Fix support for regular expressions with look-behind assertions and case insensitive matching. Ebuild by Arfrever.

Package-Manager: portage-2.1.10.57/cvs/Linux x86_64

3 files changed, 227 insertions, 1 deletions

diff --git a/dev-libs/icu/ChangeLog b/dev-libs/icu/ChangeLog
index 278c3de78ee4..0700a1041c3b 100644
--- a/dev-libs/icu/ChangeLog
+++ b/dev-libs/icu/ChangeLog
@@ -1,6 +1,13 @@
 # ChangeLog for dev-libs/icu
 # Copyright 1999-2012 Gentoo Foundation; Distributed under the GPL v2
-# $Header: /var/cvsroot/gentoo-x86/dev-libs/icu/ChangeLog,v 1.182 2012/05/05 19:38:27 hwoarang Exp $
+# $Header: /var/cvsroot/gentoo-x86/dev-libs/icu/ChangeLog,v 1.183 2012/05/06 16:26:56 hwoarang Exp $
+
+*icu-49.1.1-r1 (06 May 2012)
+
+  06 May 2012; Markos Chandras <hwoarang@gentoo.org>
+  +files/icu-49.1.1-regex.patch, +icu-49.1.1-r1.ebuild:
+  Fix support for regular expressions with look-behind assertions and case
+  insensitive matching. Ebuild by Arfrever.
 
   05 May 2012; Markos Chandras <hwoarang@gentoo.org> metadata.xml:
   remove description from metadata
diff --git a/dev-libs/icu/files/icu-49.1.1-regex.patch b/dev-libs/icu/files/icu-49.1.1-regex.patch
new file mode 100644
index 000000000000..e9d64e5c6f48
--- /dev/null
+++ b/dev-libs/icu/files/icu-49.1.1-regex.patch
@@ -0,0 +1,128 @@
+http://ssl.icu-project.org/trac/ticket/9283
+http://ssl.icu-project.org/trac/changeset/31782
+
+--- i18n/regexcmp.cpp
++++ i18n/regexcmp.cpp
+@@ -3306,10 +3306,31 @@
+             }
+ 
+         case URX_STRING_I:
+-            // TODO:  Is the case-folded string the longest?
+-            //        If so we can optimize this the same as URX_STRING.
+-            loc++;
+-            currentLen = INT32_MAX;
++            // TODO:  This code assumes that any user string that matches will be no longer
++            //        than our compiled string, with case insensitive matching.
++            //        Our compiled string has been case-folded already.
++            //
++            //        Any matching user string will have no more code points than our
++            //        compiled (folded) string.  Folding may add code points, but
++            //        not remove them.
++            //
++            //        There is a potential problem if a supplemental code point 
++            //        case-folds to a BMP code point.  In this case our compiled string
++            //        could be shorter (in code units) than a matching user string.
++            //
++            //        At this time (Unicode 6.1) there are no such characters, and this case
++            //        is not being handled.  A test, intltest regex/Bug9283, will fail if
++            //        any problematic characters are added to Unicode.
++            //
++            //        If this happens, we can make a set of the BMP chars that the
++            //        troublesome supplementals fold to, scan our string, and bump the
++            //        currentLen one extra for each that is found.
++            //
++            {
++                loc++;
++                int32_t stringLenOp = (int32_t)fRXPat->fCompiledPat->elementAti(loc);
++                currentLen = safeIncrement(currentLen, URX_VAL(stringLenOp));
++            }
+             break;
+ 
+         case URX_CTR_INIT:
+--- test/intltest/regextst.cpp
++++ test/intltest/regextst.cpp
+@@ -26,6 +26,7 @@
+ #include "unicode/regex.h"
+ #include "unicode/uchar.h"
+ #include "unicode/ucnv.h"
++#include "unicode/uniset.h"
+ #include "unicode/ustring.h"
+ #include "regextst.h"
+ #include "uvector.h"
+@@ -127,6 +128,9 @@
+         case 20: name = "CheckInvBufSize";
+             if (exec) CheckInvBufSize();
+             break;
++        case 21: name = "Bug 9283";
++            if (exec) Bug9283();
++            break;
+ 
+         default: name = "";
+             break; //needed to end loop
+@@ -5184,6 +5188,34 @@
+     delete pMatcher;
+ }
+ 
++// Bug 9283
++//   This test is checking for the existance of any supplemental characters that case-fold
++//   to a bmp character.  
++//
++//   At the time of this writing there are none. If any should appear in a subsequent release 
++//   of Unicode, the code in regular expressions compilation that determines the longest 
++//   posssible match for a literal string  will need to be enhanced.  
++//
++//   See file regexcmp.cpp, case URX_STRING_I in RegexCompile::maxMatchLength()
++//   for details on what to do in case of a failure of this test.
++//
++void RegexTest::Bug9283() {
++    UErrorCode status = U_ZERO_ERROR;
++    UnicodeSet supplementalsWithCaseFolding("[[:CWCF:]&[\\U00010000-\\U0010FFFF]]", status);
++    REGEX_CHECK_STATUS;
++    int32_t index;
++    UChar32 c;
++    for (index=0; ; index++) {
++        c = supplementalsWithCaseFolding.charAt(index);
++        if (c == -1) {
++            break;
++        }
++        UnicodeString cf = UnicodeString(c).foldCase();
++        REGEX_ASSERT(cf.length() >= 2);
++    }
++}
++
++
+ void RegexTest::CheckInvBufSize() {
+   if(inv_next>=INV_BUFSIZ) {
+     errln("%s: increase #define of INV_BUFSIZ ( is %d but needs to be at least %d )\n",
+--- test/intltest/regextst.h
++++ test/intltest/regextst.h
+@@ -1,6 +1,6 @@
+ /********************************************************************
+  * COPYRIGHT:
+- * Copyright (c) 2002-2011, International Business Machines Corporation and
++ * Copyright (c) 2002-2012, International Business Machines Corporation and
+  * others. All Rights Reserved.
+  ********************************************************************/
+ 
+@@ -45,6 +45,7 @@
+     virtual void Bug7740();
+     virtual void Bug8479();
+     virtual void Bug7029();
++    virtual void Bug9283();
+     virtual void CheckInvBufSize();
+     
+     // The following functions are internal to the regexp tests.
+--- test/testdata/regextst.txt
++++ test/testdata/regextst.txt
+@@ -1141,6 +1141,11 @@
+ "[\w]+"                           "  <0>abc\u200cdef\u200dghi</0>   "
+ "[\w]+"                         i "  <0>abc\u200cdef\u200dghi</0>   "
+ 
++# Bug 9283
++#  uregex_open fails for look-behind assertion + case-insensitive
++
++"(ab)?(?<=ab)cd|ef"             i  "<0><1>ab</1>cd</0>"
++
+ #  Random debugging, Temporary
+ #
+ #"^(?:a?b?)*$"	                  "a--"
diff --git a/dev-libs/icu/icu-49.1.1-r1.ebuild b/dev-libs/icu/icu-49.1.1-r1.ebuild
new file mode 100644
index 000000000000..212e131d47be
--- /dev/null
+++ b/dev-libs/icu/icu-49.1.1-r1.ebuild
@@ -0,0 +1,91 @@
+# Copyright 1999-2012 Gentoo Foundation
+# Distributed under the terms of the GNU General Public License v2
+# $Header: /var/cvsroot/gentoo-x86/dev-libs/icu/icu-49.1.1-r1.ebuild,v 1.1 2012/05/06 16:26:56 hwoarang Exp $
+
+EAPI="4"
+
+inherit eutils versionator
+
+MAJOR_VERSION="$(get_version_component_range 1)"
+if [[ "${PV}" =~ ^[[:digit:]]+_rc[[:digit:]]*$ ]]; then
+	MINOR_VERSION="0"
+else
+	MINOR_VERSION="$(get_version_component_range 2)"
+fi
+
+DESCRIPTION="International Components for Unicode"
+HOMEPAGE="http://www.icu-project.org/"
+
+BASE_URI="http://download.icu-project.org/files/icu4c/${PV/_/}"
+SRC_ARCHIVE="icu4c-${PV//./_}-src.tgz"
+DOCS_ARCHIVE="icu4c-${PV//./_}-docs.zip"
+
+SRC_URI="${BASE_URI}/${SRC_ARCHIVE}
+	doc? ( ${BASE_URI}/${DOCS_ARCHIVE} )"
+
+LICENSE="BSD"
+SLOT="0"
+KEYWORDS="~alpha ~amd64 ~arm ~hppa ~ia64 ~mips ~ppc ~ppc64 ~s390 ~sh ~sparc ~x86 ~x86-fbsd"
+IUSE="debug doc examples static-libs"
+
+DEPEND="doc? ( app-arch/unzip )"
+RDEPEND=""
+
+S="${WORKDIR}/${PN}/source"
+
+QA_DT_NEEDED="/usr/lib.*/libicudata\.so\.${MAJOR_VERSION}\.${MINOR_VERSION}.*"
+
+src_unpack() {
+	unpack "${SRC_ARCHIVE}"
+	if use doc; then
+		mkdir docs
+		pushd docs > /dev/null
+		unpack "${DOCS_ARCHIVE}"
+		popd > /dev/null
+	fi
+}
+
+src_prepare() {
+	# Do not hardcode flags into icu-config.
+	# https://ssl.icu-project.org/trac/ticket/6102
+	local variable
+	for variable in CFLAGS CPPFLAGS CXXFLAGS FFLAGS LDFLAGS; do
+		sed -i -e "/^${variable} =.*/s:@${variable}@::" config/Makefile.inc.in || die "sed failed"
+	done
+
+	epatch "${FILESDIR}/${PN}-4.8.1-fix_binformat_fonts.patch"
+	epatch "${FILESDIR}/${PN}-4.8.1.1-fix_ltr.patch"
+	epatch "${FILESDIR}/${P}-regex.patch"
+}
+
+src_configure() {
+	econf \
+		$(use_enable debug) \
+		$(use_enable examples samples) \
+		$(use_enable static-libs static)
+}
+
+src_test() {
+	# INTLTEST_OPTS: intltest options
+	#   -e: Exhaustive testing
+	#   -l: Reporting of memory leaks
+	#   -v: Increased verbosity
+	# IOTEST_OPTS: iotest options
+	#   -e: Exhaustive testing
+	#   -v: Increased verbosity
+	# CINTLTST_OPTS: cintltst options
+	#   -e: Exhaustive testing
+	#   -v: Increased verbosity
+	emake -j1 check
+}
+
+src_install() {
+	emake DESTDIR="${D}" install
+
+	dohtml ../readme.html
+	dodoc ../unicode-license.txt
+	if use doc; then
+		insinto /usr/share/doc/${PF}/html/api
+		doins -r "${WORKDIR}/docs/"*
+	fi
+}