Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
/* -*- C++ -*- */
|
|
|
|
/* Copyright (C) 2002 MySQL AB
|
|
|
|
|
|
|
|
This program is free software; you can redistribute it and/or modify
|
|
|
|
it under the terms of the GNU General Public License as published by
|
2006-12-23 20:17:15 +01:00
|
|
|
the Free Software Foundation; version 2 of the License.
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
|
|
|
|
This program is distributed in the hope that it will be useful,
|
|
|
|
but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
|
|
GNU General Public License for more details.
|
|
|
|
|
|
|
|
You should have received a copy of the GNU General Public License
|
|
|
|
along with this program; if not, write to the Free Software
|
|
|
|
Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */
|
|
|
|
|
|
|
|
#ifndef _SP_PCONTEXT_H_
|
|
|
|
#define _SP_PCONTEXT_H_
|
|
|
|
|
2005-05-27 12:03:37 +02:00
|
|
|
#ifdef USE_PRAGMA_INTERFACE
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
#pragma interface /* gcc class implementation */
|
|
|
|
#endif
|
|
|
|
|
|
|
|
typedef enum
|
|
|
|
{
|
|
|
|
sp_param_in,
|
|
|
|
sp_param_out,
|
|
|
|
sp_param_inout
|
|
|
|
} sp_param_mode_t;
|
|
|
|
|
2006-04-07 16:53:15 +02:00
|
|
|
typedef struct sp_variable
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
{
|
2003-03-26 12:29:58 +01:00
|
|
|
LEX_STRING name;
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
enum enum_field_types type;
|
|
|
|
sp_param_mode_t mode;
|
2005-12-07 15:01:17 +01:00
|
|
|
|
|
|
|
/*
|
2006-04-07 16:53:15 +02:00
|
|
|
offset -- this the index to the variable's value in the runtime frame.
|
|
|
|
This is calculated during parsing and used when creating sp_instr_set
|
|
|
|
instructions and Item_splocal items.
|
|
|
|
I.e. values are set/referred by array indexing in runtime.
|
2005-12-07 15:01:17 +01:00
|
|
|
*/
|
|
|
|
uint offset;
|
|
|
|
|
2004-04-06 15:48:58 +02:00
|
|
|
Item *dflt;
|
2007-06-10 12:43:57 +02:00
|
|
|
Create_field field_def;
|
2006-04-07 16:53:15 +02:00
|
|
|
} sp_variable_t;
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
|
2004-08-17 20:20:58 +02:00
|
|
|
|
2006-04-18 11:07:34 +02:00
|
|
|
#define SP_LAB_IMPL 0 // Implicit label generated by parser
|
|
|
|
#define SP_LAB_BEGIN 1 // Label at BEGIN
|
|
|
|
#define SP_LAB_ITER 2 // Label at iteration control
|
2004-08-17 20:20:58 +02:00
|
|
|
|
2006-01-16 15:22:29 +01:00
|
|
|
/*
|
|
|
|
An SQL/PSM label. Can refer to the identifier used with the
|
|
|
|
"label_name:" construct which may precede some SQL/PSM statements, or
|
|
|
|
to an implicit implementation-dependent identifier which the parser
|
|
|
|
inserts before a high-level flow control statement such as
|
|
|
|
IF/WHILE/REPEAT/LOOP, when such statement is rewritten into
|
|
|
|
a combination of low-level jump/jump_if instructions and labels.
|
|
|
|
*/
|
|
|
|
|
2002-12-16 15:40:44 +01:00
|
|
|
typedef struct sp_label
|
2002-12-11 14:24:29 +01:00
|
|
|
{
|
|
|
|
char *name;
|
|
|
|
uint ip; // Instruction index
|
2004-08-17 20:20:58 +02:00
|
|
|
int type; // begin/iter or ref/free
|
2004-11-17 11:45:05 +01:00
|
|
|
sp_pcontext *ctx; // The label's context
|
2002-12-11 14:24:29 +01:00
|
|
|
} sp_label_t;
|
|
|
|
|
2003-09-16 14:26:08 +02:00
|
|
|
typedef struct sp_cond_type
|
|
|
|
{
|
|
|
|
enum { number, state, warning, notfound, exception } type;
|
2009-09-10 11:18:29 +02:00
|
|
|
char sqlstate[SQLSTATE_LENGTH+1];
|
2003-09-16 14:26:08 +02:00
|
|
|
uint mysqlerr;
|
|
|
|
} sp_cond_type_t;
|
|
|
|
|
2006-04-07 16:53:15 +02:00
|
|
|
/*
|
|
|
|
Sanity check for SQLSTATEs. Will not check if it's really an existing
|
|
|
|
state (there are just too many), but will check length bad characters.
|
|
|
|
*/
|
2005-02-28 18:07:06 +01:00
|
|
|
extern bool
|
|
|
|
sp_cond_check(LEX_STRING *sqlstate);
|
|
|
|
|
2003-09-16 14:26:08 +02:00
|
|
|
typedef struct sp_cond
|
|
|
|
{
|
|
|
|
LEX_STRING name;
|
|
|
|
sp_cond_type_t *val;
|
|
|
|
} sp_cond_t;
|
|
|
|
|
Bug#26503 (Illegal SQL exception handler code causes the server to crash)
Before this fix, the parser would accept illegal code in SQL exceptions
handlers, that later causes the runtime to crash when executing the code,
due to memory violations in the exception handler stack.
The root cause of the problem is instructions within an exception handler
that jumps to code located outside of the handler. This is illegal according
to the SQL 2003 standard, since labels located outside the handler are not
supposed to be visible (they are "out of scope"), so any instruction that
jumps to these labels, like ITERATE or LEAVE, should not parse.
The section of the standard that is relevant for this is :
SQL:2003 SQL/PSM (ISO/IEC 9075-4:2003)
section 13.1 <compound statement>,
syntax rule 4
<quote>
The scope of the <beginning label> is CS excluding every <SQL schema
statement> contained in CS and excluding every
<local handler declaration list> contained in CS. <beginning label> shall
not be equivalent to any other <beginning label>s within that scope.
</quote>
With this fix, the C++ class sp_pcontext, which represent the "parsing
context" tree (a.k.a symbol table) of a stored procedure, has been changed
as follows:
- constructors have been cleaned up, so that only building a root node for
the tree is public; building nodes inside a tree is not public.
- a new member, m_label_scope, indicates if a given syntactic context
belongs to a DECLARE HANDLER block,
- label resolution, in the method find_label(), has been changed to
implement the restriction of scope regarding labels used in a compound
statement.
The actions in the parser, when parsing the body of a SQL exception handler,
have been changed as follows:
- the implementation of an exception handler (DECLARE HANDLER) now creates
explicitly a new sp_pcontext, to isolate the code inside the handler from
the containing compound statement context.
- registering exception handlers as a result occurs in the parent context,
see the rule sp_hcond_element
- the code in sp_hcond_list has been cleaned up, to avoid code duplication
In addition, the flags IN_SIMPLE_CASE and IN_HANDLER, declared in sp_head.h
have been removed, since they are unused and broken by design (as seen with
Bug 19194 (Right recursion in parser for CASE causes excessive stack usage,
limitation), representing a stack in a single flag is not possible.
Tests in sp-error have been added to show that illegal constructs are now
rejected.
Tests in sp have been added for code coverage, to show that ITERATE or LEAVE
statements are legal when jumping to a label in scope, inside the body of
an exception handler.
2007-03-14 19:02:32 +01:00
|
|
|
/**
|
|
|
|
The scope of a label in Stored Procedures,
|
|
|
|
for name resolution of labels in a parsing context.
|
|
|
|
*/
|
|
|
|
enum label_scope_type
|
|
|
|
{
|
|
|
|
/**
|
|
|
|
The labels declared in a parent context are in scope.
|
|
|
|
*/
|
|
|
|
LABEL_DEFAULT_SCOPE,
|
|
|
|
/**
|
|
|
|
The labels declared in a parent context are not in scope.
|
|
|
|
*/
|
|
|
|
LABEL_HANDLER_SCOPE
|
|
|
|
};
|
2005-08-25 15:34:34 +02:00
|
|
|
|
Bug#26503 (Illegal SQL exception handler code causes the server to crash)
Before this fix, the parser would accept illegal code in SQL exceptions
handlers, that later causes the runtime to crash when executing the code,
due to memory violations in the exception handler stack.
The root cause of the problem is instructions within an exception handler
that jumps to code located outside of the handler. This is illegal according
to the SQL 2003 standard, since labels located outside the handler are not
supposed to be visible (they are "out of scope"), so any instruction that
jumps to these labels, like ITERATE or LEAVE, should not parse.
The section of the standard that is relevant for this is :
SQL:2003 SQL/PSM (ISO/IEC 9075-4:2003)
section 13.1 <compound statement>,
syntax rule 4
<quote>
The scope of the <beginning label> is CS excluding every <SQL schema
statement> contained in CS and excluding every
<local handler declaration list> contained in CS. <beginning label> shall
not be equivalent to any other <beginning label>s within that scope.
</quote>
With this fix, the C++ class sp_pcontext, which represent the "parsing
context" tree (a.k.a symbol table) of a stored procedure, has been changed
as follows:
- constructors have been cleaned up, so that only building a root node for
the tree is public; building nodes inside a tree is not public.
- a new member, m_label_scope, indicates if a given syntactic context
belongs to a DECLARE HANDLER block,
- label resolution, in the method find_label(), has been changed to
implement the restriction of scope regarding labels used in a compound
statement.
The actions in the parser, when parsing the body of a SQL exception handler,
have been changed as follows:
- the implementation of an exception handler (DECLARE HANDLER) now creates
explicitly a new sp_pcontext, to isolate the code inside the handler from
the containing compound statement context.
- registering exception handlers as a result occurs in the parent context,
see the rule sp_hcond_element
- the code in sp_hcond_list has been cleaned up, to avoid code duplication
In addition, the flags IN_SIMPLE_CASE and IN_HANDLER, declared in sp_head.h
have been removed, since they are unused and broken by design (as seen with
Bug 19194 (Right recursion in parser for CASE causes excessive stack usage,
limitation), representing a stack in a single flag is not possible.
Tests in sp-error have been added to show that illegal constructs are now
rejected.
Tests in sp have been added for code coverage, to show that ITERATE or LEAVE
statements are legal when jumping to a label in scope, inside the body of
an exception handler.
2007-03-14 19:02:32 +01:00
|
|
|
/**
|
|
|
|
The parse-time context, used to keep track of declared variables/parameters,
|
2006-04-07 16:53:15 +02:00
|
|
|
conditions, handlers, cursors and labels, during parsing.
|
|
|
|
sp_contexts are organized as a tree, with one object for each begin-end
|
Bug#26503 (Illegal SQL exception handler code causes the server to crash)
Before this fix, the parser would accept illegal code in SQL exceptions
handlers, that later causes the runtime to crash when executing the code,
due to memory violations in the exception handler stack.
The root cause of the problem is instructions within an exception handler
that jumps to code located outside of the handler. This is illegal according
to the SQL 2003 standard, since labels located outside the handler are not
supposed to be visible (they are "out of scope"), so any instruction that
jumps to these labels, like ITERATE or LEAVE, should not parse.
The section of the standard that is relevant for this is :
SQL:2003 SQL/PSM (ISO/IEC 9075-4:2003)
section 13.1 <compound statement>,
syntax rule 4
<quote>
The scope of the <beginning label> is CS excluding every <SQL schema
statement> contained in CS and excluding every
<local handler declaration list> contained in CS. <beginning label> shall
not be equivalent to any other <beginning label>s within that scope.
</quote>
With this fix, the C++ class sp_pcontext, which represent the "parsing
context" tree (a.k.a symbol table) of a stored procedure, has been changed
as follows:
- constructors have been cleaned up, so that only building a root node for
the tree is public; building nodes inside a tree is not public.
- a new member, m_label_scope, indicates if a given syntactic context
belongs to a DECLARE HANDLER block,
- label resolution, in the method find_label(), has been changed to
implement the restriction of scope regarding labels used in a compound
statement.
The actions in the parser, when parsing the body of a SQL exception handler,
have been changed as follows:
- the implementation of an exception handler (DECLARE HANDLER) now creates
explicitly a new sp_pcontext, to isolate the code inside the handler from
the containing compound statement context.
- registering exception handlers as a result occurs in the parent context,
see the rule sp_hcond_element
- the code in sp_hcond_list has been cleaned up, to avoid code duplication
In addition, the flags IN_SIMPLE_CASE and IN_HANDLER, declared in sp_head.h
have been removed, since they are unused and broken by design (as seen with
Bug 19194 (Right recursion in parser for CASE causes excessive stack usage,
limitation), representing a stack in a single flag is not possible.
Tests in sp-error have been added to show that illegal constructs are now
rejected.
Tests in sp have been added for code coverage, to show that ITERATE or LEAVE
statements are legal when jumping to a label in scope, inside the body of
an exception handler.
2007-03-14 19:02:32 +01:00
|
|
|
block, one object for each exception handler,
|
|
|
|
plus a root-context for the parameters.
|
2006-04-07 16:53:15 +02:00
|
|
|
This is used during parsing for looking up defined names (e.g. declared
|
|
|
|
variables and visible labels), for error checking, and to calculate offsets
|
|
|
|
to be used at runtime. (During execution variable values, active handlers
|
|
|
|
and cursors, etc, are referred to by an index in a stack.)
|
Bug#26503 (Illegal SQL exception handler code causes the server to crash)
Before this fix, the parser would accept illegal code in SQL exceptions
handlers, that later causes the runtime to crash when executing the code,
due to memory violations in the exception handler stack.
The root cause of the problem is instructions within an exception handler
that jumps to code located outside of the handler. This is illegal according
to the SQL 2003 standard, since labels located outside the handler are not
supposed to be visible (they are "out of scope"), so any instruction that
jumps to these labels, like ITERATE or LEAVE, should not parse.
The section of the standard that is relevant for this is :
SQL:2003 SQL/PSM (ISO/IEC 9075-4:2003)
section 13.1 <compound statement>,
syntax rule 4
<quote>
The scope of the <beginning label> is CS excluding every <SQL schema
statement> contained in CS and excluding every
<local handler declaration list> contained in CS. <beginning label> shall
not be equivalent to any other <beginning label>s within that scope.
</quote>
With this fix, the C++ class sp_pcontext, which represent the "parsing
context" tree (a.k.a symbol table) of a stored procedure, has been changed
as follows:
- constructors have been cleaned up, so that only building a root node for
the tree is public; building nodes inside a tree is not public.
- a new member, m_label_scope, indicates if a given syntactic context
belongs to a DECLARE HANDLER block,
- label resolution, in the method find_label(), has been changed to
implement the restriction of scope regarding labels used in a compound
statement.
The actions in the parser, when parsing the body of a SQL exception handler,
have been changed as follows:
- the implementation of an exception handler (DECLARE HANDLER) now creates
explicitly a new sp_pcontext, to isolate the code inside the handler from
the containing compound statement context.
- registering exception handlers as a result occurs in the parent context,
see the rule sp_hcond_element
- the code in sp_hcond_list has been cleaned up, to avoid code duplication
In addition, the flags IN_SIMPLE_CASE and IN_HANDLER, declared in sp_head.h
have been removed, since they are unused and broken by design (as seen with
Bug 19194 (Right recursion in parser for CASE causes excessive stack usage,
limitation), representing a stack in a single flag is not possible.
Tests in sp-error have been added to show that illegal constructs are now
rejected.
Tests in sp have been added for code coverage, to show that ITERATE or LEAVE
statements are legal when jumping to a label in scope, inside the body of
an exception handler.
2007-03-14 19:02:32 +01:00
|
|
|
Parsing contexts for exception handlers limit the visibility of labels.
|
2006-04-07 16:53:15 +02:00
|
|
|
The pcontext tree is also kept during execution and is used for error
|
|
|
|
checking (e.g. correct number of parameters), and in the future, used by
|
|
|
|
the debugger.
|
2005-08-25 15:34:34 +02:00
|
|
|
*/
|
|
|
|
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
class sp_pcontext : public Sql_alloc
|
|
|
|
{
|
Bug#26503 (Illegal SQL exception handler code causes the server to crash)
Before this fix, the parser would accept illegal code in SQL exceptions
handlers, that later causes the runtime to crash when executing the code,
due to memory violations in the exception handler stack.
The root cause of the problem is instructions within an exception handler
that jumps to code located outside of the handler. This is illegal according
to the SQL 2003 standard, since labels located outside the handler are not
supposed to be visible (they are "out of scope"), so any instruction that
jumps to these labels, like ITERATE or LEAVE, should not parse.
The section of the standard that is relevant for this is :
SQL:2003 SQL/PSM (ISO/IEC 9075-4:2003)
section 13.1 <compound statement>,
syntax rule 4
<quote>
The scope of the <beginning label> is CS excluding every <SQL schema
statement> contained in CS and excluding every
<local handler declaration list> contained in CS. <beginning label> shall
not be equivalent to any other <beginning label>s within that scope.
</quote>
With this fix, the C++ class sp_pcontext, which represent the "parsing
context" tree (a.k.a symbol table) of a stored procedure, has been changed
as follows:
- constructors have been cleaned up, so that only building a root node for
the tree is public; building nodes inside a tree is not public.
- a new member, m_label_scope, indicates if a given syntactic context
belongs to a DECLARE HANDLER block,
- label resolution, in the method find_label(), has been changed to
implement the restriction of scope regarding labels used in a compound
statement.
The actions in the parser, when parsing the body of a SQL exception handler,
have been changed as follows:
- the implementation of an exception handler (DECLARE HANDLER) now creates
explicitly a new sp_pcontext, to isolate the code inside the handler from
the containing compound statement context.
- registering exception handlers as a result occurs in the parent context,
see the rule sp_hcond_element
- the code in sp_hcond_list has been cleaned up, to avoid code duplication
In addition, the flags IN_SIMPLE_CASE and IN_HANDLER, declared in sp_head.h
have been removed, since they are unused and broken by design (as seen with
Bug 19194 (Right recursion in parser for CASE causes excessive stack usage,
limitation), representing a stack in a single flag is not possible.
Tests in sp-error have been added to show that illegal constructs are now
rejected.
Tests in sp have been added for code coverage, to show that ITERATE or LEAVE
statements are legal when jumping to a label in scope, inside the body of
an exception handler.
2007-03-14 19:02:32 +01:00
|
|
|
public:
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
|
Bug#26503 (Illegal SQL exception handler code causes the server to crash)
Before this fix, the parser would accept illegal code in SQL exceptions
handlers, that later causes the runtime to crash when executing the code,
due to memory violations in the exception handler stack.
The root cause of the problem is instructions within an exception handler
that jumps to code located outside of the handler. This is illegal according
to the SQL 2003 standard, since labels located outside the handler are not
supposed to be visible (they are "out of scope"), so any instruction that
jumps to these labels, like ITERATE or LEAVE, should not parse.
The section of the standard that is relevant for this is :
SQL:2003 SQL/PSM (ISO/IEC 9075-4:2003)
section 13.1 <compound statement>,
syntax rule 4
<quote>
The scope of the <beginning label> is CS excluding every <SQL schema
statement> contained in CS and excluding every
<local handler declaration list> contained in CS. <beginning label> shall
not be equivalent to any other <beginning label>s within that scope.
</quote>
With this fix, the C++ class sp_pcontext, which represent the "parsing
context" tree (a.k.a symbol table) of a stored procedure, has been changed
as follows:
- constructors have been cleaned up, so that only building a root node for
the tree is public; building nodes inside a tree is not public.
- a new member, m_label_scope, indicates if a given syntactic context
belongs to a DECLARE HANDLER block,
- label resolution, in the method find_label(), has been changed to
implement the restriction of scope regarding labels used in a compound
statement.
The actions in the parser, when parsing the body of a SQL exception handler,
have been changed as follows:
- the implementation of an exception handler (DECLARE HANDLER) now creates
explicitly a new sp_pcontext, to isolate the code inside the handler from
the containing compound statement context.
- registering exception handlers as a result occurs in the parent context,
see the rule sp_hcond_element
- the code in sp_hcond_list has been cleaned up, to avoid code duplication
In addition, the flags IN_SIMPLE_CASE and IN_HANDLER, declared in sp_head.h
have been removed, since they are unused and broken by design (as seen with
Bug 19194 (Right recursion in parser for CASE causes excessive stack usage,
limitation), representing a stack in a single flag is not possible.
Tests in sp-error have been added to show that illegal constructs are now
rejected.
Tests in sp have been added for code coverage, to show that ITERATE or LEAVE
statements are legal when jumping to a label in scope, inside the body of
an exception handler.
2007-03-14 19:02:32 +01:00
|
|
|
/**
|
|
|
|
Constructor.
|
|
|
|
Builds a parsing context root node.
|
|
|
|
*/
|
|
|
|
sp_pcontext();
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
|
2003-04-02 20:42:28 +02:00
|
|
|
// Free memory
|
|
|
|
void
|
|
|
|
destroy();
|
|
|
|
|
Bug#26503 (Illegal SQL exception handler code causes the server to crash)
Before this fix, the parser would accept illegal code in SQL exceptions
handlers, that later causes the runtime to crash when executing the code,
due to memory violations in the exception handler stack.
The root cause of the problem is instructions within an exception handler
that jumps to code located outside of the handler. This is illegal according
to the SQL 2003 standard, since labels located outside the handler are not
supposed to be visible (they are "out of scope"), so any instruction that
jumps to these labels, like ITERATE or LEAVE, should not parse.
The section of the standard that is relevant for this is :
SQL:2003 SQL/PSM (ISO/IEC 9075-4:2003)
section 13.1 <compound statement>,
syntax rule 4
<quote>
The scope of the <beginning label> is CS excluding every <SQL schema
statement> contained in CS and excluding every
<local handler declaration list> contained in CS. <beginning label> shall
not be equivalent to any other <beginning label>s within that scope.
</quote>
With this fix, the C++ class sp_pcontext, which represent the "parsing
context" tree (a.k.a symbol table) of a stored procedure, has been changed
as follows:
- constructors have been cleaned up, so that only building a root node for
the tree is public; building nodes inside a tree is not public.
- a new member, m_label_scope, indicates if a given syntactic context
belongs to a DECLARE HANDLER block,
- label resolution, in the method find_label(), has been changed to
implement the restriction of scope regarding labels used in a compound
statement.
The actions in the parser, when parsing the body of a SQL exception handler,
have been changed as follows:
- the implementation of an exception handler (DECLARE HANDLER) now creates
explicitly a new sp_pcontext, to isolate the code inside the handler from
the containing compound statement context.
- registering exception handlers as a result occurs in the parent context,
see the rule sp_hcond_element
- the code in sp_hcond_list has been cleaned up, to avoid code duplication
In addition, the flags IN_SIMPLE_CASE and IN_HANDLER, declared in sp_head.h
have been removed, since they are unused and broken by design (as seen with
Bug 19194 (Right recursion in parser for CASE causes excessive stack usage,
limitation), representing a stack in a single flag is not possible.
Tests in sp-error have been added to show that illegal constructs are now
rejected.
Tests in sp have been added for code coverage, to show that ITERATE or LEAVE
statements are legal when jumping to a label in scope, inside the body of
an exception handler.
2007-03-14 19:02:32 +01:00
|
|
|
/**
|
|
|
|
Create and push a new context in the tree.
|
|
|
|
@param label_scope label scope for the new parsing context
|
|
|
|
@return the node created
|
|
|
|
*/
|
2004-08-26 12:54:30 +02:00
|
|
|
sp_pcontext *
|
Bug#26503 (Illegal SQL exception handler code causes the server to crash)
Before this fix, the parser would accept illegal code in SQL exceptions
handlers, that later causes the runtime to crash when executing the code,
due to memory violations in the exception handler stack.
The root cause of the problem is instructions within an exception handler
that jumps to code located outside of the handler. This is illegal according
to the SQL 2003 standard, since labels located outside the handler are not
supposed to be visible (they are "out of scope"), so any instruction that
jumps to these labels, like ITERATE or LEAVE, should not parse.
The section of the standard that is relevant for this is :
SQL:2003 SQL/PSM (ISO/IEC 9075-4:2003)
section 13.1 <compound statement>,
syntax rule 4
<quote>
The scope of the <beginning label> is CS excluding every <SQL schema
statement> contained in CS and excluding every
<local handler declaration list> contained in CS. <beginning label> shall
not be equivalent to any other <beginning label>s within that scope.
</quote>
With this fix, the C++ class sp_pcontext, which represent the "parsing
context" tree (a.k.a symbol table) of a stored procedure, has been changed
as follows:
- constructors have been cleaned up, so that only building a root node for
the tree is public; building nodes inside a tree is not public.
- a new member, m_label_scope, indicates if a given syntactic context
belongs to a DECLARE HANDLER block,
- label resolution, in the method find_label(), has been changed to
implement the restriction of scope regarding labels used in a compound
statement.
The actions in the parser, when parsing the body of a SQL exception handler,
have been changed as follows:
- the implementation of an exception handler (DECLARE HANDLER) now creates
explicitly a new sp_pcontext, to isolate the code inside the handler from
the containing compound statement context.
- registering exception handlers as a result occurs in the parent context,
see the rule sp_hcond_element
- the code in sp_hcond_list has been cleaned up, to avoid code duplication
In addition, the flags IN_SIMPLE_CASE and IN_HANDLER, declared in sp_head.h
have been removed, since they are unused and broken by design (as seen with
Bug 19194 (Right recursion in parser for CASE causes excessive stack usage,
limitation), representing a stack in a single flag is not possible.
Tests in sp-error have been added to show that illegal constructs are now
rejected.
Tests in sp have been added for code coverage, to show that ITERATE or LEAVE
statements are legal when jumping to a label in scope, inside the body of
an exception handler.
2007-03-14 19:02:32 +01:00
|
|
|
push_context(label_scope_type label_scope);
|
2003-11-13 19:34:56 +01:00
|
|
|
|
Bug#26503 (Illegal SQL exception handler code causes the server to crash)
Before this fix, the parser would accept illegal code in SQL exceptions
handlers, that later causes the runtime to crash when executing the code,
due to memory violations in the exception handler stack.
The root cause of the problem is instructions within an exception handler
that jumps to code located outside of the handler. This is illegal according
to the SQL 2003 standard, since labels located outside the handler are not
supposed to be visible (they are "out of scope"), so any instruction that
jumps to these labels, like ITERATE or LEAVE, should not parse.
The section of the standard that is relevant for this is :
SQL:2003 SQL/PSM (ISO/IEC 9075-4:2003)
section 13.1 <compound statement>,
syntax rule 4
<quote>
The scope of the <beginning label> is CS excluding every <SQL schema
statement> contained in CS and excluding every
<local handler declaration list> contained in CS. <beginning label> shall
not be equivalent to any other <beginning label>s within that scope.
</quote>
With this fix, the C++ class sp_pcontext, which represent the "parsing
context" tree (a.k.a symbol table) of a stored procedure, has been changed
as follows:
- constructors have been cleaned up, so that only building a root node for
the tree is public; building nodes inside a tree is not public.
- a new member, m_label_scope, indicates if a given syntactic context
belongs to a DECLARE HANDLER block,
- label resolution, in the method find_label(), has been changed to
implement the restriction of scope regarding labels used in a compound
statement.
The actions in the parser, when parsing the body of a SQL exception handler,
have been changed as follows:
- the implementation of an exception handler (DECLARE HANDLER) now creates
explicitly a new sp_pcontext, to isolate the code inside the handler from
the containing compound statement context.
- registering exception handlers as a result occurs in the parent context,
see the rule sp_hcond_element
- the code in sp_hcond_list has been cleaned up, to avoid code duplication
In addition, the flags IN_SIMPLE_CASE and IN_HANDLER, declared in sp_head.h
have been removed, since they are unused and broken by design (as seen with
Bug 19194 (Right recursion in parser for CASE causes excessive stack usage,
limitation), representing a stack in a single flag is not possible.
Tests in sp-error have been added to show that illegal constructs are now
rejected.
Tests in sp have been added for code coverage, to show that ITERATE or LEAVE
statements are legal when jumping to a label in scope, inside the body of
an exception handler.
2007-03-14 19:02:32 +01:00
|
|
|
/**
|
|
|
|
Pop a node from the parsing context tree.
|
|
|
|
@return the parent node
|
|
|
|
*/
|
2004-08-26 12:54:30 +02:00
|
|
|
sp_pcontext *
|
|
|
|
pop_context();
|
2004-08-17 20:20:58 +02:00
|
|
|
|
2004-08-26 12:54:30 +02:00
|
|
|
sp_pcontext *
|
|
|
|
parent_context()
|
2004-08-17 20:20:58 +02:00
|
|
|
{
|
2004-08-26 12:54:30 +02:00
|
|
|
return m_parent;
|
2004-08-17 20:20:58 +02:00
|
|
|
}
|
|
|
|
|
2006-02-15 12:11:29 +01:00
|
|
|
/*
|
|
|
|
Number of handlers/cursors to pop between this context and 'ctx'.
|
|
|
|
If 'exclusive' is true, don't count the last block we are leaving;
|
|
|
|
this is used for LEAVE where we will jump to the cpop/hpop instructions.
|
|
|
|
*/
|
2004-08-26 12:54:30 +02:00
|
|
|
uint
|
2006-02-15 12:11:29 +01:00
|
|
|
diff_handlers(sp_pcontext *ctx, bool exclusive);
|
2004-08-26 12:54:30 +02:00
|
|
|
uint
|
2006-02-15 12:11:29 +01:00
|
|
|
diff_cursors(sp_pcontext *ctx, bool exclusive);
|
2004-08-17 20:20:58 +02:00
|
|
|
|
2003-11-13 19:34:56 +01:00
|
|
|
|
2003-09-16 14:26:08 +02:00
|
|
|
//
|
|
|
|
// Parameters and variables
|
|
|
|
//
|
|
|
|
|
2006-04-07 16:53:15 +02:00
|
|
|
/*
|
|
|
|
The maximum number of variables used in this and all child contexts
|
|
|
|
In the root, this gives us the number of slots needed for variables
|
|
|
|
during execution.
|
|
|
|
*/
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
inline uint
|
2006-04-07 16:53:15 +02:00
|
|
|
max_var_index()
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
{
|
2006-04-07 16:53:15 +02:00
|
|
|
return m_max_var_index;
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
}
|
|
|
|
|
2006-04-07 16:53:15 +02:00
|
|
|
/*
|
|
|
|
The current number of variables used in the parents (from the root),
|
|
|
|
including this context.
|
|
|
|
*/
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
inline uint
|
2006-04-07 16:53:15 +02:00
|
|
|
current_var_count()
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
{
|
2006-04-07 16:53:15 +02:00
|
|
|
return m_var_offset + m_vars.elements;
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
}
|
|
|
|
|
2006-04-07 16:53:15 +02:00
|
|
|
/* The number of variables in this context alone */
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
inline uint
|
2006-04-07 16:53:15 +02:00
|
|
|
context_var_count()
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
{
|
2006-04-07 16:53:15 +02:00
|
|
|
return m_vars.elements;
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
}
|
|
|
|
|
2006-04-07 16:53:15 +02:00
|
|
|
/* Map index in this pcontext to runtime offset */
|
2004-08-26 12:54:30 +02:00
|
|
|
inline uint
|
2006-04-07 16:53:15 +02:00
|
|
|
var_context2runtime(uint i)
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
{
|
2006-04-07 16:53:15 +02:00
|
|
|
return m_var_offset + i;
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
}
|
|
|
|
|
2006-04-07 16:53:15 +02:00
|
|
|
/* Set type of variable. 'i' is the offset from the top */
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
inline void
|
|
|
|
set_type(uint i, enum enum_field_types type)
|
|
|
|
{
|
2006-04-07 16:53:15 +02:00
|
|
|
sp_variable_t *p= find_variable(i);
|
2003-04-02 20:42:28 +02:00
|
|
|
|
|
|
|
if (p)
|
|
|
|
p->type= type;
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
}
|
|
|
|
|
2006-04-07 16:53:15 +02:00
|
|
|
/* Set default value of variable. 'i' is the offset from the top */
|
2004-04-06 15:48:58 +02:00
|
|
|
inline void
|
|
|
|
set_default(uint i, Item *it)
|
|
|
|
{
|
2006-04-07 16:53:15 +02:00
|
|
|
sp_variable_t *p= find_variable(i);
|
2004-04-06 15:48:58 +02:00
|
|
|
|
|
|
|
if (p)
|
|
|
|
p->dflt= it;
|
|
|
|
}
|
|
|
|
|
2006-04-07 16:53:15 +02:00
|
|
|
sp_variable_t *
|
|
|
|
push_variable(LEX_STRING *name, enum enum_field_types type,
|
|
|
|
sp_param_mode_t mode);
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
|
2005-12-07 15:01:17 +01:00
|
|
|
/*
|
|
|
|
Retrieve definitions of fields from the current context and its
|
|
|
|
children.
|
|
|
|
*/
|
|
|
|
void
|
2007-06-10 12:43:57 +02:00
|
|
|
retrieve_field_definitions(List<Create_field> *field_def_lst);
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
|
|
|
|
// Find by name
|
2006-04-07 16:53:15 +02:00
|
|
|
sp_variable_t *
|
|
|
|
find_variable(LEX_STRING *name, my_bool scoped=0);
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
|
2006-04-07 16:53:15 +02:00
|
|
|
// Find by offset (from the top)
|
|
|
|
sp_variable_t *
|
|
|
|
find_variable(uint offset);
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
|
2005-12-02 14:30:42 +01:00
|
|
|
/*
|
2005-12-07 15:01:17 +01:00
|
|
|
Set the current scope boundary (for default values).
|
2005-12-02 14:30:42 +01:00
|
|
|
The argument is the number of variables to skip.
|
|
|
|
*/
|
2005-11-01 14:58:52 +01:00
|
|
|
inline void
|
|
|
|
declare_var_boundary(uint n)
|
|
|
|
{
|
2005-12-02 14:30:42 +01:00
|
|
|
m_pboundary= n;
|
2005-11-01 14:58:52 +01:00
|
|
|
}
|
|
|
|
|
2005-12-07 15:01:17 +01:00
|
|
|
/*
|
|
|
|
CASE expressions support.
|
|
|
|
*/
|
|
|
|
|
|
|
|
inline int
|
|
|
|
register_case_expr()
|
|
|
|
{
|
|
|
|
return m_num_case_exprs++;
|
|
|
|
}
|
|
|
|
|
|
|
|
inline int
|
|
|
|
get_num_case_exprs() const
|
|
|
|
{
|
|
|
|
return m_num_case_exprs;
|
|
|
|
}
|
|
|
|
|
|
|
|
inline bool
|
|
|
|
push_case_expr_id(int case_expr_id)
|
|
|
|
{
|
WL#3817: Simplify string / memory area types and make things more consistent (first part)
The following type conversions was done:
- Changed byte to uchar
- Changed gptr to uchar*
- Change my_string to char *
- Change my_size_t to size_t
- Change size_s to size_t
Removed declaration of byte, gptr, my_string, my_size_t and size_s.
Following function parameter changes was done:
- All string functions in mysys/strings was changed to use size_t
instead of uint for string lengths.
- All read()/write() functions changed to use size_t (including vio).
- All protocoll functions changed to use size_t instead of uint
- Functions that used a pointer to a string length was changed to use size_t*
- Changed malloc(), free() and related functions from using gptr to use void *
as this requires fewer casts in the code and is more in line with how the
standard functions work.
- Added extra length argument to dirname_part() to return the length of the
created string.
- Changed (at least) following functions to take uchar* as argument:
- db_dump()
- my_net_write()
- net_write_command()
- net_store_data()
- DBUG_DUMP()
- decimal2bin() & bin2decimal()
- Changed my_compress() and my_uncompress() to use size_t. Changed one
argument to my_uncompress() from a pointer to a value as we only return
one value (makes function easier to use).
- Changed type of 'pack_data' argument to packfrm() to avoid casts.
- Changed in readfrm() and writefrom(), ha_discover and handler::discover()
the type for argument 'frmdata' to uchar** to avoid casts.
- Changed most Field functions to use uchar* instead of char* (reduced a lot of
casts).
- Changed field->val_xxx(xxx, new_ptr) to take const pointers.
Other changes:
- Removed a lot of not needed casts
- Added a few new cast required by other changes
- Added some cast to my_multi_malloc() arguments for safety (as string lengths
needs to be uint, not size_t).
- Fixed all calls to hash-get-key functions to use size_t*. (Needed to be done
explicitely as this conflict was often hided by casting the function to
hash_get_key).
- Changed some buffers to memory regions to uchar* to avoid casts.
- Changed some string lengths from uint to size_t.
- Changed field->ptr to be uchar* instead of char*. This allowed us to
get rid of a lot of casts.
- Some changes from true -> TRUE, false -> FALSE, unsigned char -> uchar
- Include zlib.h in some files as we needed declaration of crc32()
- Changed MY_FILE_ERROR to be (size_t) -1.
- Changed many variables to hold the result of my_read() / my_write() to be
size_t. This was needed to properly detect errors (which are
returned as (size_t) -1).
- Removed some very old VMS code
- Changed packfrm()/unpackfrm() to not be depending on uint size
(portability fix)
- Removed windows specific code to restore cursor position as this
causes slowdown on windows and we should not mix read() and pread()
calls anyway as this is not thread safe. Updated function comment to
reflect this. Changed function that depended on original behavior of
my_pwrite() to itself restore the cursor position (one such case).
- Added some missing checking of return value of malloc().
- Changed definition of MOD_PAD_CHAR_TO_FULL_LENGTH to avoid 'long' overflow.
- Changed type of table_def::m_size from my_size_t to ulong to reflect that
m_size is the number of elements in the array, not a string/memory
length.
- Moved THD::max_row_length() to table.cc (as it's not depending on THD).
Inlined max_row_length_blob() into this function.
- More function comments
- Fixed some compiler warnings when compiled without partitions.
- Removed setting of LEX_STRING() arguments in declaration (portability fix).
- Some trivial indentation/variable name changes.
- Some trivial code simplifications:
- Replaced some calls to alloc_root + memcpy to use
strmake_root()/strdup_root().
- Changed some calls from memdup() to strmake() (Safety fix)
- Simpler loops in client-simple.c
2007-05-10 11:59:39 +02:00
|
|
|
return insert_dynamic(&m_case_expr_id_lst, (uchar*) &case_expr_id);
|
2005-12-07 15:01:17 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
inline void
|
|
|
|
pop_case_expr_id()
|
|
|
|
{
|
|
|
|
pop_dynamic(&m_case_expr_id_lst);
|
|
|
|
}
|
|
|
|
|
|
|
|
inline int
|
|
|
|
get_current_case_expr_id() const
|
|
|
|
{
|
|
|
|
int case_expr_id;
|
|
|
|
|
WL#3817: Simplify string / memory area types and make things more consistent (first part)
The following type conversions was done:
- Changed byte to uchar
- Changed gptr to uchar*
- Change my_string to char *
- Change my_size_t to size_t
- Change size_s to size_t
Removed declaration of byte, gptr, my_string, my_size_t and size_s.
Following function parameter changes was done:
- All string functions in mysys/strings was changed to use size_t
instead of uint for string lengths.
- All read()/write() functions changed to use size_t (including vio).
- All protocoll functions changed to use size_t instead of uint
- Functions that used a pointer to a string length was changed to use size_t*
- Changed malloc(), free() and related functions from using gptr to use void *
as this requires fewer casts in the code and is more in line with how the
standard functions work.
- Added extra length argument to dirname_part() to return the length of the
created string.
- Changed (at least) following functions to take uchar* as argument:
- db_dump()
- my_net_write()
- net_write_command()
- net_store_data()
- DBUG_DUMP()
- decimal2bin() & bin2decimal()
- Changed my_compress() and my_uncompress() to use size_t. Changed one
argument to my_uncompress() from a pointer to a value as we only return
one value (makes function easier to use).
- Changed type of 'pack_data' argument to packfrm() to avoid casts.
- Changed in readfrm() and writefrom(), ha_discover and handler::discover()
the type for argument 'frmdata' to uchar** to avoid casts.
- Changed most Field functions to use uchar* instead of char* (reduced a lot of
casts).
- Changed field->val_xxx(xxx, new_ptr) to take const pointers.
Other changes:
- Removed a lot of not needed casts
- Added a few new cast required by other changes
- Added some cast to my_multi_malloc() arguments for safety (as string lengths
needs to be uint, not size_t).
- Fixed all calls to hash-get-key functions to use size_t*. (Needed to be done
explicitely as this conflict was often hided by casting the function to
hash_get_key).
- Changed some buffers to memory regions to uchar* to avoid casts.
- Changed some string lengths from uint to size_t.
- Changed field->ptr to be uchar* instead of char*. This allowed us to
get rid of a lot of casts.
- Some changes from true -> TRUE, false -> FALSE, unsigned char -> uchar
- Include zlib.h in some files as we needed declaration of crc32()
- Changed MY_FILE_ERROR to be (size_t) -1.
- Changed many variables to hold the result of my_read() / my_write() to be
size_t. This was needed to properly detect errors (which are
returned as (size_t) -1).
- Removed some very old VMS code
- Changed packfrm()/unpackfrm() to not be depending on uint size
(portability fix)
- Removed windows specific code to restore cursor position as this
causes slowdown on windows and we should not mix read() and pread()
calls anyway as this is not thread safe. Updated function comment to
reflect this. Changed function that depended on original behavior of
my_pwrite() to itself restore the cursor position (one such case).
- Added some missing checking of return value of malloc().
- Changed definition of MOD_PAD_CHAR_TO_FULL_LENGTH to avoid 'long' overflow.
- Changed type of table_def::m_size from my_size_t to ulong to reflect that
m_size is the number of elements in the array, not a string/memory
length.
- Moved THD::max_row_length() to table.cc (as it's not depending on THD).
Inlined max_row_length_blob() into this function.
- More function comments
- Fixed some compiler warnings when compiled without partitions.
- Removed setting of LEX_STRING() arguments in declaration (portability fix).
- Some trivial indentation/variable name changes.
- Some trivial code simplifications:
- Replaced some calls to alloc_root + memcpy to use
strmake_root()/strdup_root().
- Changed some calls from memdup() to strmake() (Safety fix)
- Simpler loops in client-simple.c
2007-05-10 11:59:39 +02:00
|
|
|
get_dynamic((DYNAMIC_ARRAY*)&m_case_expr_id_lst, (uchar*) &case_expr_id,
|
2005-12-07 15:01:17 +01:00
|
|
|
m_case_expr_id_lst.elements - 1);
|
|
|
|
|
|
|
|
return case_expr_id;
|
|
|
|
}
|
|
|
|
|
2003-09-16 14:26:08 +02:00
|
|
|
//
|
|
|
|
// Labels
|
|
|
|
//
|
|
|
|
|
2002-12-16 15:40:44 +01:00
|
|
|
sp_label_t *
|
2002-12-11 14:24:29 +01:00
|
|
|
push_label(char *name, uint ip);
|
|
|
|
|
|
|
|
sp_label_t *
|
|
|
|
find_label(char *name);
|
|
|
|
|
2002-12-12 13:14:23 +01:00
|
|
|
inline sp_label_t *
|
|
|
|
last_label()
|
|
|
|
{
|
2004-08-26 12:54:30 +02:00
|
|
|
sp_label_t *lab= m_label.head();
|
|
|
|
|
|
|
|
if (!lab && m_parent)
|
|
|
|
lab= m_parent->last_label();
|
|
|
|
return lab;
|
2002-12-12 13:14:23 +01:00
|
|
|
}
|
|
|
|
|
2002-12-16 15:40:44 +01:00
|
|
|
inline sp_label_t *
|
2002-12-11 14:24:29 +01:00
|
|
|
pop_label()
|
|
|
|
{
|
2002-12-16 15:40:44 +01:00
|
|
|
return m_label.pop();
|
2002-12-11 14:24:29 +01:00
|
|
|
}
|
|
|
|
|
2003-09-16 14:26:08 +02:00
|
|
|
//
|
|
|
|
// Conditions
|
|
|
|
//
|
|
|
|
|
2008-11-21 14:38:42 +01:00
|
|
|
int
|
2003-09-16 14:26:08 +02:00
|
|
|
push_cond(LEX_STRING *name, sp_cond_type_t *val);
|
|
|
|
|
|
|
|
inline void
|
|
|
|
pop_cond(uint num)
|
|
|
|
{
|
|
|
|
while (num--)
|
2006-04-07 16:53:15 +02:00
|
|
|
pop_dynamic(&m_conds);
|
2003-09-16 14:26:08 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
sp_cond_type_t *
|
2003-11-13 19:34:56 +01:00
|
|
|
find_cond(LEX_STRING *name, my_bool scoped=0);
|
2003-09-16 14:26:08 +02:00
|
|
|
|
|
|
|
//
|
|
|
|
// Handlers
|
|
|
|
//
|
|
|
|
|
2005-04-08 19:58:04 +02:00
|
|
|
inline void
|
|
|
|
push_handler(sp_cond_type_t *cond)
|
|
|
|
{
|
WL#3817: Simplify string / memory area types and make things more consistent (first part)
The following type conversions was done:
- Changed byte to uchar
- Changed gptr to uchar*
- Change my_string to char *
- Change my_size_t to size_t
- Change size_s to size_t
Removed declaration of byte, gptr, my_string, my_size_t and size_s.
Following function parameter changes was done:
- All string functions in mysys/strings was changed to use size_t
instead of uint for string lengths.
- All read()/write() functions changed to use size_t (including vio).
- All protocoll functions changed to use size_t instead of uint
- Functions that used a pointer to a string length was changed to use size_t*
- Changed malloc(), free() and related functions from using gptr to use void *
as this requires fewer casts in the code and is more in line with how the
standard functions work.
- Added extra length argument to dirname_part() to return the length of the
created string.
- Changed (at least) following functions to take uchar* as argument:
- db_dump()
- my_net_write()
- net_write_command()
- net_store_data()
- DBUG_DUMP()
- decimal2bin() & bin2decimal()
- Changed my_compress() and my_uncompress() to use size_t. Changed one
argument to my_uncompress() from a pointer to a value as we only return
one value (makes function easier to use).
- Changed type of 'pack_data' argument to packfrm() to avoid casts.
- Changed in readfrm() and writefrom(), ha_discover and handler::discover()
the type for argument 'frmdata' to uchar** to avoid casts.
- Changed most Field functions to use uchar* instead of char* (reduced a lot of
casts).
- Changed field->val_xxx(xxx, new_ptr) to take const pointers.
Other changes:
- Removed a lot of not needed casts
- Added a few new cast required by other changes
- Added some cast to my_multi_malloc() arguments for safety (as string lengths
needs to be uint, not size_t).
- Fixed all calls to hash-get-key functions to use size_t*. (Needed to be done
explicitely as this conflict was often hided by casting the function to
hash_get_key).
- Changed some buffers to memory regions to uchar* to avoid casts.
- Changed some string lengths from uint to size_t.
- Changed field->ptr to be uchar* instead of char*. This allowed us to
get rid of a lot of casts.
- Some changes from true -> TRUE, false -> FALSE, unsigned char -> uchar
- Include zlib.h in some files as we needed declaration of crc32()
- Changed MY_FILE_ERROR to be (size_t) -1.
- Changed many variables to hold the result of my_read() / my_write() to be
size_t. This was needed to properly detect errors (which are
returned as (size_t) -1).
- Removed some very old VMS code
- Changed packfrm()/unpackfrm() to not be depending on uint size
(portability fix)
- Removed windows specific code to restore cursor position as this
causes slowdown on windows and we should not mix read() and pread()
calls anyway as this is not thread safe. Updated function comment to
reflect this. Changed function that depended on original behavior of
my_pwrite() to itself restore the cursor position (one such case).
- Added some missing checking of return value of malloc().
- Changed definition of MOD_PAD_CHAR_TO_FULL_LENGTH to avoid 'long' overflow.
- Changed type of table_def::m_size from my_size_t to ulong to reflect that
m_size is the number of elements in the array, not a string/memory
length.
- Moved THD::max_row_length() to table.cc (as it's not depending on THD).
Inlined max_row_length_blob() into this function.
- More function comments
- Fixed some compiler warnings when compiled without partitions.
- Removed setting of LEX_STRING() arguments in declaration (portability fix).
- Some trivial indentation/variable name changes.
- Some trivial code simplifications:
- Replaced some calls to alloc_root + memcpy to use
strmake_root()/strdup_root().
- Changed some calls from memdup() to strmake() (Safety fix)
- Simpler loops in client-simple.c
2007-05-10 11:59:39 +02:00
|
|
|
insert_dynamic(&m_handlers, (uchar*)&cond);
|
2005-04-08 19:58:04 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
bool
|
|
|
|
find_handler(sp_cond_type *cond);
|
|
|
|
|
2003-09-16 14:26:08 +02:00
|
|
|
inline uint
|
2006-04-07 16:53:15 +02:00
|
|
|
max_handler_index()
|
2003-09-16 14:26:08 +02:00
|
|
|
{
|
2006-04-07 16:53:15 +02:00
|
|
|
return m_max_handler_index + m_context_handlers;
|
2003-09-16 14:26:08 +02:00
|
|
|
}
|
|
|
|
|
2004-08-17 20:20:58 +02:00
|
|
|
inline void
|
2005-04-13 16:57:49 +02:00
|
|
|
add_handlers(uint n)
|
2004-08-17 20:20:58 +02:00
|
|
|
{
|
2006-04-07 16:53:15 +02:00
|
|
|
m_context_handlers+= n;
|
2004-08-17 20:20:58 +02:00
|
|
|
}
|
|
|
|
|
2003-10-10 16:57:21 +02:00
|
|
|
//
|
|
|
|
// Cursors
|
|
|
|
//
|
|
|
|
|
2008-11-21 14:38:42 +01:00
|
|
|
int
|
2003-10-10 16:57:21 +02:00
|
|
|
push_cursor(LEX_STRING *name);
|
|
|
|
|
|
|
|
my_bool
|
2003-11-13 19:34:56 +01:00
|
|
|
find_cursor(LEX_STRING *name, uint *poff, my_bool scoped=0);
|
2003-10-10 16:57:21 +02:00
|
|
|
|
2005-11-18 16:30:27 +01:00
|
|
|
/* Find by offset (for debugging only) */
|
2005-11-17 11:11:48 +01:00
|
|
|
my_bool
|
2005-11-18 16:30:27 +01:00
|
|
|
find_cursor(uint offset, LEX_STRING *n);
|
2005-11-17 11:11:48 +01:00
|
|
|
|
2004-08-26 12:54:30 +02:00
|
|
|
inline uint
|
2006-04-07 16:53:15 +02:00
|
|
|
max_cursor_index()
|
2003-10-10 16:57:21 +02:00
|
|
|
{
|
2006-04-07 16:53:15 +02:00
|
|
|
return m_max_cursor_index + m_cursors.elements;
|
2003-10-10 16:57:21 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
inline uint
|
2006-04-07 16:53:15 +02:00
|
|
|
current_cursor_count()
|
2003-10-10 16:57:21 +02:00
|
|
|
{
|
2006-04-07 16:53:15 +02:00
|
|
|
return m_cursor_offset + m_cursors.elements;
|
2003-10-10 16:57:21 +02:00
|
|
|
}
|
|
|
|
|
2004-08-26 12:54:30 +02:00
|
|
|
protected:
|
2004-08-17 20:20:58 +02:00
|
|
|
|
Bug#26503 (Illegal SQL exception handler code causes the server to crash)
Before this fix, the parser would accept illegal code in SQL exceptions
handlers, that later causes the runtime to crash when executing the code,
due to memory violations in the exception handler stack.
The root cause of the problem is instructions within an exception handler
that jumps to code located outside of the handler. This is illegal according
to the SQL 2003 standard, since labels located outside the handler are not
supposed to be visible (they are "out of scope"), so any instruction that
jumps to these labels, like ITERATE or LEAVE, should not parse.
The section of the standard that is relevant for this is :
SQL:2003 SQL/PSM (ISO/IEC 9075-4:2003)
section 13.1 <compound statement>,
syntax rule 4
<quote>
The scope of the <beginning label> is CS excluding every <SQL schema
statement> contained in CS and excluding every
<local handler declaration list> contained in CS. <beginning label> shall
not be equivalent to any other <beginning label>s within that scope.
</quote>
With this fix, the C++ class sp_pcontext, which represent the "parsing
context" tree (a.k.a symbol table) of a stored procedure, has been changed
as follows:
- constructors have been cleaned up, so that only building a root node for
the tree is public; building nodes inside a tree is not public.
- a new member, m_label_scope, indicates if a given syntactic context
belongs to a DECLARE HANDLER block,
- label resolution, in the method find_label(), has been changed to
implement the restriction of scope regarding labels used in a compound
statement.
The actions in the parser, when parsing the body of a SQL exception handler,
have been changed as follows:
- the implementation of an exception handler (DECLARE HANDLER) now creates
explicitly a new sp_pcontext, to isolate the code inside the handler from
the containing compound statement context.
- registering exception handlers as a result occurs in the parent context,
see the rule sp_hcond_element
- the code in sp_hcond_list has been cleaned up, to avoid code duplication
In addition, the flags IN_SIMPLE_CASE and IN_HANDLER, declared in sp_head.h
have been removed, since they are unused and broken by design (as seen with
Bug 19194 (Right recursion in parser for CASE causes excessive stack usage,
limitation), representing a stack in a single flag is not possible.
Tests in sp-error have been added to show that illegal constructs are now
rejected.
Tests in sp have been added for code coverage, to show that ITERATE or LEAVE
statements are legal when jumping to a label in scope, inside the body of
an exception handler.
2007-03-14 19:02:32 +01:00
|
|
|
/**
|
|
|
|
Constructor for a tree node.
|
|
|
|
@param prev the parent parsing context
|
|
|
|
@param label_scope label_scope for this parsing context
|
|
|
|
*/
|
|
|
|
sp_pcontext(sp_pcontext *prev, label_scope_type label_scope);
|
|
|
|
|
2005-12-07 15:01:17 +01:00
|
|
|
/*
|
2006-04-07 16:53:15 +02:00
|
|
|
m_max_var_index -- number of variables (including all types of arguments)
|
2005-12-07 15:01:17 +01:00
|
|
|
in this context including all children contexts.
|
|
|
|
|
2006-04-07 16:53:15 +02:00
|
|
|
m_max_var_index >= m_vars.elements.
|
2005-12-07 15:01:17 +01:00
|
|
|
|
2006-04-07 16:53:15 +02:00
|
|
|
m_max_var_index of the root parsing context contains number of all
|
2005-12-07 15:01:17 +01:00
|
|
|
variables (including arguments) in all enclosed contexts.
|
|
|
|
*/
|
2006-04-07 16:53:15 +02:00
|
|
|
uint m_max_var_index;
|
2005-12-07 15:01:17 +01:00
|
|
|
|
2004-08-26 12:54:30 +02:00
|
|
|
// The maximum sub context's framesizes
|
2006-04-07 16:53:15 +02:00
|
|
|
uint m_max_cursor_index;
|
|
|
|
uint m_max_handler_index;
|
|
|
|
uint m_context_handlers; // No. of handlers in this context
|
2004-08-17 20:20:58 +02:00
|
|
|
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
private:
|
|
|
|
|
2004-08-26 12:54:30 +02:00
|
|
|
sp_pcontext *m_parent; // Parent context
|
|
|
|
|
2005-12-07 15:01:17 +01:00
|
|
|
/*
|
2006-04-07 16:53:15 +02:00
|
|
|
m_var_offset -- this is an index of the first variable in this
|
|
|
|
parsing context.
|
2005-12-07 15:01:17 +01:00
|
|
|
|
2006-04-07 16:53:15 +02:00
|
|
|
m_var_offset is 0 for root context.
|
2005-12-07 15:01:17 +01:00
|
|
|
|
|
|
|
Since now each variable is stored in separate place, no reuse is done,
|
2006-04-07 16:53:15 +02:00
|
|
|
so m_var_offset is different for all enclosed contexts.
|
2005-12-07 15:01:17 +01:00
|
|
|
*/
|
2006-04-07 16:53:15 +02:00
|
|
|
uint m_var_offset;
|
2005-12-07 15:01:17 +01:00
|
|
|
|
2006-04-07 16:53:15 +02:00
|
|
|
uint m_cursor_offset; // Cursor offset for this context
|
2005-12-07 15:01:17 +01:00
|
|
|
|
2005-11-01 14:58:52 +01:00
|
|
|
/*
|
2005-12-02 14:30:42 +01:00
|
|
|
Boundary for finding variables in this context. This is the number
|
|
|
|
of variables currently "invisible" to default clauses.
|
|
|
|
This is normally 0, but will be larger during parsing of
|
|
|
|
DECLARE ... DEFAULT, to get the scope right for DEFAULT values.
|
2005-11-01 14:58:52 +01:00
|
|
|
*/
|
|
|
|
uint m_pboundary;
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
|
2005-12-07 15:01:17 +01:00
|
|
|
int m_num_case_exprs;
|
|
|
|
|
2006-04-07 16:53:15 +02:00
|
|
|
DYNAMIC_ARRAY m_vars; // Parameters/variables
|
2005-12-07 15:01:17 +01:00
|
|
|
DYNAMIC_ARRAY m_case_expr_id_lst; /* Stack of CASE expression ids. */
|
2006-04-07 16:53:15 +02:00
|
|
|
DYNAMIC_ARRAY m_conds; // Conditions
|
|
|
|
DYNAMIC_ARRAY m_cursors; // Cursors
|
|
|
|
DYNAMIC_ARRAY m_handlers; // Handlers, for checking for duplicates
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
|
2002-12-11 14:24:29 +01:00
|
|
|
List<sp_label_t> m_label; // The label list
|
|
|
|
|
2004-08-26 12:54:30 +02:00
|
|
|
List<sp_pcontext> m_children; // Children contexts, used for destruction
|
|
|
|
|
Bug#26503 (Illegal SQL exception handler code causes the server to crash)
Before this fix, the parser would accept illegal code in SQL exceptions
handlers, that later causes the runtime to crash when executing the code,
due to memory violations in the exception handler stack.
The root cause of the problem is instructions within an exception handler
that jumps to code located outside of the handler. This is illegal according
to the SQL 2003 standard, since labels located outside the handler are not
supposed to be visible (they are "out of scope"), so any instruction that
jumps to these labels, like ITERATE or LEAVE, should not parse.
The section of the standard that is relevant for this is :
SQL:2003 SQL/PSM (ISO/IEC 9075-4:2003)
section 13.1 <compound statement>,
syntax rule 4
<quote>
The scope of the <beginning label> is CS excluding every <SQL schema
statement> contained in CS and excluding every
<local handler declaration list> contained in CS. <beginning label> shall
not be equivalent to any other <beginning label>s within that scope.
</quote>
With this fix, the C++ class sp_pcontext, which represent the "parsing
context" tree (a.k.a symbol table) of a stored procedure, has been changed
as follows:
- constructors have been cleaned up, so that only building a root node for
the tree is public; building nodes inside a tree is not public.
- a new member, m_label_scope, indicates if a given syntactic context
belongs to a DECLARE HANDLER block,
- label resolution, in the method find_label(), has been changed to
implement the restriction of scope regarding labels used in a compound
statement.
The actions in the parser, when parsing the body of a SQL exception handler,
have been changed as follows:
- the implementation of an exception handler (DECLARE HANDLER) now creates
explicitly a new sp_pcontext, to isolate the code inside the handler from
the containing compound statement context.
- registering exception handlers as a result occurs in the parent context,
see the rule sp_hcond_element
- the code in sp_hcond_list has been cleaned up, to avoid code duplication
In addition, the flags IN_SIMPLE_CASE and IN_HANDLER, declared in sp_head.h
have been removed, since they are unused and broken by design (as seen with
Bug 19194 (Right recursion in parser for CASE causes excessive stack usage,
limitation), representing a stack in a single flag is not possible.
Tests in sp-error have been added to show that illegal constructs are now
rejected.
Tests in sp have been added for code coverage, to show that ITERATE or LEAVE
statements are legal when jumping to a label in scope, inside the body of
an exception handler.
2007-03-14 19:02:32 +01:00
|
|
|
/**
|
|
|
|
Scope of labels for this parsing context.
|
|
|
|
*/
|
|
|
|
label_scope_type m_label_scope;
|
|
|
|
|
|
|
|
private:
|
|
|
|
sp_pcontext(const sp_pcontext &); /* Prevent use of these */
|
|
|
|
void operator=(sp_pcontext &);
|
Simplistic, experimental framework for Stored Procedures (SPs).
Implements creation and dropping of PROCEDUREs, IN, OUT, and INOUT parameters,
single-statement procedures, rudimentary multi-statement (begin-end) prodedures
(when the client can handle it), and local variables.
Missing most of the embedded SQL language, all attributes, FUNCTIONs, error handling,
reparses procedures at each call (no caching), etc, etc.
Certainly buggy too, but procedures can actually be created and called....
2002-12-08 19:59:22 +01:00
|
|
|
}; // class sp_pcontext : public Sql_alloc
|
|
|
|
|
|
|
|
|
|
|
|
#endif /* _SP_PCONTEXT_H_ */
|