Populating Tables with New Data
Populating a table with data is simply the process of entering new data into a table, whether through a manual process using individual commands or through batch processes using programs or other related software.
Many factors can affect what data and how much data can be put into a table when populating tables with data. Some major factors include existing table constraints, the physical table size, column data types, the length of columns, and other integrity constraints, such as primary and foreign keys. The following sections help you learn the basics of inserting new data into a table, in addition to offering some Dos and Don'ts.
Note
Do not forget that SQL statements can be in upper- or lowercase. The data, depending on how it is stored in the database, is not case-sensitive. These examples use both lower- and uppercases just to show that it does not affect the outcome.
Inserting Data into a Table
Use the INSERT statement to insert new data into a table. There are a few options with the INSERT statement; look at the following basic syntax to begin:
insert into schema.table_name
VALUES ('value1', 'value2', [ NULL ] );Using this INSERT statement syntax, you must include every column in the specified table in the VALUES list. Notice that each value in this list is separated by a comma. The values inserted into the table must be enclosed by quotation marks for character and date data types. Quotation marks are not required for numeric data types or NULL values using the NULL keyword. A value should be present for each column in the table.
In the following example, you insert a new record into the PRODUCTS_TBL table.
Table structure:
products_tbl COLUMN Name Null? DATA Type ------------------------------ -------- ------------- PROD_ID NOT NULL VARCHAR2(10) PROD_DESC NOT NULL VARCHAR2(25) COST NOT NULL NUMBER(6,2)
INSERT INTO PRODUCTS_TBL VALUES ('7725','LEATHER GLOVES',24.99);
001 001 1 row created. 002 002
In this example, you insert three values into a table with three columns. The inserted values are in the same order as the columns listed in the table. The first two values are inserted using quotation marks, because the data types of the corresponding columns are of character type. The third value's associated column, COST, is a numeric data type and does not require quotation marks, although they can be used.
Note
The schema name, or table owner, has not been specified as part of the table name, as it was shown in the syntax. The schema name is not required if you are connected to the database as the user who owns the table.
Inserting Data into Limited Columns of a Table
There is a way you can insert data into a table's limited columns. For instance, suppose you want to insert all values for an employee except a pager number. You must, in this case, specify a column list as well as a VALUES list in your INSERT statement.
INSERT INTO EMPLOYEE_TBL (EMP_ID, LAST_NAME, FIRST_NAME, MIDDLE_NAME, ADDRESS, CITY, STATE, ZIP, PHONE) VALUES ('123456789', 'SMITH', 'JOHN', 'JAY', '12 BEACON CT', 'INDIANAPOLIS', 'IN', '46222', '3172996868'),
001 001 1 row created.
The syntax for inserting values into a limited number of columns in a table is as follows:
INSERT INTO SCHEMA TABLE_NAME ('COLUMN1', 'COLUMN2')
VALUES ('VALUE1', 'VALUE2'),You use ORDERS_TBL and insert values into only specified columns in the following example.
ORDERS_TBL COLUMN NAME Null? DATA TYPE ------------------------------ --------- ------------ ORD_NUM NOT NULL VARCHAR2(10) CUST_ID NOT NULL VARCHAR2(10) PROD_ID NOT NULL VARCHAR2(10) QTY NOT NULL NUMBER(4) ORD_DATE DATE
Sample INSERT statement:
insert into orders_tbl (ord_num,cust_id,prod_id,qty) values ('23A16','109','7725',2);
001 001 1 row created. 002 002
You have specified a column list enclosed by parentheses after the table name in the INSERT statement. You have listed all columns into which you want to insert data. ORD_DATE is the only excluded column. You can see, if you look at the table definition, that ORD_DATE does not require data for every record in the table. You know that ORD_DATE does not require data because NOT NULL is not specified in the table definition. NOT NULL tells us that NULL values are not allowed in the column. Furthermore, the list of values must appear in the order in which you want to insert them according to the column list.
Note
The column list in the INSERT statement does not have to reflect the same order of columns as in the definition of the associated table, but the list of values must be in the order of the associated columns in the column list.
Inserting Data from Another Table
You can insert data into a table based on the results of a query from another table using a combination of the INSERT statement and the SELECT statement. Briefly, a query is an inquiry to the database that expects data to be returned. See Hour 7 for more information on queries. A query is a question that the user asks the database, and the data returned is the answer. In the case of combining the INSERT statement with the SELECT statement, you are able to insert the data retrieved from a query into a table.
The syntax for inserting data from another table is
insert into schema.table_name [('column1', 'column2')]
select [*|('column1', 'column2')]
from table_name
[where condition(s)];
You see three new keywords in this syntax, which are covered here briefly. These keywords are SELECT, FROM, and WHERE. SELECT is the main command used to initiate a query in SQL. FROM is a clause in the query that specifies the names of tables in which the target data should be found. The WHERE clause, also part of the query, is used to place conditions on the query itself. An example condition may state: WHERE NAME = 'SMITH'. These three keywords are covered extensively during Hour 7 and Hour 8, "Using Operators to Categorize Data."
A condition is a way of placing criteria on data affected by a SQL statement.
The following example uses a simple query to view all data in the PRODUCTS_TBL table. SELECT * tells the database server that you want information on all columns of the table. Because there is no WHERE clause, you want to see all records in the table as well.
select * from products_tbl;
PROD_ID PROD_DESC COST ---------- ------------------------------ ----- 11235 WITCHES COSTUME 29.99 222 PLASTIC PUMPKIN 18 INCH 7.75 13 FALSE PARAFFIN TEETH 1.1 90 LIGHTED LANTERNS 14.5 15 ASSORTED COSTUMES 10 9 CANDY CORN 1.35 6 PUMPKIN CANDY 1.45 87 PLASTIC SPIDERS 1.05 119 ASSORTED MASKS 4.95 1234 KEY CHAIN 5.95 2345 OAK BOOKSHELF 59.99 11 rows selected.
Now, insert values into the PRODUCTS_TMP table based on the preceding query. You can see that 11 rows are created in the temporary table.
INSERT INTO PRODUCTS_TMP SELECT * FROM PRODUCTS_TBL;
001 001 11 rows created.
The following query shows all data in the PRODUCTS_TMP table that you just inserted:
SELECT * FROM PRODUCTS_TMP;
PROD_ID PROD_DESC COST ---------- ------------------------------ ----- 11235 WITCHES COSTUME 29.99 222 PLASTIC PUMPKIN 18 INCH 7.75 13 FALSE PARAFFIN TEETH 1.1 90 LIGHTED LANTERNS 14.5 15 ASSORTED COSTUMES 10 9 CANDY CORN 1.35 6 PUMPKIN CANDY 1.45 87 PLASTIC SPIDERS 1.05 119 ASSORTED MASKS 4.95 1234 KEY CHAIN 5.95 2345 OAK BOOKSHELF 59.99 11 rows selected.
Inserting NULL Values
Inserting a NULL value into a column of a table is a simple matter. You might want to insert a NULL value into a column if the value of the column in question is unknown. For instance, not every person carries a pager, so it would be inaccurate to enter an erroneous pager number—not to mention, you would not be budgeting space. A NULL value can be inserted into a column of a table using the keyword NULL.
The syntax for inserting a NULL value follows:
insert into schema.table_name values
('column1', NULL, 'column3'),The NULL keyword should be used in the associated column that exists in the table. That column will not have data in it for that row if you enter NULL. In the syntax, a NULL value is being entered in the place of COLUMN2.
Study the two following examples:
INSERT INTO ORDERS_TBL (ORD_NUM,CUST_ID,PROD_ID,QTY,ORD_DATE) VALUES ('23A16','109','7725',2,NULL);
001 001 1 row created.
In the first example, all columns in which to insert values are listed, which also happen to be every column in the ORDERS_TBL table. You insert a NULL value for the ORD_DATE column, meaning that you either do not know the order date, or there is no order date at this time.
INSERT INTO ORDERS_TBL VALUES ('23A16','109','7725',2, ''),
001 001 1 row created.
There are two differences from the first statement in the second example, but the results are the same. First, there is not a column list. Remember that a column list is not required if you are inserting data into all columns of a table. Second, instead of inserting the value NULL into the ORD_DATE column, you insert '' (two single quotation marks together), which also symbolizes a NULL value (because there is nothing between them) .