InlineParameterMapParser maintains the character string without the necessity
by using StringTokenizer.
---------------------------------------------------------------------------------------------------------
Key: IBATIS-549
URL: https://issues.apache.org/jira/browse/IBATIS-549
Project: iBatis for Java
Issue Type: Wish
Components: SQL Maps
Reporter: Takahiro Kido
Priority: Minor
To maintain the property value, InlineParameterMapParser uses StringTokenizer.
StringTokenizer#nextToken uses String#substring(int start, int end).
String#substring(int start, int end) uses String(int start, int length, char[]
data).
ex) INSERT INTO TABLE_A(COL_1,COL_2,....COL_N)
VALUES(#VALUE_1#,#VALUE_2#,....,#VALUE_N#);
char[] data keeps "INSERT INTO TABLE_A(COL_1,COL_2,....COL_N)
VALUES(#VALUE_1#,#VALUE_2#,....,#VALUE_N#)".
"INSERT INTO TABLE_A(COL_1,COL_2,....COL_N) VALUES("
This character string need not be maintained.
When String passed to StringTokenizer is made the character string since '#' of
the first token,
it is settled with less amount of the heap use.
char[] data keeps "#VALUE_1#,#VALUE_2#,....,#VALUE_N#)".
If the amount of the heap use is valued, you should make a new character string
since '#' of the first token with StringBuilder or StringBuffer beforehand.
Because a lot of heaps are consumed oppositely when it is tried to make a
character string new in each token, this attempt is not suitable.
The effect of reducing the amount of the heap use concerning select and insert
statement can be chiefly expected.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
