oracle 删除重复内容（Oracle删除重复内容）.doc

oracle 删除重复内容（Oracle删除重复内容） In normal work, you may encounter an attempt to create a unique index on a column or column in a library table. System prompt ora-01452: no unique index can be created and duplicate records found. The principle of deleting duplicate records: (1) in Oracle, each record has a ROWID, and the ROWID is unique throughout the database. ROWID determines which data file, block, or row on which each record is in the oracle. (2) in repeated records, all columns may have the same contents, but the ROWID will not be the same, so as long as the ROWID is the largest in the duplicate record, the rest will be deleted. ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- As a Oracle database developer, you often need to create indexes for columns or columns of some tables, providing direct fast access to rows. But when you create, you often encounter such hints: ORA-01452: you cannot create unique indexes and find duplicate records. The Oracle system hints that a unique index cannot be created on the table because duplicate records are found in the system discovery table. A unique index can be created only if the duplicate record is found in the table and the record is deleted. Here, in table table_name, three different methods are introduced to determine duplicate records in a library table. [b]1.[/b][b] uses the grouping function to find duplicate rows [/b][b] in a table [/b] makes it easy to identify duplicate rows by using the grouping function GROUP select in the BY/HAVING statement. Suppose you want to create a unique index of the column column, column for group by group statistics and return the number of each group, if the group records more than 1, there is a duplicate line. The commands are as follows: SQLSelect column from table_name Group by column Having count (column) 1; This query method is simple and f