see on GitHub

Conventions related to querying

FindPropertyNameForIndex and FindPropertyNameForDynamicIndex

These two conventions specify functions that are used to find the indexed property name for a static index or a dynamic one. This can be useful when you are indexing nested properties of complex types. Their default implementations are:

Conventions.FindPropertyNameForIndex = (indexedType, indexName, path, prop) => (path + prop).Replace(",", "_").Replace(".", "_");
Conventions.FindPropertyNameForDynamicIndex = (indexedType, indexName, path, prop) => path + prop;

The arguments in the order of appearance: an indexed document type, an index name, a current path and a property path.


The consistency options used when querying a database. Its default value is:

Conventions.DefaultQueryingConsistency = ConsistencyOptions.None;

This option ensures that after querying an index at time T, you will never see the results of the index at a time prior to T. This is ensured by the server and require no action from the client. Note that because of the indexing process done by the server in the background you might get stale index results.

If you really need your query results to include documents you have just changed or added, you can use the option:

Conventions.DefaultQueryingConsistency = ConsistencyOptions.AlwaysWaitForNonStaleResultsAsOfLastWrite;

Then the client will wait 15 seconds for non-stale results, using an etag of a last written document within a session as a cut-off point. If after that time the results are still stale, then TimeoutException will be thrown.

Beware of AlwaysWaitForNonStaleResultsAsOfLastWrite overuse

The indexing mechanism in RavenDB is built on a BASE model. In order to avoid querying consistency pitfalls in the future you need to consider this at the data modeling phase.

The usage of WaitForNonStaleResultsAsOfLastWrite at a query level is usually reasonable on only very rare occassions. The need of taking advantage of AlwaysWaitForNonStaleResultsAsOfLastWrite convention is very often a symptom of deeper issues in an application model and misunderstanding of querying concepts in RavenDB.


If you work in a sharded environment then ShardedDocumentStore needs to merge results of a map-reduce index that come from different shards. So they need to be reduced on the client side. The behavior of the client side reduction can by overwritten by the following convention:

Conventions.ApplyReduceFunction = (type, resultType, results, transformResults) => 
	// carry out the reduction over results from different shards


It determines whether queries on document id are allowed. By default, queries on id are disabled, because it is far more efficient to do a Load() than a Query() if you already know the id.

Conventions.AllowQueriesOnId = false;

Backward compatibility only

This convention is provided for backward compatibility purposes only. It is NOT recommended to set its value to true.