File Types (Magic xpa 3.x)
Every file defined in Magic xpa creates a physical file that contains data and is an SQL table on IBM i. Physical file data can be in the form of a subset. Each subset has its own member name. At least one member name must exist in the file.
Indexes can be specified for both physical and logical files. The file record structure for each file is known as the record format and must have a record format name assigned to it.
Every key defined in Magic xpa becomes a logical file that is an SQL index on IBM i.
Physical files contain the actual file data. The record format in a physical file describes the field order and details of each record, including:
Logical files can be classified as:
-
Simple logical files, which define an index to be used when accessing a physical file.
-
Unconditional views, which can extract specific fields from the physical file record format, change the field position in the record, and define additional fields not found in the original structure.
-
Conditional views, which contain instructions to conditionally select data from physical files, such as Select, Omit, and Compare.
-
Joined views, which combine up to 32 physical record formats into one logical record view definition.
|
Note:
|
Only simple logical files can be used for a Magic xpa key definition. Other types of logical files must be defined separately in the Magic xpa Data repository.
|
In this section, the terms file and table, as well as the terms key and index, are used interchangeably. Because the IBM i gateway is an ISAM gateway, IBM i products and documentation use the ISAM terms, file and key. Magic xpa products and documentation, however, follow SQL terminology, which uses the terms table (or data source) and index.
