Add IPPREFIX

velox/docs/develop/types.rst

@@ -137,6 +137,7 @@ JSON                      VARCHAR
 TIMESTAMP WITH TIME ZONE  BIGINT
 UUID                      HUGEINT
 IPADDRESS                 HUGEINT
+IPPREFIX                  VARBINARY
 ========================  =====================
 
 TIMESTAMP WITH TIME ZONE represents a time point in milliseconds precision
@@ -155,6 +156,23 @@ used in IPADDRESS/IPPREFIX related functions. This type can be used to
 create IPPREFIX networks as well as to check IPADDRESS validity within
 IPPREFIX networks.
 
+IPPREFIX represents an IPV6 or IPV4 formatted IPV6 address along with a one byte
+prefix length. Its physical type is VARBINARY but has a fixed length of 17 bytes.
+The format that the address is stored in is defined as part of `(RFC 4291#section-2.5.5.2) <https://datatracker.ietf.org/doc/html/rfc4291.html#section-2.5.5.2>`_.
+The prefix length is stored in the last byte of the VARBINARY array.
+The IP address stored is the canonical(smallest) IP address in the
+subnet range. This type can be used in IP subnet functions.
+
+Example:
+
+In this example the first 32 bits(*FFFF:FFFF*) represents the network prefix.
+As a result the IPPREFIX object stores *FFFF:FFFF::* and the length 32 for both of these IPPREFIX objects. 
+
+::
+
+   IPPREFIX 'FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF/32' -- IPPREFIX 'FFFF:FFFF:0000:0000:0000:0000:0000:0000/32'
+   IPPREFIX 'FFFF:FFFF:4455:6677:8899:AABB:CCDD:EEFF/32' -- IPPREFIX 'FFFF:FFFF:0000:0000:0000:0000:0000:0000/32'
+
 Spark Types
 ~~~~~~~~~~~~
 The `data types <https://spark.apache.org/docs/latest/sql-ref-datatypes.html>`_ in Spark have some semantic differences compared to those in 

velox/docs/functions/presto/conversion.rst

@@ -30,7 +30,7 @@ are supported if the conversion of their element types are supported. In additio
 supported conversions to/from JSON are listed in :doc:`json`.
 
 .. list-table::
-   :widths: 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25
+   :widths: 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25 25
    :header-rows: 1
 
    * -
@@ -49,6 +49,7 @@ supported conversions to/from JSON are listed in :doc:`json`.
      - interval day to second
      - decimal
      - ipaddress
+     - ipprefix
    * - tinyint
      - Y
      - Y
@@ -65,6 +66,7 @@ supported conversions to/from JSON are listed in :doc:`json`.
      -
      - Y
      - 
+     -
    * - smallint
      - Y
      - Y
@@ -81,6 +83,7 @@ supported conversions to/from JSON are listed in :doc:`json`.
      -
      - Y
      - 
+     -
    * - integer
      - Y
      - Y
@@ -97,6 +100,7 @@ supported conversions to/from JSON are listed in :doc:`json`.
      -
      - Y
      - 
+     -
    * - bigint
      - Y
      - Y
@@ -113,6 +117,7 @@ supported conversions to/from JSON are listed in :doc:`json`.
      -
      - Y
      - 
+     -
    * - boolean
      - Y
      - Y
@@ -129,6 +134,7 @@ supported conversions to/from JSON are listed in :doc:`json`.
      -
      - Y
      - 
+     -
    * - real
      - Y
      - Y
@@ -145,6 +151,7 @@ supported conversions to/from JSON are listed in :doc:`json`.
      -
      - Y
      - 
+     -
    * - double
      - Y
      - Y
@@ -161,6 +168,7 @@ supported conversions to/from JSON are listed in :doc:`json`.
      -
      - Y
      - 
+     -
    * - varchar
      - Y
      - Y
@@ -177,6 +185,7 @@ supported conversions to/from JSON are listed in :doc:`json`.
      -
      - Y
      - Y
+     - Y
    * - varbinary
      - 
      - 
@@ -193,6 +202,7 @@ supported conversions to/from JSON are listed in :doc:`json`.
      -
      - 
      - Y
+     -
    * - timestamp
      -
      -
@@ -209,6 +219,7 @@ supported conversions to/from JSON are listed in :doc:`json`.
      -
      -
      - 
+     -
    * - timestamp with time zone
      -
      -
@@ -225,6 +236,7 @@ supported conversions to/from JSON are listed in :doc:`json`.
      -
      -
      -
+     -
    * - date
      -
      -
@@ -241,6 +253,7 @@ supported conversions to/from JSON are listed in :doc:`json`.
      -
      -
      -
+     -
    * - interval day to second
      -
      -
@@ -257,6 +270,7 @@ supported conversions to/from JSON are listed in :doc:`json`.
      -
      -
      -
+     -
    * - decimal
      - Y
      - Y
@@ -273,6 +287,7 @@ supported conversions to/from JSON are listed in :doc:`json`.
      -
      - Y
      -
+     -
    * - ipaddress
      - 
      - 
@@ -288,7 +303,25 @@ supported conversions to/from JSON are listed in :doc:`json`.
      -
      -
      - 
+     - Y
+     - Y
+   * - ipprefix
+     - 
+     - 
+     - 
+     - 
+     - 
      - 
+     - 
+     - Y
+     - 
+     -
+     -
+     -
+     -
+     - 
+     - Y
+     - Y
 
 Cast to Integral Types
 ----------------------
@@ -689,6 +722,33 @@ IPV4 mapped IPV6:
 
   SELECT cast(ipaddress '::ffff:ffff:ffff' as varchar); -- '255.255.255.255'
 
+From IPPrefix
+^^^^^^^^^^^^^^
+
+Casting from IPPREFIX to VARCHAR returns a string formatted as *x.x.x.x/<prefix-length>* for IPV4 formatted IPV6 addresses.
+
+For all other IPV6 addresses it will be formatted in compressed alternate form IPV6 defined in `RFC 4291#section-2.2 <https://datatracker.ietf.org/doc/html/rfc4291.html#section-2.2>`_
+followed by */<prefix-length>*. `[RFC 4291#section-2.3] <https://datatracker.ietf.org/doc/html/rfc4291.html#section-2.3>`_ 
+
+IPV4:
+
+::
+
+  SELECT cast(ipprefix '1.2.0.0/16' as varchar); -- '1.2.0.0/16'
+
+IPV6:
+
+::
+
+  SELECT cast(ipprefix '2001:db8::ff00:42:8329/128' as varchar); -- '2001:db8::ff00:42:8329/128'
+  SELECT cast(ipprefix '0:0:0:0:0:0:13.1.68.3/32' as varchar); -- '::/32'
+
+IPV4 mapped IPV6:
+
+::
+
+  SELECT cast(ipaddress '::ffff:ffff:0000/16' as varchar); -- '255.255.0.0/16'
+
 Cast to VARBINARY
 -----------------
 
@@ -1036,6 +1096,8 @@ Invalid example
 Cast to IPADDRESS
 -----------------
 
+.. _ipaddress-varchar:
+
 From VARCHAR
 ^^^^^^^^^^^^
 
@@ -1128,6 +1190,67 @@ Invalid examples:
 
   SELECT cast(from_hex('f000001100') as ipaddress); -- Invalid IP address binary length: 5
 
+From IPPREFIX
+^^^^^^^^^^^^^
+
+Returns the canonical(lowest) IPADDRESS in the subnet range.
+
+Examples:
+
+::
+
+  SELECT cast(ipprefix '1.2.3.4/24' as ipaddress) -- ipaddress '1.2.3.0'
+  SELECT cast(ipprefix '2001:db8::ff00:42:8329/64' as ipaddress) -- ipaddress '2001:db8::'
+
+Cast to IPPREFIX
+----------------
+
+From VARCHAR
+^^^^^^^^^^^^
+
+The IPPREFIX string must be in the form of *<ip_address>/<ip_prefix>* as defined in `(RFC 4291#section-2.3) <https://datatracker.ietf.org/doc/html/rfc4291.html#section-2.3>`_
+The IPADDRESS portion of the IPPREFIX follows the same rules as casting
+`IPADDRESS to VARCHAR <#ipaddress-varchar>`_.
+
+The prefix portion must be <= 32 if the IP is an IPV4 address or <= 128 for an IPV6 address.
+As with IPADDRESS, any IPV6 address in the form of an IPV4 mapped IPV6 address will be
+interpreted as an IPV4 address. Only the canonical(smallest) IP address will be stored
+in the IPPREFIX.
+
+Examples:
+
+Valid examples:
+
+::
+
+  SELECT cast('2001:0db8:0000:0000:0000:ff00:0042:8329/32' as ipprefix); -- ipprefix '2001:0db8::/32'
+  SELECT cast('1.2.3.4/24' as ipprefix); -- ipprefix '1.2.3.0/24'
+  SELECT cast('::ffff:ffff:ffff/16' as ipprefix); -- ipprefix '255.255.0.0/16'
+
+Invalid examples:
+
+::
+
+  SELECT cast('2001:db8::1::1/1' as ipprefix); -- Invalid IP address '2001:db8::1::1'
+  SELECT cast('2001:0db8:0000:0000:0000:ff00:0042:8329/129' as ipprefix); -- CIDR value '129' is > network bit count '128'
+  SELECT cast('2001:0db8:0000:0000:0000:ff00:0042:8329/-1' as ipprefix); -- Mask value '-1' not a valid mask
+  SELECT cast('255.2.3.4/33' as ipprefix); -- CIDR value '33' is > network bit count '32'
+  SELECT cast('::ffff:ffff:ffff/33' as ipprefix); -- CIDR value '33' is > network bit count '32'
+
+
+From IPADDRESS
+^^^^^^^^^^^^^^
+
+Returns an IPPREFIX where the prefix length is the length of the entire IP Address.
+Prefix length for IPV4 is 32 and for IPV6 it is 128. 
+
+Examples:
+
+::
+
+  SELECT cast(ipaddress '1.2.3.4' as ipprefix) -- ipprefix '1.2.3.4/32'
+  SELECT cast(ipaddress '2001:db8::ff00:42:8329' as ipprefix) -- ipprefix '2001:db8::/128'
+
 Miscellaneous
 -------------
 

velox/expression/tests/CustomTypeTest.cpp

@@ -217,6 +217,7 @@ TEST_F(CustomTypeTest, getCustomTypeNames) {
           "TIMESTAMP WITH TIME ZONE",
           "UUID",
           "IPADDRESS",
+          "IPPREFIX",
       }),
       names);
 
@@ -231,6 +232,7 @@ TEST_F(CustomTypeTest, getCustomTypeNames) {
           "TIMESTAMP WITH TIME ZONE",
           "UUID",
           "IPADDRESS",
+          "IPPREFIX",
           "FANCY_INT",
       }),
       names);

velox/functions/prestosql/IPAddressFunctions.h

@@ -16,11 +16,13 @@
 #pragma once
 
 #include "velox/functions/prestosql/types/IPAddressType.h"
+#include "velox/functions/prestosql/types/IPPrefixType.h"
 
 namespace facebook::velox::functions {
 
 void registerIPAddressFunctions(const std::string& prefix) {
   registerIPAddressType();
+  registerIPPrefixType();
 }
 
 } // namespace facebook::velox::functions

velox/functions/prestosql/TypeOf.cpp

@@ -16,6 +16,7 @@
 #include "velox/expression/VectorFunction.h"
 #include "velox/functions/prestosql/types/HyperLogLogType.h"
 #include "velox/functions/prestosql/types/IPAddressType.h"
+#include "velox/functions/prestosql/types/IPPrefixType.h"
 #include "velox/functions/prestosql/types/JsonType.h"
 #include "velox/functions/prestosql/types/TimestampWithTimeZoneType.h"
 #include "velox/functions/prestosql/types/UuidType.h"
@@ -78,6 +79,8 @@ std::string typeName(const TypePtr& type) {
     case TypeKind::VARBINARY:
       if (isHyperLogLogType(type)) {
         return "HyperLogLog";
+      } else if (isIPPrefixType(type)) {
+        return "ipaddress";
       }
       return "varbinary";
     case TypeKind::TIMESTAMP:

velox/functions/prestosql/tests/CMakeLists.txt

@@ -65,6 +65,7 @@ add_executable(
   HyperLogLogFunctionsTest.cpp
   InPredicateTest.cpp
   IPAddressCastTest.cpp
+  IPPrefixCastTest.cpp
   JsonCastTest.cpp
   JsonExtractScalarTest.cpp
   JsonFunctionsTest.cpp

velox/functions/prestosql/tests/IPAddressCastTest.cpp

@@ -24,28 +24,28 @@ namespace {
 class IPAddressCastTest : public functions::test::FunctionBaseTest {
  protected:
   std::optional<std::string> castToVarchar(
-      const std::optional<std::string> input) {
+      const std::optional<std::string>& input) {
     auto result = evaluateOnce<std::string>(
-        "cast(cast(c0 as ipaddress) as varchar)", input);
+        "cast(cast(cast(c0 as ipaddress) as ipaddress) as varchar)", input);
     return result;
   }
 
   std::optional<int128_t> castFromVarbinary(
-      const std::optional<std::string> input) {
+      const std::optional<std::string>& input) {
     auto result =
         evaluateOnce<int128_t>("cast(from_hex(c0) as ipaddress)", input);
     return result;
   }
 
-  std::optional<std::string> allCasts(const std::optional<std::string> input) {
+  std::optional<std::string> allCasts(const std::optional<std::string>& input) {
     auto result = evaluateOnce<std::string>(
         "cast(cast(cast(cast(c0 as ipaddress) as varbinary) as ipaddress) as varchar)",
         input);
     return result;
   }
 };
 
-int128_t stringToInt128(std::string value) {
+int128_t stringToInt128(const std::string& value) {
   int128_t res = 0;
   for (char c : value) {
     res = res * 10 + c - '0';