Azure DocumentDB Adapter Limitations

Supported Types
Precision of Number Type
Special Item Names
Consistency

Supported Types

Correspondence of the JSON types of Document DB supported by this adapter and the internal data types (schema types) is as follows.

DocumentDB data type name	Supported	Schema type name	Remarks
string	○	String
number	○	BigDecimal
true/false	○	boolean
null	○	-	Dealt as null regardless of schema type.
object	×	-
array	×	-

○:Supported.
×:Not supported.

Precision of Number Type

Number type of Azure DocumentDB is IEEE754 double precision.

When writing numerical value which is over IEEE754 double precision in DocumentDB as BigDecimal type, the value is rounded. (without error)
If you would like to avoid rounding, set the schema type as String type when writing document.

However, SQL use for numerical values is not available because of saving it as String type in DocumentDB.

For data types which is supported by DocumentDB, refer to the following page.

"Frequently asked questions about DocumentDB" - (https://azure.microsoft.com/en-us/documentation/articles/documentdb-faq/)

Special Item Names

There are some items which DocumentDB generates for documents, such as "_rid", "_ts", "_self", "_etag", or "_attachments". No value can be written in these items.

Avoid to put underscore(_) at the beginning of item names defined as schema as much as possible, because the beginning of item names which DocumentDB generates is underscore(_).

For items which DocumentDB system generates, refer to the following page.

"System vs. user defined resources" - (https://azure.microsoft.com/en-us/documentation/articles/documentdb-resources/#system-vs-user-defined-resources)
"Operations on DocumentDB Documents" - (https://msdn.microsoft.com/en-us/library/azure/dn782247.aspx)

Consistency

In DocumentDB adapter, "Session" is used among the 4 types of consistencies ("Strong", "Bounded-Staleness", "Session", and "Eventual").

For consistency of DocumentDB, refer to the following page.

"Using consistency levels to maximize availability and performance in DocumentDB" ( https://azure.microsoft.com/en-us/documentation/articles/documentdb-consistency-levels/)