mirror/postgresql
2
0
Fork 0
mirror of https://git.postgresql.org/git/postgresql.git synced 2024-12-21 08:29:39 +08:00
Code Issues Packages Projects Releases Wiki Activity

Convert the core lexer and parser into fully reentrant code, by making use

Browse Source 
of features added to flex and bison since this code was originally written.
This change doesn't in itself offer any new capability, but it's needed
infrastructure for planned improvements in plpgsql.

Another feature now available in flex is the ability to make it use palloc
instead of malloc, so do that to avoid possible memory leaks.  (We should
at some point change the other lexers likewise, but this commit doesn't
touch them.)
This commit is contained in:
Tom Lane 2009-07-13 02:02:20 +00:00
parent da4b900176
commit 91e71929ba
5 changed files with 503 additions and 328 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
src/backend/nls.mk
View File

@ -1,8 +1,8 @@
# $PostgreSQL: pgsql/src/backend/nls.mk,v 1.27 2009/06/26 19:33:43 petere Exp $
# $PostgreSQL: pgsql/src/backend/nls.mk,v 1.28 2009/07/13 02:02:19 tgl Exp $
CATALOG_NAME := postgres
AVAIL_LANGUAGES := de es fr ja pt_BR tr
GETTEXT_FILES := + gettext-files
GETTEXT_TRIGGERS:= _ errmsg errmsg_plural:1,2 errdetail errdetail_log errdetail_plural:1,2 errhint errcontext write_stderr yyerror
GETTEXT_TRIGGERS:= _ errmsg errmsg_plural:1,2 errdetail errdetail_log errdetail_plural:1,2 errhint errcontext write_stderr yyerror parser_yyerror
gettext-files: distprep
find $(srcdir)/ $(srcdir)/../port/ -name '*.c' -print >$@

229
src/backend/parser/gram.y
View File

@ -11,7 +11,7 @@
*
*
* IDENTIFICATION
* $PostgreSQL: pgsql/src/backend/parser/gram.y,v 2.667 2009/07/12 17:12:33 tgl Exp $
* $PostgreSQL: pgsql/src/backend/parser/gram.y,v 2.668 2009/07/13 02:02:20 tgl Exp $
*
* HISTORY
* AUTHOR DATE MAJOR EVENT
@ -92,10 +92,6 @@
#define YYMALLOC palloc
#define YYFREE pfree
extern List *parsetree; /* final parse result is delivered here */
static bool QueryIsRule = FALSE;
/* Private struct for the result of privilege_target production */
typedef struct PrivTarget
{
@ -103,14 +99,14 @@ typedef struct PrivTarget
List *objs;
} PrivTarget;
/*
* If you need access to certain yacc-generated variables and find that
* they're static by default, uncomment the next line. (this is not a
* problem, yet.)
*/
/*#define __YYSCLASS*/
static Node *makeColumnRef(char *colname, List *indirection, int location);
#define parser_yyerror(msg) scanner_yyerror(msg, yyscanner)
#define parser_errposition(pos) scanner_errposition(pos, yyscanner)
static void base_yyerror(YYLTYPE *yylloc, base_yyscan_t yyscanner,
const char *msg);
static Node *makeColumnRef(char *colname, List *indirection,
int location, base_yyscan_t yyscanner);
static Node *makeTypeCast(Node *arg, TypeName *typename, int location);
static Node *makeStringConst(char *str, int location);
static Node *makeStringConstCast(char *str, int location, TypeName *typename);
@ -120,16 +116,18 @@ static Node *makeBitStringConst(char *str, int location);
static Node *makeNullAConst(int location);
static Node *makeAConst(Value *v, int location);
static Node *makeBoolAConst(bool state, int location);
static FuncCall *makeOverlaps(List *largs, List *rargs, int location);
static void check_qualified_name(List *names);
static List *check_func_name(List *names);
static List *check_indirection(List *indirection);
static FuncCall *makeOverlaps(List *largs, List *rargs,
int location, base_yyscan_t yyscanner);
static void check_qualified_name(List *names, base_yyscan_t yyscanner);
static List *check_func_name(List *names, base_yyscan_t yyscanner);
static List *check_indirection(List *indirection, base_yyscan_t yyscanner);
static List *extractArgTypes(List *parameters);
static SelectStmt *findLeftmostSelect(SelectStmt *node);
static void insertSelectOptions(SelectStmt *stmt,
List *sortClause, List *lockingClause,
Node *limitOffset, Node *limitCount,
WithClause *withClause);
WithClause *withClause,
base_yyscan_t yyscanner);
static Node *makeSetOp(SetOperation op, bool all, Node *larg, Node *rarg);
static Node *doNegate(Node *n, int location);
static void doNegateFloat(Value *v);
@ -141,10 +139,14 @@ static TypeName *TableFuncTypeName(List *columns);
%}
%pure-parser
%expect 0
%name-prefix="base_yy"
%locations
%parse-param {base_yyscan_t yyscanner}
%lex-param {base_yyscan_t yyscanner}
%union
{
int ival;
@ -576,26 +578,29 @@ static TypeName *TableFuncTypeName(List *columns);
%%
/*
* Handle comment-only lines, and ;; SELECT * FROM pg_class ;;;
* psql already handles such cases, but other interfaces don't.
* bjm 1999/10/05
* The target production for the whole parse.
*/
stmtblock: stmtmulti { parsetree = $1; }
stmtblock: stmtmulti
{
pg_yyget_extra(yyscanner)->parsetree = $1;
}
;
/* the thrashing around here is to discard "empty" statements... */
stmtmulti: stmtmulti ';' stmt
{ if ($3 != NULL)
$$ = lappend($1, $3);
else
$$ = $1;
{
if ($3 != NULL)
$$ = lappend($1, $3);
else
$$ = $1;
}
| stmt
{ if ($1 != NULL)
{
if ($1 != NULL)
$$ = list_make1($1);
else
else
$$ = NIL;
}
}
;
stmt :
@ -1190,7 +1195,7 @@ set_rest: /* Generic SET syntaxes: */
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("current database cannot be changed"),
scanner_errposition(@2)));
parser_errposition(@2)));
$$ = NULL; /*not reached*/
}
| SCHEMA Sconst
@ -1305,7 +1310,7 @@ zone_value:
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("time zone interval must be HOUR or HOUR TO MINUTE"),
scanner_errposition(@3)));
parser_errposition(@3)));
}
t->typmods = $3;
$$ = makeStringConstCast($2, @2, t);
@ -1320,12 +1325,12 @@ zone_value:
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("time zone interval must be HOUR or HOUR TO MINUTE"),
scanner_errposition(@6)));
parser_errposition(@6)));
if (list_length($6) != 1)
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("interval precision specified twice"),
scanner_errposition(@1)));
parser_errposition(@1)));
t->typmods = lappend($6, makeIntConst($3, @3));
}
else
@ -2428,7 +2433,7 @@ key_match: MATCH FULL
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("MATCH PARTIAL not yet implemented"),
scanner_errposition(@1)));
parser_errposition(@1)));
$$ = FKCONSTR_MATCH_PARTIAL;
}
| MATCH SIMPLE
@ -2521,7 +2526,7 @@ CreateAsStmt:
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("CREATE TABLE AS cannot specify INTO"),
scanner_errposition(exprLocation((Node *) n->intoClause))));
parser_errposition(exprLocation((Node *) n->intoClause))));
$4->rel->istemp = $2;
n->intoClause = $4;
/* Implement WITH NO DATA by forcing top-level LIMIT 0 */
@ -3174,7 +3179,7 @@ TriggerEvents:
| TriggerEvents OR TriggerOneEvent
{
if ($1 & $3)
yyerror("duplicate trigger events specified");
parser_yyerror("duplicate trigger events specified");
$$ = $1 | $3;
}
;
@ -3245,7 +3250,7 @@ ConstraintAttributeSpec:
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("constraint declared INITIALLY DEFERRED must be DEFERRABLE"),
scanner_errposition(@1)));
parser_errposition(@1)));
$$ = $1 | $2;
}
| ConstraintTimeSpec
@ -3261,7 +3266,7 @@ ConstraintAttributeSpec:
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("constraint declared INITIALLY DEFERRED must be DEFERRABLE"),
scanner_errposition(@1)));
parser_errposition(@1)));
$$ = $1 | $2;
}
| /*EMPTY*/
@ -3434,7 +3439,7 @@ DefineStmt:
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("improper qualified name (too many dotted names): %s",
NameListToString($3)),
scanner_errposition(@3)));
parser_errposition(@3)));
break;
}
r->location = @3;
@ -3638,7 +3643,7 @@ opt_recheck: RECHECK
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("RECHECK is no longer required"),
errhint("Update your data type."),
scanner_errposition(@1)));
parser_errposition(@1)));
$$ = TRUE;
}
| /*EMPTY*/ { $$ = FALSE; }
@ -5043,7 +5048,7 @@ oper_argtypes:
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("missing argument"),
errhint("Use NONE to denote the missing argument of a unary operator."),
scanner_errposition(@3)));
parser_errposition(@3)));
}
| '(' Typename ',' Typename ')'
{ $$ = list_make2($2, $4); }
@ -5600,7 +5605,7 @@ AlterOwnerStmt: ALTER AGGREGATE func_name aggr_args OWNER TO RoleId
*****************************************************************************/
RuleStmt: CREATE opt_or_replace RULE name AS
{ QueryIsRule=TRUE; }
{ pg_yyget_extra(yyscanner)->QueryIsRule = TRUE; }
ON event TO qualified_name where_clause
DO opt_instead RuleActionList
{
@ -5613,7 +5618,7 @@ RuleStmt: CREATE opt_or_replace RULE name AS
n->instead = $13;
n->actions = $14;
$$ = (Node *)n;
QueryIsRule=FALSE;
pg_yyget_extra(yyscanner)->QueryIsRule = FALSE;
}
;
@ -6605,7 +6610,7 @@ insert_column_item:
{
$$ = makeNode(ResTarget);
$$->name = $1;
$$->indirection = check_indirection($2);
$$->indirection = check_indirection($2, yyscanner);
$$->val = NULL;
$$->location = @1;
}
@ -6735,7 +6740,7 @@ multiple_set_clause:
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("number of columns does not match number of values"),
scanner_errposition(@1)));
parser_errposition(@1)));
forboth(col_cell, $2, val_cell, $5)
{
ResTarget *res_col = (ResTarget *) lfirst(col_cell);
@ -6753,7 +6758,7 @@ set_target:
{
$$ = makeNode(ResTarget);
$$->name = $1;
$$->indirection = check_indirection($2);
$$->indirection = check_indirection($2, yyscanner);
$$->val = NULL; /* upper production sets this */
$$->location = @1;
}
@ -6863,49 +6868,56 @@ select_no_parens:
| select_clause sort_clause
{
insertSelectOptions((SelectStmt *) $1, $2, NIL,
NULL, NULL, NULL);
NULL, NULL, NULL,
yyscanner);
$$ = $1;
}
| select_clause opt_sort_clause for_locking_clause opt_select_limit
{
insertSelectOptions((SelectStmt *) $1, $2, $3,
list_nth($4, 0), list_nth($4, 1),
NULL);
NULL,
yyscanner);
$$ = $1;
}
| select_clause opt_sort_clause select_limit opt_for_locking_clause
{
insertSelectOptions((SelectStmt *) $1, $2, $4,
list_nth($3, 0), list_nth($3, 1),
NULL);
NULL,
yyscanner);
$$ = $1;
}
| with_clause select_clause
{
insertSelectOptions((SelectStmt *) $2, NULL, NIL,
NULL, NULL,
$1);
$1,
yyscanner);
$$ = $2;
}
| with_clause select_clause sort_clause
{
insertSelectOptions((SelectStmt *) $2, $3, NIL,
NULL, NULL,
$1);
$1,
yyscanner);
$$ = $2;
}
| with_clause select_clause opt_sort_clause for_locking_clause opt_select_limit
{
insertSelectOptions((SelectStmt *) $2, $3, $4,
list_nth($5, 0), list_nth($5, 1),
$1);
$1,
yyscanner);
$$ = $2;
}
| with_clause select_clause opt_sort_clause select_limit opt_for_locking_clause
{
insertSelectOptions((SelectStmt *) $2, $3, $5,
list_nth($4, 0), list_nth($4, 1),
$1);
$1,
yyscanner);
$$ = $2;
}
;
@ -7160,7 +7172,7 @@ select_limit:
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("LIMIT #,# syntax is not supported"),
errhint("Use separate LIMIT and OFFSET clauses."),
scanner_errposition(@1)));
parser_errposition(@1)));
}
/* SQL:2008 syntax variants */
| OFFSET select_offset_value2 row_or_rows
@ -7382,13 +7394,13 @@ table_ref: relation_expr
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("VALUES in FROM must have an alias"),
errhint("For example, FROM (VALUES ...) [AS] foo."),
scanner_errposition(@1)));
parser_errposition(@1)));
else
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("subquery in FROM must have an alias"),
errhint("For example, FROM (SELECT ...) [AS] foo."),
scanner_errposition(@1)));
parser_errposition(@1)));
$$ = NULL;
}
| select_with_parens alias_clause
@ -7743,7 +7755,7 @@ SimpleTypename:
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("interval precision specified twice"),
scanner_errposition(@1)));
parser_errposition(@1)));
$$->typmods = lappend($5, makeIntConst($3, @3));
}
else
@ -7869,7 +7881,7 @@ opt_float: '(' Iconst ')'
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("precision for type float must be at least 1 bit"),
scanner_errposition(@2)));
parser_errposition(@2)));
else if ($2 <= 24)
$$ = SystemTypeName("float4");
else if ($2 <= 53)
@ -7878,7 +7890,7 @@ opt_float: '(' Iconst ')'
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("precision for type float must be less than 54 bits"),
scanner_errposition(@2)));
parser_errposition(@2)));
}
| /*EMPTY*/
{
@ -8394,7 +8406,7 @@ a_expr: c_expr { $$ = $1; }
}
| row OVERLAPS row
{
$$ = (Node *)makeOverlaps($1, $3, @2);
$$ = (Node *)makeOverlaps($1, $3, @2, yyscanner);
}
| a_expr IS TRUE_P
{
@ -8574,7 +8586,7 @@ a_expr: c_expr { $$ = $1; }
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("UNIQUE predicate is not yet implemented"),
scanner_errposition(@1)));
parser_errposition(@1)));
}
| a_expr IS DOCUMENT_P %prec IS
{
@ -8681,7 +8693,7 @@ c_expr: columnref { $$ = $1; }
{
A_Indirection *n = makeNode(A_Indirection);
n->arg = (Node *) p;
n->indirection = check_indirection($2);
n->indirection = check_indirection($2, yyscanner);
$$ = (Node *) n;
}
else
@ -8693,7 +8705,7 @@ c_expr: columnref { $$ = $1; }
{
A_Indirection *n = makeNode(A_Indirection);
n->arg = $2;
n->indirection = check_indirection($4);
n->indirection = check_indirection($4, yyscanner);
$$ = (Node *)n;
}
else
@ -9413,12 +9425,12 @@ frame_extent: frame_bound
ereport(ERROR,
(errcode(ERRCODE_WINDOWING_ERROR),
errmsg("frame start cannot be UNBOUNDED FOLLOWING"),
scanner_errposition(@1)));
parser_errposition(@1)));
if ($1 & FRAMEOPTION_START_CURRENT_ROW)
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("frame start at CURRENT ROW is not implemented"),
scanner_errposition(@1)));
parser_errposition(@1)));
$$ = $1 | FRAMEOPTION_END_CURRENT_ROW;
}
| BETWEEN frame_bound AND frame_bound
@ -9428,17 +9440,17 @@ frame_extent: frame_bound
ereport(ERROR,
(errcode(ERRCODE_WINDOWING_ERROR),
errmsg("frame start cannot be UNBOUNDED FOLLOWING"),
scanner_errposition(@2)));
parser_errposition(@2)));
if ($2 & FRAMEOPTION_START_CURRENT_ROW)
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("frame start at CURRENT ROW is not implemented"),
scanner_errposition(@2)));
parser_errposition(@2)));
if ($4 & FRAMEOPTION_START_UNBOUNDED_PRECEDING)
ereport(ERROR,
(errcode(ERRCODE_WINDOWING_ERROR),
errmsg("frame end cannot be UNBOUNDED PRECEDING"),
scanner_errposition(@4)));
parser_errposition(@4)));
/* shift converts START_ options to END_ options */
$$ = FRAMEOPTION_BETWEEN | $2 | ($4 << 1);
}
@ -9742,11 +9754,11 @@ case_arg: a_expr { $$ = $1; }
*/
columnref: relation_name
{
$$ = makeColumnRef($1, NIL, @1);
$$ = makeColumnRef($1, NIL, @1, yyscanner);
}
| relation_name indirection
{
$$ = makeColumnRef($1, $2, @1);
$$ = makeColumnRef($1, $2, @1, yyscanner);
}
;
@ -9912,7 +9924,7 @@ qualified_name:
}
| relation_name indirection
{
check_qualified_name($2);
check_qualified_name($2, yyscanner);
$$ = makeNode(RangeVar);
switch (list_length($2))
{
@ -9931,7 +9943,7 @@ qualified_name:
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("improper qualified name (too many dotted names): %s",
NameListToString(lcons(makeString($1), $2))),
scanner_errposition(@1)));
parser_errposition(@1)));
break;
}
$$->location = @1;
@ -9970,7 +9982,10 @@ file_name: Sconst { $$ = $1; };
func_name: type_function_name
{ $$ = list_make1(makeString($1)); }
| relation_name indirection
{ $$ = check_func_name(lcons(makeString($1), $2)); }
{
$$ = check_func_name(lcons(makeString($1), $2),
yyscanner);
}
;
@ -10036,7 +10051,7 @@ AexprConst: Iconst
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("interval precision specified twice"),
scanner_errposition(@1)));
parser_errposition(@1)));
t->typmods = lappend($6, makeIntConst($3, @3));
}
else
@ -10552,30 +10567,42 @@ reserved_keyword:
SpecialRuleRelation:
OLD
{
if (QueryIsRule)
if (pg_yyget_extra(yyscanner)->QueryIsRule)
$$ = "*OLD*";
else
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("OLD used in query that is not in a rule"),
scanner_errposition(@1)));
parser_errposition(@1)));
}
| NEW
{
if (QueryIsRule)
if (pg_yyget_extra(yyscanner)->QueryIsRule)
$$ = "*NEW*";
else
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("NEW used in query that is not in a rule"),
scanner_errposition(@1)));
parser_errposition(@1)));
}
;
%%
/*
* The signature of this function is required by bison. However, we
* ignore the passed yylloc and instead use the last token position
* available from the scanner.
*/
static void
base_yyerror(YYLTYPE *yylloc, base_yyscan_t yyscanner, const char *msg)
{
parser_yyerror(msg);
}
static Node *
makeColumnRef(char *colname, List *indirection, int location)
makeColumnRef(char *colname, List *indirection,
int location, base_yyscan_t yyscanner)
{
/*
* Generate a ColumnRef node, with an A_Indirection node added if there
@ -10598,13 +10625,14 @@ makeColumnRef(char *colname, List *indirection, int location)
{
/* easy case - all indirection goes to A_Indirection */
c->fields = list_make1(makeString(colname));
i->indirection = check_indirection(indirection);
i->indirection = check_indirection(indirection, yyscanner);
}
else
{
/* got to split the list in two */
i->indirection = check_indirection(list_copy_tail(indirection,
nfields));
nfields),
yyscanner);
indirection = list_truncate(indirection, nfields);
c->fields = lcons(makeString(colname), indirection);
}
@ -10615,7 +10643,7 @@ makeColumnRef(char *colname, List *indirection, int location)
{
/* We only allow '*' at the end of a ColumnRef */
if (lnext(l) != NULL)
yyerror("improper use of \"*\"");
parser_yyerror("improper use of \"*\"");
}
nfields++;
}
@ -10744,7 +10772,7 @@ makeBoolAConst(bool state, int location)
* Create and populate a FuncCall node to support the OVERLAPS operator.
*/
static FuncCall *
makeOverlaps(List *largs, List *rargs, int location)
makeOverlaps(List *largs, List *rargs, int location, base_yyscan_t yyscanner)
{
FuncCall *n = makeNode(FuncCall);
@ -10755,14 +10783,14 @@ makeOverlaps(List *largs, List *rargs, int location)
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("wrong number of parameters on left side of OVERLAPS expression"),
scanner_errposition(location)));
parser_errposition(location)));
if (list_length(rargs) == 1)
rargs = lappend(rargs, rargs);
else if (list_length(rargs) != 2)
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("wrong number of parameters on right side of OVERLAPS expression"),
scanner_errposition(location)));
parser_errposition(location)));
n->args = list_concat(largs, rargs);
n->agg_star = FALSE;
n->agg_distinct = FALSE;
@ -10778,14 +10806,14 @@ makeOverlaps(List *largs, List *rargs, int location)
* subscripts and '*', which we then must reject here.
*/
static void
check_qualified_name(List *names)
check_qualified_name(List *names, base_yyscan_t yyscanner)
{
ListCell *i;
foreach(i, names)
{
if (!IsA(lfirst(i), String))
yyerror("syntax error");
parser_yyerror("syntax error");
}
}
@ -10795,14 +10823,14 @@ check_qualified_name(List *names)
* and '*', which we then must reject here.
*/
static List *
check_func_name(List *names)
check_func_name(List *names, base_yyscan_t yyscanner)
{
ListCell *i;
foreach(i, names)
{
if (!IsA(lfirst(i), String))
yyerror("syntax error");
parser_yyerror("syntax error");
}
return names;
}
@ -10813,7 +10841,7 @@ check_func_name(List *names)
* in the grammar, so do it here.
*/
static List *
check_indirection(List *indirection)
check_indirection(List *indirection, base_yyscan_t yyscanner)
{
ListCell *l;
@ -10822,7 +10850,7 @@ check_indirection(List *indirection)
if (IsA(lfirst(l), A_Star))
{
if (lnext(l) != NULL)
yyerror("improper use of \"*\"");
parser_yyerror("improper use of \"*\"");
}
}
return indirection;
@ -10871,7 +10899,8 @@ static void
insertSelectOptions(SelectStmt *stmt,
List *sortClause, List *lockingClause,
Node *limitOffset, Node *limitCount,
WithClause *withClause)
WithClause *withClause,
base_yyscan_t yyscanner)
{
Assert(IsA(stmt, SelectStmt));
@ -10885,7 +10914,7 @@ insertSelectOptions(SelectStmt *stmt,
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("multiple ORDER BY clauses not allowed"),
scanner_errposition(exprLocation((Node *) sortClause))));
parser_errposition(exprLocation((Node *) sortClause))));
stmt->sortClause = sortClause;
}
/* We can handle multiple locking clauses, though */
@ -10896,7 +10925,7 @@ insertSelectOptions(SelectStmt *stmt,
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("multiple OFFSET clauses not allowed"),
scanner_errposition(exprLocation(limitOffset))));
parser_errposition(exprLocation(limitOffset))));
stmt->limitOffset = limitOffset;
}
if (limitCount)
@ -10905,7 +10934,7 @@ insertSelectOptions(SelectStmt *stmt,
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("multiple LIMIT clauses not allowed"),
scanner_errposition(exprLocation(limitCount))));
parser_errposition(exprLocation(limitCount))));
stmt->limitCount = limitCount;
}
if (withClause)
@ -10914,7 +10943,7 @@ insertSelectOptions(SelectStmt *stmt,
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("multiple WITH clauses not allowed"),
scanner_errposition(exprLocation((Node *) withClause))));
parser_errposition(exprLocation((Node *) withClause))));
stmt->withClause = withClause;
}
}
@ -11046,9 +11075,10 @@ makeXmlExpr(XmlExprOp op, char *name, List *named_args, List *args,
* Initialize to parse one query string
*/
void
parser_init(void)
parser_init(base_yy_extra_type *yyext)
{
QueryIsRule = FALSE;
yyext->parsetree = NIL; /* in case grammar forgets to set it */
yyext->QueryIsRule = FALSE;
}
/*
@ -11102,4 +11132,9 @@ TableFuncTypeName(List *columns)
*/
#undef base_yylex
/* Undefine some other stuff that would conflict in scan.c, too */
#undef yyerror
#undef yylval
#undef yylloc
#include "scan.c"

98
src/backend/parser/parser.c
View File

@ -14,7 +14,7 @@
* Portions Copyright (c) 1994, Regents of the University of California
*
* IDENTIFICATION
* $PostgreSQL: pgsql/src/backend/parser/parser.c,v 1.79 2009/07/12 17:12:34 tgl Exp $
* $PostgreSQL: pgsql/src/backend/parser/parser.c,v 1.80 2009/07/13 02:02:20 tgl Exp $
*
*-------------------------------------------------------------------------
*/
@ -25,14 +25,6 @@
#include "parser/parser.h"
List *parsetree; /* result of parsing is left here */
static bool have_lookahead; /* is lookahead info valid? */
static int lookahead_token; /* one-token lookahead */
static YYSTYPE lookahead_yylval; /* yylval for lookahead token */
static YYLTYPE lookahead_yylloc; /* yylloc for lookahead token */
/*
* raw_parser
* Given a query in string form, do lexical and grammatical analysis.
@ -42,22 +34,29 @@ static YYLTYPE lookahead_yylloc; /* yylloc for lookahead token */
List *
raw_parser(const char *str)
{
base_yyscan_t yyscanner;
base_yy_extra_type yyextra;
int yyresult;
parsetree = NIL; /* in case grammar forgets to set it */
have_lookahead = false;
/* initialize the flex scanner */
yyscanner = scanner_init(str, &yyextra);
scanner_init(str);
parser_init();
/* filtered_base_yylex() only needs this much initialization */
yyextra.have_lookahead = false;
yyresult = base_yyparse();
/* initialize the bison parser */
parser_init(&yyextra);
scanner_finish();
/* Parse! */
yyresult = base_yyparse(yyscanner);
/* Clean up (release memory) */
scanner_finish(yyscanner);
if (yyresult) /* error */
return NIL;
return parsetree;
return yyextra.parsetree;
}
@ -69,25 +68,27 @@ raw_parser(const char *str)
* passed string does represent one single string literal.
*
* We export this function to avoid having plpgsql depend on internal details
* of the core grammar (such as the token code assigned to SCONST). Note
* that since the scanner isn't presently re-entrant, this cannot be used
* during use of the main parser/scanner.
* of the core grammar (such as the token code assigned to SCONST).
*/
char *
pg_parse_string_token(const char *token)
{
base_yyscan_t yyscanner;
base_yy_extra_type yyextra;
int ctoken;
YYSTYPE yylval;
YYLTYPE yylloc;
scanner_init(token);
yyscanner = scanner_init(token, &yyextra);
ctoken = base_yylex();
ctoken = base_yylex(&yylval, &yylloc, yyscanner);
if (ctoken != SCONST) /* caller error */
elog(ERROR, "expected string constant, got token code %d", ctoken);
scanner_finish();
scanner_finish(yyscanner);
return base_yylval.str;
return yylval.str;
}
@ -105,23 +106,24 @@ pg_parse_string_token(const char *token)
* layer does.
*/
int
filtered_base_yylex(void)
filtered_base_yylex(YYSTYPE *lvalp, YYLTYPE *llocp, base_yyscan_t yyscanner)
{
base_yy_extra_type *yyextra = pg_yyget_extra(yyscanner);
int cur_token;
int next_token;
YYSTYPE cur_yylval;
YYLTYPE cur_yylloc;
/* Get next token --- we might already have it */
if (have_lookahead)
if (yyextra->have_lookahead)
{
cur_token = lookahead_token;
base_yylval = lookahead_yylval;
base_yylloc = lookahead_yylloc;
have_lookahead = false;
cur_token = yyextra->lookahead_token;
*lvalp = yyextra->lookahead_yylval;
*llocp = yyextra->lookahead_yylloc;
yyextra->have_lookahead = false;
}
else
cur_token = base_yylex();
cur_token = base_yylex(lvalp, llocp, yyscanner);
/* Do we need to look ahead for a possible multiword token? */
switch (cur_token)
@ -131,9 +133,9 @@ filtered_base_yylex(void)
/*
* NULLS FIRST and NULLS LAST must be reduced to one token
*/
cur_yylval = base_yylval;
cur_yylloc = base_yylloc;
next_token = base_yylex();
cur_yylval = *lvalp;
cur_yylloc = *llocp;
next_token = base_yylex(lvalp, llocp, yyscanner);
switch (next_token)
{
case FIRST_P:
@ -144,13 +146,13 @@ filtered_base_yylex(void)
break;
default:
/* save the lookahead token for next time */
lookahead_token = next_token;
lookahead_yylval = base_yylval;
lookahead_yylloc = base_yylloc;
have_lookahead = true;
yyextra->lookahead_token = next_token;
yyextra->lookahead_yylval = *lvalp;
yyextra->lookahead_yylloc = *llocp;
yyextra->have_lookahead = true;
/* and back up the output info to cur_token */
base_yylval = cur_yylval;
base_yylloc = cur_yylloc;
*lvalp = cur_yylval;
*llocp = cur_yylloc;
break;
}
break;
@ -160,9 +162,9 @@ filtered_base_yylex(void)
/*
* WITH TIME must be reduced to one token
*/
cur_yylval = base_yylval;
cur_yylloc = base_yylloc;
next_token = base_yylex();
cur_yylval = *lvalp;
cur_yylloc = *llocp;
next_token = base_yylex(lvalp, llocp, yyscanner);
switch (next_token)
{
case TIME:
@ -170,13 +172,13 @@ filtered_base_yylex(void)
break;
default:
/* save the lookahead token for next time */
lookahead_token = next_token;
lookahead_yylval = base_yylval;
lookahead_yylloc = base_yylloc;
have_lookahead = true;
yyextra->lookahead_token = next_token;
yyextra->lookahead_yylval = *lvalp;
yyextra->lookahead_yylloc = *llocp;
yyextra->have_lookahead = true;
/* and back up the output info to cur_token */
base_yylval = cur_yylval;
base_yylloc = cur_yylloc;
*lvalp = cur_yylval;
*llocp = cur_yylloc;
break;
}
break;

416
src/backend/parser/scan.l
View File

@ -24,7 +24,7 @@
* Portions Copyright (c) 1994, Regents of the University of California
*
* IDENTIFICATION
* $PostgreSQL: pgsql/src/backend/parser/scan.l,v 1.154 2009/07/12 17:12:34 tgl Exp $
* $PostgreSQL: pgsql/src/backend/parser/scan.l,v 1.155 2009/07/13 02:02:20 tgl Exp $
*
*-------------------------------------------------------------------------
*/
@ -43,9 +43,6 @@
#undef fprintf
#define fprintf(file, fmt, msg) ereport(ERROR, (errmsg_internal("%s", msg)))
static int xcdepth = 0; /* depth of nesting in slash-star comments */
static char *dolqstart; /* current $foo$ quote start string */
/*
* GUC variables. This is a DIRECT violation of the warning given at the
* head of gram.y, ie flex/bison code must not depend on any GUC variables;
@ -57,29 +54,11 @@ int backslash_quote = BACKSLASH_QUOTE_SAFE_ENCODING;
bool escape_string_warning = true;
bool standard_conforming_strings = false;
static bool warn_on_first_escape;
static bool saw_non_ascii = false;
/*
* literalbuf is used to accumulate literal values when multiple rules
* are needed to parse a single literal. Call startlit to reset buffer
* to empty, addlit to add text. Note that the buffer is palloc'd and
* starts life afresh on every parse cycle.
* Set the type of yyextra. All state variables used by the scanner should
* be in yyextra, *not* statically allocated.
*/
static char *literalbuf; /* expandable buffer */
static int literallen; /* actual current length */
static int literalalloc; /* current allocated buffer size */
#define startlit() (literalbuf[0] = '\0', literallen = 0)
static void addlit(char *ytext, int yleng);
static void addlitchar(unsigned char ychar);
static char *litbufdup(void);
static char *litbuf_udeescape(unsigned char escape);
#define lexer_errposition() scanner_errposition(yylloc)
static void check_escape_warning(void);
static void check_string_escape_warning(unsigned char ychar);
#define YY_EXTRA_TYPE base_yy_extra_type *
/*
* Each call to yylex must set yylloc to the location of the found token
@ -88,22 +67,51 @@ static void check_string_escape_warning(unsigned char ychar);
* this should be done in the first such rule, else yylloc will point
* into the middle of the token.
*/
#define SET_YYLLOC() (yylloc = yytext - scanbuf)
#define SET_YYLLOC() (*(yylloc) = yytext - yyextra->scanbuf)
/* Handles to the buffer that the lexer uses internally */
static YY_BUFFER_STATE scanbufhandle;
static char *scanbuf;
/*
* Advance yylloc by the given number of bytes.
*/
#define ADVANCE_YYLLOC(delta) ( *(yylloc) += (delta) )
static unsigned char unescape_single_char(unsigned char c);
#define startlit() ( yyextra->literallen = 0 )
static void addlit(char *ytext, int yleng, base_yyscan_t yyscanner);
static void addlitchar(unsigned char ychar, base_yyscan_t yyscanner);
static char *litbufdup(base_yyscan_t yyscanner);
static char *litbuf_udeescape(unsigned char escape, base_yyscan_t yyscanner);
static unsigned char unescape_single_char(unsigned char c, base_yyscan_t yyscanner);
#define yyerror(msg) scanner_yyerror(msg, yyscanner)
#define lexer_errposition() scanner_errposition(*(yylloc), yyscanner)
static void check_string_escape_warning(unsigned char ychar, base_yyscan_t yyscanner);
static void check_escape_warning(base_yyscan_t yyscanner);
/*
* Work around a bug in flex 2.5.35: it emits a couple of functions that
* it forgets to emit declarations for. Since we use -Wmissing-prototypes,
* this would cause warnings. Providing our own declarations should be
* harmless even when the bug gets fixed.
*/
extern int base_yyget_column(yyscan_t yyscanner);
extern void base_yyset_column(int column_no, yyscan_t yyscanner);
%}
%option reentrant
%option bison-bridge
%option bison-locations
%option 8bit
%option never-interactive
%option nodefault
%option noinput
%option nounput
%option noyywrap
%option noyyalloc
%option noyyrealloc
%option noyyfree
%option warn
%option prefix="base_yy"
/*
@ -350,23 +358,23 @@ other .
{xcstart} {
/* Set location in case of syntax error in comment */
SET_YYLLOC();
xcdepth = 0;
yyextra->xcdepth = 0;
BEGIN(xc);
/* Put back any characters past slash-star; see above */
yyless(2);
}
<xc>{xcstart} {
xcdepth++;
(yyextra->xcdepth)++;
/* Put back any characters past slash-star; see above */
yyless(2);
}
<xc>{xcstop} {
if (xcdepth <= 0)
if (yyextra->xcdepth <= 0)
BEGIN(INITIAL);
else
xcdepth--;
(yyextra->xcdepth)--;
}
<xc>{xcinside} {
@ -393,18 +401,18 @@ other .
SET_YYLLOC();
BEGIN(xb);
startlit();
addlitchar('b');
addlitchar('b', yyscanner);
}
<xb>{quotestop} |
<xb>{quotefail} {
yyless(1);
BEGIN(INITIAL);
yylval.str = litbufdup();
yylval->str = litbufdup(yyscanner);
return BCONST;
}
<xh>{xhinside} |
<xb>{xbinside} {
addlit(yytext, yyleng);
addlit(yytext, yyleng, yyscanner);
}
<xh>{quotecontinue} |
<xb>{quotecontinue} {
@ -422,13 +430,13 @@ other .
SET_YYLLOC();
BEGIN(xh);
startlit();
addlitchar('x');
addlitchar('x', yyscanner);
}
<xh>{quotestop} |
<xh>{quotefail} {
yyless(1);
BEGIN(INITIAL);
yylval.str = litbufdup();
yylval->str = litbufdup(yyscanner);
return XCONST;
}
<xh><<EOF>> { yyerror("unterminated hexadecimal string literal"); }
@ -445,13 +453,13 @@ other .
/* nchar had better be a keyword! */
keyword = ScanKeywordLookup("nchar");
Assert(keyword != NULL);
yylval.keyword = keyword->name;
yylval->keyword = keyword->name;
return keyword->value;
}
{xqstart} {
warn_on_first_escape = true;
saw_non_ascii = false;
yyextra->warn_on_first_escape = true;
yyextra->saw_non_ascii = false;
SET_YYLLOC();
if (standard_conforming_strings)
BEGIN(xq);
@ -460,8 +468,8 @@ other .
startlit();
}
{xestart} {
warn_on_first_escape = false;
saw_non_ascii = false;
yyextra->warn_on_first_escape = false;
yyextra->saw_non_ascii = false;
SET_YYLLOC();
BEGIN(xe);
startlit();
@ -485,31 +493,33 @@ other .
* check that the data remains valid if it might have been
* made invalid by unescaping any chars.
*/
if (saw_non_ascii)
pg_verifymbstr(literalbuf, literallen, false);
yylval.str = litbufdup();
if (yyextra->saw_non_ascii)
pg_verifymbstr(yyextra->literalbuf,
yyextra->literallen,
false);
yylval->str = litbufdup(yyscanner);
return SCONST;
}
<xus>{xusstop1} {
/* throw back all but the quote */
yyless(1);
BEGIN(INITIAL);
yylval.str = litbuf_udeescape('\\');
yylval->str = litbuf_udeescape('\\', yyscanner);
return SCONST;
}
<xus>{xusstop2} {
BEGIN(INITIAL);
yylval.str = litbuf_udeescape(yytext[yyleng-2]);
yylval->str = litbuf_udeescape(yytext[yyleng-2], yyscanner);
return SCONST;
}
<xq,xe,xus>{xqdouble} {
addlitchar('\'');
addlitchar('\'', yyscanner);
}
<xq,xus>{xqinside} {
addlit(yytext, yyleng);
addlit(yytext, yyleng, yyscanner);
}
<xe>{xeinside} {
addlit(yytext, yyleng);
addlit(yytext, yyleng, yyscanner);
}
<xe>{xeescape} {
if (yytext[1] == '\'')
@ -523,37 +533,38 @@ other .
errhint("Use '' to write quotes in strings. \\' is insecure in client-only encodings."),
lexer_errposition()));
}
check_string_escape_warning(yytext[1]);
addlitchar(unescape_single_char(yytext[1]));
check_string_escape_warning(yytext[1], yyscanner);
addlitchar(unescape_single_char(yytext[1], yyscanner),
yyscanner);
}
<xe>{xeoctesc} {
unsigned char c = strtoul(yytext+1, NULL, 8);
check_escape_warning();
addlitchar(c);
check_escape_warning(yyscanner);
addlitchar(c, yyscanner);
if (c == '\0' || IS_HIGHBIT_SET(c))
saw_non_ascii = true;
yyextra->saw_non_ascii = true;
}
<xe>{xehexesc} {
unsigned char c = strtoul(yytext+2, NULL, 16);
check_escape_warning();
addlitchar(c);
check_escape_warning(yyscanner);
addlitchar(c, yyscanner);
if (c == '\0' || IS_HIGHBIT_SET(c))
saw_non_ascii = true;
yyextra->saw_non_ascii = true;
}
<xq,xe,xus>{quotecontinue} {
/* ignore */
}
<xe>. {
/* This is only needed for \ just before EOF */
addlitchar(yytext[0]);
addlitchar(yytext[0], yyscanner);
}
<xq,xe,xus><<EOF>> { yyerror("unterminated quoted string"); }
{dolqdelim} {
SET_YYLLOC();
dolqstart = pstrdup(yytext);
yyextra->dolqstart = pstrdup(yytext);
BEGIN(xdolq);
startlit();
}
@ -565,11 +576,12 @@ other .
return yytext[0];
}
<xdolq>{dolqdelim} {
if (strcmp(yytext, dolqstart) == 0)
if (strcmp(yytext, yyextra->dolqstart) == 0)
{
pfree(dolqstart);
pfree(yyextra->dolqstart);
yyextra->dolqstart = NULL;
BEGIN(INITIAL);
yylval.str = litbufdup();
yylval->str = litbufdup(yyscanner);
return SCONST;
}
else
@ -579,19 +591,19 @@ other .
* the $... part to the output, but put back the final
* $ for rescanning. Consider $delim$...$junk$delim$
*/
addlit(yytext, yyleng-1);
addlit(yytext, yyleng-1, yyscanner);
yyless(yyleng-1);
}
}
<xdolq>{dolqinside} {
addlit(yytext, yyleng);
addlit(yytext, yyleng, yyscanner);
}
<xdolq>{dolqfailed} {
addlit(yytext, yyleng);
addlit(yytext, yyleng, yyscanner);
}
<xdolq>. {
/* This is only needed for $ inside the quoted text */
addlitchar(yytext[0]);
addlitchar(yytext[0], yyscanner);
}
<xdolq><<EOF>> { yyerror("unterminated dollar-quoted string"); }
@ -609,24 +621,24 @@ other .
char *ident;
BEGIN(INITIAL);
if (literallen == 0)
if (yyextra->literallen == 0)
yyerror("zero-length delimited identifier");
ident = litbufdup();
if (literallen >= NAMEDATALEN)
truncate_identifier(ident, literallen, true);
yylval.str = ident;
ident = litbufdup(yyscanner);
if (yyextra->literallen >= NAMEDATALEN)
truncate_identifier(ident, yyextra->literallen, true);
yylval->str = ident;
return IDENT;
}
<xui>{xuistop1} {
char *ident;
BEGIN(INITIAL);
if (literallen == 0)
if (yyextra->literallen == 0)
yyerror("zero-length delimited identifier");
ident = litbuf_udeescape('\\');
if (literallen >= NAMEDATALEN)
truncate_identifier(ident, literallen, true);
yylval.str = ident;
ident = litbuf_udeescape('\\', yyscanner);
if (yyextra->literallen >= NAMEDATALEN)
truncate_identifier(ident, yyextra->literallen, true);
yylval->str = ident;
/* throw back all but the quote */
yyless(1);
return IDENT;
@ -635,19 +647,19 @@ other .
char *ident;
BEGIN(INITIAL);
if (literallen == 0)
if (yyextra->literallen == 0)
yyerror("zero-length delimited identifier");
ident = litbuf_udeescape(yytext[yyleng - 2]);
if (literallen >= NAMEDATALEN)
truncate_identifier(ident, literallen, true);
yylval.str = ident;
ident = litbuf_udeescape(yytext[yyleng - 2], yyscanner);
if (yyextra->literallen >= NAMEDATALEN)
truncate_identifier(ident, yyextra->literallen, true);
yylval->str = ident;
return IDENT;
}
<xd,xui>{xddouble} {
addlitchar('"');
addlitchar('"', yyscanner);
}
<xd,xui>{xdinside} {
addlit(yytext, yyleng);
addlit(yytext, yyleng, yyscanner);
}
<xd,xui><<EOF>> { yyerror("unterminated quoted identifier"); }
@ -659,7 +671,7 @@ other .
yyless(1);
/* and treat it as {identifier} */
ident = downcase_truncate_identifier(yytext, yyleng, true);
yylval.str = ident;
yylval->str = ident;
return IDENT;
}
@ -747,15 +759,15 @@ other .
/* Convert "!=" operator to "<>" for compatibility */
if (strcmp(yytext, "!=") == 0)
yylval.str = pstrdup("<>");
yylval->str = pstrdup("<>");
else
yylval.str = pstrdup(yytext);
yylval->str = pstrdup(yytext);
return Op;
}
{param} {
SET_YYLLOC();
yylval.ival = atol(yytext + 1);
yylval->ival = atol(yytext + 1);
return PARAM;
}
@ -774,20 +786,20 @@ other .
)
{
/* integer too large, treat it as a float */
yylval.str = pstrdup(yytext);
yylval->str = pstrdup(yytext);
return FCONST;
}
yylval.ival = val;
yylval->ival = val;
return ICONST;
}
{decimal} {
SET_YYLLOC();
yylval.str = pstrdup(yytext);
yylval->str = pstrdup(yytext);
return FCONST;
}
{real} {
SET_YYLLOC();
yylval.str = pstrdup(yytext);
yylval->str = pstrdup(yytext);
return FCONST;
}
{realfail1} {
@ -799,14 +811,14 @@ other .
*/
yyless(yyleng-1);
SET_YYLLOC();
yylval.str = pstrdup(yytext);
yylval->str = pstrdup(yytext);
return FCONST;
}
{realfail2} {
/* throw back the [Ee][+-], and proceed as above */
yyless(yyleng-2);
SET_YYLLOC();
yylval.str = pstrdup(yytext);
yylval->str = pstrdup(yytext);
return FCONST;
}
@ -821,7 +833,7 @@ other .
keyword = ScanKeywordLookup(yytext);
if (keyword != NULL)
{
yylval.keyword = keyword->name;
yylval->keyword = keyword->name;
return keyword->value;
}
@ -830,7 +842,7 @@ other .
* if necessary.
*/
ident = downcase_truncate_identifier(yytext, yyleng, true);
yylval.str = ident;
yylval->str = ident;
return IDENT;
}
@ -846,6 +858,22 @@ other .
%%
/*
* Arrange access to yyextra for subroutines of the main yylex() function.
* We expect each subroutine to have a yyscanner parameter. Rather than
* use the yyget_xxx functions, which might or might not get inlined by the
* compiler, we cheat just a bit and cast yyscanner to the right type.
*/
#undef yyextra
#define yyextra (((struct yyguts_t *) yyscanner)->yyextra_r)
/* Likewise for a couple of other things we need. */
#undef yylloc
#define yylloc (((struct yyguts_t *) yyscanner)->yylloc_r)
#undef yyleng
#define yyleng (((struct yyguts_t *) yyscanner)->yyleng_r)
/*
* scanner_errposition
* Report a lexer or grammar error cursor position, if possible.
@ -854,38 +882,39 @@ other .
* is a dummy (always 0, in fact).
*
* Note that this can only be used for messages emitted during raw parsing
* (essentially, scan.l and gram.y), since it requires scanbuf to still be
* valid.
* (essentially, scan.l and gram.y), since it requires the yyscanner struct
* to still be available.
*/
int
scanner_errposition(int location)
scanner_errposition(int location, base_yyscan_t yyscanner)
{
int pos;
Assert(scanbuf != NULL); /* else called from wrong place */
if (location < 0)
return 0; /* no-op if location is unknown */
/* Convert byte offset to character number */
pos = pg_mbstrlen_with_len(scanbuf, location) + 1;
pos = pg_mbstrlen_with_len(yyextra->scanbuf, location) + 1;
/* And pass it to the ereport mechanism */
return errposition(pos);
}
/*
* yyerror
* scanner_yyerror
* Report a lexer or grammar error.
*
* The message's cursor position identifies the most recently lexed token.
* The message's cursor position is whatever YYLLOC was last set to,
* ie, the start of the current token if called within yylex(), or the
* most recently lexed token if called from the grammar.
* This is OK for syntax error messages from the Bison parser, because Bison
* parsers report error as soon as the first unparsable token is reached.
* Beware of using yyerror for other purposes, as the cursor position might
* be misleading!
*/
void
yyerror(const char *message)
scanner_yyerror(const char *message, base_yyscan_t yyscanner)
{
const char *loc = scanbuf + yylloc;
const char *loc = yyextra->scanbuf + *yylloc;
if (*loc == YY_END_OF_BUFFER_CHAR)
{
@ -909,31 +938,32 @@ yyerror(const char *message)
/*
* Called before any actual parsing is done
*/
void
scanner_init(const char *str)
base_yyscan_t
scanner_init(const char *str, base_yy_extra_type *yyext)
{
Size slen = strlen(str);
Size slen = strlen(str);
yyscan_t scanner;
/*
* Might be left over after ereport()
*/
if (YY_CURRENT_BUFFER)
yy_delete_buffer(YY_CURRENT_BUFFER);
if (yylex_init(&scanner) != 0)
elog(ERROR, "yylex_init() failed: %m");
base_yyset_extra(yyext, scanner);
/*
* Make a scan buffer with special termination needed by flex.
*/
scanbuf = palloc(slen + 2);
memcpy(scanbuf, str, slen);
scanbuf[slen] = scanbuf[slen + 1] = YY_END_OF_BUFFER_CHAR;
scanbufhandle = yy_scan_buffer(scanbuf, slen + 2);
yyext->scanbuf = (char *) palloc(slen + 2);
yyext->scanbuflen = slen;
memcpy(yyext->scanbuf, str, slen);
yyext->scanbuf[slen] = yyext->scanbuf[slen + 1] = YY_END_OF_BUFFER_CHAR;
yy_scan_buffer(yyext->scanbuf, slen + 2, scanner);
/* initialize literal buffer to a reasonable but expansible size */
literalalloc = 1024;
literalbuf = (char *) palloc(literalalloc);
startlit();
yyext->literalalloc = 1024;
yyext->literalbuf = (char *) palloc(yyext->literalalloc);
yyext->literallen = 0;
BEGIN(INITIAL);
return scanner;
}
@ -941,60 +971,71 @@ scanner_init(const char *str)
* Called after parsing is done to clean up after scanner_init()
*/
void
scanner_finish(void)
scanner_finish(base_yyscan_t yyscanner)
{
yy_delete_buffer(scanbufhandle);
pfree(scanbuf);
scanbuf = NULL;
/*
* We don't bother to call yylex_destroy(), because all it would do
* is pfree a small amount of control storage. It's cheaper to leak
* the storage until the parsing context is destroyed. The amount of
* space involved is usually negligible compared to the output parse
* tree anyway.
*
* We do bother to pfree the scanbuf and literal buffer, but only if they
* represent a nontrivial amount of space. The 8K cutoff is arbitrary.
*/
if (yyextra->scanbuflen >= 8192)
pfree(yyextra->scanbuf);
if (yyextra->literalalloc >= 8192)
pfree(yyextra->literalbuf);
}
static void
addlit(char *ytext, int yleng)
addlit(char *ytext, int yleng, base_yyscan_t yyscanner)
{
/* enlarge buffer if needed */
if ((literallen+yleng) >= literalalloc)
if ((yyextra->literallen + yleng) >= yyextra->literalalloc)
{
do {
literalalloc *= 2;
} while ((literallen+yleng) >= literalalloc);
literalbuf = (char *) repalloc(literalbuf, literalalloc);
yyextra->literalalloc *= 2;
} while ((yyextra->literallen + yleng) >= yyextra->literalalloc);
yyextra->literalbuf = (char *) repalloc(yyextra->literalbuf,
yyextra->literalalloc);
}
/* append new data, add trailing null */
memcpy(literalbuf+literallen, ytext, yleng);
literallen += yleng;
literalbuf[literallen] = '\0';
/* append new data */
memcpy(yyextra->literalbuf + yyextra->literallen, ytext, yleng);
yyextra->literallen += yleng;
}
static void
addlitchar(unsigned char ychar)
addlitchar(unsigned char ychar, base_yyscan_t yyscanner)
{
/* enlarge buffer if needed */
if ((literallen+1) >= literalalloc)
if ((yyextra->literallen + 1) >= yyextra->literalalloc)
{
literalalloc *= 2;
literalbuf = (char *) repalloc(literalbuf, literalalloc);
yyextra->literalalloc *= 2;
yyextra->literalbuf = (char *) repalloc(yyextra->literalbuf,
yyextra->literalalloc);
}
/* append new data, add trailing null */
literalbuf[literallen] = ychar;
literallen += 1;
literalbuf[literallen] = '\0';
/* append new data */
yyextra->literalbuf[yyextra->literallen] = ychar;
yyextra->literallen += 1;
}
/*
* One might be tempted to write pstrdup(literalbuf) instead of this,
* but for long literals this is much faster because the length is
* already known.
* Create a palloc'd copy of literalbuf, adding a trailing null.
*/
static char *
litbufdup(void)
litbufdup(base_yyscan_t yyscanner)
{
char *new;
int llen = yyextra->literallen;
char *new;
new = palloc(literallen + 1);
memcpy(new, literalbuf, literallen+1);
new = palloc(llen + 1);
memcpy(new, yyextra->literalbuf, llen);
new[llen] = '\0';
return new;
}
@ -1012,23 +1053,23 @@ hexval(unsigned char c)
}
static void
check_unicode_value(pg_wchar c, char * loc)
check_unicode_value(pg_wchar c, char *loc, base_yyscan_t yyscanner)
{
if (GetDatabaseEncoding() == PG_UTF8)
return;
if (c > 0x7F)
{
yylloc += (char *) loc - literalbuf + 3; /* 3 for U&" */
ADVANCE_YYLLOC(loc - yyextra->literalbuf + 3); /* 3 for U&" */
yyerror("Unicode escape values cannot be used for code point values above 007F when the server encoding is not UTF8");
}
}
static char *
litbuf_udeescape(unsigned char escape)
litbuf_udeescape(unsigned char escape, base_yyscan_t yyscanner)
{
char *new;
char *in, *out;
char *litbuf, *in, *out;
if (isxdigit(escape)
|| escape == '+'
@ -1036,17 +1077,21 @@ litbuf_udeescape(unsigned char escape)
|| escape == '"'
|| scanner_isspace(escape))
{
yylloc += literallen + yyleng + 1;
ADVANCE_YYLLOC(yyextra->literallen + yyleng + 1);
yyerror("invalid Unicode escape character");
}
/* Make literalbuf null-terminated to simplify the scanning loop */
litbuf = yyextra->literalbuf;
litbuf[yyextra->literallen] = '\0';
/*
* This relies on the subtle assumption that a UTF-8 expansion
* cannot be longer than its escaped representation.
*/
new = palloc(literallen + 1);
new = palloc(yyextra->literallen + 1);
in = literalbuf;
in = litbuf;
out = new;
while (*in)
{
@ -1060,7 +1105,7 @@ litbuf_udeescape(unsigned char escape)
else if (isxdigit(in[1]) && isxdigit(in[2]) && isxdigit(in[3]) && isxdigit(in[4]))
{
pg_wchar unicode = hexval(in[1]) * 16*16*16 + hexval(in[2]) * 16*16 + hexval(in[3]) * 16 + hexval(in[4]);
check_unicode_value(unicode, in);
check_unicode_value(unicode, in, yyscanner);
unicode_to_utf8(unicode, (unsigned char *) out);
in += 5;
out += pg_mblen(out);
@ -1072,14 +1117,14 @@ litbuf_udeescape(unsigned char escape)
{
pg_wchar unicode = hexval(in[2]) * 16*16*16*16*16 + hexval(in[3]) * 16*16*16*16 + hexval(in[4]) * 16*16*16
+ hexval(in[5]) * 16*16 + hexval(in[6]) * 16 + hexval(in[7]);
check_unicode_value(unicode, in);
check_unicode_value(unicode, in, yyscanner);
unicode_to_utf8(unicode, (unsigned char *) out);
in += 8;
out += pg_mblen(out);
}
else
{
yylloc += in - literalbuf + 3; /* 3 for U&" */
ADVANCE_YYLLOC(in - litbuf + 3); /* 3 for U&" */
yyerror("invalid Unicode escape value");
}
}
@ -1098,7 +1143,7 @@ litbuf_udeescape(unsigned char escape)
}
static unsigned char
unescape_single_char(unsigned char c)
unescape_single_char(unsigned char c, base_yyscan_t yyscanner)
{
switch (c)
{
@ -1115,47 +1160,74 @@ unescape_single_char(unsigned char c)
default:
/* check for backslash followed by non-7-bit-ASCII */
if (c == '\0' || IS_HIGHBIT_SET(c))
saw_non_ascii = true;
yyextra->saw_non_ascii = true;
return c;
}
}
static void
check_string_escape_warning(unsigned char ychar)
check_string_escape_warning(unsigned char ychar, base_yyscan_t yyscanner)
{
if (ychar == '\'')
{
if (warn_on_first_escape && escape_string_warning)
if (yyextra->warn_on_first_escape && escape_string_warning)
ereport(WARNING,
(errcode(ERRCODE_NONSTANDARD_USE_OF_ESCAPE_CHARACTER),
errmsg("nonstandard use of \\' in a string literal"),
errhint("Use '' to write quotes in strings, or use the escape string syntax (E'...')."),
lexer_errposition()));
warn_on_first_escape = false; /* warn only once per string */
yyextra->warn_on_first_escape = false; /* warn only once per string */
}
else if (ychar == '\\')
{
if (warn_on_first_escape && escape_string_warning)
if (yyextra->warn_on_first_escape && escape_string_warning)
ereport(WARNING,
(errcode(ERRCODE_NONSTANDARD_USE_OF_ESCAPE_CHARACTER),
errmsg("nonstandard use of \\\\ in a string literal"),
errhint("Use the escape string syntax for backslashes, e.g., E'\\\\'."),
lexer_errposition()));
warn_on_first_escape = false; /* warn only once per string */
yyextra->warn_on_first_escape = false; /* warn only once per string */
}
else
check_escape_warning();
check_escape_warning(yyscanner);
}
static void
check_escape_warning(void)
check_escape_warning(base_yyscan_t yyscanner)
{
if (warn_on_first_escape && escape_string_warning)
if (yyextra->warn_on_first_escape && escape_string_warning)
ereport(WARNING,
(errcode(ERRCODE_NONSTANDARD_USE_OF_ESCAPE_CHARACTER),
errmsg("nonstandard use of escape in a string literal"),
errhint("Use the escape string syntax for escapes, e.g., E'\\r\\n'."),
lexer_errposition()));
warn_on_first_escape = false; /* warn only once per string */
yyextra->warn_on_first_escape = false; /* warn only once per string */
}
/*
* Interface functions to make flex use palloc() instead of malloc().
* It'd be better to make these static, but flex insists otherwise.
*/
void *
base_yyalloc(size_t bytes, base_yyscan_t yyscanner)
{
return palloc(bytes);
}
void *
base_yyrealloc(void *ptr, size_t bytes, base_yyscan_t yyscanner)
{
if (ptr)
return repalloc(ptr, bytes);
else
return palloc(bytes);
}
void
base_yyfree(void *ptr, base_yyscan_t yyscanner)
{
if (ptr)
pfree(ptr);
}

84
src/include/parser/gramparse.h
View File

@ -11,7 +11,7 @@
* Portions Copyright (c) 1996-2009, PostgreSQL Global Development Group
* Portions Copyright (c) 1994, Regents of the University of California
*
* $PostgreSQL: pgsql/src/include/parser/gramparse.h,v 1.45 2009/07/12 17:12:34 tgl Exp $
* $PostgreSQL: pgsql/src/include/parser/gramparse.h,v 1.46 2009/07/13 02:02:20 tgl Exp $
*
*-------------------------------------------------------------------------
*/
@ -36,19 +36,85 @@
*/
#include "parser/gram.h"
/*
* The YY_EXTRA data that a flex scanner allows us to pass around. Private
* state needed for raw parsing/lexing goes here.
*/
typedef struct base_yy_extra_type
{
/*
* The string the lexer is physically scanning. We keep this mainly so
* that we can cheaply compute the offset of the current token (yytext).
*/
char *scanbuf;
Size scanbuflen;
/*
* literalbuf is used to accumulate literal values when multiple rules
* are needed to parse a single literal. Call startlit() to reset buffer
* to empty, addlit() to add text. NOTE: the string in literalbuf is
* NOT necessarily null-terminated, but there always IS room to add a
* trailing null at offset literallen. We store a null only when we
* need it.
*/
char *literalbuf; /* palloc'd expandable buffer */
int literallen; /* actual current string length */
int literalalloc; /* current allocated buffer size */
int xcdepth; /* depth of nesting in slash-star comments */
char *dolqstart; /* current $foo$ quote start string */
/* state variables for literal-lexing warnings */
bool warn_on_first_escape;
bool saw_non_ascii;
/*
* State variables for filtered_base_yylex().
*/
bool have_lookahead; /* is lookahead info valid? */
int lookahead_token; /* one-token lookahead */
YYSTYPE lookahead_yylval; /* yylval for lookahead token */
YYLTYPE lookahead_yylloc; /* yylloc for lookahead token */
/*
* State variables that belong to the grammar, not the lexer. It's
* simpler to keep these here than to invent a separate structure.
* These fields are unused/undefined if the lexer is invoked on its own.
*/
List *parsetree; /* final parse result is delivered here */
bool QueryIsRule; /* signals we are parsing CREATE RULE */
} base_yy_extra_type;
/*
* The type of yyscanner is opaque outside scan.l.
*/
typedef void *base_yyscan_t;
/*
* In principle we should use yyget_extra() to fetch the yyextra field
* from a yyscanner struct. However, flex always puts that field first,
* and this is sufficiently performance-critical to make it seem worth
* cheating a bit to use an inline macro.
*/
#define pg_yyget_extra(yyscanner) (*((base_yy_extra_type **) (yyscanner)))
/* from parser.c */
extern int filtered_base_yylex(void);
extern int filtered_base_yylex(YYSTYPE *lvalp, YYLTYPE *llocp,
base_yyscan_t yyscanner);
/* from scan.l */
extern void scanner_init(const char *str);
extern void scanner_finish(void);
extern int base_yylex(void);
extern int scanner_errposition(int location);
extern void base_yyerror(const char *message);
extern base_yyscan_t scanner_init(const char *str, base_yy_extra_type *yyext);
extern void scanner_finish(base_yyscan_t yyscanner);
extern int base_yylex(YYSTYPE *lvalp, YYLTYPE *llocp,
base_yyscan_t yyscanner);
extern int scanner_errposition(int location, base_yyscan_t yyscanner);
extern void scanner_yyerror(const char *message, base_yyscan_t yyscanner);
/* from gram.y */
extern void parser_init(void);
extern int base_yyparse(void);
extern void parser_init(base_yy_extra_type *yyext);
extern int base_yyparse(base_yyscan_t yyscanner);
#endif /* GRAMPARSE_H */