From 66eeb9f9902e7c271b2cd40f6e7608d50ae6b4b0 Mon Sep 17 00:00:00 2001
From: TShapinsky <tobiasshapinsky@gmail.com>
Date: Thu, 9 May 2024 10:14:12 -0600
Subject: [PATCH] commit hashing notebook

---
 notebooks/hashing.ipynb | 157 ++++++++++++++++++++++++++++++++++++++++
 1 file changed, 157 insertions(+)
 create mode 100644 notebooks/hashing.ipynb

diff --git a/notebooks/hashing.ipynb b/notebooks/hashing.ipynb
new file mode 100644
index 000000000..ddaadfb0f
--- /dev/null
+++ b/notebooks/hashing.ipynb
@@ -0,0 +1,157 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from rdflib import Graph, Namespace\n",
+    "import time\n",
+    "from buildingmotif.namespaces import BRICK, A\n",
+    "from buildingmotif.utils import approximate_graph_hash\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "g = Graph().parse('../tests/unit/fixtures/smallOffice_brick.ttl')\n",
+    "MODEL = Namespace(\"urn:model#\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 31,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "ttl serialize time: 0.02885890007019043\n",
+      "n3 serialize time: 0.02724599838256836\n",
+      "xml serialize time: 0.009903907775878906\n",
+      "nt serialize time: 0.0031800270080566406\n",
+      "trig serialize time: 0.0222320556640625\n",
+      "json-ld serialize time: 0.01989889144897461\n",
+      "hext serialize time: 0.0049059391021728516\n"
+     ]
+    }
+   ],
+   "source": [
+    "begin = time.time()\n",
+    "g.serialize(format=\"ttl\")\n",
+    "end = time.time()\n",
+    "print(f\"ttl serialize time: {end - begin}\")\n",
+    "\n",
+    "begin = time.time()\n",
+    "g.serialize(format=\"n3\")\n",
+    "end = time.time()\n",
+    "print(f\"n3 serialize time: {end - begin}\")\n",
+    "\n",
+    "begin = time.time()\n",
+    "g.serialize(format=\"xml\")\n",
+    "end = time.time()\n",
+    "print(f\"xml serialize time: {end - begin}\")\n",
+    "\n",
+    "begin = time.time()\n",
+    "g.serialize(format=\"nt\")\n",
+    "end = time.time()\n",
+    "print(f\"nt serialize time: {end - begin}\")\n",
+    "\n",
+    "begin = time.time()\n",
+    "g.serialize(format=\"trig\")\n",
+    "end = time.time()\n",
+    "print(f\"trig serialize time: {end - begin}\")\n",
+    "\n",
+    "begin = time.time()\n",
+    "g.serialize(format=\"json-ld\")\n",
+    "end = time.time()\n",
+    "print(f\"json-ld serialize time: {end - begin}\")\n",
+    "\n",
+    "begin = time.time()\n",
+    "g.serialize(format=\"hext\")\n",
+    "end = time.time()\n",
+    "print(f\"hext serialize time: {end - begin}\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 32,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "6727799026531355694\n",
+      "-4781114786451684549\n",
+      "5257158006352705709\n"
+     ]
+    }
+   ],
+   "source": [
+    "g.add((MODEL[\"a\"], A, BRICK[\"AHU\"]))\n",
+    "g.serialize('before.nt', format='nt')\n",
+    "# print(hash(g.serialize(format='nt')))\n",
+    "print(approximate_graph_hash(g))\n",
+    "triple = (MODEL[\"b\"], A, BRICK[\"Sensor\"])\n",
+    "g.add(triple)\n",
+    "# g.serialize('after.nt', format='nt')\n",
+    "# print(hash(g.serialize(format='nt')))\n",
+    "print(approximate_graph_hash(g))\n",
+    "g.remove(triple)\n",
+    "g.serialize('before_after.nt', format='nt')\n",
+    "# print(hash(g.serialize(format='nt')))\n",
+    "print(approximate_graph_hash(g))\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "triple = (MODEL[\"b\"], A, BRICK[\"Sensor\"])\n",
+    "g.add((MODEL[\"a\"], A, BRICK[\"AHU\"]))\n",
+    "before = approximate_graph_hash(g)\n",
+    "while True:\n",
+    "    g.add(triple)\n",
+    "    approximate_graph_hash(g)\n",
+    "    g.remove(triple)\n",
+    "    assert before == approximate_graph_hash(g)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": ".venv",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.11.4"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}