnToLineStringsConverter<Point>, LeftConverter> || std::is_same_v<ColumnToLineStringsConverter<Point>, RightConverter>)
                throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Any argument of function {} must not be LineString", getName());
+            else if constexpr (std::is_same_v<ColumnToMultiLineStringsConverter<Point>, LeftConverter> || std::is_same_v<ColumnToMultiLineStringsConverter<Point>, RightConverter>)
+                throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Any argument of function {} must not be MultiLineString", getName());
            else
            {
                auto first = LeftConverter::convert(arguments[0].column->convertToFullColumnIfConst());
--- a/src/Functions/polygonsSymDifference.cpp
+++ b/src/Functions/polygonsSymDifference.cpp
@ -73,6 +73,8 @@ public:
                throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Any argument of function {} must not be Point", getName());
            else if constexpr (std::is_same_v<ColumnToLineStringsConverter<Point>, LeftConverter> || std::is_same_v<ColumnToLineStringsConverter<Point>, RightConverter>)
                throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Any argument of function {} must not be LineString", getName());
+            else if constexpr (std::is_same_v<ColumnToMultiLineStringsConverter<Point>, LeftConverter> || std::is_same_v<ColumnToMultiLineStringsConverter<Point>, RightConverter>)
+                throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Any argument of function {} must not be MultiLineString", getName());
            else
            {
                auto first = LeftConverter::convert(arguments[0].column->convertToFullColumnIfConst());
--- a/src/Functions/polygonsUnion.cpp
+++ b/src/Functions/polygonsUnion.cpp
@ -73,6 +73,8 @@ public:
                throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Any argument of function {} must not be Point", getName());
            else if constexpr (std::is_same_v<ColumnToLineStringsConverter<Point>, LeftConverter> || std::is_same_v<ColumnToLineStringsConverter<Point>, RightConverter>)
                throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Any argument of function {} must not be LineString", getName());
+            else if constexpr (std::is_same_v<ColumnToMultiLineStringsConverter<Point>, LeftConverter> || std::is_same_v<ColumnToMultiLineStringsConverter<Point>, RightConverter>)
+                throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Any argument of function {} must not be MultiLineString", getName());
            else
            {
                auto first = LeftConverter::convert(arguments[0].column->convertToFullColumnIfConst());
--- a/src/Functions/polygonsWithin.cpp
+++ b/src/Functions/polygonsWithin.cpp
@ -77,6 +77,8 @@ public:
                throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Any argument of function {} must not be Point", getName());
            else if constexpr (std::is_same_v<ColumnToLineStringsConverter<Point>, LeftConverter> || std::is_same_v<ColumnToLineStringsConverter<Point>, RightConverter>)
                throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Any argument of function {} must not be LineString", getName());
+            else if constexpr (std::is_same_v<ColumnToMultiLineStringsConverter<Point>, LeftConverter> || std::is_same_v<ColumnToMultiLineStringsConverter<Point>, RightConverter>)
+                throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Any argument of function {} must not be MultiLineString", getName());
            else
            {
                auto first = LeftConverter::convert(arguments[0].column->convertToFullColumnIfConst());
--- a/src/Functions/readWkt.cpp
+++ b/src/Functions/readWkt.cpp
@ -87,6 +87,11 @@ struct ReadWKTLineStringNameHolder
    static constexpr const char * name = "readWKTLineString";
 };

+struct ReadWKTMultiLineStringNameHolder
+{
+    static constexpr const char * name = "readWKTMultiLineString";
+};
+
 struct ReadWKTRingNameHolder
 {
    static constexpr const char * name = "readWKTRing";
@ -131,6 +136,31 @@ Parses a Well-Known Text (WKT) representation of a LineString geometry and retur
        },
        .categories{"Unique identifiers"}
    });
+    factory.registerFunction<FunctionReadWKT<DataTypeMultiLineStringName, CartesianMultiLineString, MultiLineStringSerializer<CartesianPoint>, ReadWKTMultiLineStringNameHolder>>(FunctionDocumentation
+    {
+        .description=R"(
+Parses a Well-Known Text (WKT) representation of a MultiLineString geometry and returns it in the internal ClickHouse format.
+)",
+        .syntax = "readWKTMultiLineString(wkt_string)",
+        .arguments{
+            {"wkt_string", "The input WKT string representing a MultiLineString geometry."}
+        },
+        .returned_value = "The function returns a ClickHouse internal representation of the multilinestring geometry.",
+        .examples{
+            {"first call", "SELECT readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3), (4 4, 5 5, 6 6))');", R"(
+┌─readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3), (4 4, 5 5, 6 6))')─┐
+│ [[(1,1),(2,2),(3,3)],[(4,4),(5,5),(6,6)]]                                    │
+└──────────────────────────────────────────────────────────────────────────────┘
+
+            )"},
+            {"second call", "SELECT toTypeName(readWKTLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1))'));", R"(
+┌─toTypeName(readWKTLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1))'))─┐
+│ MultiLineString                                                         │
+└─────────────────────────────────────────────────────────────────────────┘
+            )"},
+        },
+        .categories{"Unique identifiers"}
+    });
    factory.registerFunction<FunctionReadWKT<DataTypeRingName, CartesianRing, RingSerializer<CartesianPoint>, ReadWKTRingNameHolder>>();
    factory.registerFunction<FunctionReadWKT<DataTypePolygonName, CartesianPolygon, PolygonSerializer<CartesianPoint>, ReadWKTPolygonNameHolder>>();
    factory.registerFunction<FunctionReadWKT<DataTypeMultiPolygonName, CartesianMultiPolygon, MultiPolygonSerializer<CartesianPoint>, ReadWKTMultiPolygonNameHolder>>();
--- a/tests/queries/0_stateless/03215_multilinestring_geometry.reference
+++ b/tests/queries/0_stateless/03215_multilinestring_geometry.reference
@ -0,0 +1,17 @@
+-- { echoOn }
+SELECT readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1))');
+[[(1,1),(2,2),(3,3),(1,1)]]
+SELECT toTypeName(readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1))'));
+MultiLineString
+SELECT wkt(readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1))'));
+MULTILINESTRING((1 1,2 2,3 3,1 1))
+SELECT readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1), (1 0, 2 0, 3 0))');
+[[(1,1),(2,2),(3,3),(1,1)],[(1,0),(2,0),(3,0)]]
+SELECT toTypeName(readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1), (1 0, 2 0, 3 0))'));
+MultiLineString
+SELECT wkt(readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1), (1 0, 2 0, 3 0))'));
+MULTILINESTRING((1 1,2 2,3 3,1 1),(1 0,2 0,3 0))
+-- Native Array(Array(Tuple(Float64, Float64))) is treated as Polygon, not as MultiLineString.
+WITH wkt(CAST([[(1, 1), (2, 2), (3, 3), (1, 1)]], 'Array(Array(Tuple(Float64, Float64)))')) as x
+SELECT x, toTypeName(x), readWKTPolygon(x) as y, toTypeName(y);
+POLYGON((1 1,2 2,3 3,1 1))	String	[[(1,1),(2,2),(3,3),(1,1)]]	Polygon
--- a/tests/queries/0_stateless/03215_multilinestring_geometry.sql
+++ b/tests/queries/0_stateless/03215_multilinestring_geometry.sql
@ -0,0 +1,12 @@
+-- { echoOn }
+SELECT readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1))');
+SELECT toTypeName(readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1))'));
+SELECT wkt(readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1))'));
+
+SELECT readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1), (1 0, 2 0, 3 0))');
+SELECT toTypeName(readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1), (1 0, 2 0, 3 0))'));
+SELECT wkt(readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1), (1 0, 2 0, 3 0))'));
+
+-- Native Array(Array(Tuple(Float64, Float64))) is treated as Polygon, not as MultiLineString.
+WITH wkt(CAST([[(1, 1), (2, 2), (3, 3), (1, 1)]], 'Array(Array(Tuple(Float64, Float64)))')) as x
+SELECT x, toTypeName(x), readWKTPolygon(x) as y, toTypeName(y);
--- a/utils/check-style/aspell-ignore/en/aspell-dict.txt
+++ b/utils/check-style/aspell-ignore/en/aspell-dict.txt
@ -1,4 +1,4 @@
-personal_ws-1.1 en 2942 
+personal_ws-1.1 en 2942
 AArch
 ACLs
 ALTERs
@ -561,6 +561,7 @@ MindsDB
 Mongodb
 Monotonicity
 MsgPack
+MultiLineString
 MultiPolygon
 Multiline
 Multiqueries
@ -2361,6 +2362,7 @@ rankCorr
 rapidjson
 rawblob
 readWKTLineString
+readWKTMultiLineString
 readWKTMultiPolygon
 readWKTPoint
 readWKTPolygon