home *** CD-ROM | disk | FTP | other *** search
-
- Indexing Scheme
-
- An index is a sorted list that is used as a reference. In
- Family Scrapbook, all records are stored on disk in the order of
- record number. Indexes are used because you will not always want
- to access the records in record number order.
-
- If you wanted to print a list of persons in the order of
- birthdates, the program would use the birthdate index as a reference
- to find the records. The index would tell the program what order to
- read the records from your disk so that they would be in birthdate
- order.
-
- Indexes are only updated by the reindex utility. This means
- that when you add, change, or delete a record, you must then reindex
- the database so that the indexes are correct. The reindex utility
- reads all of the records from the database and then sorts them in the
- order that the index requires. It then creates the index from this list.
- This is done for every index during the reindex process.
-
- Many of the reports that Family Scrapbook produces make
- extensive use of the indexes, so it is wise to make sure that the
- database is properly indexed, especially before printing reports. To
- remind you when to reindex, the program will put the letter R in the
- lower right of the screen. It only does this when you add or delete a
- record. Sometimes changing a record can also require reindexing,
- such as changing a person's birthdate. In this situation the R will
- not appear. The Reindex Database selection is found in the Utilities
- Menu, but can be run from many other areas of the program by
- pressing Alt-R.
-
- Note: The reindex utility is contained in an external
- executable file named REINDEX.EXE. This file can be run directly
- from DOS. You may wish to do this if your database grows very large
- (600 or more persons). The reindex utility attempts to do all sorting
- in memory, but will sort to disk if there is not enough memory. By
- running it independently of Family Scrapbook, it will have more
- memory to operate in and, therefore, will operate faster on large
- databases.
-
-