Fix matching characters above 255 when a negative character type was used without enabled UCP in a positive class

pcre-8.39-Fix-character-class-bug-when-a-Unicode-property-was-.patch

@@ -0,0 +1,269 @@
+From 1767e8be04f300ad10f2b358c281a23f0759e3d3 Mon Sep 17 00:00:00 2001
+From: ph10 <ph10@2f5784b3-3f2a-0410-8824-cb99058d5e15>
+Date: Thu, 4 Aug 2016 17:15:38 +0000
+Subject: [PATCH] Fix character class bug when a Unicode property was present
+ with \D etc in a non-ucp character class in a wide character mode.
+MIME-Version: 1.0
+Content-Type: text/plain; charset=UTF-8
+Content-Transfer-Encoding: 8bit
+
+git-svn-id: svn://vcs.exim.org/pcre/code/trunk@1664 2f5784b3-3f2a-0410-8824-cb99058d5e15
+
+Petr Písař: Ported to 8.39.
+
+diff --git a/pcre_compile.c b/pcre_compile.c
+index 7cd3950..fb80ed1 100644
+--- a/pcre_compile.c
++++ b/pcre_compile.c
+@@ -5579,6 +5579,34 @@ for (;; ptr++)
+ #endif
+ #if defined SUPPORT_UTF || !defined COMPILE_PCRE8
+       {
++      /* For non-UCP wide characters, in a non-negative class containing \S or
++      similar (should_flip_negation is set), all characters greater than 255
++      must be in the class. */
++
++      if (
++#if defined COMPILE_PCRE8
++           utf &&
++#endif
++           should_flip_negation && !negate_class && (options & PCRE_UCP) == 0)
++        {
++        *class_uchardata++ = XCL_RANGE;
++        if (utf)   /* Will always be utf in the 8-bit library */
++          {
++          class_uchardata += PRIV(ord2utf)(0x100, class_uchardata);
++          class_uchardata += PRIV(ord2utf)(0x10ffff, class_uchardata);
++          }
++        else       /* Can only happen for the 16-bit & 32-bit libraries */
++          {
++#if defined COMPILE_PCRE16
++          *class_uchardata++ = 0x100;
++          *class_uchardata++ = 0xffffu;
++#elif defined COMPILE_PCRE32
++          *class_uchardata++ = 0x100;
++          *class_uchardata++ = 0xffffffffu;
++#endif
++          }
++        }
++
+       *class_uchardata++ = XCL_END;    /* Marks the end of extra data */
+       *code++ = OP_XCLASS;
+       code += LINK_SIZE;
+diff --git a/testdata/testinput16 b/testdata/testinput16
+index 15419e6..7ccde0a 100644
+--- a/testdata/testinput16
++++ b/testdata/testinput16
+@@ -38,4 +38,30 @@
+ /s+/i8SI
+     SSss\x{17f}
+ 
++/[\W\p{Any}]/BZ
++    abc
++    123 
++
++/[\W\pL]/BZ
++    abc
++    ** Failers 
++    123
++    
++/[\D]/8
++    \x{1d7cf}
++
++/[\D\P{Nd}]/8
++    \x{1d7cf}
++
++/[^\D]/8
++    a9b
++    ** Failers
++    \x{1d7cf}
++
++/[^\D\P{Nd}]/8
++    a9b
++    \x{1d7cf}
++    ** Failers
++    \x{10000}
++
+ /-- End of testinput16 --/
+diff --git a/testdata/testinput19 b/testdata/testinput19
+index ce45afc..dfe8c7b 100644
+--- a/testdata/testinput19
++++ b/testdata/testinput19
+@@ -25,4 +25,21 @@
+ /s+/i8SI
+     SSss\x{17f}
+ 
++/[\D]/8
++    \x{1d7cf}
++
++/[\D\P{Nd}]/8
++    \x{1d7cf}
++
++/[^\D]/8
++    a9b
++    ** Failers
++    \x{1d7cf}
++
++/[^\D\P{Nd}]/8
++    a9b
++    \x{1d7cf}
++    ** Failers
++    \x{10000}
++
+ /-- End of testinput19 --/ 
+diff --git a/testdata/testinput7 b/testdata/testinput7
+index 00b9738..f44a810 100644
+--- a/testdata/testinput7
++++ b/testdata/testinput7
+@@ -838,15 +838,6 @@ of case for anything other than the ASCII letters. --/
+ /^s?c/mi8I
+     scat
+ 
+-/[\W\p{Any}]/BZ
+-    abc
+-    123 
+-
+-/[\W\pL]/BZ
+-    abc
+-    ** Failers 
+-    123     
+-
+ /a[[:punct:]b]/WBZ
+ 
+ /a[[:punct:]b]/8WBZ
+diff --git a/testdata/testoutput16 b/testdata/testoutput16
+index fd184cd..e6ba26a 100644
+--- a/testdata/testoutput16
++++ b/testdata/testoutput16
+@@ -138,4 +138,56 @@ Starting chars: S s \xc5
+     SSss\x{17f}
+  0: SSss\x{17f}
+ 
++/[\W\p{Any}]/BZ
++------------------------------------------------------------------
++        Bra
++        [\x00-/:-@[-^`{-\xff\p{Any}]
++        Ket
++        End
++------------------------------------------------------------------
++    abc
++ 0: a
++    123 
++ 0: 1
++
++/[\W\pL]/BZ
++------------------------------------------------------------------
++        Bra
++        [\x00-/:-@[-^`{-\xff\p{L}]
++        Ket
++        End
++------------------------------------------------------------------
++    abc
++ 0: a
++    ** Failers 
++ 0: *
++    123
++No match
++    
++/[\D]/8
++    \x{1d7cf}
++ 0: \x{1d7cf}
++
++/[\D\P{Nd}]/8
++    \x{1d7cf}
++ 0: \x{1d7cf}
++
++/[^\D]/8
++    a9b
++ 0: 9
++    ** Failers
++No match
++    \x{1d7cf}
++No match
++
++/[^\D\P{Nd}]/8
++    a9b
++ 0: 9
++    \x{1d7cf}
++ 0: \x{1d7cf}
++    ** Failers
++No match
++    \x{10000}
++No match
++
+ /-- End of testinput16 --/
+diff --git a/testdata/testoutput19 b/testdata/testoutput19
+index eb8a8f6..982bea4 100644
+--- a/testdata/testoutput19
++++ b/testdata/testoutput19
+@@ -105,4 +105,30 @@ Starting chars: S s \xff
+     SSss\x{17f}
+  0: SSss\x{17f}
+ 
++/[\D]/8
++    \x{1d7cf}
++ 0: \x{1d7cf}
++
++/[\D\P{Nd}]/8
++    \x{1d7cf}
++ 0: \x{1d7cf}
++
++/[^\D]/8
++    a9b
++ 0: 9
++    ** Failers
++No match
++    \x{1d7cf}
++No match
++
++/[^\D\P{Nd}]/8
++    a9b
++ 0: 9
++    \x{1d7cf}
++ 0: \x{1d7cf}
++    ** Failers
++No match
++    \x{10000}
++No match
++
+ /-- End of testinput19 --/ 
+diff --git a/testdata/testoutput7 b/testdata/testoutput7
+index fdfff64..2b167b2 100644
+--- a/testdata/testoutput7
++++ b/testdata/testoutput7
+@@ -2295,32 +2295,6 @@ Need char = 'c' (caseless)
+     scat
+  0: sc
+ 
+-/[\W\p{Any}]/BZ
+-------------------------------------------------------------------
+-        Bra
+-        [\x00-/:-@[-^`{-\xff\p{Any}]
+-        Ket
+-        End
+-------------------------------------------------------------------
+-    abc
+- 0: a
+-    123 
+- 0: 1
+-
+-/[\W\pL]/BZ
+-------------------------------------------------------------------
+-        Bra
+-        [\x00-/:-@[-^`{-\xff\p{L}]
+-        Ket
+-        End
+-------------------------------------------------------------------
+-    abc
+- 0: a
+-    ** Failers 
+- 0: *
+-    123     
+-No match
+-
+ /a[[:punct:]b]/WBZ
+ ------------------------------------------------------------------
+         Bra
+-- 
+2.5.5
+

pcre.spec

@@ -41,6 +41,10 @@ Patch2:     pcre-8.39-Fix-bad-interaction-between-o-and-M-in-pcregrep.patch
 # Fix register overwite in JIT when SSE2 acceleration is enabled,
 # in upstream after 8.39
 Patch3:     pcre-8.39-Fix-register-overwite-in-JIT-when-SSE2-acceleration-.patch
+# Fix matching characters above 255 when a negative character type was used
+# without enabled UCP in a positive class, in upstream after 8.39,
+# upstream bug #1866
+Patch4:     pcre-8.39-Fix-character-class-bug-when-a-Unicode-property-was-.patch
 BuildRequires:  readline-devel
 BuildRequires:  autoconf
 BuildRequires:  automake
@@ -132,6 +136,7 @@ Utilities demonstrating PCRE capabilities like pcregrep or pcretest.
 %patch1 -p1
 %patch2 -p1
 %patch3 -p1
+%patch4 -p1
 # Because of rpath patch
 libtoolize --copy --force
 autoreconf -vif
@@ -230,6 +235,8 @@ make %{?_smp_mflags} check VERBOSE=yes
 %changelog
 * Tue Aug 30 2016 Petr Pisar <ppisar@redhat.com> - 8.39-3
 - Fix register overwite in JIT when SSE2 acceleration is enabled
+- Fix matching characters above 255 when a negative character type was used
+  without enabled UCP in a positive class (upstream bug #1866)
 
 * Mon Jun 20 2016 Petr Pisar <ppisar@redhat.com> - 8.39-2
 - Fix repeated pcregrep output if -o with -M options were used and the match