Merge pull request #56 from wey-gu/example_algo_from_ngql

feat: add example of pyspark to scan data from ngql
wey-gu · Feb 23, 2023 · de2ea0c · de2ea0c
2 parents d605e58 + bcd3c6a
commit de2ea0c
Showing 1 changed file with 23 additions and 2 deletions.
diff --git a/spark/pagerank_example.ipynb b/spark/pagerank_example.ipynb
@@ -38,6 +38,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "# option 0: read graph data for one type of edge, scan from all storaged instances\n",
     "df = spark.read.format(\n",
     "  \"com.vesoft.nebula.connector.NebulaDataSource\").option(\n",
     "    \"type\", \"edge\").option(\n",
@@ -46,9 +47,29 @@
     "    \"returnCols\", \"degree\").option(\n",
     "    \"metaAddress\", \"metad0:9559\").option(\n",
     "    \"partitionNumber\", 1).load()\n",
+    "\n",
+    "# option 1: read graph data with ngql, get data from graphd\n",
+    "df = spark.read.format(\n",
+    "  \"com.vesoft.nebula.connector.NebulaDataSource\").option(\n",
+    "    \"type\", \"edge\").option(\n",
+    "    \"spaceName\", \"basketballplayer\").option(\n",
+    "    \"label\", \"follow\").option(\n",
+    "    \"returnCols\", \"degree\").option(\n",
+    "    \"metaAddress\", \"metad0:9559\").option(\n",
+    "    \"graphAddress\", \"graphd:9669\").option(\n",
+    "    \"ngql\", \"MATCH ()-[e:follow]->() return e LIMIT 1000\").option(\n",
+    "    \"partitionNumber\", 1).load()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "31871548",
+   "metadata": {},
+   "outputs": [],
+   "source": [
     "# prConfig = PRConfig(3, 0.85)\n",
-    "prConfig = spark._jvm.PRConfig(3, 0.85)\n",
-    "\n"
+    "prConfig = spark._jvm.PRConfig(3, 0.85)"
    ]
   },
   {