Google Cloud Bigtable与Google Cloud Datastore(Google Cloud Bigtable vs Google Cloud Datastore)

 
 
 v_cnt未初始化，即在第一次调用dbms_output.put_line IS NULL ，因此您在输出中看到一个空行而不是行号。 
 
 
 最后一次获取不执行任何操作，因为游标中没有更多行，因此变量i, a, j在最后一次获取后不会更改。 您可以根据需要获取，但是在从游标中取出最后一行之后， FETCH INTO运算符不会生成新数据。 
 
 
 为了得到你需要的东西（标准循环行为），抓取循环应该被组织为 
 
LOOP
    FETCH c1 INTO i, a, j;
    EXIT WHEN c1%NOTFOUND;
    v_cnt := c1%ROWCOUNT; 

    dbms_output.put_line(i || ' ' || a || ' SALARY IS ' || j || ' row count is ' || v_cnt);

END LOOP;
 
 为了避免与获取光标相关，可以使用“for loop”来完成。 在大多数情况下，这是一个合适的解决方案，绝对是您在问题中提供的示例。 
 
FOR l_rec IN c1
LOOP

  i := l_rec.first_name;
  a := l_rec.last_name;
  j := l_rec.salary;
  v_cnt := c1%ROWCOUNT;

  dbms_output.put_line(i || ' ' || a || ' SALARY IS ' || j || ' row count is ' || v_cnt);

END LOOP;

 
 
v_cnt is not initialized i.e. IS NULL on the first call of dbms_output.put_line, so you see an empty line instead of row number in the output.
 
 
The last fetch makes no action, because there are no more rows in cursor, so variables i, a, j doesn't change after last fetch. You may fetch as long as you want, but after the last row has been fetched out of cursor the FETCH INTO operator produces no new data.
 
 
To get what you need (standard loop behavior) the fetching loop shall be organized as
 
LOOP
    FETCH c1 INTO i, a, j;
    EXIT WHEN c1%NOTFOUND;
    v_cnt := c1%ROWCOUNT; 

    dbms_output.put_line(i || ' ' || a || ' SALARY IS ' || j || ' row count is ' || v_cnt);

END LOOP;
 
To avoid compliactions with fetching a cursor you can do it using "for loop". This is an appropriate solution in the most cases, and definitely in the case of example you gave in the question.
 
FOR l_rec IN c1
LOOP

  i := l_rec.first_name;
  a := l_rec.last_name;
  j := l_rec.salary;
  v_cnt := c1%ROWCOUNT;

  dbms_output.put_line(i || ' ' || a || ' SALARY IS ' || j || ' row count is ' || v_cnt);

END LOOP;