Revision: 292
http://www.exim.org/viewvc/pcre2?view=rev&revision=292
Author: ph10
Date: 2015-06-19 17:41:22 +0100 (Fri, 19 Jun 2015)
Log Message:
-----------
Fix EBCDIC 0x41 omission and give error for \p and \P in a class when not
supported.
Modified Paths:
--------------
code/trunk/ChangeLog
code/trunk/src/pcre2_compile.c
code/trunk/src/pcre2_internal.h
code/trunk/testdata/testinputEBC
code/trunk/testdata/testoutputEBC
Modified: code/trunk/ChangeLog
===================================================================
--- code/trunk/ChangeLog 2015-06-19 10:16:33 UTC (rev 291)
+++ code/trunk/ChangeLog 2015-06-19 16:41:22 UTC (rev 292)
@@ -1,7 +1,7 @@
Change Log for PCRE2
--------------------
- Version 10.20 16-June-2015
+Version 10.20 16-June-2015
--------------------------
1. Callouts with string arguments have been added.
@@ -167,7 +167,15 @@
43. Single character repetition in JIT has been improved. 20-30% speedup
was achieved on certain patterns.
+44. The EBCDIC character 0x41 is a non-breaking space, equivalent to 0xa0 in
+ASCII/Unicode. This has now been added to the list of characters that are
+recognized as white space in EBCDIC.
+
+45. When PCRE2 was compiled without Unicode support, the use of \p and \P gave
+an error (correctly) when used outside a class, but did not give an error
+within a class.
+
Version 10.10 06-March-2015
---------------------------
Modified: code/trunk/src/pcre2_compile.c
===================================================================
--- code/trunk/src/pcre2_compile.c 2015-06-19 10:16:33 UTC (rev 291)
+++ code/trunk/src/pcre2_compile.c 2015-06-19 16:41:22 UTC (rev 292)
@@ -4397,9 +4397,9 @@
cb, PRIV(vspace_list));
break;
-#ifdef SUPPORT_UNICODE
case ESC_p:
case ESC_P:
+#ifdef SUPPORT_UNICODE
{
BOOL negated;
unsigned int ptype = 0, pdata = 0;
@@ -4413,6 +4413,9 @@
class_has_8bitchar--; /* Undo! */
}
break;
+#else
+ *errorcodeptr = ERR45;
+ goto FAILED;
#endif
/* Unrecognized escapes are faulted. */
Modified: code/trunk/src/pcre2_internal.h
===================================================================
--- code/trunk/src/pcre2_internal.h 2015-06-19 10:16:33 UTC (rev 291)
+++ code/trunk/src/pcre2_internal.h 2015-06-19 16:41:22 UTC (rev 292)
@@ -382,7 +382,7 @@
PCRE for a long time. */
#define HSPACE_LIST \
- CHAR_HT, CHAR_SPACE, 0xa0, \
+ CHAR_HT, CHAR_SPACE, CHAR_NBSP, \
0x1680, 0x180e, 0x2000, 0x2001, 0x2002, 0x2003, 0x2004, 0x2005, \
0x2006, 0x2007, 0x2008, 0x2009, 0x200A, 0x202f, 0x205f, 0x3000, \
NOTACHAR
@@ -408,7 +408,7 @@
#define HSPACE_BYTE_CASES \
case CHAR_HT: \
case CHAR_SPACE: \
- case 0xa0 /* NBSP */
+ case CHAR_NBSP
#define HSPACE_CASES \
HSPACE_BYTE_CASES: \
@@ -435,11 +435,12 @@
/* -------------- EBCDIC environments -------------- */
#else
-#define HSPACE_LIST CHAR_HT, CHAR_SPACE
+#define HSPACE_LIST CHAR_HT, CHAR_SPACE, CHAR_NBSP
#define HSPACE_BYTE_CASES \
case CHAR_HT: \
- case CHAR_SPACE
+ case CHAR_SPACE: \
+ case CHAR_NBSP
#define HSPACE_CASES HSPACE_BYTE_CASES
@@ -633,6 +634,7 @@
#define CHAR_ESC '\047'
#define CHAR_DEL '\007'
+#define CHAR_NBSP ((unsigned char)'\x41')
#define STR_ESC "\047"
#define STR_DEL "\007"
@@ -647,6 +649,7 @@
#define CHAR_NEL ((unsigned char)'\x85')
#define CHAR_ESC '\033'
#define CHAR_DEL '\177'
+#define CHAR_NBSP ((unsigned char)'\xa0')
#define STR_LF "\n"
#define STR_NL STR_LF
@@ -1029,6 +1032,7 @@
#define CHAR_VERTICAL_LINE '\174'
#define CHAR_RIGHT_CURLY_BRACKET '\175'
#define CHAR_TILDE '\176'
+#define CHAR_NBSP ((unsigned char)'\xa0')
#define STR_HT "\011"
#define STR_VT "\013"
Modified: code/trunk/testdata/testinputEBC
===================================================================
--- code/trunk/testdata/testinputEBC 2015-06-19 10:16:33 UTC (rev 291)
+++ code/trunk/testdata/testinputEBC 2015-06-19 16:41:22 UTC (rev 292)
@@ -29,13 +29,16 @@
/^A\\x88/
A B
+ A\x41B
# Test \H
/^A\\xC8/
AB
+ A\x42B
** Fail
A B
+ A\x41B
# Test \R
Modified: code/trunk/testdata/testoutputEBC
===================================================================
--- code/trunk/testdata/testoutputEBC 2015-06-19 10:16:33 UTC (rev 291)
+++ code/trunk/testdata/testoutputEBC 2015-06-19 16:41:22 UTC (rev 292)
@@ -41,6 +41,8 @@
/^A\\x88/
A B
0: A\x20
+ A\x41B
+ 0: AA
# Test \H
@@ -47,10 +49,14 @@
/^A\\xC8/
AB
0: AB
+ A\x42B
+ 0: AB
** Fail
No match
A B
No match
+ A\x41B
+No match
# Test \R
