解析 dbmsSql.sqlNULL 博文链接：https://hwhuang.iteye.com/

文件名称: 解析 dbmsSql.sql

所属分类: 其它

开发工具:

文件大小: 83kb

下载次数: 0

上传时间: 2019-03-17

提供者: weixin_********

下载 (83kb)

不能下载？报告错误

详细说明：NULL 博文链接：https://hwhuang.iteye.com/blog/554177二一 RULES AND LIMITATIONS Bind variables of a SQL statement are identified by their names. When binding a value to a hind variable, the string identifying the bind variable in the statement may optionally contain the eading colon. For example, if the parsed sQl statement is SELECT ENAME FROM EMP WHERE SAL > X", on binding the variable to a value, it can be identified using either of the strings: X and X Columns of the row being selected in a select statement are identified by their relative positions (1, 2, 3,.)as they appear on the select list from left to right. Privileges are associated with the caller of the procedures/functions in this package as follows If the caller is an anonymous PL/SQL block, the procedures/ functions are run using the privileges of the current user. If the caller is a stored procedure, the procedures/ functions are run using the privileges of the owner of the stored procedure WARNING: Using the package to dynamically execute DDL statements can results in the program hanging For example, a call to a procedure in a - package will result in the package being locked until the execution returns to the user side. Any operation that results in a conflicting lock, such as dynamically trying to drop the package, before the first lock is released will result in a hang The flow of procedure calls will typically look like this open cursor parse bind variabl yes no execute define column able value execute etch rows column valuc I close cursor I CONSTANTS v6 constant integer =0 native constant integer I v7 cOnstant integer: =2 TYPES type varchar2a is table of varchar 2(32767) index by binary integer bug 2410688: for users who require larger than varchar2(256) this type has been introduced together with parse overloads hat take this type. type varchar2s is table of varchar2(256) index by binary integer Note that with the introduction of varchar 2a we will deprecate this type, with phase out over a number of releases. For DateTime types, the field col scale is used to denote the fractional seconds precision. For Interval types, the field col precision is used to denote the leading field precision and the field col scale is used to denote the fractional seconds precision. type desc rec is record binary integer =0 col max len binary integer: =0 CoI name varchar col name len binary integer =0 col schema name archar2(32) .=,, col schema name len binary integer: =0 col precision binary integer :0 col scale binary integer =0 col charsets binary integer =0, col charsetform binary integer :=0 col null ok ocean TRUE type desc tab is table of desc rec index by binary integer bug 702903 reveals that col name can be of any length, not just 32 which can be resolved by changing the maximum size above from 32 to 32767. However, this will affect the signature of the package and to avoid that side effect, the current APl describe columns is left unchanged but a new Pi describe columns is added at the end of this package specification he new APi relies on a table type desc tab2 whose array element is a new record type desc rec2, and desc rec2 contains the variable col name with a - maximum size of 32.767. If the original api describe co lumns is used and col name encounters an overflow, an error will be raised type desc rec2 is record col type binary integer : 0, col max len binary integer =0, col name archar2(32767):= co name en binary integer:=0 col schema name varchar2 (32) col schema name len binary integer =0, col precision binary integer -0 co scale binary integer =0 col charsetid binary integer :=0 col charsetform binary integer :0 col null ok boolean TRUE) type desc tab2 is table of desc rec 2 index by binary integer Bulk sql types type Number table is table of number Index by binary integer type Varchar2 Table is table of varchar2(2000) index by binary integer type date Table is table of date index by binary integer type blob table is table of blob index by binary integer type Clob Table is table of clob index by binary integer type bfile table is table of bfile ndex by binary integer TYPE Urowid Table Is table Of urowid INdEX BY binary integer TYPE time table IS tabLe oF time unconstrained NDEⅩBY binary integer TYPe timestamp table Is table or timestamp unconstrained INDEX BY binary integer TYPe time with time zone Tablc IS TABLE OF TIME TZ UNCONSTRAINED INDEX BY binary integer tYPe timestamp with time zone table Is table oF TIMESTAMP TZ UNCONSTRAINED INDEX BY binary integer TYPE timestamp with ltz Table Is TABLE OF TIMESTAMP LTZ UNCONSTRAiNED INDEX BY binary integer TYPE interval year to MONTII Table Is TAble or yminterval unconstrained INdeX BY binary integer TYPE interval day to second table is table of dsinterval unconstrained INDEX BY binary integer type binary float Table is table of binary float index by binary integer type binary double table is table of binary double index by binary integer type Cfile table is table of cfile index by binary integer EXCEPTIONS Inconsistent type exception pragma exception init(inconsistent type, -6562) This exception is raised by procedure column value or variable value if the type of the given out argument where to put the requested value is di fferent from the type of the value PROCEDURES AND FUNCTIONS function open cursor return integer pragma restrict references(open cursor, RNDS, WNDS) Open a new cursor. When no longer needed, this cursor must be closed explicitly by Iling "close cursor Return value Cursor id number of the new cursor function is open(c in integer) return boolean pragma restrict references(is open, RNDS, WNDS) Return true is the given cursor is currently open Input parameters Cursor id number of the cursor to check Return value TRUE if the given cursor is open, tALSE if it is not procedure close cursor(c in out integer) pragma restrict references(close cursor, RNDS, WNDS) Close the given cursor. Input parameters Cursor id number of the cursor to close Output parameters Will be nulled procedure parse(c in integer, statement in varchar2 language flag in integer) Parse the given statement in the given cursor. NOTE THAT PARSING AND - EXECUTING DDL STATEMENTS CAN CALSE HANGS Currently, the deferred parsing feature of the Oracle Call Interface is not used. As a result, statements are parsed immediately. In addition DDL statements are executed immediately when parsed. However, the behavior may change in the future so that the actual parsing (and execution of DDL statement) do not occur until the cursor is executed with execute DO NOT RELY ON THE CURRENT TIMING OF THE ACTUAL PARSING Input parameters: Cursor id number of the cursor in where to parse the statement statement Statement to parse. language flag Specifies behavior for statement. Valid values are v6, v7 and NATIVE v6 and v7 specifies behavior according to Version 6 and ORACLE7, respectively. NATiVe specifies behavior according to the version of the database the program is connected to. procedure parse(c in integer, statement in varchar2a Ib in integer, ub in integer lfflg in boolean, language flag in integer) Description:(copied from parse for varchar2s) Parse the given statement in the given cursor. The statement is not in one piece but resides in little pieces in the PL/SQL table statement Conceptually what happens is that the SQL string is put together as follows String: statement(1b) statement(1b+ 1) statement (ub) Then a regular parse follows. If lfflg is true then a newline is inserted after each piece For further information and for documentation on the rest of the arguments see the regular parse procedure below. procedure parse(c in integer, statement in varchar2 Ib in integer, ub in integer Ifflg in boolean, language flag in integer) Parse the given statement in the given cursor. The statement is not in one piece but resides in little piccos in the PL /SQL table statement Conceptually what happens is that the SQL string is put together as follows String: statement(1b) statement(1b+1) statement(ub) Then a regular parse follows. If"lfflg is true then a newline is inserted after cach piece For further information and for documentation on the rest of the arguments see the regular parse procedure below. procedure bind variable(c in integer, name in varchar2, value in number) pragma restrict references(bind variable, WNDS procedure bind variable(c in integer, name in varchar2, value in varchar2 character set any cs) pragma restrict references (bind variable, WNDS) procedure bind variable(c in integer, name in varchar2 value in varchar2 character set any Cs It value size in integer) pragma restrict references(bind variable, WNDS) procedure bind variable(c in integer, name in varchar2, value in date) pragma restrict references(bind variable, WNDS) procedure bind variable(c in integer, name in varchar2, value in blob pragma restrict references (bind variable, WNDS) procedure bind variable(c in integer, name in varchar 2, value in clob character set any cs) pragma restrict references(bind variable, WNDS procedure bind variable(c in integer, name in varchar2, value in bfile pragma restrict references (bind variable, WNDS procedure bind variable char(c in integer, name in varchar2 lue in char character set any cs) pragma restrict references(bind variable char, WNDS) procedure bind variable char(c in integer, name in varchar2 value in char character set any cs, out value size in integer) pragma restrict references(bind variable char, WNDS procedure bind variable raw(c in integer, name in varchar2 value in raw) gma restrict references(bind variable raw, WNDS) procedure bind variable raw(c in integer, name in varchar 2 value in raw, out value size in integer) pragma restrict references (bind variable raw, WNDS) procedure bind variable rowid(c in integer, name in varchar2 value in rowid) pragma restrict references(bind variable rowid, WNDS) procedure bind array(c in integer, name in varchar2 n tab in Number Table) pragma restrict references (bind array, WNDS) procedure bind array(c in integer, name in varchar2 c tab in Varchar2 Table) pragma restrict references(bind array, WNDS) procedure bind array(c in integer, name in varchar2 d tab in Date Table pragma restrict references(bind array, WNDS procedure bind array(c in integer, name in varchar2 b1 tab in Blob Table) pragma restrict references(bind array, WNDS procedure bind array(c in integer, name in varchar2 cl tab in Clob Table pragma restrict references(bind array, WNDS procedure bind array(c in integer, name in varchar2 bf tab in Bfile Table) pragma restrict references(bind array, WNDS) procedure bind array(c in integer, name in varchar2 n tab in Number Table index in integer, index 2 in integer pragma restrict references(bind array, WNDS procedure bind array(c in integer, name in varchar2 c tab in varchar2 tabl dexi in integer dex2 in integer); pragma restrict references (bind array, WNDS) procedure bind array(c in integer, name in varchar2 d tab in date Table index1 in integer, index2 in integer) pragma restrict references(bind array, WNDS procedure bind array(c in integer, name in varchar2 bl tab in blob table indexl in integer, index2 in integer pragma restrict references(bind array, WNDS) procedure bind array(c in integer, name in varchar2 cl tab in Clob Table index1 in integer, index2 in integer) pragma restrict references (bind array, WNDS procedure bind array(c in integer, name in varchar2 bf tab in Bfile Table, index in integer, index2 in integer) pragma restrict references (bind array, WNDS Bind the given value to the variable identified by its name in the parsed statement in the given cursor. If the variable is an in or in/out variable, the given bind value should be a valid one. If the variable is an out variable, the given bind value is ignored Input parameters Cursor id number of the cursor to bind name

(系统自动生成,下载前可以参看下载内容)