Allow using Glib for UTF8 support

8 files changed, 236 insertions, 54 deletions

diff --git a/Makefile.am b/Makefile.am
index 87def30b2..17d3dd6af 100644
--- a/Makefile.am
+++ b/Makefile.am
@@ -35,6 +35,8 @@ pubconfpath = @pubconfpath@
 pkgconfigdir = $(libdir)/pkgconfig
 krb5rcachedir = @krb5rcachedir@
 
+UNICODE_LIBS=@UNICODE_LIBS@
+
 AM_CFLAGS =
 if WANT_AUX_INFO
     AM_CFLAGS += -aux-info $@.X
@@ -118,7 +120,8 @@ krb5plugin_LTLIBRARIES = \
 endif
 
 noinst_LTLIBRARIES = \
-    libsss_crypt.la
+    libsss_crypt.la \
+    libsss_utf8.la
 
 if HAVE_NSS
     SSS_CRYPT_SOURCES = src/util/crypto/nss/nss_sha512crypt.c \
@@ -141,6 +144,9 @@ libsss_crypt_la_CFLAGS = \
 libsss_crypt_la_LIBADD = \
     $(SSS_CRYPT_LIBS)
 
+libsss_utf8_la_SOURCES = src/util/sss_utf8.c
+libsss_utf8_la_LIBADD = $(UNICODE_LIBS)
+
 if BUILD_PYTHON_BINDINGS
 pyexec_LTLIBRARIES = \
     pysss.la \
@@ -188,6 +194,7 @@ AM_CPPFLAGS = \
     $(INI_CONFIG_CFLAGS) \
     $(DHASH_CFLAGS) \
     $(LIBNL_CFLAGS) \
+    $(GLIB2_CFLAGS) \
     -DLIBDIR=\"$(libdir)\" \
     -DVARDIR=\"$(localstatedir)\" \
     -DSHLIBEXT=\"$(SHLIBEXT)\" \
@@ -268,7 +275,8 @@ SSSD_LIBS = \
     $(SSS_CRYPT_LIBS) \
     $(OPENLDAP_LIBS) \
     $(TDB_LIBS) \
-    libsss_crypt.la
+    libsss_crypt.la \
+    libsss_utf8.la
 
 PYTHON_BINDINGS_LIBS = \
     $(TALLOC_LIBS) \
@@ -293,8 +301,9 @@ TOOLS_LIBS = \
     $(INI_CONFIG_LIBS) \
     $(COLLECTION_LIBS) \
     $(DHASH_LIBS) \
-	$(OPENLDAP_LIBS) \
-	$(TDB_LIBS) \
+    $(OPENLDAP_LIBS) \
+    $(TDB_LIBS) \
+    $(UNICODE_LIBS) \
     libsss_crypt.la
 
 if BUILD_SELINUX
@@ -315,6 +324,7 @@ dist_noinst_HEADERS = \
     src/util/sss_ldap.h \
     src/util/sss_python.h \
     src/util/sss_krb5.h \
+    src/util/sss_utf8.h \
     src/util/refcount.h \
     src/util/find_uid.h \
     src/util/user_info_msg.h \
@@ -367,14 +377,13 @@ if HAVE_NSS
     dist_noinst_HEADERS += src/util/crypto/nss/nss_util.h
 endif
 
-
 lib_LTLIBRARIES = libipa_hbac.la
 dist_pkgconfig_DATA += src/providers/ipa/ipa_hbac.pc
 libipa_hbac_la_SOURCES = \
     src/providers/ipa/hbac_evaluator.c
 libipa_hbac_la_LDFLAGS = \
-    -version 1:0:1 \
-    -lunistring
+    -version 1:0:1
+libipa_hbac_la_LIBADD = libsss_utf8.la
 
 include_HEADERS = \
     src/providers/ipa/ipa_hbac.h
@@ -399,8 +408,7 @@ sssd_nss_SOURCES = \
     $(SSSD_RESPONDER_OBJ)
 sssd_nss_LDADD = \
     $(TDB_LIBS) \
-    $(SSSD_LIBS) \
-    -lunistring
+    $(SSSD_LIBS)
 
 sssd_pam_SOURCES = \
     src/responder/pam/pam_LOCAL_domain.c \
@@ -411,8 +419,7 @@ sssd_pam_SOURCES = \
     $(SSSD_RESPONDER_OBJ)
 sssd_pam_LDADD = \
     $(TDB_LIBS) \
-    $(SSSD_LIBS) \
-    -lunistring
+    $(SSSD_LIBS)
 
 sssd_be_SOURCES = \
     src/providers/data_provider_be.c \
diff --git a/configure.ac b/configure.ac
index 5e0878bca..5417d53a0 100644
--- a/configure.ac
+++ b/configure.ac
@@ -123,7 +123,19 @@ m4_include([src/external/nsupdate.m4])
 m4_include([src/external/libkeyutils.m4])
 m4_include([src/external/libnl.m4])
 m4_include([src/util/signal.m4])
-m4_include([src/external/libunistring.m4])
+
+WITH_UNICODE_LIB
+if test x$unicode_lib = xlibunistring; then
+    m4_include([src/external/libunistring.m4])
+            AC_DEFINE_UNQUOTED(HAVE_LIBUNISTRING, 1, [Using libunistring for unicode])
+            UNICODE_LIBS=-lunistring
+            AC_SUBST(UNICODE_LIBS)
+else
+    m4_include([src/external/glib.m4])
+            AC_DEFINE_UNQUOTED(HAVE_GLIB2, 1, [Using libunistring for unicode])
+            UNICODE_LIBS=$GLIB2_LIBS
+            AC_SUBST(UNICODE_LIBS)
+fi
 
 PKG_CHECK_MODULES([DBUS],[dbus-1])
 dnl if test -n "`$PKG_CONFIG --modversion dbus-1 | grep '^0\.'`" ; then
diff --git a/src/conf_macros.m4 b/src/conf_macros.m4
index 2a109b5f8..81d945eda 100644
--- a/src/conf_macros.m4
+++ b/src/conf_macros.m4
@@ -280,3 +280,23 @@ AC_DEFUN([WITH_NOLOGIN_SHELL],
     fi
     AC_DEFINE_UNQUOTED(NOLOGIN_SHELL, "$nologin_shell", [The shell used to deny access to users])
   ])
+
+AC_DEFUN([WITH_UNICODE_LIB],
+  [ AC_ARG_WITH([unicode-lib],
+                [AC_HELP_STRING([--with-unicode-lib=<library>],
+                                [Which library to use for unicode processing (libunistring, glib2) [libunistring]]
+                               )
+                ]
+               )
+    unicode_lib="libunistring"
+    if test x"$with_unicode_lib" != x; then
+        unicode_lib=$with_unicode_lib
+    fi
+
+    if test x"$unicode_lib" != x"libunistring" -a x"$unicode_lib" != x"glib2"; then
+        AC_MSG_ERROR([Unsupported unicode library])
+    fi
+
+    AM_CONDITIONAL([WITH_LIBUNISTRING], test x"$unicode_lib" = x"libunistring")
+    AM_CONDITIONAL([WITH_GLIB], test x"$unicode_lib" = x"glib2")
+  ])
diff --git a/src/external/glib.m4 b/src/external/glib.m4
new file mode 100644
index 000000000..fe9a8272e
--- /dev/null
+++ b/src/external/glib.m4
@@ -0,0 +1,11 @@
+PKG_CHECK_MODULES([GLIB2],[glib-2.0])
+
+if test x$has_glib2 != xno; then
+    SAFE_LIBS="$LIBS"
+    LIBS="$GLIB2_LIBS"
+
+    AC_CHECK_FUNC([g_utf8_validate],
+                  AC_DEFINE([HAVE_G_UTF8_VALIDATE], [1],
+                            [Define if g_utf8_validate exists]))
+    LIBS="$SAFE_LIBS"
+fi
\ No newline at end of file
diff --git a/src/providers/ipa/hbac_evaluator.c b/src/providers/ipa/hbac_evaluator.c
index 476ad6482..a41aa5bb6 100644
--- a/src/providers/ipa/hbac_evaluator.c
+++ b/src/providers/ipa/hbac_evaluator.c
@@ -25,10 +25,9 @@
 
 #include <stdlib.h>
 #include <string.h>
-#include <unistr.h>
-#include <unicase.h>
 #include <errno.h>
 #include "providers/ipa/ipa_hbac.h"
+#include "util/sss_utf8.h"
 
 #ifndef HAVE_ERRNO_T
 #define HAVE_ERRNO_T
@@ -240,7 +239,6 @@ static errno_t hbac_evaluate_element(struct hbac_rule_element *rule_el,
     size_t i, j;
     const uint8_t *rule_name;
     const uint8_t *req_name;
-    int result;
     int ret;
 
     if (rule_el->category & HBAC_CATEGORY_ALL) {
@@ -255,21 +253,11 @@ static errno_t hbac_evaluate_element(struct hbac_rule_element *rule_el,
                 rule_name = (const uint8_t *) rule_el->names[i];
                 req_name = (const uint8_t *) req_el->name;
 
-                /* Do a case-insensitive comparison.
-                 * The input must be encoded in UTF8.
-                 * We have no way of knowing the language,
-                 * so we'll pass NULL for the language and
-                 * hope for the best.
-                 */
-                errno = 0;
-                ret = u8_casecmp(rule_name, u8_strlen(rule_name),
-                                 req_name, u8_strlen(req_name),
-                                 NULL, NULL, &result);
-                if (ret < 0) {
-                    return errno;
-                }
-
-                if (result == 0) {
+                /* Do a case-insensitive comparison. */
+                ret = sss_utf8_case_eq(rule_name, req_name);
+                if (ret != EOK && ret != ENOMATCH) {
+                    return ret;
+                } else if (ret == EOK) {
                     *matched = true;
                     return EOK;
                 }
@@ -287,21 +275,11 @@ static errno_t hbac_evaluate_element(struct hbac_rule_element *rule_el,
             for (j = 0; req_el->groups[j]; j++) {
                 req_name = (const uint8_t *) req_el->groups[j];
 
-                /* Do a case-insensitive comparison.
-                 * The input must be encoded in UTF8.
-                 * We have no way of knowing the language,
-                 * so we'll pass NULL for the language and
-                 * hope for the best.
-                 */
-                errno = 0;
-                ret = u8_casecmp(rule_name, u8_strlen(rule_name),
-                                 req_name, u8_strlen(req_name),
-                                 NULL, NULL, &result);
-                if (ret < 0) {
-                    return errno;
-                }
-
-                if (result == 0) {
+                /* Do a case-insensitive comparison. */
+                ret = sss_utf8_case_eq(rule_name, req_name);
+                if (ret != EOK && ret != ENOMATCH) {
+                    return ret;
+                } else if (ret == EOK) {
                     *matched = true;
                     return EOK;
                 }
diff --git a/src/responder/common/responder_common.c b/src/responder/common/responder_common.c
index 5389031bc..af7a99efd 100644
--- a/src/responder/common/responder_common.c
+++ b/src/responder/common/responder_common.c
@@ -33,9 +33,9 @@
 #include <sys/time.h>
 #include <errno.h>
 #include <popt.h>
-#include <unistr.h>
 #include "config.h"
 #include "util/util.h"
+#include "util/sss_utf8.h"
 #include "db/sysdb.h"
 #include "confdb/confdb.h"
 #include "dbus/dbus.h"
@@ -683,11 +683,3 @@ int sss_dp_get_domain_conn(struct resp_ctx *rctx, const char *domain,
 
     return EOK;
 }
-
-bool sss_utf8_check(const uint8_t *s, size_t n)
-{
-    if (u8_check(s, n) == NULL) {
-        return true;
-    }
-    return false;
-}
diff --git a/src/util/sss_utf8.c b/src/util/sss_utf8.c
new file mode 100644
index 000000000..4a98233bb
--- /dev/null
+++ b/src/util/sss_utf8.c
@@ -0,0 +1,119 @@
+/*
+    SSSD
+
+    Authors:
+        Stephen Gallagher <sgallagh@redhat.com>
+
+    Copyright (C) 2011 Red Hat
+
+    This program is free software; you can redistribute it and/or modify
+    it under the terms of the GNU General Public License as published by
+    the Free Software Foundation; either version 3 of the License, or
+    (at your option) any later version.
+
+    This program is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+    GNU General Public License for more details.
+
+    You should have received a copy of the GNU General Public License
+    along with this program.  If not, see <http://www.gnu.org/licenses/>.
+*/
+
+#include "util/util.h"
+#include "sss_utf8.h"
+
+#ifdef HAVE_LIBUNISTRING
+bool sss_utf8_check(const uint8_t *s, size_t n)
+{
+    if (u8_check(s, n) == NULL) {
+        return true;
+    }
+    return false;
+}
+
+#elif HAVE_GLIB2
+bool sss_utf8_check(const uint8_t *s, size_t n)
+{
+    return g_utf8_validate((const gchar *)s, n, NULL);
+}
+
+#else
+#error No unicode library
+#endif
+
+/* Returns EOK on match, ENOTUNIQ if comparison succeeds but
+ * does not match.
+ * May return other errno error codes on failure
+ */
+#ifdef HAVE_LIBUNISTRING
+errno_t sss_utf8_case_eq(const uint8_t *s1, const uint8_t *s2)
+{
+
+    /* Do a case-insensitive comparison.
+     * The input must be encoded in UTF8.
+     * We have no way of knowing the language,
+     * so we'll pass NULL for the language and
+     * hope for the best.
+     */
+    int ret;
+    int resultp;
+    size_t n1, n2;
+    errno = 0;
+
+    n1 = u8_strlen(s1);
+    n2 = u8_strlen(s2);
+
+    ret = u8_casecmp(s1, n1,
+                     s2, n2,
+                     NULL, NULL,
+                     &resultp);
+    if (ret < 0) {
+        /* An error occurred */
+        return errno;
+    }
+
+    if (resultp == 0) {
+        return EOK;
+    }
+    return ENOMATCH;
+}
+
+#elif HAVE_GLIB2
+errno_t sss_utf8_case_eq(const uint8_t *s1, const uint8_t *s2)
+{
+    gchar *gs1;
+    gchar *gs2;
+    gssize n1, n2;
+    gint gret;
+    errno_t ret;
+
+    n1 = g_utf8_strlen((const gchar *)s1, -1);
+    n2 = g_utf8_strlen((const gchar *)s2, -1);
+
+    gs1 = g_utf8_casefold((const gchar *)s1, n1);
+    if (gs1 == NULL) {
+        return ENOMEM;
+    }
+
+    gs2 = g_utf8_casefold((const gchar *)s2, n2);
+    if (gs2 == NULL) {
+        return ENOMEM;
+    }
+
+    gret = g_utf8_collate(gs1, gs2);
+    if (gret == 0) {
+        ret = EOK;
+    } else {
+        ret = ENOMATCH;
+    }
+
+    g_free(gs1);
+    g_free(gs2);
+
+    return ret;
+}
+
+#else
+#error No unicode library
+#endif
diff --git a/src/util/sss_utf8.h b/src/util/sss_utf8.h
new file mode 100644
index 000000000..37dcff959
--- /dev/null
+++ b/src/util/sss_utf8.h
@@ -0,0 +1,43 @@
+/*
+    SSSD
+
+    Authors:
+        Stephen Gallagher <sgallagh@redhat.com>
+
+    Copyright (C) 2011 Red Hat
+
+    This program is free software; you can redistribute it and/or modify
+    it under the terms of the GNU General Public License as published by
+    the Free Software Foundation; either version 3 of the License, or
+    (at your option) any later version.
+
+    This program is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+    GNU General Public License for more details.
+
+    You should have received a copy of the GNU General Public License
+    along with this program.  If not, see <http://www.gnu.org/licenses/>.
+*/
+
+#ifndef SSS_UTF8_H_
+#define SSS_UTF8_H_
+
+#ifdef HAVE_LIBUNISTRING
+#include <unistr.h>
+#include <unicase.h>
+#elif HAVE_GLIB2
+#include <glib.h>
+#endif
+#include "util/util.h"
+
+#ifndef ENOMATCH
+#define ENOMATCH -1
+#endif
+
+bool sss_utf8_check(const uint8_t *s, size_t n);
+
+errno_t sss_utf8_case_eq(const uint8_t *s1, const uint8_t *s2);
+
+
+#endif /* SSS_UTF8_H_ */