Neo4j problems with type comparisons in “WHERE”

Question

I am getting the following error when executing my Neo4j queries through java:

org.neo4j.graphdb.QueryExecutionException: Don't know how to compare that. Left: "0" (String); Right: 0 (Long)
at org.neo4j.kernel.impl.query.QueryExecutionKernelException.asUserException(QueryExecutionKernelException.java:35)
at org.neo4j.cypher.internal.javacompat.ExecutionResult.converted(ExecutionResult.java:399)
at org.neo4j.cypher.internal.javacompat.ExecutionResult.hasNext(ExecutionResult.java:232)
at main.java.com.bag.server.database.Neo4jDatabaseAccess.readObject(Neo4jDatabaseAccess.java:172)
at main.java.com.bag.server.TestServer.handleNodeRead(TestServer.java:259)
at main.java.com.bag.server.TestServer.appExecuteUnordered(TestServer.java:153)
at bftsmart.tom.server.defaultservices.DefaultRecoverable.executeUnordered(DefaultRecoverable.java:417)
at bftsmart.tom.ServiceReplica.receiveReadonlyMessage(ServiceReplica.java:214)
at bftsmart.tom.core.DeliveryThread.deliverUnordered(DeliveryThread.java:289)
at bftsmart.tom.core.TOMLayer.requestReceived(TOMLayer.java:290)
at bftsmart.communication.client.netty.NettyClientServerCommunicationSystemServerSide.channelRead0(NettyClientServerCommunicationSystemServerSide.java:184)
at bftsmart.communication.client.netty.NettyClientServerCommunicationSystemServerSide.channelRead0(NettyClientServerCommunicationSystemServerSide.java:61)
at io.netty.channel.SimpleChannelInboundHandler.channelRead(SimpleChannelInboundHandler.java:105)
at io.netty.channel.AbstractChannelHandlerContext.invokeChannelRead(AbstractChannelHandlerContext.java:292)
at io.netty.channel.AbstractChannelHandlerContext.fireChannelRead(AbstractChannelHandlerContext.java:278)
at io.netty.handler.codec.ByteToMessageDecoder.fireChannelRead(ByteToMessageDecoder.java:277)
at io.netty.handler.codec.ByteToMessageDecoder.channelRead(ByteToMessageDecoder.java:264)
at io.netty.channel.AbstractChannelHandlerContext.invokeChannelRead(AbstractChannelHandlerContext.java:292)
at io.netty.channel.AbstractChannelHandlerContext.fireChannelRead(AbstractChannelHandlerContext.java:278)
at io.netty.channel.DefaultChannelPipeline.fireChannelRead(DefaultChannelPipeline.java:962)
at io.netty.channel.nio.AbstractNioByteChannel$NioByteUnsafe.read(AbstractNioByteChannel.java:131)
at io.netty.channel.nio.NioEventLoop.processSelectedKey(NioEventLoop.java:528)
at io.netty.channel.nio.NioEventLoop.processSelectedKeysOptimized(NioEventLoop.java:485)
at io.netty.channel.nio.NioEventLoop.processSelectedKeys(NioEventLoop.java:399)
at io.netty.channel.nio.NioEventLoop.run(NioEventLoop.java:371)
at io.netty.util.concurrent.SingleThreadEventExecutor$2.run(SingleThreadEventExecutor.java:112)
at io.netty.util.concurrent.DefaultThreadFactory$DefaultRunnableDecorator.run(DefaultThreadFactory.java:137)
at java.lang.Thread.run(Thread.java:745)

But I am sure that I have the "0" as a String inside the Database. And that I am entering it as a string in the query:

String.format(" WHERE r.%s <= %s OR n.%s IS NULL", "snapshotId", Long.toString(0), "snapshotId")

Answer 1

You're using string mangling instead of parameters, so you lose the type info (because it's just putting it all in a string). If you type the character '0' without quotes in Cypher, it is going to be interpreted as an Int . So instead of your current query,

WHERE r.snapshotId <= 0 or n.snapshotId IS NULL

you actually want it to say

WHERE r.snapshotId <= '0' or n.snapshotId IS NULL

with the quotes around the '0', so it treats it as a string.

The real solution, though, is instead of string mangling, write a query to say:

WHERE r.snapshotId <= {zero_string} or n.snapshotId IS NULL

and then pass a parameter zero_string that is set to Long.toString(0) . That way the driver will handle types for you while packing, unpacking, and interpreting the data.

EDIT: Or if you really need the property name to be dynamic as well, pass it as a parameter too:

WHERE r[{zero_param}] <= {zero_string} or n[{zero_param}] IS NULL

UPDATE: You can modify this to work for multiple key-value pairs by passing in a Map and doing some iterative work. The naive way to do it would be like so:

WHERE ALL(k IN KEYS({map_param}) WHERE r[k] <= {map_param}[k] OR n[k] IS NULL)

but this is likely to be very slow on any scale, as I don't think the query planner will be able to optimize it. Try to narrow down your r matches on some other criteria before you apply this filter.

Answer 2

The string generated by your format statement is:

" WHERE r.snapshotId <= 0 OR n.snapshotId IS NULL"

The Long.toString(0) value still causes a numeric 0 to be output, as far as Cypher is concerned, since the literal value is not surrounded by quotes.

To fix this, you have to make sure the values being compared have compatible types.

1. Comparing as strings

One way to do this is to make sure the numeric value is represented as a string (by quoting it):

String.format(" WHERE r.%s <= '%s' OR n.%s IS NULL", "snapshotId", 0, "snapshotId")

2. Comparing as numeric values

However, if you expect the result of the comparison to make sense arithmetically, it generally does not make sense to use a string comparison. For example, (".1" <= "0") is true , whereas (.1 <= 0) is false . Therefore, for comparisons that make sense arithmetically, you should probably convert r.snapshotId to either a float or an integer.

For example:

String.format(" WHERE TOFLOAT(r.%s) <= %s OR n.%s IS NULL", "snapshotId", 0, "snapshotId")

Use parameters

Finally, as @ToreEschliman suggested, you should be passing parameters instead of using string mangling.