Movatterモバイル変換

[0]ホーム
URL:

Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

 Sign up
Appearance settings

Commit043f9eb

Browse files
author
Thomas G. Lockhart
committed
Implement hex literal conversion to bit string literal.
 May not be the long-term solution (some continuing discussion with Peter E.) but better than the current mapping of a conversion to integer which I'd put in years ago before we had any bit string types at all.This is already supported in the bit string implementation elsewhere.
1 parentce5dc56 commit043f9eb

File tree

1 file changed

+40
-34
lines changed

1 file changed

+40
-34
lines changed

‎src/backend/parser/scan.l

Lines changed: 40 additions & 34 deletions
Original file line numberDiff line numberDiff line change
@@ -9,7 +9,7 @@
99
*
1010
*
1111
* IDENTIFICATION
12-
* $Header: /cvsroot/pgsql/src/backend/parser/scan.l,v 1.97 2002/06/22 02:04:45 thomas Exp $
12+
* $Header: /cvsroot/pgsql/src/backend/parser/scan.l,v 1.98 2002/08/04 06:36:18 thomas Exp $
1313
*
1414
*-------------------------------------------------------------------------
1515
*/
@@ -60,7 +60,7 @@ static char *litbufdup(void);
6060
* When we parse a token that requires multiple lexer rules to process,
6161
* we set token_start to point at the true start of the token, for use
6262
* by yyerror(). yytext will point at just the text consumed by the last
63-
* rule, so it's not very helpful (eg, it might contain just the last
63+
* rule, so it's not very helpful (e.g., it might contain just the last
6464
* quote mark of a quoted identifier). But to avoid cluttering every rule
6565
* with setting token_start, we allow token_start = NULL to denote that
6666
* it's okay to use yytext.
@@ -93,10 +93,10 @@ unsigned char unescape_single_char(unsigned char c);
9393
* and to eliminate parsing troubles for numeric strings.
9494
* Exclusive states:
9595
* <xb> bit string literal
96-
* <xc> extended C-style comments - thomas 1997-07-12
97-
* <xd> delimited identifiers (double-quoted identifiers) - thomas 1997-10-27
98-
* <xh> hexadecimal numeric string - thomas 1997-11-16
99-
* <xq> quoted strings - thomas 1997-07-30
96+
* <xc> extended C-style comments
97+
* <xd> delimited identifiers (double-quoted identifiers)
98+
* <xh> hexadecimal numeric string
99+
* <xq> quoted strings
100100
*/
101101

102102
%xxb
@@ -106,6 +106,13 @@ unsigned char unescape_single_char(unsigned char c);
106106
%xxq
107107

108108
/* Bit string
109+
* It is tempting to scan the string for only those characters
110+
* which are allowed. However, this leads to silently swallowed
111+
* characters if illegal characters are included in the string.
112+
* For example, if xbinside is [01] then B'ABCD' is interpreted
113+
* as a zero-length string, and the ABCD' is lost!
114+
* Better to pass the string forward and let the input routines
115+
* validate the contents.
109116
*/
110117
xbstart[bB]{quote}
111118
xbstop{quote}
@@ -116,7 +123,7 @@ xbcat{quote}{whitespace_with_newline}{quote}
116123
*/
117124
xhstart[xX]{quote}
118125
xhstop{quote}
119-
xhinside[^']+
126+
xhinside[^']*
120127
xhcat{quote}{whitespace_with_newline}{quote}
121128

122129
/* National character
@@ -244,7 +251,7 @@ other.
244251
* style of two adjacent single quotes "''" and in the Postgres/Java style
245252
* of escaped-quote "\'".
246253
* Other embedded escaped characters are matched explicitly and the leading
247-
* backslash is dropped from the string. - thomas 1997-09-24
254+
* backslash is dropped from the string.
248255
* Note that xcstart must appear before operator, as explained above!
249256
* Also whitespace (comment) must appear before operator.
250257
*/
@@ -291,8 +298,10 @@ other.
291298

292299
{xbstart}{
293300
/* Binary bit type.
294-
* Should be passing the type forward into the parser
295-
* rather than trying to embed it into the string.
301+
* At some point we should simply pass the string
302+
* forward to the parser and label it there.
303+
* In the meantime, place a leading "b" on the string
304+
* to mark it for the input routine as a binary string.
296305
*/
297306
token_start = yytext;
298307
BEGIN(xb);
@@ -301,10 +310,8 @@ other.
301310
}
302311
<xb>{xbstop}{
303312
BEGIN(INITIAL);
304-
if (literalbuf[strspn(literalbuf +1,"01") +1] !='\0')
305-
yyerror("invalid bit string input");
306313
yylval.str =litbufdup();
307-
returnBITCONST;
314+
returnBCONST;
308315
}
309316
<xh>{xhinside}|
310317
<xb>{xbinside}{
@@ -314,44 +321,43 @@ other.
314321
<xb>{xbcat}{
315322
/* ignore */
316323
}
317-
<xb><<EOF>>{yyerror("unterminated bit string literal"); }
318-
324+
<xb><<EOF>>{
325+
yyerror("unterminated bit string literal");
326+
}
319327
{xhstart}{
320328
/* Hexadecimal bit type.
321-
* Should be passing the type forward into the parser
322-
* rather than trying to embed it into the string.
329+
* At some point we should simply pass the string
330+
* forward to the parser and label it there.
331+
* In the meantime, place a leading "x" on the string
332+
* to mark it for the input routine as a hex string.
323333
*/
324334
token_start = yytext;
325335
BEGIN(xh);
326336
startlit();
337+
addlitchar('x');
327338
}
328339
<xh>{xhstop}{
329-
long val;
330-
char* endptr;
331-
332340
BEGIN(INITIAL);
333-
errno =0;
334-
val =strtol(literalbuf, &endptr,16);
335-
if (*endptr !='\0' || errno == ERANGE
336-
#ifdef HAVE_LONG_INT_64
337-
/* if long > 32 bits, check for overflow of int4 */
338-
|| val != (long) ((int32) val)
339-
#endif
340-
)
341-
yyerror("bad hexadecimal integer input");
342-
yylval.ival = val;
343-
return ICONST;
341+
yylval.str =litbufdup();
342+
return XCONST;
344343
}
345-
<xh><<EOF>>{yyerror("unterminated hexadecimalinteger"); }
344+
<xh><<EOF>>{yyerror("unterminated hexadecimalstring literal"); }
346345

347346
{xnstart}{
348347
/* National character.
349-
*Need to remember type info to flow it forward into the parser.
350-
*Not yet implemented. - thomas 2002-06-17
348+
*We will pass this along as a normal character string,
349+
*but preceded with an internally-generated "NCHAR".
351350
*/
351+
const ScanKeyword *keyword;
352+
353+
/* This had better be a keyword! */
354+
keyword =ScanKeywordLookup("nchar");
355+
Assert(keyword !=NULL);
356+
yylval.keyword = keyword->name;
352357
token_start = yytext;
353358
BEGIN(xq);
354359
startlit();
360+
return keyword->value;
355361
}
356362

357363

0 commit comments

Comments
 (0)
[8]ページ先頭
©2009-2025 Movatter.jp