mattwil Tue May 5 01:35:13 2009 UTC
Modified files:
/ZendEngine2 zend_highlight.c zend_language_scanner.l
/php-src/ext/standard/tests/strings highlight_file.phpt
Log:
Implemented manual scanning for strings/comments, plus misc. fixes
For details, see http://news.php.net/php.internals/43808
http://cvs.php.net/viewvc.cgi/ZendEngine2/zend_highlight.c?r1=1.65&r2=1.66&diff_format=u
Index: ZendEngine2/zend_highlight.c
diff -u ZendEngine2/zend_highlight.c:1.65 ZendEngine2/zend_highlight.c:1.66
--- ZendEngine2/zend_highlight.c:1.65 Wed Dec 31 11:12:29 2008
+++ ZendEngine2/zend_highlight.c Tue May 5 01:35:13 2009
@@ -17,7 +17,7 @@
+----------------------------------------------------------------------+
*/
-/* $Id: zend_highlight.c,v 1.65 2008/12/31 11:12:29 sebastian Exp $ */
+/* $Id: zend_highlight.c,v 1.66 2009/05/05 01:35:13 mattwil Exp $ */
#include "zend.h"
#include <zend_language_parser.h>
@@ -127,14 +127,8 @@
zend_printf("<span style=\"color: %s\">",
last_color);
}
}
- switch (token_type) {
- case T_END_HEREDOC:
- zend_html_puts(Z_STRVAL(token), Z_STRLEN(token)
TSRMLS_CC);
- break;
- default:
- zend_html_puts(LANG_SCNG(yy_text),
LANG_SCNG(yy_leng) TSRMLS_CC);
- break;
- }
+
+ zend_html_puts(LANG_SCNG(yy_text), LANG_SCNG(yy_leng)
TSRMLS_CC);
if (Z_TYPE(token) == IS_STRING ||
Z_TYPE(token) == IS_UNICODE) {
@@ -156,19 +150,6 @@
Z_TYPE(token) = 0;
}
- /* handler for trailing comments, see bug #42767 */
- if (LANG_SCNG(yy_leng) && LANG_SCNG(yy_text) < LANG_SCNG(yy_limit)) {
- if (last_color != syntax_highlighter_ini->highlight_comment) {
- if (last_color !=
syntax_highlighter_ini->highlight_html) {
- zend_printf("</span>");
- }
- if (syntax_highlighter_ini->highlight_comment !=
syntax_highlighter_ini->highlight_html) {
- zend_printf("<span style=\"color: %s\">",
syntax_highlighter_ini->highlight_comment);
- }
- }
- zend_html_puts(LANG_SCNG(yy_text), (LANG_SCNG(yy_limit) -
LANG_SCNG(yy_text)) TSRMLS_CC);
- }
-
if (last_color != syntax_highlighter_ini->highlight_html) {
zend_printf("</span>\n");
}
http://cvs.php.net/viewvc.cgi/ZendEngine2/zend_language_scanner.l?r1=1.205&r2=1.206&diff_format=u
Index: ZendEngine2/zend_language_scanner.l
diff -u ZendEngine2/zend_language_scanner.l:1.205
ZendEngine2/zend_language_scanner.l:1.206
--- ZendEngine2/zend_language_scanner.l:1.205 Thu Mar 26 20:01:38 2009
+++ ZendEngine2/zend_language_scanner.l Tue May 5 01:35:13 2009
@@ -21,7 +21,7 @@
+----------------------------------------------------------------------+
*/
-/* $Id: zend_language_scanner.l,v 1.205 2009/03/26 20:01:38 felipe Exp $ */
+/* $Id: zend_language_scanner.l,v 1.206 2009/05/05 01:35:13 mattwil Exp $ */
#if 0
# define YYDEBUG(s, c) printf("state: %d char: %c\n", s, c)
@@ -115,13 +115,19 @@
} \
}
+/* To save initial string length after scanning to first variable,
CG(doc_comment_len) can be reused */
+#define SET_DOUBLE_QUOTES_SCANNED_LENGTH(len) CG(doc_comment_len) = (len)
+#define GET_DOUBLE_QUOTES_SCANNED_LENGTH() CG(doc_comment_len)
+
+#define IS_LABEL_START(c) (((c) >= 'a' && (c) <= 'z') || ((c) >= 'A' && (c) <=
'Z') || (c) == '_' || (c) >= 0x7F)
+
#define ZEND_IS_OCT(c) ((c)>='0' && (c)<='7')
#define ZEND_IS_HEX(c) (((c)>='0' && (c)<='9') || ((c)>='a' && (c)<='f') ||
((c)>='A' && (c)<='F'))
BEGIN_EXTERN_C()
static void _yy_push_state(int new_state TSRMLS_DC)
- {
+{
zend_stack_push(&SCNG(state_stack), (void *) &YYGETCONDITION(),
sizeof(int));
YYSETCONDITION(new_state);
}
@@ -1324,63 +1330,8 @@
WHITESPACE [ \n\r\t]+
TABS_AND_SPACES [ \t]*
TOKENS [;:,.\[\]()|^&+-/*=%!~$<>?...@]
-ANY_CHAR [^\x00]
+ANY_CHAR [^]
NEWLINE ("\r"|"\n"|"\r\n")
-NULL [\x00]{1}
-
-/*
- * LITERAL_DOLLAR matches unescaped $ that aren't followed by a label character
- * or a { and therefore will be taken literally. The case of literal $ before
- * a variable or "${" is handled in a rule for each string type
- */
-DOUBLE_QUOTES_LITERAL_DOLLAR ("$"+([^a-zA-Z_\x7f-\xff$"\\{]|("\\"{ANY_CHAR})))
-BACKQUOTE_LITERAL_DOLLAR ("$"+([^a-zA-Z_\x7f-\xff$`\\{]|("\\"{ANY_CHAR})))
-HEREDOC_LITERAL_DOLLAR ("$"+([^a-zA-Z_\x7f-\xff$\n\r\\{]|("\\"[^\n\r])))
-
-/*
- * Usually, HEREDOC_NEWLINE will just function like a simple NEWLINE, but some
- * special cases need to be handled. HEREDOC_CHARS doesn't allow a line to
- * match when { or $, and/or \ is at the end. (("{"*|"$"*)"\\"?) handles that,
- * along with cases where { or $, and/or \ is the ONLY thing on a line
- *
- * The other case is when a line contains a label, followed by ONLY
- * { or $, and/or \ Handled by ({LABEL}";"?((("{"+|"$"+)"\\"?)|"\\"))
- */
-HEREDOC_NEWLINE
((({LABEL}";"?((("{"+|"$"+)"\\"?)|"\\"))|(("{"*|"$"*)"\\"?)){NEWLINE})
-
-/*
- * This pattern is just used in the next 2 for matching { or literal $, and/or
- * \ escape sequence immediately at the beginning of a line or after a label
- */
-HEREDOC_CURLY_OR_ESCAPE_OR_DOLLAR
(("{"+[^$\n\r\\{])|("{"*"\\"[^\n\r])|{HEREDOC_LITERAL_DOLLAR})
-
-/*
- * These 2 label-related patterns allow HEREDOC_CHARS to continue "regular"
- * matching after a newline that starts with either a non-label character or a
- * label that isn't followed by a newline. Like HEREDOC_CHARS, they won't match
- * a variable or "{$" Matching a newline, and possibly label, up TO a variable
- * or "{$", is handled in the heredoc rules
- *
- * The HEREDOC_LABEL_NO_NEWLINE pattern (";"[^$\n\r\\{]) handles cases where ;
- * follows a label. [^a-zA-Z0-9_\x7f-\xff;$\n\r\\{] is needed to prevent a
label
- * character or ; from matching on a possible (real) ending label
- */
-HEREDOC_NON_LABEL
([^a-zA-Z_\x7f-\xff$\n\r\\{]|{HEREDOC_CURLY_OR_ESCAPE_OR_DOLLAR})
-HEREDOC_LABEL_NO_NEWLINE
({LABEL}([^a-zA-Z0-9_\x7f-\xff;$\n\r\\{]|(";"[^$\n\r\\{])|(";"?{HEREDOC_CURLY_OR_ESCAPE_OR_DOLLAR})))
-
-/*
- * CHARS matches everything up to a variable or "{$"
- * {'s are matched as long as they aren't followed by a $
- * The case of { before "{$" is handled in a rule for each string type
- *
- * For heredocs, matching continues across/after newlines if/when it's known
- * that the next line doesn't contain a possible ending label
- */
-DOUBLE_QUOTES_CHARS
("{"*([^$"\\{]|("\\"{ANY_CHAR}))|{DOUBLE_QUOTES_LITERAL_DOLLAR})
-BACKQUOTE_CHARS
("{"*([^$`\\{]|("\\"{ANY_CHAR}))|{BACKQUOTE_LITERAL_DOLLAR})
-HEREDOC_CHARS
("{"*([^$\n\r\\{]|("\\"[^\n\r]))|{HEREDOC_LITERAL_DOLLAR}|({HEREDOC_NEWLINE}+({HEREDOC_NON_LABEL}|{HEREDOC_LABEL_NO_NEWLINE})))
-
-NOWDOC_CHARS
([^\n\r]|{NEWLINE}+([^a-zA-Z_\x7f-\xff\n\r]|({LABEL}([^a-zA-Z0-9_\x7f-\xff;\n\r]|(";"[^\n\r])))))
/* compute yyleng before each rule */
<!*> := yyleng = YYCURSOR - SCNG(yy_text);
@@ -2037,6 +1988,14 @@
}
<INITIAL>"<script"{WHITESPACE}+"language"{WHITESPACE}*"="{WHITESPACE}*("php"|"\"php\""|"'php'"){WHITESPACE}*">"
{
+ YYCTYPE *bracket = zend_memrchr(yytext, '<', yyleng - (sizeof("script
language=php>") - 1));
+
+ if (bracket != SCNG(yy_text)) {
+ /* Handle previously scanned HTML, as possible <script> tags
found are assumed to not be PHP's */
+ YYCURSOR = bracket;
+ goto inline_html;
+ }
+
HANDLE_NEWLINES(yytext, yyleng);
Z_STRVAL_P(zendlval) = yytext; /* no copying - intentional */
Z_STRLEN_P(zendlval) = yyleng;
@@ -2107,29 +2066,48 @@
}
<INITIAL>{ANY_CHAR} {
+ if (YYCURSOR > YYLIMIT) {
+ return 0;
+ }
inline_char_handler:
while (1) {
YYCTYPE *ptr = memchr(YYCURSOR, '<', YYLIMIT - YYCURSOR);
- if (ptr == NULL) {
- YYCURSOR = YYLIMIT;
- yyleng = YYCURSOR - SCNG(yy_text);
- break;
-
- } else {
- YYCURSOR = ptr + 1;
+ YYCURSOR = ptr ? ptr + 1 : YYLIMIT;
- /* stop if it may be an opening tag (<?, <%, <script>).
this condition is not optimal though */
- if (YYCURSOR < YYLIMIT && (*YYCURSOR == '?' ||
*YYCURSOR == '%' || *YYCURSOR == 's')) {
- --YYCURSOR;
- yyleng = YYCURSOR - SCNG(yy_text);
- break;
+ if (YYCURSOR < YYLIMIT) {
+ switch (*YYCURSOR) {
+ case '?':
+ if (CG(short_tags) ||
!strncasecmp(YYCURSOR + 1, "php", 3)) { /* Assume [ \t\n\r] follows "php" */
+ break;
+ }
+ continue;
+ case '%':
+ if (CG(asp_tags)) {
+ break;
+ }
+ continue;
+ case 's':
+ case 'S':
+ /* Probably NOT an opening PHP <script>
tag, so don't end the HTML chunk yet
+ * If it is, the PHP <script> tag rule
checks for any HTML scanned before it */
+ YYCURSOR--;
+ yymore();
+ default:
+ continue;
}
+
+ YYCURSOR--;
}
+
+ break;
}
+inline_html:
+ yyleng = YYCURSOR - SCNG(yy_text);
+
Z_STRVAL_P(zendlval) = (char *) estrndup(yytext, yyleng);
Z_STRLEN_P(zendlval) = yyleng;
Z_TYPE_P(zendlval) = IS_STRING;
@@ -2192,7 +2170,6 @@
/* Invalid rule to return a more explicit parse error with proper line
number */
yyless(0);
yy_pop_state(TSRMLS_C);
- ZVAL_EMPTY_TEXT(zendlval); /* Empty since it won't be used */
return T_ENCAPSED_AND_WHITESPACE;
}
@@ -2215,98 +2192,78 @@
<ST_IN_SCRIPTING>"#"|"//" {
- BEGIN(ST_ONE_LINE_COMMENT);
- yymore();
-}
-
-<ST_ONE_LINE_COMMENT>"?"|"%"|">" {
- yymore();
-}
+ while (YYCURSOR < YYLIMIT) {
+ switch (*YYCURSOR++) {
+ case '\r':
+ if (*YYCURSOR == '\n') {
+ YYCURSOR++;
+ }
+ /* fall through */
+ case '\n':
+ CG(zend_lineno)++;
+ break;
+ case '%':
+ if (!CG(asp_tags)) {
+ continue;
+ }
+ /* fall through */
+ case '?':
+ if (*YYCURSOR == '>') {
+ YYCURSOR--;
+ break;
+ }
+ /* fall through */
+ default:
+ continue;
+ }
-<ST_ONE_LINE_COMMENT>[^\n\r?%>]*{ANY_CHAR} {
- switch (yytext[yyleng-1]) {
- case '?': case '%': case '>':
- yyless(yyleng-1);
- yymore();
- break;
- case '\n':
- CG(zend_lineno)++;
- /* intentional fall through */
- default:
- Z_STRVAL_P(zendlval) = yytext; /* no copying -
intentional */
- Z_STRLEN_P(zendlval) = yyleng;
- Z_TYPE_P(zendlval) = IS_STRING;
- BEGIN(ST_IN_SCRIPTING);
- return T_COMMENT;
+ break;
}
-}
-<ST_ONE_LINE_COMMENT>{NEWLINE} {
- Z_STRVAL_P(zendlval) = yytext; /* no copying - intentional */
- Z_STRLEN_P(zendlval) = yyleng;
- Z_TYPE_P(zendlval) = IS_STRING;
- BEGIN(ST_IN_SCRIPTING);
- CG(zend_lineno)++;
+ yyleng = YYCURSOR - SCNG(yy_text);
+
return T_COMMENT;
}
-<ST_ONE_LINE_COMMENT>"?>"|"%>" {
- if (CG(asp_tags) || yytext[yyleng-2] != '%') { /* asp comment? */
- Z_STRVAL_P(zendlval) = yytext; /* no copying - intentional */
- Z_STRLEN_P(zendlval) = yyleng-2;
- Z_TYPE_P(zendlval) = IS_STRING;
- yyless(yyleng - 2);
- BEGIN(ST_IN_SCRIPTING);
- return T_COMMENT;
+<ST_IN_SCRIPTING>"/*"|"/**"{WHITESPACE} {
+ int doc_com;
+
+ if (yyleng > 2) {
+ doc_com = 1;
+ RESET_DOC_COMMENT();
} else {
- yymore();
+ doc_com = 0;
}
-}
-
-<ST_IN_SCRIPTING>"/**"{WHITESPACE} {
- RESET_DOC_COMMENT();
- BEGIN(ST_DOC_COMMENT);
- yymore();
-}
-
-<ST_COMMENT,ST_DOC_COMMENT>{NULL} {
- zend_error(E_COMPILE_WARNING, "Unterminated comment starting line %d",
CG(zend_lineno));
- return 0;
-}
-<ST_IN_SCRIPTING>"/*" {
- BEGIN(ST_COMMENT);
- yymore();
-}
+ while (YYCURSOR < YYLIMIT) {
+ if (*YYCURSOR++ == '*' && *YYCURSOR == '/') {
+ break;
+ }
+ }
+ if (YYCURSOR < YYLIMIT) {
+ YYCURSOR++;
+ } else {
+ zend_error(E_COMPILE_WARNING, "Unterminated comment starting
line %d", CG(zend_lineno));
+ }
-<ST_COMMENT,ST_DOC_COMMENT>[^*]+ {
- yymore();
-}
+ yyleng = YYCURSOR - SCNG(yy_text);
+ HANDLE_NEWLINES(yytext, yyleng);
-<ST_DOC_COMMENT>"*/" {
- zval temp;
+ if (doc_com) {
+ zval tmp;
- HANDLE_NEWLINES(yytext, yyleng);
- if (!zend_copy_scanner_string(&temp, yytext, yyleng, IS_UNICODE,
SCNG(output_conv) TSRMLS_CC)) {
- return 0;
+ if (!zend_copy_scanner_string(&tmp, yytext, yyleng, IS_UNICODE,
SCNG(output_conv) TSRMLS_CC)) {
+ return 0;
+ }
+ CG(doc_comment) = tmp.value.uni.val;
+ CG(doc_comment_len) = tmp.value.uni.len;
+ return T_DOC_COMMENT;
}
- CG(doc_comment) = temp.value.uni.val;
- CG(doc_comment_len) = temp.value.uni.len;
- BEGIN(ST_IN_SCRIPTING);
- return T_DOC_COMMENT;
-}
-<ST_COMMENT>"*/" {
- HANDLE_NEWLINES(yytext, yyleng);
- BEGIN(ST_IN_SCRIPTING);
return T_COMMENT;
}
-<ST_COMMENT,ST_DOC_COMMENT>"*" {
- yymore();
-}
-
<ST_IN_SCRIPTING>("?>"|"</script"{WHITESPACE}*">"){NEWLINE}? {
Z_STRVAL_P(zendlval) = yytext; /* no copying - intentional */
Z_STRLEN_P(zendlval) = yyleng;
@@ -2330,40 +2287,85 @@
}
-/* ("{"*|"$"*) handles { or $ at the end of a string (or the entire contents)
- */
-<ST_IN_SCRIPTING>(["]{DOUBLE_QUOTES_CHARS}*("{"*|"$"*)["]) {
- return zend_scan_unicode_escape_string(zendlval, yytext+1, yyleng-2,
0x22 /*'"'*/, T_CONSTANT_ENCAPSED_STRING TSRMLS_CC);
-}
+<ST_IN_SCRIPTING>b?['] {
+ int bprefix = (yytext[0] != '\'') ? 1 : 0;
+ while (1) {
+ if (YYCURSOR < YYLIMIT) {
+ if (*YYCURSOR == '\'') {
+ YYCURSOR++;
+ yyleng = YYCURSOR - SCNG(yy_text);
-<ST_IN_SCRIPTING>(b["]{DOUBLE_QUOTES_CHARS}*("{"*|"$"*)["]) {
- zend_scan_binary_escape_string(zendlval, yytext+2, yyleng-3, '"'
TSRMLS_CC);
- return T_CONSTANT_ENCAPSED_STRING;
-}
+ break;
+ } else if (*YYCURSOR++ == '\\' && YYCURSOR < YYLIMIT) {
+ YYCURSOR++;
+ }
+ } else {
+ yyleng = YYLIMIT - SCNG(yy_text);
+ /* Unclosed single quotes; treat similar to double
quotes, but without a separate token
+ * for ' (unrecognized by parser), instead of old flex
fallback to "Unexpected character..."
+ * rule, which continued in ST_IN_SCRIPTING state after
the quote */
+ return T_ENCAPSED_AND_WHITESPACE;
+ }
+ }
-<ST_IN_SCRIPTING>([']([^'\\]|("\\"{ANY_CHAR}))*[']) {
- return zend_scan_unicode_single_string(zendlval TSRMLS_CC);
+ if (bprefix) {
+ zend_scan_binary_single_string(zendlval, yytext+2, yyleng-3
TSRMLS_CC);
+ return T_CONSTANT_ENCAPSED_STRING;
+ } else {
+ return zend_scan_unicode_single_string(zendlval TSRMLS_CC);
+ }
}
-<ST_IN_SCRIPTING>("b'"([^'\\]|("\\"{ANY_CHAR}))*[']) {
- zend_scan_binary_single_string(zendlval, yytext+2, yyleng-3 TSRMLS_CC);
- return T_CONSTANT_ENCAPSED_STRING;
-}
+<ST_IN_SCRIPTING>b?["] {
+ int bprefix = (yytext[0] != '"') ? 1 : 0;
+ while (YYCURSOR < YYLIMIT) {
+ switch (*YYCURSOR++) {
+ case '"':
+ yyleng = YYCURSOR - SCNG(yy_text);
-<ST_IN_SCRIPTING>["] {
- BEGIN(ST_DOUBLE_QUOTES);
- return '"';
-}
+ if (bprefix) {
+
zend_scan_binary_escape_string(zendlval, yytext+2, yyleng-3, '"' TSRMLS_CC);
+ return T_CONSTANT_ENCAPSED_STRING;
+ } else {
+ return
zend_scan_unicode_escape_string(zendlval, yytext+1, yyleng-2, 0x22 /*'"'*/,
T_CONSTANT_ENCAPSED_STRING TSRMLS_CC);
+ }
+ case '$':
+ if (IS_LABEL_START(*YYCURSOR) || *YYCURSOR ==
'{') {
+ break;
+ }
+ continue;
+ case '{':
+ if (*YYCURSOR == '$') {
+ break;
+ }
+ continue;
+ case '\\':
+ if (YYCURSOR < YYLIMIT) {
+ YYCURSOR++;
+ }
+ /* fall through */
+ default:
+ continue;
+ }
+
+ YYCURSOR--;
+ break;
+ }
+
+ /* Remember how much was scanned to save rescanning */
+ SET_DOUBLE_QUOTES_SCANNED_LENGTH(YYCURSOR - SCNG(yy_text) - yyleng);
+
+ YYCURSOR = SCNG(yy_text) + yyleng;
-<ST_IN_SCRIPTING>b["] {
BEGIN(ST_DOUBLE_QUOTES);
- return T_BINARY_DOUBLE;
+ return bprefix ? T_BINARY_DOUBLE : '"';
}
+
<ST_IN_SCRIPTING>b?"<<<"{TABS_AND_SPACES}({LABEL}|([']{LABEL}['])|(["]{LABEL}["])){NEWLINE}
{
char *s;
int bprefix = (yytext[0] != '<') ? 1 : 0;
@@ -2398,7 +2400,7 @@
/* Check for ending label on the next line */
if (CG(heredoc_len) < YYLIMIT - YYCURSOR && !memcmp(YYCURSOR, s,
CG(heredoc_len))) {
- unsigned char *end = YYCURSOR + CG(heredoc_len);
+ YYCTYPE *end = YYCURSOR + CG(heredoc_len);
if (*end == ';') {
end++;
@@ -2419,54 +2421,6 @@
}
-/* Match everything up to and including a possible ending label, so if the
label
- * doesn't match, it's kept with the rest of the string
- *
- * {HEREDOC_NEWLINE}+ handles the case of more than one newline sequence that
- * couldn't be matched with HEREDOC_CHARS, because of the following label
- */
-<ST_HEREDOC>{HEREDOC_CHARS}*{HEREDOC_NEWLINE}+{LABEL}";"?[\n\r] {
- char *end = yytext + yyleng - 1;
-
- if (end[-1] == ';') {
- end--;
- yyleng--;
- }
-
- if (yyleng > CG(heredoc_len) && !memcmp(end - CG(heredoc_len),
CG(heredoc), CG(heredoc_len))) {
- int len = yyleng - CG(heredoc_len) - 2; /* 2 for newline before
and after label */
-
- /* May have matched fooLABEL; make sure there's a newline
before it */
- if (yytext[len] != '\n') {
- if (yytext[len] != '\r') {
- yyless(yyleng - 1);
- yymore();
- }
- } else if (len > 0 && yytext[len - 1] == '\r') {
- len--; /* Windows newline */
- }
-
- /* Go back before label, to match in ST_END_HEREDOC state.
yytext will include
- * newline before label, for zend_highlight/strip, tokenizer,
etc. */
- yyless(yyleng - CG(heredoc_len) - 1); /* 1 for newline after
label */
-
- CG(increment_lineno) = 1; /* For newline before label */
- BEGIN(ST_END_HEREDOC);
-
- if (CG(literal_type) == IS_UNICODE) {
- return zend_scan_unicode_escape_string(zendlval,
yytext, len, 0, T_ENCAPSED_AND_WHITESPACE TSRMLS_CC);
- } else {
- zend_scan_binary_escape_string(zendlval, yytext, len, 0
TSRMLS_CC);
- return T_ENCAPSED_AND_WHITESPACE;
- }
- } else {
- /* Go back to end of label, so the next match works correctly
in case of
- * a variable or another label at the beginning of the next
line */
- yyless(yyleng - 1);
- yymore();
- }
-}
-
<ST_END_HEREDOC>{ANY_CHAR} {
YYCURSOR += CG(heredoc_len) - 1;
yyleng = CG(heredoc_len);
@@ -2480,31 +2434,69 @@
}
-/* Will only match when $ follows: "{$" */
-<ST_DOUBLE_QUOTES,ST_BACKQUOTE,ST_HEREDOC>"{" {
+<ST_DOUBLE_QUOTES,ST_BACKQUOTE,ST_HEREDOC>"{$" {
Z_LVAL_P(zendlval) = (long) '{';
yy_push_state(ST_IN_SCRIPTING TSRMLS_CC);
+ yyless(1);
return T_CURLY_OPEN;
}
-<ST_DOUBLE_QUOTES>{DOUBLE_QUOTES_CHARS}+ {
- if (CG(literal_type) == IS_UNICODE) {
- return zend_scan_unicode_escape_string(zendlval, yytext,
yyleng, 0x22 /*'"'*/, T_ENCAPSED_AND_WHITESPACE TSRMLS_CC);
- } else {
- zend_scan_binary_escape_string(zendlval, yytext, yyleng, '"'
TSRMLS_CC);
- return T_ENCAPSED_AND_WHITESPACE;
- }
+<ST_DOUBLE_QUOTES>["] {
+ BEGIN(ST_IN_SCRIPTING);
+ return '"';
}
-/* "{"{2,}|"$"{2,} handles { before "{$" or literal $ before a variable or "${"
- * (("{"+|"$"+)["]) handles { or $ at the end of a string
- *
- * Same for backquotes and heredocs, except the second case doesn't apply to
- * heredocs. yyless(yyleng - 1) is used to correct taking one character too
many
- */
-<ST_DOUBLE_QUOTES>{DOUBLE_QUOTES_CHARS}*("{"{2,}|"$"{2,}|(("{"+|"$"+)["])) {
- yyless(yyleng - 1);
+<ST_BACKQUOTE>[`] {
+ BEGIN(ST_IN_SCRIPTING);
+ return '`';
+}
+
+
+<ST_DOUBLE_QUOTES>{ANY_CHAR} {
+ if (GET_DOUBLE_QUOTES_SCANNED_LENGTH()) {
+ YYCURSOR += GET_DOUBLE_QUOTES_SCANNED_LENGTH() - 1;
+ SET_DOUBLE_QUOTES_SCANNED_LENGTH(0);
+
+ goto double_quotes_scan_done;
+ }
+
+ if (YYCURSOR > YYLIMIT) {
+ return 0;
+ }
+ if (yytext[0] == '\\' && YYCURSOR < YYLIMIT) {
+ YYCURSOR++;
+ }
+
+ while (YYCURSOR < YYLIMIT) {
+ switch (*YYCURSOR++) {
+ case '"':
+ break;
+ case '$':
+ if (IS_LABEL_START(*YYCURSOR) || *YYCURSOR ==
'{') {
+ break;
+ }
+ continue;
+ case '{':
+ if (*YYCURSOR == '$') {
+ break;
+ }
+ continue;
+ case '\\':
+ if (YYCURSOR < YYLIMIT) {
+ YYCURSOR++;
+ }
+ /* fall through */
+ default:
+ continue;
+ }
+
+ YYCURSOR--;
+ break;
+ }
+
+double_quotes_scan_done:
+ yyleng = YYCURSOR - SCNG(yy_text);
if (CG(literal_type) == IS_UNICODE) {
return zend_scan_unicode_escape_string(zendlval, yytext,
yyleng, 0x22 /*'"'*/, T_ENCAPSED_AND_WHITESPACE TSRMLS_CC);
@@ -2515,17 +2507,42 @@
}
-<ST_BACKQUOTE>{BACKQUOTE_CHARS}+ {
- if (CG(literal_type) == IS_UNICODE) {
- return zend_scan_unicode_escape_string(zendlval, yytext,
yyleng, 0x60 /*'`'*/, T_ENCAPSED_AND_WHITESPACE TSRMLS_CC);
- } else {
- zend_scan_binary_escape_string(zendlval, yytext, yyleng, '`'
TSRMLS_CC);
- return T_ENCAPSED_AND_WHITESPACE;
+<ST_BACKQUOTE>{ANY_CHAR} {
+ if (YYCURSOR > YYLIMIT) {
+ return 0;
+ }
+ if (yytext[0] == '\\' && YYCURSOR < YYLIMIT) {
+ YYCURSOR++;
}
-}
-<ST_BACKQUOTE>{BACKQUOTE_CHARS}*("{"{2,}|"$"{2,}|(("{"+|"$"+)[`])) {
- yyless(yyleng - 1);
+ while (YYCURSOR < YYLIMIT) {
+ switch (*YYCURSOR++) {
+ case '`':
+ break;
+ case '$':
+ if (IS_LABEL_START(*YYCURSOR) || *YYCURSOR ==
'{') {
+ break;
+ }
+ continue;
+ case '{':
+ if (*YYCURSOR == '$') {
+ break;
+ }
+ continue;
+ case '\\':
+ if (YYCURSOR < YYLIMIT) {
+ YYCURSOR++;
+ }
+ /* fall through */
+ default:
+ continue;
+ }
+
+ YYCURSOR--;
+ break;
+ }
+
+ yyleng = YYCURSOR - SCNG(yy_text);
if (CG(literal_type) == IS_UNICODE) {
return zend_scan_unicode_escape_string(zendlval, yytext,
yyleng, 0x60 /*'`'*/, T_ENCAPSED_AND_WHITESPACE TSRMLS_CC);
@@ -2536,90 +2553,144 @@
}
-/* ({HEREDOC_NEWLINE}+({LABEL}";"?)?)? handles the possible case of newline
- * sequences, possibly followed by a label, that couldn't be matched with
- * HEREDOC_CHARS because of a following variable or "{$"
- *
- * This doesn't affect real ending labels, as they are followed by a newline,
- * which will result in a longer match for the correct rule if present
- */
-<ST_HEREDOC>{HEREDOC_CHARS}*({HEREDOC_NEWLINE}+({LABEL}";"?)?)? {
- if (CG(literal_type) == IS_UNICODE) {
- return zend_scan_unicode_escape_string(zendlval, yytext,
yyleng, 0, T_ENCAPSED_AND_WHITESPACE TSRMLS_CC);
- } else {
- zend_scan_binary_escape_string(zendlval, yytext, yyleng, 0
TSRMLS_CC);
- return T_ENCAPSED_AND_WHITESPACE;
+<ST_HEREDOC>{ANY_CHAR} {
+ int newline = 0;
+
+ if (YYCURSOR > YYLIMIT) {
+ return 0;
}
-}
-<ST_HEREDOC>{HEREDOC_CHARS}*({HEREDOC_NEWLINE}+({LABEL}";"?)?)?("{"{2,}|"$"{2,})
{
- yyless(yyleng - 1);
+ YYCURSOR--;
+
+ while (YYCURSOR < YYLIMIT) {
+ switch (*YYCURSOR++) {
+ case '\r':
+ if (*YYCURSOR == '\n') {
+ YYCURSOR++;
+ }
+ /* fall through */
+ case '\n':
+ /* Check for ending label on the next line */
+ if (IS_LABEL_START(*YYCURSOR) &&
CG(heredoc_len) < YYLIMIT - YYCURSOR && !memcmp(YYCURSOR, CG(heredoc),
CG(heredoc_len))) {
+ YYCTYPE *end = YYCURSOR +
CG(heredoc_len);
+
+ if (*end == ';') {
+ end++;
+ }
+
+ if (*end == '\n' || *end == '\r') {
+ /* newline before label will be
subtracted from returned text, but
+ * yyleng/yytext will include
it, for zend_highlight/strip, tokenizer, etc. */
+ if (YYCURSOR[-2] == '\r' &&
YYCURSOR[-1] == '\n') {
+ newline = 2; /* Windows
newline */
+ } else {
+ newline = 1;
+ }
+
+ CG(increment_lineno) = 1; /*
For newline before label */
+ BEGIN(ST_END_HEREDOC);
+
+ goto heredoc_scan_done;
+ }
+ }
+ continue;
+ case '$':
+ if (IS_LABEL_START(*YYCURSOR) || *YYCURSOR ==
'{') {
+ break;
+ }
+ continue;
+ case '{':
+ if (*YYCURSOR == '$') {
+ break;
+ }
+ continue;
+ case '\\':
+ if (YYCURSOR < YYLIMIT && *YYCURSOR != '\n' &&
*YYCURSOR != '\r') {
+ YYCURSOR++;
+ }
+ /* fall through */
+ default:
+ continue;
+ }
+
+ YYCURSOR--;
+ break;
+ }
+
+heredoc_scan_done:
+ yyleng = YYCURSOR - SCNG(yy_text);
if (CG(literal_type) == IS_UNICODE) {
- return zend_scan_unicode_escape_string(zendlval, yytext,
yyleng, 0, T_ENCAPSED_AND_WHITESPACE TSRMLS_CC);
+ return zend_scan_unicode_escape_string(zendlval, yytext, yyleng
- newline, 0, T_ENCAPSED_AND_WHITESPACE TSRMLS_CC);
} else {
- zend_scan_binary_escape_string(zendlval, yytext, yyleng, 0
TSRMLS_CC);
+ zend_scan_binary_escape_string(zendlval, yytext, yyleng -
newline, 0 TSRMLS_CC);
return T_ENCAPSED_AND_WHITESPACE;
}
}
-<ST_NOWDOC>({NOWDOC_CHARS}+{NEWLINE}+|{NEWLINE}+){LABEL}";"?[\n\r] {
- char *end = yytext + yyleng - 1;
+<ST_NOWDOC>{ANY_CHAR} {
+ int newline = 0;
- if (end[-1] == ';') {
- end--;
- yyleng--;
- }
-
- if (yyleng > CG(heredoc_len) && !memcmp(end - CG(heredoc_len),
CG(heredoc), CG(heredoc_len))) {
- int len = yyleng - CG(heredoc_len) - 2; /* 2 for newline before
and after label */
-
- /* May have matched fooLABEL; make sure there's a newline
before it */
- if (yytext[len] != '\n') {
- if (yytext[len] != '\r') {
- yyless(yyleng - 1);
- yymore();
- }
- } else if (len > 0 && yytext[len - 1] == '\r') {
- len--; /* Windows newline */
- }
+ if (YYCURSOR > YYLIMIT) {
+ return 0;
+ }
- /* Go back before label, to match in ST_END_HEREDOC state.
yytext will include
- * newline before label, for zend_highlight/strip, tokenizer,
etc. */
- yyless(yyleng - CG(heredoc_len) - 1); /* 1 for newline after
label */
+ YYCURSOR--;
- CG(increment_lineno) = 1; /* For newline before label */
- BEGIN(ST_END_HEREDOC);
+ while (YYCURSOR < YYLIMIT) {
+ switch (*YYCURSOR++) {
+ case '\r':
+ if (*YYCURSOR == '\n') {
+ YYCURSOR++;
+ }
+ /* fall through */
+ case '\n':
+ /* Check for ending label on the next line */
+ if (IS_LABEL_START(*YYCURSOR) &&
CG(heredoc_len) < YYLIMIT - YYCURSOR && !memcmp(YYCURSOR, CG(heredoc),
CG(heredoc_len))) {
+ YYCTYPE *end = YYCURSOR +
CG(heredoc_len);
- if (!zend_copy_scanner_string(zendlval, yytext, len,
CG(literal_type), SCNG(output_conv) TSRMLS_CC)) {
- return 0;
- }
- HANDLE_NEWLINES(yytext, len);
- return T_ENCAPSED_AND_WHITESPACE;
- } else {
- /* Go back to end of label, so the next match works correctly
in case of
- * another label at the beginning of the next line */
- yyless(yyleng - 1);
- yymore();
- }
-}
+ if (*end == ';') {
+ end++;
+ }
+ if (*end == '\n' || *end == '\r') {
+ /* newline before label will be
subtracted from returned text, but
+ * yyleng/yytext will include
it, for zend_highlight/strip, tokenizer, etc. */
+ if (YYCURSOR[-2] == '\r' &&
YYCURSOR[-1] == '\n') {
+ newline = 2; /* Windows
newline */
+ } else {
+ newline = 1;
+ }
-<ST_DOUBLE_QUOTES>["] {
- BEGIN(ST_IN_SCRIPTING);
- return '"';
-}
+ CG(increment_lineno) = 1; /*
For newline before label */
+ BEGIN(ST_END_HEREDOC);
+ goto nowdoc_scan_done;
+ }
+ }
+ /* fall through */
+ default:
+ continue;
+ }
+ }
-<ST_BACKQUOTE>[`] {
- BEGIN(ST_IN_SCRIPTING);
- return '`';
+nowdoc_scan_done:
+ yyleng = YYCURSOR - SCNG(yy_text);
+
+ if (!zend_copy_scanner_string(zendlval, yytext, yyleng - newline,
CG(literal_type), SCNG(output_conv) TSRMLS_CC)) {
+ return 0;
+ }
+ HANDLE_NEWLINES(yytext, yyleng - newline);
+ return T_ENCAPSED_AND_WHITESPACE;
}
-<*>{NULL} { return 0; } /* EOF */
<ST_IN_SCRIPTING,ST_VAR_OFFSET>{ANY_CHAR} {
+ if (YYCURSOR > YYLIMIT) {
+ return 0;
+ }
+
zend_error(E_COMPILE_WARNING,"Unexpected character in input: '%c'
(ASCII=%d) state=%d", yytext[0], yytext[0], YYSTATE);
goto restart;
}
http://cvs.php.net/viewvc.cgi/php-src/ext/standard/tests/strings/highlight_file.phpt?r1=1.8&r2=1.9&diff_format=u
Index: php-src/ext/standard/tests/strings/highlight_file.phpt
diff -u php-src/ext/standard/tests/strings/highlight_file.phpt:1.8
php-src/ext/standard/tests/strings/highlight_file.phpt:1.9
--- php-src/ext/standard/tests/strings/highlight_file.phpt:1.8 Mon Mar 16
01:40:01 2009
+++ php-src/ext/standard/tests/strings/highlight_file.phpt Tue May 5
01:35:13 2009
@@ -49,7 +49,7 @@
</span>
</code>bool(true)
<code><span style="color: #000000">
-<span style="color: #0000BB"><?php </span><span style="color:
#007700">echo </span><span style="color: #FF9900">"test ?></span>
+<span style="color: #0000BB"><?php </span><span style="color:
#007700">echo </span><span style="color: #DD0000">"test ?></span>
</span>
</code>bool(true)
<code><span style="color: #000000">
--
PHP CVS Mailing List (http://www.php.net/)
To unsubscribe, visit: http://www.php.net/unsub.php
