From 9ef184fd7c00ce9524537cc551bbbe02e86a2f49 Mon Sep 17 00:00:00 2001
From: simon <me@simon987.net>
Date: Sat, 20 Jul 2019 09:52:31 -0400
Subject: [PATCH] Initial commit

---
 .gitignore     |  8 ++++++
 CMakeLists.txt | 10 +++++++
 README.md      | 13 +++++++++
 hamming.c      | 78 ++++++++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 109 insertions(+)
 create mode 100644 .gitignore
 create mode 100644 CMakeLists.txt
 create mode 100644 README.md
 create mode 100644 hamming.c

diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000..f5b17b3
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1,8 @@
+cmake-build-debug/
+CMakeFiles
+cmake_install.cmake
+Makefile
+CMakeCache.txt
+*.so
+*.cbp
+.idea/
\ No newline at end of file
diff --git a/CMakeLists.txt b/CMakeLists.txt
new file mode 100644
index 0000000..f98a47c
--- /dev/null
+++ b/CMakeLists.txt
@@ -0,0 +1,10 @@
+cmake_minimum_required(VERSION 3.14)
+include_directories("/usr/include/postgresql/server/")
+project(hamming C)
+
+set(CMAKE_C_STANDARD 99)
+
+set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -march=native")
+
+find_package("PostgreSQL REQUIRED")
+add_library(hamming SHARED hamming.c)
\ No newline at end of file
diff --git a/README.md b/README.md
new file mode 100644
index 0000000..4f21340
--- /dev/null
+++ b/README.md
@@ -0,0 +1,13 @@
+# PostgreSQL hamming distance
+
+Hamming distance for fixed-length `bytea` datatype.
+
+
+### Compiling from source (Cmake)
+
+```bash
+cmake .
+make
+```
+
+See [hamming.c](hamming.c) for more information
\ No newline at end of file
diff --git a/hamming.c b/hamming.c
new file mode 100644
index 0000000..2b8c440
--- /dev/null
+++ b/hamming.c
@@ -0,0 +1,78 @@
+#include "postgresql/server/postgres.h"
+#include "postgresql/server/fmgr.h"
+
+PG_MODULE_MAGIC;
+
+int const HASH_SIZE = 12;
+int const HASH_BITS = HASH_SIZE * HASH_SIZE;
+int const BYTEA_LEN = HASH_BITS / 8;
+
+PG_FUNCTION_INFO_V1(hash_is_within_distance);
+
+/**
+ * Check if the hamming distance of the two raw byte arrays
+ * is within the specified distance
+ *
+ * It is assumed that: the two arrays are exactly
+ *  BYTEA_LEN bytes long
+ *
+ * Import with
+    CREATE OR REPLACE FUNCTION hash_is_within_distance(bytea, bytea, integer) RETURNS boolean
+     AS '/path/to/libhamming.so', 'hash_is_within_distance'
+     LANGUAGE C STRICT;'
+ *
+ * @return the hamming distance between the two arrays
+ */
+Datum hash_is_within_distance(PG_FUNCTION_ARGS) {
+
+    bytea *hash1 = PG_GETARG_BYTEA_P(0);
+    bytea *hash2 = PG_GETARG_BYTEA_P(1);
+    int32 max_distance = PG_GETARG_INT32(2);
+
+    int distance = 0;
+
+    char *h1 = hash1->vl_len_;
+    char *h2 = hash2->vl_len_;
+
+    for (int i = BYTEA_LEN; i >= 0; i--) {
+        distance += __builtin_popcount(h1[i] ^ h2[i]);
+
+        if (distance > max_distance) {
+            PG_RETURN_BOOL(false);
+        }
+    }
+
+    PG_RETURN_BOOL(true);
+}
+
+PG_FUNCTION_INFO_V1(hash_distance);
+
+/**
+ * Hamming distance of two raw byte arrays
+ *
+ * It is assumed that: the two arrays are exactly
+ *  BYTEA_LEN bytes long
+ *
+ * Import with
+    CREATE OR REPLACE FUNCTION hash_distance(bytea, bytea) RETURNS integer
+     AS '/path/to/libhamming.so', 'hash_distance'
+     LANGUAGE C STRICT;'
+ *
+ * @return the hamming distance between the two arrays
+ */
+Datum hash_distance(PG_FUNCTION_ARGS) {
+
+    bytea *hash1 = PG_GETARG_BYTEA_P(0);
+    bytea *hash2 = PG_GETARG_BYTEA_P(1);
+
+    int distance = 0;
+
+    char *h1 = hash1->vl_len_;
+    char *h2 = hash2->vl_len_;
+
+    for (int i = BYTEA_LEN; i >= 0; i--) {
+        distance += __builtin_popcount(h1[i] ^ h2[i]);
+    }
+
+    PG_RETURN_INT32(distance);
+}