database/postgres
1
0
Fork 0
mirror of https://github.com/postgres/postgres.git synced 2025-09-02 04:21:28 +03:00
Code Activity

Improve parser so that we can show an error cursor position for errors

Browse Source 
during parse analysis, not only errors detected in the flex/bison stages.
This is per my earlier proposal.  This commit includes all the basic
infrastructure, but locations are only tracked and reported for errors
involving column references, function calls, and operators.  More could
be done later but this seems like a good set to start with.  I've also
moved the ReportSyntaxErrorPosition logic out of psql and into libpq,
which should make it available to more people --- even within psql this
is an improvement because warnings weren't handled by ReportSyntaxErrorPosition.
This commit is contained in:
Tom Lane
2006-03-14 22:48:25 +00:00
parent 48fb696753
commit 20ab467d76
80 changed files with 1347 additions and 997 deletions
Download Patch File Download Diff File Expand all files Collapse all files

103
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 1.132 2006/03/07 01:00:17 tgl Exp $
* $PostgreSQL: pgsql/src/backend/parser/scan.l,v 1.133 2006/03/14 22:48:21 tgl Exp $
*
*-------------------------------------------------------------------------
*/
@@ -74,20 +74,19 @@ static int literalalloc; /* current allocated buffer size */
static void addlit(char *ytext, int yleng);
static void addlitchar(unsigned char ychar);
static char *litbufdup(void);
static int pg_err_position(void);
static int lexer_errposition(void);
static void check_escape_warning(void);
static void check_string_escape_warning(unsigned char ychar);
/*
* 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,
* we set token_start to point at the true start of the token, for use
* by yyerror(). yytext will point at just the text consumed by the last
* rule, so it's not very helpful (e.g., it might contain just the last
* quote mark of a quoted identifier). But to avoid cluttering every rule
* with setting token_start, we allow token_start = NULL to denote that
* it's okay to use yytext.
* this should be done in the first such rule, else yylloc will point
* into the middle of the token.
*/
static char *token_start;
#define SET_YYLLOC() (yylloc = yytext - scanbuf)
/* Handles to the buffer that the lexer uses internally */
static YY_BUFFER_STATE scanbufhandle;
@@ -316,17 +315,13 @@ other .
%%
%{
/* code to execute during start of each call of yylex() */
token_start = NULL;
%}
{whitespace} {
/* ignore */
}
{xcstart} {
token_start = yytext;
/* Set location in case of syntax error in comment */
SET_YYLLOC();
xcdepth = 0;
BEGIN(xc);
/* Put back any characters past slash-star; see above */
@@ -341,11 +336,7 @@ other .
<xc>{xcstop} {
if (xcdepth <= 0)
{
BEGIN(INITIAL);
/* reset token_start for next token */
token_start = NULL;
}
else
xcdepth--;
}
@@ -371,7 +362,7 @@ other .
* In the meantime, place a leading "b" on the string
* to mark it for the input routine as a binary string.
*/
token_start = yytext;
SET_YYLLOC();
BEGIN(xb);
startlit();
addlitchar('b');
@@ -400,7 +391,7 @@ other .
* In the meantime, place a leading "x" on the string
* to mark it for the input routine as a hex string.
*/
token_start = yytext;
SET_YYLLOC();
BEGIN(xh);
startlit();
addlitchar('x');
@@ -421,6 +412,7 @@ other .
*/
const ScanKeyword *keyword;
SET_YYLLOC();
yyless(1); /* eat only 'n' this time */
/* nchar had better be a keyword! */
keyword = ScanKeywordLookup("nchar");
@@ -431,7 +423,7 @@ other .
{xqstart} {
warn_on_first_escape = true;
token_start = yytext;
SET_YYLLOC();
if (standard_conforming_strings)
BEGIN(xq);
else
@@ -440,7 +432,7 @@ other .
}
{xestart} {
warn_on_first_escape = false;
token_start = yytext;
SET_YYLLOC();
BEGIN(xe);
startlit();
}
@@ -490,7 +482,7 @@ other .
<xq,xe><<EOF>> { yyerror("unterminated quoted string"); }
{dolqdelim} {
token_start = yytext;
SET_YYLLOC();
dolqstart = pstrdup(yytext);
BEGIN(xdolq);
startlit();
@@ -533,7 +525,7 @@ other .
<xdolq><<EOF>> { yyerror("unterminated dollar-quoted string"); }
{xdstart} {
token_start = yytext;
SET_YYLLOC();
BEGIN(xd);
startlit();
}
@@ -558,10 +550,12 @@ other .
<xd><<EOF>> { yyerror("unterminated quoted identifier"); }
{typecast} {
SET_YYLLOC();
return TYPECAST;
}
{self} {
SET_YYLLOC();
return yytext[0];
}
@@ -611,6 +605,8 @@ other .
nchars--; /* else remove the +/-, and check again */
}
SET_YYLLOC();
if (nchars < yyleng)
{
/* Strip the unwanted chars from the token */
@@ -644,6 +640,7 @@ other .
}
{param} {
SET_YYLLOC();
yylval.ival = atol(yytext + 1);
return PARAM;
}
@@ -652,6 +649,7 @@ other .
long val;
char* endptr;
SET_YYLLOC();
errno = 0;
val = strtol(yytext, &endptr, 10);
if (*endptr != '\0' || errno == ERANGE
@@ -669,10 +667,12 @@ other .
return ICONST;
}
{decimal} {
SET_YYLLOC();
yylval.str = pstrdup(yytext);
return FCONST;
}
{real} {
SET_YYLLOC();
yylval.str = pstrdup(yytext);
return FCONST;
}
@@ -684,12 +684,14 @@ other .
* syntax error anyway, we don't bother to distinguish.
*/
yyless(yyleng-1);
SET_YYLLOC();
yylval.str = pstrdup(yytext);
return FCONST;
}
{realfail2} {
/* throw back the [Ee][+-], and proceed as above */
yyless(yyleng-2);
SET_YYLLOC();
yylval.str = pstrdup(yytext);
return FCONST;
}
@@ -699,6 +701,8 @@ other .
const ScanKeyword *keyword;
char *ident;
SET_YYLLOC();
/* Is it a keyword? */
keyword = ScanKeywordLookup(yytext);
if (keyword != NULL)
@@ -717,25 +721,52 @@ other .
}
{other} {
SET_YYLLOC();
return yytext[0];
}
<<EOF>> {
SET_YYLLOC();
yyterminate();
}
%%
/*
* lexer_errposition
* Report a lexical-analysis-time cursor position, if possible.
*
* This is expected to be used within an ereport() call. The return value
* is a dummy (always 0, in fact).
*
* Note that this can only be used for messages from the lexer itself,
* since it depends on scanbuf to still be valid.
*/
static int
pg_err_position(void)
lexer_errposition(void)
{
const char *loc = token_start ? token_start : yytext;
int pos;
/* in multibyte encodings, return index in characters not bytes */
return pg_mbstrlen_with_len(scanbuf, loc - scanbuf) + 1;
/* Convert byte offset to character number */
pos = pg_mbstrlen_with_len(scanbuf, yylloc) + 1;
/* And pass it to the ereport mechanism */
return errposition(pos);
}
/*
* yyerror
* Report a lexer or grammar error.
*
* The message's cursor position identifies the most recently lexed token.
* 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)
{
const char *loc = token_start ? token_start : yytext;
int cursorpos = pg_err_position();
const char *loc = scanbuf + yylloc;
if (*loc == YY_END_OF_BUFFER_CHAR)
{
@@ -743,7 +774,7 @@ yyerror(const char *message)
(errcode(ERRCODE_SYNTAX_ERROR),
/* translator: %s is typically "syntax error" */
errmsg("%s at end of input", _(message)),
errposition(cursorpos)));
lexer_errposition()));
}
else
{
@@ -751,7 +782,7 @@ yyerror(const char *message)
(errcode(ERRCODE_SYNTAX_ERROR),
/* translator: first %s is typically "syntax error" */
errmsg("%s at or near \"%s\"", _(message), loc),
errposition(cursorpos)));
lexer_errposition()));
}
}
@@ -878,7 +909,7 @@ check_string_escape_warning(unsigned char ychar)
(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'...')."),
errposition(pg_err_position())));
lexer_errposition()));
warn_on_first_escape = false; /* warn only once per string */
}
else if (ychar == '\\')
@@ -888,7 +919,7 @@ check_string_escape_warning(unsigned char ychar)
(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'\\\\'."),
errposition(pg_err_position())));
lexer_errposition()));
warn_on_first_escape = false; /* warn only once per string */
}
else
@@ -903,6 +934,6 @@ check_escape_warning(void)
(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'."),
errposition(pg_err_position())));
lexer_errposition()));
warn_on_first_escape = false; /* warn only once per string */
}