Weibin Zeng

cc/ @lixueclaire @Thespica @SemyonSinchenko @sighingnow

done

Maybe this can be link to a new issue in fastFFI

> I'm sorry that I've been busy applying for jobs recently. But I will continue to push forward with this task. the dataset for performance evaluation has been upload here:...

May need to integrate some tool like: [gcov](https://gcc.gnu.org/onlinedocs/gcc/Gcov.html) for c++ reference: https://github.com/catchorg/Catch2/issues/839

Maybe we can use the [APOC procedure](https://neo4j.com/docs/spark/current/architecture/#_apoc), but it only available in enterprise edition

And anther solution is change the option [schema.flatten.limit](https://neo4j.com/docs/spark/current/reading/#_neo4j_read_options) to 1(default is 10). I change the code https://github.com/alibaba/GraphAr/blob/fe4ebb9b2dbcf30e63cd3895a51d4d614fe80df3/spark/src/main/scala/com/alibaba/graphar/example/Neo4j2GraphAr.scala#L75-L78 to ```java val person_df = spark.read .format("org.neo4j.spark.DataSource") .option("schema.flatten.limit", 1) .option("query", "MATCH (n:Person) RETURN...

And I think this is not the bug of GraphAr, so I change the title and label to `improvement`.

Thanks to @liuxiaocs7, we also has a [discussion](https://github.com/alibaba/GraphAr/discussions/343) about integrate GraphAr through `Loader`, we can discuss more details in the future.

> > Thanks to @liuxiaocs7, we also has a [discussion](https://github.com/alibaba/GraphAr/discussions/321) about integrate GraphAr through `Loader`, we can discuss more details in the future. > > Hi, you mean this =>...