The ENTRYID structure contains an entry identifier for a MAPI object.
Header file: | MAPIDEFS.H |
Related macros: | CbNewENTRYID SizedENTRYID |
typedef struct
{
BYTE abFlags[4];
BYTE ab[MAPI_DIM];
} ENTRYID, FAR *LPENTRYID;
The following flags can be set in abFlags[0]:
The ENTRYID structure is used by message store and address book providers to construct unique identifiers for their objects. Entry identifiers are used to identify the following types of objects:
Message stores
Folders
Messages
Address book containers
Distribution lists
Messaging users
Status objects
Profile sections
Each provider uses a format for the ENTRYID structure that makes sense for that provider.
Entry identifiers cannot be compared directly because one object can be represented by two different binary values. To determine if two entry identifiers represent the same object, call the IMAPISession::CompareEntryIDs method.
When a client calls an object's IMAPIProp::GetProps method to retrieve its entry identifier, the object returns the most permanent form of the entry identifier. A client can verify that an entry identifier is long-term by checking that none of the flags are set in the first byte of the abFlags member.
When a client accesses an entry identifier through a column in a table, most likely this entry identifier is short-term rather than long-term. Short-term entry identifiers can be used to open their corresponding objects only in the current MAPI session. A client can verify that an entry identifier is short-term by checking that all of the flags are set in the first byte of the abFlags member.
Some entry identifiers are short-term, but have long-term use. Such an entry identifier will have one or more of the appropriate flags set in the first byte of its abFlags member.
An ENTRYID structure is similar to a FLATENTRY structure. However, there are some differences:
Clients should always pass in naturally aligned entry identifiers. Although providers should handle arbitrarily aligned entry identifiers, clients should not expect this behavior. Failure to pass a suitable aligned entry identifier to a method can result in an alignment fault on RISC processors.
The natural alignment factor, typically 8 bytes, is the largest data type supported by the CPU, and usually the same alignment factor used by the system memory allocator. A naturally aligned memory address allows the CPU to access any data type it supports at that address without generating an alignment fault. For RISC CPUs, a data type of size N bytes must usually be aligned on an even multiple of N bytes, with the address being an even multiple of N.
For more information, see Entry Identifiers.
ID Structures, IMAPISupport::CompareEntryIDs, PR_RECORD_KEY