scan.l

Go to the documentation of this file.

%top{
/*-------------------------------------------------------------------------
 *
 * scan.l
 *    lexical scanner for PostgreSQL
 *
 * NOTE NOTE NOTE:
 *
 * The rules in this file must be kept in sync with src/fe_utils/psqlscan.l
 * and src/interfaces/ecpg/preproc/pgc.l!
 *
 * The rules are designed so that the scanner never has to backtrack,
 * in the sense that there is always a rule that can match the input
 * consumed so far (the rule action may internally throw back some input
 * with yyless(), however).  As explained in the flex manual, this makes
 * for a useful speed increase --- several percent faster when measuring
 * raw parsing (Flex + Bison).  The extra complexity is mostly in the rules
 * for handling float numbers and continued string literals.  If you change
 * the lexical rules, verify that you haven't broken the no-backtrack
 * property by running flex with the "-b" option and checking that the
 * resulting "lex.backup" file says that no backing up is needed.  (As of
 * Postgres 9.2, this check is made automatically by the Makefile.)
 *
 *
 * Portions Copyright (c) 1996-2026, PostgreSQL Global Development Group
 * Portions Copyright (c) 1994, Regents of the University of California
 *
 * IDENTIFICATION
 *    src/backend/parser/scan.l
 *
 *-------------------------------------------------------------------------
 */
#include "postgres.h"
 
#include <ctype.h>
#include <unistd.h>
 
#include "common/string.h"
#include "gramparse.h"
#include "nodes/miscnodes.h"
#include "parser/parser.h"      /* only needed for GUC variables */
#include "parser/scansup.h"
#include "port/pg_bitutils.h"
#include "mb/pg_wchar.h"
#include "utils/builtins.h"
}
 
%{
 
/* LCOV_EXCL_START */
 
/* Avoid exit() on fatal scanner errors (a bit ugly -- see yy_fatal_error) */
#undef fprintf
#define fprintf(file, fmt, msg)  fprintf_to_ereport(fmt, msg)
 
static void
fprintf_to_ereport(const char *fmt, const char *msg)
{
    ereport(ERROR, (errmsg_internal("%s", msg)));
}
 
/*
 * GUC variable.  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;
 * as such, changing its value can induce very unintuitive behavior.
 * In practice, backslash_quote is not too awful since it only controls
 * whether to throw an error: it cannot change non-error results.
 */
int         backslash_quote = BACKSLASH_QUOTE_SAFE_ENCODING;
 
/*
 * Constant data exported from this file.  This array maps from the
 * zero-based keyword numbers returned by ScanKeywordLookup to the
 * Bison token numbers needed by gram.y.  This is exported because
 * callers need to pass it to scanner_init, if they are using the
 * standard keyword list ScanKeywords.
 */
#define PG_KEYWORD(kwname, value, category, collabel) value,
 
const uint16 ScanKeywordTokens[] = {
#include "parser/kwlist.h"
};
 
#undef PG_KEYWORD
 
/*
 * Set the type of YYSTYPE.
 */
#define YYSTYPE core_YYSTYPE
 
/*
 * Each call to yylex must set yylloc to the location of the found token
 * (expressed as a byte offset from the start of the input text).
 * When we parse a token that requires multiple lexer rules to process,
 * this should be done in the first such rule, else yylloc will point
 * into the middle of the token.
 */
#define SET_YYLLOC()  (*(yylloc) = yytext - yyextra->scanbuf)
 
/*
 * Advance yylloc by the given number of bytes.
 */
#define ADVANCE_YYLLOC(delta)  ( *(yylloc) += (delta) )
 
/*
 * Sometimes, we do want yylloc to point into the middle of a token; this is
 * useful for instance to throw an error about an escape sequence within a
 * string literal.  But if we find no error there, we want to revert yylloc
 * to the token start, so that that's the location reported to the parser.
 * Use PUSH_YYLLOC/POP_YYLLOC to save/restore yylloc around such code.
 * (Currently the implied "stack" is just one location, but someday we might
 * need to nest these.)
 */
#define PUSH_YYLLOC()   (yyextra->save_yylloc = *(yylloc))
#define POP_YYLLOC()    (*(yylloc) = yyextra->save_yylloc)
 
#define startlit()  ( yyextra->literallen = 0 )
static void addlit(char *ytext, int yleng, core_yyscan_t yyscanner);
static void addlitchar(unsigned char ychar, core_yyscan_t yyscanner);
static char *litbufdup(core_yyscan_t yyscanner);
static unsigned char unescape_single_char(unsigned char c, core_yyscan_t yyscanner);
static int  process_integer_literal(const char *token, YYSTYPE *lval, int base);
static void addunicode(char32_t c, yyscan_t yyscanner);
 
#define yyerror(msg)  scanner_yyerror(msg, yyscanner)
 
#define lexer_errposition()  scanner_errposition(*(yylloc), 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="core_yy"
%option extra-type="core_yy_extra_type *"
 
/*
 * OK, here is a short description of lex/flex rules behavior.
 * The longest pattern which matches an input string is always chosen.
 * For equal-length patterns, the first occurring in the rules list is chosen.
 * INITIAL is the starting state, to which all non-conditional rules apply.
 * Exclusive states change parsing rules while the state is active.  When in
 * an exclusive state, only those rules defined for that state apply.
 *
 * We use exclusive states for quoted strings, extended comments,
 * and to eliminate parsing troubles for numeric strings.
 * Exclusive states:
 *  <xb> bit string literal
 *  <xc> extended C-style comments
 *  <xd> delimited identifiers (double-quoted identifiers)
 *  <xh> hexadecimal byte string
 *  <xq> standard quoted strings
 *  <xqs> quote stop (detect continued strings)
 *  <xe> extended quoted strings (support backslash escape sequences)
 *  <xdolq> $foo$ quoted strings
 *  <xui> quoted identifier with Unicode escapes
 *  <xus> quoted string with Unicode escapes
 *  <xeu> Unicode surrogate pair in extended quoted string
 *
 * Remember to add an <<EOF>> case whenever you add a new exclusive state!
 * The default one is probably not the right thing.
 */
 
%x xb
%x xc
%x xd
%x xh
%x xq
%x xqs
%x xe
%x xdolq
%x xui
%x xus
%x xeu
 
/*
 * In order to make the world safe for Windows and Mac clients as well as
 * Unix ones, we accept either \n or \r as a newline.  A DOS-style \r\n
 * sequence will be seen as two successive newlines, but that doesn't cause
 * any problems.  Comments that start with -- and extend to the next
 * newline are treated as equivalent to a single whitespace character.
 *
 * NOTE a fine point: if there is no newline following --, we will absorb
 * everything to the end of the input as a comment.  This is correct.  Older
 * versions of Postgres failed to recognize -- as a comment if the input
 * did not end with a newline.
 *
 * non_newline_space tracks all the other space characters except newlines.
 *
 * XXX if you change the set of whitespace characters, fix scanner_isspace()
 * to agree.
 */
 
space               [ \t\n\r\f\v]
non_newline_space   [ \t\f\v]
newline             [\n\r]
non_newline         [^\n\r]
 
comment         ("--"{non_newline}*)
 
whitespace      ({space}+|{comment})
 
/*
 * SQL requires at least one newline in the whitespace separating
 * string literals that are to be concatenated.  Silly, but who are we
 * to argue?  Note that {whitespace_with_newline} should not have * after
 * it, whereas {whitespace} should generally have a * after it...
 */
 
special_whitespace      ({space}+|{comment}{newline})
non_newline_whitespace  ({non_newline_space}|{comment})
whitespace_with_newline ({non_newline_whitespace}*{newline}{special_whitespace}*)
 
quote           '
/* If we see {quote} then {quotecontinue}, the quoted string continues */
quotecontinue   {whitespace_with_newline}{quote}
 
/*
 * {quotecontinuefail} is needed to avoid lexer backup when we fail to match
 * {quotecontinue}.  It might seem that this could just be {whitespace}*,
 * but if there's a dash after {whitespace_with_newline}, it must be consumed
 * to see if there's another dash --- which would start a {comment} and thus
 * allow continuation of the {quotecontinue} token.
 */
quotecontinuefail   {whitespace}*"-"?
 
/* Bit string
 * It is tempting to scan the string for only those characters
 * which are allowed. However, this leads to silently swallowed
 * characters if illegal characters are included in the string.
 * For example, if xbinside is [01] then B'ABCD' is interpreted
 * as a zero-length string, and the ABCD' is lost!
 * Better to pass the string forward and let the input routines
 * validate the contents.
 */
xbstart         [bB]{quote}
xbinside        [^']*
 
/* Hexadecimal byte string */
xhstart         [xX]{quote}
xhinside        [^']*
 
/* National character */
xnstart         [nN]{quote}
 
/* Quoted string that allows backslash escapes */
xestart         [eE]{quote}
xeinside        [^\\']+
xeescape        [\\][^0-7]
xeoctesc        [\\][0-7]{1,3}
xehexesc        [\\]x[0-9A-Fa-f]{1,2}
xeunicode       [\\](u[0-9A-Fa-f]{4}|U[0-9A-Fa-f]{8})
xeunicodefail   [\\](u[0-9A-Fa-f]{0,3}|U[0-9A-Fa-f]{0,7})
 
/* Extended quote
 * xqdouble implements embedded quote, ''''
 */
xqstart         {quote}
xqdouble        {quote}{quote}
xqinside        [^']+
 
/* $foo$ style quotes ("dollar quoting")
 * The quoted string starts with $foo$ where "foo" is an optional string
 * in the form of an identifier, except that it may not contain "$",
 * and extends to the first occurrence of an identical string.
 * There is *no* processing of the quoted text.
 *
 * {dolqfailed} is an error rule to avoid scanner backup when {dolqdelim}
 * fails to match its trailing "$".
 */
dolq_start      [A-Za-z\200-\377_]
dolq_cont       [A-Za-z\200-\377_0-9]
dolqdelim       \$({dolq_start}{dolq_cont}*)?\$
dolqfailed      \${dolq_start}{dolq_cont}*
dolqinside      [^$]+
 
/* Double quote
 * Allows embedded spaces and other special characters into identifiers.
 */
dquote          \"
xdstart         {dquote}
xdstop          {dquote}
xddouble        {dquote}{dquote}
xdinside        [^"]+
 
/* Quoted identifier with Unicode escapes */
xuistart        [uU]&{dquote}
 
/* Quoted string with Unicode escapes */
xusstart        [uU]&{quote}
 
/* error rule to avoid backup */
xufailed        [uU]&
 
 
/* C-style comments
 *
 * The "extended comment" syntax closely resembles allowable operator syntax.
 * The tricky part here is to get lex to recognize a string starting with
 * slash-star as a comment, when interpreting it as an operator would produce
 * a longer match --- remember lex will prefer a longer match!  Also, if we
 * have something like plus-slash-star, lex will think this is a 3-character
 * operator whereas we want to see it as a + operator and a comment start.
 * The solution is two-fold:
 * 1. append {op_chars}* to xcstart so that it matches as much text as
 *    {operator} would. Then the tie-breaker (first matching rule of same
 *    length) ensures xcstart wins.  We put back the extra stuff with yyless()
 *    in case it contains a star-slash that should terminate the comment.
 * 2. In the operator rule, check for slash-star within the operator, and
 *    if found throw it back with yyless().  This handles the plus-slash-star
 *    problem.
 * Dash-dash comments have similar interactions with the operator rule.
 */
xcstart         \/\*{op_chars}*
xcstop          \*+\/
xcinside        [^*/]+
 
ident_start     [A-Za-z\200-\377_]
ident_cont      [A-Za-z\200-\377_0-9\$]
 
identifier      {ident_start}{ident_cont}*
 
/* Assorted special-case operators and operator-like tokens */
typecast        "::"
dot_dot         \.\.
colon_equals    ":="
 
/*
 * These operator-like tokens (unlike the above ones) also match the {operator}
 * rule, which means that they might be overridden by a longer match if they
 * are followed by a comment start or a + or - character. Accordingly, if you
 * add to this list, you must also add corresponding code to the {operator}
 * block to return the correct token in such cases. (This is not needed in
 * psqlscan.l since the token value is ignored there.)
 */
equals_greater  "=>"
less_equals     "<="
greater_equals  ">="
less_greater    "<>"
not_equals      "!="
/* Note there is no need for left_arrow, since "<-" is not a single operator. */
right_arrow     "->"
 
/*
 * "self" is the set of chars that should be returned as single-character
 * tokens.  "op_chars" is the set of chars that can make up "Op" tokens,
 * which can be one or more characters long (but if a single-char token
 * appears in the "self" set, it is not to be returned as an Op).  Note
 * that the sets overlap, but each has some chars that are not in the other.
 *
 * If you change either set, adjust the character lists appearing in the
 * rule for "operator"!
 */
self            [,()\[\].;\:\|\+\-\*\/\%\^<>\=]
op_chars        [\~\!\@\#\^\&\|\`\?\+\-\*\/\%<>\=]
operator        {op_chars}+
 
/*
 * Numbers
 *
 * Unary minus is not part of a number here.  Instead we pass it separately to
 * the parser, and there it gets coerced via doNegate().
 *
 * {numericfail} is used because we would like "1..10" to lex as 1, dot_dot, 10.
 *
 * {realfail} is added to prevent the need for scanner
 * backup when the {real} rule fails to match completely.
 */
decdigit        [0-9]
hexdigit        [0-9A-Fa-f]
octdigit        [0-7]
bindigit        [0-1]
 
decinteger      {decdigit}(_?{decdigit})*
hexinteger      0[xX](_?{hexdigit})+
octinteger      0[oO](_?{octdigit})+
bininteger      0[bB](_?{bindigit})+
 
hexfail         0[xX]_?
octfail         0[oO]_?
binfail         0[bB]_?
 
numeric         (({decinteger}\.{decinteger}?)|(\.{decinteger}))
numericfail     {decinteger}\.\.
 
real            ({decinteger}|{numeric})[Ee][-+]?{decinteger}
realfail        ({decinteger}|{numeric})[Ee][-+]
 
/* Positional parameters don't accept underscores. */
param           \${decdigit}+
 
/*
 * An identifier immediately following an integer literal is disallowed because
 * in some cases it's ambiguous what is meant: for example, 0x1234 could be
 * either a hexinteger or a decinteger "0" and an identifier "x1234".  We can
 * detect such problems by seeing if integer_junk matches a longer substring
 * than any of the XXXinteger patterns (decinteger, hexinteger, octinteger,
 * bininteger).  One "junk" pattern is sufficient because
 * {decinteger}{identifier} will match all the same strings we'd match with
 * {hexinteger}{identifier} etc.
 *
 * Note that the rule for integer_junk must appear after the ones for
 * XXXinteger to make this work correctly: 0x1234 will match both hexinteger
 * and integer_junk, and we need hexinteger to be chosen in that case.
 *
 * Also disallow strings matched by numeric_junk, real_junk and param_junk
 * for consistency.
 */
integer_junk    {decinteger}{identifier}
numeric_junk    {numeric}{identifier}
real_junk       {real}{identifier}
param_junk      \${decdigit}+{identifier}
 
other           .
 
/*
 * Dollar quoted strings are totally opaque, and no escaping is done on them.
 * Other quoted strings must allow some special characters such as single-quote
 *  and newline.
 * Embedded single-quotes are implemented both in the SQL standard
 *  style of two adjacent single quotes "''" and in the Postgres/Java style
 *  of escaped-quote "\'".
 * Other embedded escaped characters are matched explicitly and the leading
 *  backslash is dropped from the string.
 * Note that xcstart must appear before operator, as explained above!
 *  Also whitespace (comment) must appear before operator.
 */
 
%%
 
{whitespace}    {
                    /* ignore */
                }
 
{xcstart}       {
                    /* Set location in case of syntax error in comment */
                    SET_YYLLOC();
                    yyextra->xcdepth = 0;
                    BEGIN(xc);
                    /* Put back any characters past slash-star; see above */
                    yyless(2);
                }
 
<xc>{
{xcstart}       {
                    (yyextra->xcdepth)++;
                    /* Put back any characters past slash-star; see above */
                    yyless(2);
                }
 
{xcstop}        {
                    if (yyextra->xcdepth <= 0)
                        BEGIN(INITIAL);
                    else
                        (yyextra->xcdepth)--;
                }
 
{xcinside}      {
                    /* ignore */
                }
 
{op_chars}      {
                    /* ignore */
                }
 
\*+             {
                    /* ignore */
                }
 
<<EOF>>         {
                    yyerror("unterminated /* comment");
                }
} /* <xc> */
 
{xbstart}       {
                    /* Binary bit type.
                     * At some point we should simply pass the string
                     * forward to the parser and label it there.
                     * In the meantime, place a leading "b" on the string
                     * to mark it for the input routine as a binary string.
                     */
                    SET_YYLLOC();
                    BEGIN(xb);
                    startlit();
                    addlitchar('b', yyscanner);
                }
<xh>{xhinside}  |
<xb>{xbinside}  {
                    addlit(yytext, yyleng, yyscanner);
                }
<xb><<EOF>>     { yyerror("unterminated bit string literal"); }
 
{xhstart}       {
                    /* Hexadecimal bit type.
                     * At some point we should simply pass the string
                     * forward to the parser and label it there.
                     * In the meantime, place a leading "x" on the string
                     * to mark it for the input routine as a hex string.
                     */
                    SET_YYLLOC();
                    BEGIN(xh);
                    startlit();
                    addlitchar('x', yyscanner);
                }
<xh><<EOF>>     { yyerror("unterminated hexadecimal string literal"); }
 
{xnstart}       {
                    /* National character.
                     * We will pass this along as a normal character string,
                     * but preceded with an internally-generated "NCHAR".
                     */
                    int     kwnum;
 
                    SET_YYLLOC();
                    yyless(1);  /* eat only 'n' this time */
 
                    kwnum = ScanKeywordLookup("nchar",
                                              yyextra->keywordlist);
                    if (kwnum >= 0)
                    {
                        yylval->keyword = GetScanKeyword(kwnum,
                                                         yyextra->keywordlist);
                        return yyextra->keyword_tokens[kwnum];
                    }
                    else
                    {
                        /* If NCHAR isn't a keyword, just return "n" */
                        yylval->str = pstrdup("n");
                        return IDENT;
                    }
                }
 
{xqstart}       {
                    yyextra->saw_non_ascii = false;
                    SET_YYLLOC();
                    BEGIN(xq);
                    startlit();
                }
{xestart}       {
                    yyextra->saw_non_ascii = false;
                    SET_YYLLOC();
                    BEGIN(xe);
                    startlit();
                }
{xusstart}      {
                    SET_YYLLOC();
                    BEGIN(xus);
                    startlit();
                }
 
<xb,xh,xq,xe,xus>{quote} {
                    /*
                     * When we are scanning a quoted string and see an end
                     * quote, we must look ahead for a possible continuation.
                     * If we don't see one, we know the end quote was in fact
                     * the end of the string.  To reduce the lexer table size,
                     * we use a single "xqs" state to do the lookahead for all
                     * types of strings.
                     */
                    yyextra->state_before_str_stop = YYSTATE;
                    BEGIN(xqs);
                }
<xqs>{quotecontinue} {
                    /*
                     * Found a quote continuation, so return to the in-quote
                     * state and continue scanning the literal.  Nothing is
                     * added to the literal's contents.
                     */
                    BEGIN(yyextra->state_before_str_stop);
                }
<xqs>{quotecontinuefail} |
<xqs>{other} |
<xqs><<EOF>>    {
                    /*
                     * Failed to see a quote continuation.  Throw back
                     * everything after the end quote, and handle the string
                     * according to the state we were in previously.
                     */
                    yyless(0);
                    BEGIN(INITIAL);
 
                    switch (yyextra->state_before_str_stop)
                    {
                        case xb:
                            yylval->str = litbufdup(yyscanner);
                            return BCONST;
                        case xh:
                            yylval->str = litbufdup(yyscanner);
                            return XCONST;
                        case xq:
                        case xe:
                            /*
                             * Check that the data remains valid, if it might
                             * have been made invalid by unescaping any chars.
                             */
                            if (yyextra->saw_non_ascii)
                                pg_verifymbstr(yyextra->literalbuf,
                                               yyextra->literallen,
                                               false);
                            yylval->str = litbufdup(yyscanner);
                            return SCONST;
                        case xus:
                            yylval->str = litbufdup(yyscanner);
                            return USCONST;
                        default:
                            yyerror("unhandled previous state in xqs");
                    }
                }
 
<xq,xe,xus>{xqdouble} {
                    addlitchar('\'', yyscanner);
                }
<xq,xus>{xqinside}  {
                    addlit(yytext, yyleng, yyscanner);
                }
<xe>{xeinside}  {
                    addlit(yytext, yyleng, yyscanner);
                }
<xe>{xeunicode} {
                    char32_t    c = strtoul(yytext + 2, NULL, 16);
 
                    /* Remember start of overall string token ... */
                    PUSH_YYLLOC();
                    /* ... and set the error cursor to point at this esc seq */
                    SET_YYLLOC();
 
                    if (is_utf16_surrogate_first(c))
                    {
                        yyextra->utf16_first_part = c;
                        BEGIN(xeu);
                    }
                    else if (is_utf16_surrogate_second(c))
                        yyerror("invalid Unicode surrogate pair");
                    else
                        addunicode(c, yyscanner);
 
                    /* Restore yylloc to be start of string token */
                    POP_YYLLOC();
                }
<xeu>{xeunicode} {
                    char32_t    c = strtoul(yytext + 2, NULL, 16);
 
                    /* Remember start of overall string token ... */
                    PUSH_YYLLOC();
                    /* ... and set the error cursor to point at this esc seq */
                    SET_YYLLOC();
 
                    if (!is_utf16_surrogate_second(c))
                        yyerror("invalid Unicode surrogate pair");
 
                    c = surrogate_pair_to_codepoint(yyextra->utf16_first_part, c);
 
                    addunicode(c, yyscanner);
 
                    /* Restore yylloc to be start of string token */
                    POP_YYLLOC();
 
                    BEGIN(xe);
                }
<xeu>. |
<xeu>\n |
<xeu><<EOF>>    {
                    /* Set the error cursor to point at missing esc seq */
                    SET_YYLLOC();
                    yyerror("invalid Unicode surrogate pair");
                }
<xe,xeu>{xeunicodefail} {
                    /* Set the error cursor to point at malformed esc seq */
                    SET_YYLLOC();
                    ereport(ERROR,
                            (errcode(ERRCODE_INVALID_ESCAPE_SEQUENCE),
                             errmsg("invalid Unicode escape"),
                             errhint("Unicode escapes must be \\uXXXX or \\UXXXXXXXX."),
                             lexer_errposition()));
                }
<xe>{xeescape}  {
                    if (yytext[1] == '\'')
                    {
                        if (yyextra->backslash_quote == BACKSLASH_QUOTE_OFF ||
                            (yyextra->backslash_quote == BACKSLASH_QUOTE_SAFE_ENCODING &&
                             PG_ENCODING_IS_CLIENT_ONLY(pg_get_client_encoding())))
                            ereport(ERROR,
                                    (errcode(ERRCODE_NONSTANDARD_USE_OF_ESCAPE_CHARACTER),
                                     errmsg("unsafe use of \\' in a string literal"),
                                     errhint("Use '' to write quotes in strings. \\' is insecure in client-only encodings."),
                                     lexer_errposition()));
                    }
                    addlitchar(unescape_single_char(yytext[1], yyscanner),
                               yyscanner);
                }
<xe>{xeoctesc}  {
                    unsigned char c = strtoul(yytext + 1, NULL, 8);
 
                    addlitchar(c, yyscanner);
                    if (c == '\0' || IS_HIGHBIT_SET(c))
                        yyextra->saw_non_ascii = true;
                }
<xe>{xehexesc}  {
                    unsigned char c = strtoul(yytext + 2, NULL, 16);
 
                    addlitchar(c, yyscanner);
                    if (c == '\0' || IS_HIGHBIT_SET(c))
                        yyextra->saw_non_ascii = true;
                }
<xe>.           {
                    /* This is only needed for \ just before EOF */
                    addlitchar(yytext[0], yyscanner);
                }
<xq,xe,xus><<EOF>>      { yyerror("unterminated quoted string"); }
 
{dolqdelim}     {
                    SET_YYLLOC();
                    yyextra->dolqstart = pstrdup(yytext);
                    BEGIN(xdolq);
                    startlit();
                }
{dolqfailed}    {
                    SET_YYLLOC();
                    /* throw back all but the initial "$" */
                    yyless(1);
                    /* and treat it as {other} */
                    return yytext[0];
                }
<xdolq>{dolqdelim} {
                    if (strcmp(yytext, yyextra->dolqstart) == 0)
                    {
                        pfree(yyextra->dolqstart);
                        yyextra->dolqstart = NULL;
                        BEGIN(INITIAL);
                        yylval->str = litbufdup(yyscanner);
                        return SCONST;
                    }
                    else
                    {
                        /*
                         * When we fail to match $...$ to dolqstart, transfer
                         * the $... part to the output, but put back the final
                         * $ for rescanning.  Consider $delim$...$junk$delim$
                         */
                        addlit(yytext, yyleng - 1, yyscanner);
                        yyless(yyleng - 1);
                    }
                }
<xdolq>{dolqinside} {
                    addlit(yytext, yyleng, yyscanner);
                }
<xdolq>{dolqfailed} {
                    addlit(yytext, yyleng, yyscanner);
                }
<xdolq>.        {
                    /* This is only needed for $ inside the quoted text */
                    addlitchar(yytext[0], yyscanner);
                }
<xdolq><<EOF>>  { yyerror("unterminated dollar-quoted string"); }
 
{xdstart}       {
                    SET_YYLLOC();
                    BEGIN(xd);
                    startlit();
                }
{xuistart}      {
                    SET_YYLLOC();
                    BEGIN(xui);
                    startlit();
                }
<xd>{xdstop}    {
                    char       *ident;
 
                    BEGIN(INITIAL);
                    if (yyextra->literallen == 0)
                        yyerror("zero-length delimited identifier");
                    ident = litbufdup(yyscanner);
                    if (yyextra->literallen >= NAMEDATALEN)
                        truncate_identifier(ident, yyextra->literallen, true);
                    yylval->str = ident;
                    return IDENT;
                }
<xui>{dquote}   {
                    BEGIN(INITIAL);
                    if (yyextra->literallen == 0)
                        yyerror("zero-length delimited identifier");
                    /* can't truncate till after we de-escape the ident */
                    yylval->str = litbufdup(yyscanner);
                    return UIDENT;
                }
<xd,xui>{xddouble}  {
                    addlitchar('"', yyscanner);
                }
<xd,xui>{xdinside}  {
                    addlit(yytext, yyleng, yyscanner);
                }
<xd,xui><<EOF>>     { yyerror("unterminated quoted identifier"); }
 
{xufailed}  {
                    char       *ident;
 
                    SET_YYLLOC();
                    /* throw back all but the initial u/U */
                    yyless(1);
                    /* and treat it as {identifier} */
                    ident = downcase_truncate_identifier(yytext, yyleng, true);
                    yylval->str = ident;
                    return IDENT;
                }
 
{typecast}      {
                    SET_YYLLOC();
                    return TYPECAST;
                }
 
{dot_dot}       {
                    SET_YYLLOC();
                    return DOT_DOT;
                }
 
{colon_equals}  {
                    SET_YYLLOC();
                    return COLON_EQUALS;
                }
 
{equals_greater} {
                    SET_YYLLOC();
                    return EQUALS_GREATER;
                }
 
{less_equals}   {
                    SET_YYLLOC();
                    return LESS_EQUALS;
                }
 
{greater_equals} {
                    SET_YYLLOC();
                    return GREATER_EQUALS;
                }
 
{less_greater}  {
                    /* We accept both "<>" and "!=" as meaning NOT_EQUALS */
                    SET_YYLLOC();
                    return NOT_EQUALS;
                }
 
{not_equals}    {
                    /* We accept both "<>" and "!=" as meaning NOT_EQUALS */
                    SET_YYLLOC();
                    return NOT_EQUALS;
                }
 
{right_arrow}   {
                    SET_YYLLOC();
                    return RIGHT_ARROW;
                }
 
{self}          {
                    SET_YYLLOC();
                    return yytext[0];
                }
 
{operator}      {
                    /*
                     * Check for embedded slash-star or dash-dash; those
                     * are comment starts, so operator must stop there.
                     * Note that slash-star or dash-dash at the first
                     * character will match a prior rule, not this one.
                     */
                    int         nchars = yyleng;
                    char       *slashstar = strstr(yytext, "/*");
                    char       *dashdash = strstr(yytext, "--");
 
                    if (slashstar && dashdash)
                    {
                        /* if both appear, take the first one */
                        if (slashstar > dashdash)
                            slashstar = dashdash;
                    }
                    else if (!slashstar)
                        slashstar = dashdash;
                    if (slashstar)
                        nchars = slashstar - yytext;
 
                    /*
                     * For SQL compatibility, '+' and '-' cannot be the
                     * last char of a multi-char operator unless the operator
                     * contains chars that are not in SQL operators.
                     * The idea is to lex '=-' as two operators, but not
                     * to forbid operator names like '?-' that could not be
                     * sequences of SQL operators.
                     */
                    if (nchars > 1 &&
                        (yytext[nchars - 1] == '+' ||
                         yytext[nchars - 1] == '-'))
                    {
                        int         ic;
 
                        for (ic = nchars - 2; ic >= 0; ic--)
                        {
                            char c = yytext[ic];
                            if (c == '~' || c == '!' || c == '@' ||
                                c == '#' || c == '^' || c == '&' ||
                                c == '|' || c == '`' || c == '?' ||
                                c == '%')
                                break;
                        }
                        if (ic < 0)
                        {
                            /*
                             * didn't find a qualifying character, so remove
                             * all trailing [+-]
                             */
                            do {
                                nchars--;
                            } while (nchars > 1 &&
                                 (yytext[nchars - 1] == '+' ||
                                  yytext[nchars - 1] == '-'));
                        }
                    }
 
                    SET_YYLLOC();
 
                    if (nchars < yyleng)
                    {
                        /* Strip the unwanted chars from the token */
                        yyless(nchars);
                        /*
                         * If what we have left is only one char, and it's
                         * one of the characters matching "self", then
                         * return it as a character token the same way
                         * that the "self" rule would have.
                         */
                        if (nchars == 1 &&
                            strchr(",()[].;:|+-*/%^<>=", yytext[0]))
                            return yytext[0];
                        /*
                         * Likewise, if what we have left is two chars, and
                         * those match the tokens ">=", "<=", "=>", "<>" or
                         * "!=", then we must return the appropriate token
                         * rather than the generic Op.
                         */
                        if (nchars == 2)
                        {
                            if (yytext[0] == '=' && yytext[1] == '>')
                                return EQUALS_GREATER;
                            if (yytext[0] == '>' && yytext[1] == '=')
                                return GREATER_EQUALS;
                            if (yytext[0] == '<' && yytext[1] == '=')
                                return LESS_EQUALS;
                            if (yytext[0] == '<' && yytext[1] == '>')
                                return NOT_EQUALS;
                            if (yytext[0] == '!' && yytext[1] == '=')
                                return NOT_EQUALS;
                            if (yytext[0] == '-' && yytext[1] == '>')
                                return RIGHT_ARROW;
                        }
                    }
 
                    /*
                     * Complain if operator is too long.  Unlike the case
                     * for identifiers, we make this an error not a notice-
                     * and-truncate, because the odds are we are looking at
                     * a syntactic mistake anyway.
                     */
                    if (nchars >= NAMEDATALEN)
                        yyerror("operator too long");
 
                    yylval->str = pstrdup(yytext);
                    return Op;
                }
 
{param}         {
                    ErrorSaveContext escontext = {T_ErrorSaveContext};
                    int32       val;
 
                    SET_YYLLOC();
                    val = pg_strtoint32_safe(yytext + 1, (Node *) &escontext);
                    if (escontext.error_occurred)
                        yyerror("parameter number too large");
                    yylval->ival = val;
                    return PARAM;
                }
{param_junk}    {
                    SET_YYLLOC();
                    yyerror("trailing junk after parameter");
                }
 
{decinteger}    {
                    SET_YYLLOC();
                    return process_integer_literal(yytext, yylval, 10);
                }
{hexinteger}    {
                    SET_YYLLOC();
                    return process_integer_literal(yytext, yylval, 16);
                }
{octinteger}    {
                    SET_YYLLOC();
                    return process_integer_literal(yytext, yylval, 8);
                }
{bininteger}    {
                    SET_YYLLOC();
                    return process_integer_literal(yytext, yylval, 2);
                }
{hexfail}       {
                    SET_YYLLOC();
                    yyerror("invalid hexadecimal integer");
                }
{octfail}       {
                    SET_YYLLOC();
                    yyerror("invalid octal integer");
                }
{binfail}       {
                    SET_YYLLOC();
                    yyerror("invalid binary integer");
                }
{numeric}       {
                    SET_YYLLOC();
                    yylval->str = pstrdup(yytext);
                    return FCONST;
                }
{numericfail}   {
                    /* throw back the .., and treat as integer */
                    yyless(yyleng - 2);
                    SET_YYLLOC();
                    return process_integer_literal(yytext, yylval, 10);
                }
{real}          {
                    SET_YYLLOC();
                    yylval->str = pstrdup(yytext);
                    return FCONST;
                }
{realfail}      {
                    SET_YYLLOC();
                    yyerror("trailing junk after numeric literal");
                }
{integer_junk}  {
                    SET_YYLLOC();
                    yyerror("trailing junk after numeric literal");
                }
{numeric_junk}  {
                    SET_YYLLOC();
                    yyerror("trailing junk after numeric literal");
                }
{real_junk}     {
                    SET_YYLLOC();
                    yyerror("trailing junk after numeric literal");
                }
 
 
{identifier}    {
                    int         kwnum;
                    char       *ident;
 
                    SET_YYLLOC();
 
                    /* Is it a keyword? */
                    kwnum = ScanKeywordLookup(yytext,
                                              yyextra->keywordlist);
                    if (kwnum >= 0)
                    {
                        yylval->keyword = GetScanKeyword(kwnum,
                                                         yyextra->keywordlist);
                        return yyextra->keyword_tokens[kwnum];
                    }
 
                    /*
                     * No.  Convert the identifier to lower case, and truncate
                     * if necessary.
                     */
                    ident = downcase_truncate_identifier(yytext, yyleng, true);
                    yylval->str = ident;
                    return IDENT;
                }
 
{other}         {
                    SET_YYLLOC();
                    return yytext[0];
                }
 
<<EOF>>         {
                    SET_YYLLOC();
                    yyterminate();
                }
 
%%
 
/* LCOV_EXCL_STOP */
 
/*
 * 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.
 *
 * This is expected to be used within an ereport() call, or via an error
 * callback such as setup_scanner_errposition_callback().  The return value
 * is a dummy (always 0, in fact).
 *
 * Note that this can only be used for messages emitted during raw parsing
 * (essentially, scan.l, parser.c, and gram.y), since it requires the
 * yyscanner struct to still be available.
 */
int
scanner_errposition(int location, core_yyscan_t yyscanner)
{
    int         pos;
 
    if (location < 0)
        return 0;               /* no-op if location is unknown */
 
    /* Convert byte offset to character number */
    pos = pg_mbstrlen_with_len(yyextra->scanbuf, location) + 1;
    /* And pass it to the ereport mechanism */
    return errposition(pos);
}
 
/*
 * Error context callback for inserting scanner error location.
 *
 * Note that this will be called for *any* error occurring while the
 * callback is installed.  We avoid inserting an irrelevant error location
 * if the error is a query cancel --- are there any other important cases?
 */
static void
scb_error_callback(void *arg)
{
    ScannerCallbackState *scbstate = (ScannerCallbackState *) arg;
 
    if (geterrcode() != ERRCODE_QUERY_CANCELED)
        (void) scanner_errposition(scbstate->location, scbstate->yyscanner);
}
 
/*
 * setup_scanner_errposition_callback
 *      Arrange for non-scanner errors to report an error position
 *
 * Sometimes the scanner calls functions that aren't part of the scanner
 * subsystem and can't reasonably be passed the yyscanner pointer; yet
 * we would like any errors thrown in those functions to be tagged with an
 * error location.  Use this function to set up an error context stack
 * entry that will accomplish that.  Usage pattern:
 *
 *      declare a local variable "ScannerCallbackState scbstate"
 *      ...
 *      setup_scanner_errposition_callback(&scbstate, yyscanner, location);
 *      call function that might throw error;
 *      cancel_scanner_errposition_callback(&scbstate);
 */
void
setup_scanner_errposition_callback(ScannerCallbackState *scbstate,
                                   core_yyscan_t yyscanner,
                                   int location)
{
    /* Setup error traceback support for ereport() */
    scbstate->yyscanner = yyscanner;
    scbstate->location = location;
    scbstate->errcallback.callback = scb_error_callback;
    scbstate->errcallback.arg = scbstate;
    scbstate->errcallback.previous = error_context_stack;
    error_context_stack = &scbstate->errcallback;
}
 
/*
 * Cancel a previously-set-up errposition callback.
 */
void
cancel_scanner_errposition_callback(ScannerCallbackState *scbstate)
{
    /* Pop the error context stack */
    error_context_stack = scbstate->errcallback.previous;
}
 
/*
 * scanner_yyerror
 *      Report a lexer or grammar error.
 *
 * 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
scanner_yyerror(const char *message, core_yyscan_t yyscanner)
{
    const char *loc = yyextra->scanbuf + *yylloc;
 
    if (*loc == YY_END_OF_BUFFER_CHAR)
    {
        ereport(ERROR,
                (errcode(ERRCODE_SYNTAX_ERROR),
        /* translator: %s is typically the translation of "syntax error" */
                 errmsg("%s at end of input", _(message)),
                 lexer_errposition()));
    }
    else
    {
        ereport(ERROR,
                (errcode(ERRCODE_SYNTAX_ERROR),
        /* translator: first %s is typically the translation of "syntax error" */
                 errmsg("%s at or near \"%s\"", _(message), loc),
                 lexer_errposition()));
    }
}
 
 
/*
 * Called before any actual parsing is done
 */
core_yyscan_t
scanner_init(const char *str,
             core_yy_extra_type *yyext,
             const ScanKeywordList *keywordlist,
             const uint16 *keyword_tokens)
{
    Size        slen = strlen(str);
    yyscan_t    scanner;
 
    if (yylex_init(&scanner) != 0)
        elog(ERROR, "yylex_init() failed: %m");
 
    core_yyset_extra(yyext, scanner);
 
    yyext->keywordlist = keywordlist;
    yyext->keyword_tokens = keyword_tokens;
 
    yyext->backslash_quote = backslash_quote;
 
    /*
     * Make a scan buffer with special termination needed by flex.
     */
    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 */
    yyext->literalalloc = 1024;
    yyext->literalbuf = (char *) palloc(yyext->literalalloc);
    yyext->literallen = 0;
 
    return scanner;
}
 
 
/*
 * Called after parsing is done to clean up after scanner_init()
 */
void
scanner_finish(core_yyscan_t yyscanner)
{
    /*
     * 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, core_yyscan_t yyscanner)
{
    /* enlarge buffer if needed */
    if ((yyextra->literallen + yleng) >= yyextra->literalalloc)
    {
        yyextra->literalalloc = pg_nextpower2_32(yyextra->literallen + yleng + 1);
        yyextra->literalbuf = (char *) repalloc(yyextra->literalbuf,
                                                yyextra->literalalloc);
    }
    /* append new data */
    memcpy(yyextra->literalbuf + yyextra->literallen, ytext, yleng);
    yyextra->literallen += yleng;
}
 
 
static void
addlitchar(unsigned char ychar, core_yyscan_t yyscanner)
{
    /* enlarge buffer if needed */
    if ((yyextra->literallen + 1) >= yyextra->literalalloc)
    {
        yyextra->literalalloc *= 2;
        yyextra->literalbuf = (char *) repalloc(yyextra->literalbuf,
                                                yyextra->literalalloc);
    }
    /* append new data */
    yyextra->literalbuf[yyextra->literallen] = ychar;
    yyextra->literallen += 1;
}
 
 
/*
 * Create a palloc'd copy of literalbuf, adding a trailing null.
 */
static char *
litbufdup(core_yyscan_t yyscanner)
{
    int         llen = yyextra->literallen;
    char       *new;
 
    new = palloc(llen + 1);
    memcpy(new, yyextra->literalbuf, llen);
    new[llen] = '\0';
    return new;
}
 
/*
 * Process {decinteger}, {hexinteger}, etc.  Note this will also do the right
 * thing with {numeric}, ie digits and a decimal point.
 */
static int
process_integer_literal(const char *token, YYSTYPE *lval, int base)
{
    ErrorSaveContext escontext = {T_ErrorSaveContext};
    int32       val;
 
    val = pg_strtoint32_safe(token, (Node *) &escontext);
    if (escontext.error_occurred)
    {
        /* integer too large (or contains decimal pt), treat it as a float */
        lval->str = pstrdup(token);
        return FCONST;
    }
    lval->ival = val;
    return ICONST;
}
 
static void
addunicode(char32_t c, core_yyscan_t yyscanner)
{
    ScannerCallbackState scbstate;
    char        buf[MAX_UNICODE_EQUIVALENT_STRING + 1];
 
    if (!is_valid_unicode_codepoint(c))
        yyerror("invalid Unicode escape value");
 
    /*
     * We expect that pg_unicode_to_server() will complain about any
     * unconvertible code point, so we don't have to set saw_non_ascii.
     */
    setup_scanner_errposition_callback(&scbstate, yyscanner, *(yylloc));
    pg_unicode_to_server(c, (unsigned char *) buf);
    cancel_scanner_errposition_callback(&scbstate);
    addlit(buf, strlen(buf), yyscanner);
}
 
static unsigned char
unescape_single_char(unsigned char c, core_yyscan_t yyscanner)
{
    switch (c)
    {
        case 'b':
            return '\b';
        case 'f':
            return '\f';
        case 'n':
            return '\n';
        case 'r':
            return '\r';
        case 't':
            return '\t';
        case 'v':
            return '\v';
        default:
            /* check for backslash followed by non-7-bit-ASCII */
            if (c == '\0' || IS_HIGHBIT_SET(c))
                yyextra->saw_non_ascii = true;
 
            return c;
    }
}
 
/*
 * Interface functions to make flex use palloc() instead of malloc().
 * It'd be better to make these static, but flex insists otherwise.
 */
 
void *
core_yyalloc(yy_size_t bytes, core_yyscan_t yyscanner)
{
    return palloc(bytes);
}
 
void *
core_yyrealloc(void *ptr, yy_size_t bytes, core_yyscan_t yyscanner)
{
    if (ptr)
        return repalloc(ptr, bytes);
    else
        return palloc(bytes);
}
 
void
core_yyfree(void *ptr, core_yyscan_t yyscanner)
{
    if (ptr)
        pfree(ptr);
}