100 important knowledge points that SQL must master: using cursors

This lesson will teach you what a cursor is , How to use cursors .

21.1 The cursor
SQL The retrieval operation returns a set of rows called the result set , This group of returned lines are all related to SQL sentence
Matching lines （ Zero line to multiple lines ）. Simply use SELECT sentence , There is no way to get the number
a line 、 Next line or before 10 That's ok . But this is the relationship DBMS Functional components .

Result set （result set）
SQL The results retrieved by the query .
Sometimes , Need to move forward or backward one or more lines in the retrieved lines , This is the use of cursors
Where to go . The cursor （cursor） Is a store in DBMS Database query on server ,
It's not one SELECT sentence , But the result set retrieved by the statement . In storage
After cursor , The application can scroll or browse the data as needed .

explain ：SQLite Support
SQLite Supported cursors are called steps （step）, The basic concepts described in this lesson apply to
SQLite Steps for , But the grammar may be completely different .

Different DBMS Support different cursor options and features . Some common options and features are as follows .
   Can mark cursor as read-only , Enable data to be read , But you can't update and delete .
   Can control the directional operation that can be performed （ forward 、 backward 、 First of all 、 Last 、 Absolute position and
Relative position, etc ）.
   Can mark some columns as editable , Some columns are not editable .
   Prescribed scope , Make the cursor respond to the specific request to create it （ Like stored procedures ） Or for all requests
Accessible .
   instructions DBMS For the retrieved data （ Instead of pointing out the activity data in the table ） replicate ,
Keep data unchanged during cursor opening and access .
Cursors are mainly used for interactive applications , The user needs to scroll the data on the screen , And the data
Browse or make changes .

21.2 Use cursors
Using cursors involves several explicit steps .
   Before using the cursor , You must declare （ Definition ） it . This process does not actually retrieve data ,
It just defines what to use SELECT Statement and cursor options .
   Once declared , You must open the cursor for use . This process uses the SELECT
Statement to actually retrieve the data .
   For data filled cursors , Remove as needed （ retrieval ） All walks of life .
   When ending cursor usage , Cursor must be closed , If possible , Release cursor （ Rely on tools
Somatic DBMS）.
After declaring the cursor , The cursor can be opened and closed frequently as needed . When the cursor is open , Rootable
Perform fetch operations as often as needed .

21.2.1 Create cursors
Use DECLARE Statement create cursor , This statement is in different DBMS It's different .
DECLARE Named cursor , And define the corresponding SELECT sentence , Bring... As needed WHERE and
Other clauses . To illustrate , We create a cursor to retrieve all places without email addresses
There are customers , As part of the application , Help operators find out the vacant email
Address .
The following is how to create this cursor DB2、MariaDB、MySQL and SQL Server edition .
Input ▼
DECLARE CustCursor CURSOR
FOR
SELECT * FROM Customers
WHERE cust_email IS NULL;
Here is Oracle and PostgreSQL edition ：
Input ▼
DECLARE CURSOR CustCursor
IS
SELECT * FROM Customers
WHERE cust_email IS NULL;
analysis ▼
In the above two versions , DECLARE Statement is used to define and name cursors , Here for
CustCursor . SELECT Statement defines a containing no email address （ NULL value ）
Cursor of all customers .
After defining the cursor , You can open it .

21.2.2 Use cursors
Use OPEN CURSOR Statement open cursor , This statement is simple , In most DBMS
The syntax in is the same ：
Input ▼
OPEN CURSOR CustCursor
analysis ▼
Processing OPEN CURSOR When the sentence is , Execute the query , Store the retrieved data for browsing and
rolling .
Now we can use FETCH Statement accesses cursor data . FETCH Indicate which rows to retrieve , from
Where to retrieve them and where to put them （ Such as variable name ）. The first example uses Oracle
Syntax retrieves a row from the cursor （ first line ）：
Input ▼
DECLARE TYPE CustCursor IS REF CURSOR
RETURN Customers%ROWTYPE;
DECLARE CustRecord Customers%ROWTYPE
BEGIN
OPEN CustCursor;
FETCH CustCursor INTO CustRecord;
CLOSE CustCursor;
END;
analysis ▼
In this case , FETCH Used to retrieve the current line （ Automatically start with the first line ）, Put it in the statement
The variable of CustRecord in . The retrieved data will not be processed .
The next example （ Also used Oracle grammar ） in , From the first line to the last line , Cycle the retrieved data ：
Input ▼
DECLARE TYPE CustCursor IS REF CURSOR
RETURN Customers%ROWTYPE;
DECLARE CustRecord Customers%ROWTYPE
BEGIN
OPEN CustCursor;
LOOP
FETCH CustCursor INTO CustRecord;
EXIT WHEN CustCursor%NOTFOUND;
...
END LOOP;
CLOSE CustCursor;
END;
analysis ▼
As in the previous example , This example uses FETCH Retrieve the current row , Put it in a place called
CustRecord Variables in . But the difference is , there FETCH be located LOOP Inside , because
This is repeated . Code EXIT WHEN CustCursor%NOTFOUND Make it impossible to take out more
Terminate processing when there are more rows （ Exit loop ）. This example does not do the actual processing , Practical example
The ellipsis can be replaced by the specific processing code in the sub .
Here's another example , This use Microsoft SQL Server grammar ：
Input ▼
 

DECLARE @cust_id CHAR(10),
@cust_name CHAR(50),
@cust_address CHAR(50),
@cust_city CHAR(50),
@cust_state CHAR(5),
@cust_zip CHAR(10),
@cust_country CHAR(50),
@cust_contact CHAR(50),
@cust_email CHAR(255)
21.2  Use cursors  | |  201
OPEN CustCursor
FETCH NEXT FROM CustCursor
INTO @cust_id, @cust_name, @cust_address,
@cust_city, @cust_state, @cust_zip,
@cust_country, @cust_contact, @cust_email
...
WHILE @@FETCH_STATUS = 0
BEGIN
FETCH NEXT FROM CustCursor
INTO @cust_id, @cust_name, @cust_address,
@cust_city, @cust_state, @cust_zip,
@cust_country, @cust_contact, @cust_email
...
END
CLOSE CustCursor

analysis ▼
In this case , Declare a variable for each retrieved column , FETCH Statement to retrieve a row and save
Value into these variables . Use WHILE Cycle through each row , Conditions WHILE @@FETCH_
STATUS = 0 Terminate processing when no more rows can be fetched （ Exit loop ）. This example is not
Carry out specific treatment , In the actual code , You should replace them with specific processing code ... .
21.2.3 Close cursor
As mentioned in the previous examples , The cursor needs to be closed when it is used . Besides ,SQL Server
etc. DBMS It is required to explicitly release the resources occupied by the cursor . Here is DB2、Oracle and
PostgreSQL The grammar of .
Input ▼
CLOSE CustCursor
Here is Microsoft SQL Server Version of .

Input ▼
CLOSE CustCursor
DEALLOCATE CURSOR CustCursor
analysis ▼
CLOSE Statement to close the cursor . Once the cursor is closed , If you don't open it again , Will not make
use . The second time you use it, you don't need to declare , Just use OPEN Open it and .