Mailing-List: contact commits-help@subversion.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@subversion.apache.org
Content-Type: text/plain; charset="utf-8"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Subject: svn commit: r1731300 - in /subversion/trunk/subversion:
 include/private/svn_utf_private.h libsvn_repos/dump.c
 libsvn_subr/utf8proc.c svn/cl-log.h svn/log-cmd.c svn/svn.c
 tests/cmdline/log_tests.py tests/libsvn_subr/utf-test.c
Date: Fri, 19 Feb 2016 22:11:11 -0000
To: commits@subversion.apache.org
From: kotkov@apache.org
Message-Id: <20160219221112.017573A0185@svn01-us-west.apache.org>

Author: kotkov
Date: Fri Feb 19 22:11:11 2016
New Revision: 1731300

URL: http://svn.apache.org/viewvc?rev=1731300&view=rev
Log:
Make svn log --search case-insensitive.

Use utf8proc to do the normalization and locale-independent case folding
(UTF8PROC_CASEFOLD) for both the search pattern and the input strings.

Related discussion is in http://svn.haxx.se/dev/archive-2013-04/0374.shtml
(Subject: "log --search test failures on trunk and 1.8.x").

* subversion/include/private/svn_utf_private.h
  (svn_utf__normalize): Add new boolean argument to perform case folding.

* subversion/libsvn_subr/utf8proc.c
  (normalize_cstring): Add new boolean argument to perform case folding.
   In case it is non-zero, set the flags to UTF8PROC_CASEFOLD when
   doing the Unicode decomposition.
  (svn_utf__normalize): Pass new argument to normalize_cstring().
  (svn_utf__is_normalized): Adjust call to normalize_cstring().

* subversion/libsvn_repos/dump.c
  (extract_mergeinfo_paths, verify_mergeinfo_normalization,
   check_name_collision): Update callers of svn_utf__normalize().

* subversion/svn/cl-log.h
  (): Include svn_string_private for svn_membuf_t.
  (svn_cl__log_receiver_baton): Add an svn_membuf_t for the case folding
   and normalization purposes.

* subversion/svn/log-cmd.c
  (): Include svn_utf_private.h.
  (match): New helper that normalizes, folds case of the input, and matches
   it against the specified pattern.
  (match_search_pattern): Now accepts an svn_membuf_t.  Call the new helper
   function to perform the pattern matching.
  (svn_cl__log_entry_receiver, svn_cl__log_entry_receiver_xml): Pass the
   svn_membuf_t from the baton when calling match_search_pattern().
  (svn_cl__log): Initialize the svn_membuf_t in the log receiver baton.

* subversion/svn/svn.c
  (): Include svn_utf_private.h.
  (sub_main): Normalize and fold case of --search and --search-and arguments.

* subversion/tests/cmdline/log_tests.py
  (log_search): Adjust expectations, since --search is now case-insensitive.

* subversion/tests/libsvn_subr/utf-test.c
  (test_utf_normalize): New test for svn_utf__normalize().
  (test_funcs): Add new test.

Modified:
    subversion/trunk/subversion/include/private/svn_utf_private.h
    subversion/trunk/subversion/libsvn_repos/dump.c
    subversion/trunk/subversion/libsvn_subr/utf8proc.c
    subversion/trunk/subversion/svn/cl-log.h
    subversion/trunk/subversion/svn/log-cmd.c
    subversion/trunk/subversion/svn/svn.c
    subversion/trunk/subversion/tests/cmdline/log_tests.py
    subversion/trunk/subversion/tests/libsvn_subr/utf-test.c

Modified: subversion/trunk/subversion/include/private/svn_utf_private.h
URL: http://svn.apache.org/viewvc/subversion/trunk/subversion/include/private/svn_utf_private.h?rev=1731300&r1=1731299&r2=1731300&view=diff
==============================================================================
--- subversion/trunk/subversion/include/private/svn_utf_private.h (original)
+++ subversion/trunk/subversion/include/private/svn_utf_private.h Fri Feb 19 22:11:11 2016
@@ -139,6 +139,9 @@ svn_utf__normcmp(int *result,
  * null-terminated; otherwise, consider the string only up to the
  * given length.
  *
+ * If CASEFOLD is non-zero, perform Unicode case folding, e.g., for
+ * case-insensitive string comparison.
+ *
  * Return the normalized string in *RESULT, which shares storage with
  * BUF and is valid only until the next time BUF is modified.
  *
@@ -148,6 +151,7 @@ svn_utf__normcmp(int *result,
 svn_error_t*
 svn_utf__normalize(const char **result,
                    const char *str, apr_size_t len,
+                   svn_boolean_t casefold,
                    svn_membuf_t *buf);
 
 /* Check if STRING is a valid, NFC-normalized UTF-8 string.  Note that

Modified: subversion/trunk/subversion/libsvn_repos/dump.c
URL: http://svn.apache.org/viewvc/subversion/trunk/subversion/libsvn_repos/dump.c?rev=1731300&r1=1731299&r2=1731300&view=diff
==============================================================================
--- subversion/trunk/subversion/libsvn_repos/dump.c (original)
+++ subversion/trunk/subversion/libsvn_repos/dump.c Fri Feb 19 22:11:11 2016
@@ -897,7 +897,7 @@ extract_mergeinfo_paths(void *baton, con
   if (xb->normalize)
     {
       const char *normkey;
-      SVN_ERR(svn_utf__normalize(&normkey, key, klen, &xb->buffer));
+      SVN_ERR(svn_utf__normalize(&normkey, key, klen, FALSE, &xb->buffer));
       svn_hash_sets(xb->result,
                     apr_pstrdup(xb->buffer.pool, normkey),
                     normalized_unique);
@@ -951,7 +951,7 @@ verify_mergeinfo_normalization(void *bat
   const char *normpath;
   const char *found;
 
-  SVN_ERR(svn_utf__normalize(&normpath, path, klen, &vb->buffer));
+  SVN_ERR(svn_utf__normalize(&normpath, path, klen, FALSE, &vb->buffer));
   found = svn_hash_gets(vb->normalized_paths, normpath);
   if (!found)
       svn_hash_sets(vb->normalized_paths,
@@ -2233,7 +2233,7 @@ check_name_collision(void *baton, const
   const char *name;
   const char *found;
 
-  SVN_ERR(svn_utf__normalize(&name, key, klen, &cb->buffer));
+  SVN_ERR(svn_utf__normalize(&name, key, klen, FALSE, &cb->buffer));
 
   found = svn_hash_gets(cb->normalized, name);
   if (!found)
@@ -2252,7 +2252,7 @@ check_name_collision(void *baton, const
 
       SVN_ERR(svn_utf__normalize(
                   &normpath, svn_relpath_join(db->path, name, iterpool),
-                  SVN_UTF__UNKNOWN_LENGTH, &cb->buffer));
+                  SVN_UTF__UNKNOWN_LENGTH, FALSE, &cb->buffer));
       notify_warning(iterpool, eb->notify_func, eb->notify_baton,
                      svn_repos_notify_warning_name_collision,
                      _("Duplicate representation of path '%s'"), normpath);

Modified: subversion/trunk/subversion/libsvn_subr/utf8proc.c
URL: http://svn.apache.org/viewvc/subversion/trunk/subversion/libsvn_subr/utf8proc.c?rev=1731300&r1=1731299&r2=1731300&view=diff
==============================================================================
--- subversion/trunk/subversion/libsvn_subr/utf8proc.c (original)
+++ subversion/trunk/subversion/libsvn_subr/utf8proc.c Fri Feb 19 22:11:11 2016
@@ -126,15 +126,20 @@ decompose_normalized(apr_size_t *result_
  * STRING. Upon return, BUFFER->data points at a NUL-terminated string
  * of UTF-8 characters.
  *
+ * If CASEFOLD is non-zero, perform Unicode case folding, e.g., for
+ * case-insensitive string comparison.
+ *
  * A returned error may indicate that STRING contains invalid UTF-8 or
  * invalid Unicode codepoints. Any error message comes from utf8proc.
  */
 static svn_error_t *
 normalize_cstring(apr_size_t *result_length,
                   const char *string, apr_size_t length,
+                  svn_boolean_t casefold,
                   svn_membuf_t *buffer)
 {
-  ssize_t result = unicode_decomposition(0, string, length, buffer);
+  ssize_t result = unicode_decomposition(casefold ? UTF8PROC_CASEFOLD : 0,
+                                         string, length, buffer);
   if (result >= 0)
     {
       svn_membuf__resize(buffer, result * sizeof(apr_int32_t) + 1);
@@ -199,10 +204,11 @@ svn_utf__normcmp(int *result,
 svn_error_t*
 svn_utf__normalize(const char **result,
                    const char *str, apr_size_t len,
+                   svn_boolean_t casefold,
                    svn_membuf_t *buf)
 {
   apr_size_t result_length;
-  SVN_ERR(normalize_cstring(&result_length, str, len, buf));
+  SVN_ERR(normalize_cstring(&result_length, str, len, casefold, buf));
   *result = (const char*)(buf->data);
   return SVN_NO_ERROR;
 }
@@ -359,7 +365,7 @@ svn_utf__is_normalized(const char *strin
   apr_size_t result_length;
   const apr_size_t length = strlen(string);
   svn_membuf__create(&buffer, length * sizeof(apr_int32_t), scratch_pool);
-  err = normalize_cstring(&result_length, string, length, &buffer);
+  err = normalize_cstring(&result_length, string, length, FALSE, &buffer);
   if (err)
     {
       svn_error_clear(err);

Modified: subversion/trunk/subversion/svn/cl-log.h
URL: http://svn.apache.org/viewvc/subversion/trunk/subversion/svn/cl-log.h?rev=1731300&r1=1731299&r2=1731300&view=diff
==============================================================================
--- subversion/trunk/subversion/svn/cl-log.h (original)
+++ subversion/trunk/subversion/svn/cl-log.h Fri Feb 19 22:11:11 2016
@@ -31,6 +31,8 @@
 
 #include "svn_types.h"
 
+#include "private/svn_string_private.h"
+
 #ifdef __cplusplus
 extern "C" {
 #endif /* __cplusplus */
@@ -70,6 +72,9 @@ typedef struct svn_cl__log_receiver_bato
    * the log message, or a changed path matches one of these patterns. */
   apr_array_header_t *search_patterns;
 
+  /* Buffer for Unicode normalization and case folding. */
+  svn_membuf_t buffer;
+
   /* Pool for persistent allocations. */
   apr_pool_t *pool;
 } svn_cl__log_receiver_baton;

Modified: subversion/trunk/subversion/svn/log-cmd.c
URL: http://svn.apache.org/viewvc/subversion/trunk/subversion/svn/log-cmd.c?rev=1731300&r1=1731299&r2=1731300&view=diff
==============================================================================
--- subversion/trunk/subversion/svn/log-cmd.c (original)
+++ subversion/trunk/subversion/svn/log-cmd.c Fri Feb 19 22:11:11 2016
@@ -38,6 +38,7 @@
 
 #include "private/svn_cmdline_private.h"
 #include "private/svn_sorts_private.h"
+#include "private/svn_utf_private.h"
 
 #include "cl.h"
 #include "cl-log.h"
@@ -110,6 +111,24 @@ display_diff(const svn_log_entry_t *log_
   return SVN_NO_ERROR;
 }
 
+/* Return TRUE if STR matches PATTERN. Else, return FALSE. Assumes that
+ * PATTERN is a UTF-8 string normalized to form C with case folding
+ * applied. Use BUF for temporary allocations. */
+static svn_boolean_t
+match(const char *pattern, const char *str, svn_membuf_t *buf)
+{
+  svn_error_t *err;
+
+  err = svn_utf__normalize(&str, str, strlen(str), TRUE, buf);
+  if (err)
+    {
+      /* Can't match invalid data. */
+      svn_error_clear(err);
+      return FALSE;
+    }
+
+  return apr_fnmatch(pattern, str, 0) == APR_SUCCESS;
+}
 
 /* Return TRUE if SEARCH_PATTERN matches the AUTHOR, DATE, LOG_MESSAGE,
  * or a path in the set of keys of the CHANGED_PATHS hash. Else, return FALSE.
@@ -120,22 +139,22 @@ match_search_pattern(const char *search_
                      const char *date,
                      const char *log_message,
                      apr_hash_t *changed_paths,
+                     svn_membuf_t *buf,
                      apr_pool_t *pool)
 {
   /* Match any substring containing the pattern, like UNIX 'grep' does. */
   const char *pattern = apr_psprintf(pool, "*%s*", search_pattern);
-  int flags = 0;
 
   /* Does the author match the search pattern? */
-  if (author && apr_fnmatch(pattern, author, flags) == APR_SUCCESS)
+  if (author && match(pattern, author, buf))
     return TRUE;
 
   /* Does the date the search pattern? */
-  if (date && apr_fnmatch(pattern, date, flags) == APR_SUCCESS)
+  if (date && match(pattern, date, buf))
     return TRUE;
 
   /* Does the log message the search pattern? */
-  if (log_message && apr_fnmatch(pattern, log_message, flags) == APR_SUCCESS)
+  if (log_message && match(pattern, log_message, buf))
     return TRUE;
 
   if (changed_paths)
@@ -150,15 +169,14 @@ match_search_pattern(const char *search_
           const char *path = apr_hash_this_key(hi);
           svn_log_changed_path2_t *log_item;
 
-          if (apr_fnmatch(pattern, path, flags) == APR_SUCCESS)
+          if (match(pattern, path, buf))
             return TRUE;
 
           /* Match copy-from paths, too. */
           log_item = apr_hash_this_val(hi);
           if (log_item->copyfrom_path
               && SVN_IS_VALID_REVNUM(log_item->copyfrom_rev)
-              && apr_fnmatch(pattern,
-                             log_item->copyfrom_path, flags) == APR_SUCCESS)
+              && match(pattern, log_item->copyfrom_path, buf))
             return TRUE;
         }
     }
@@ -168,13 +186,14 @@ match_search_pattern(const char *search_
 
 /* Match all search patterns in SEARCH_PATTERNS against AUTHOR, DATE, MESSAGE,
  * and CHANGED_PATHS. Return TRUE if any pattern matches, else FALSE.
- * SCRACH_POOL is used for temporary allocations. */
+ * BUF and SCRATCH_POOL are used for temporary allocations. */
 static svn_boolean_t
 match_search_patterns(apr_array_header_t *search_patterns,
                       const char *author,
                       const char *date,
                       const char *message,
                       apr_hash_t *changed_paths,
+                      svn_membuf_t *buf,
                       apr_pool_t *scratch_pool)
 {
   int i;
@@ -197,7 +216,7 @@ match_search_patterns(apr_array_header_t
 
           pattern = APR_ARRAY_IDX(pattern_group, j, const char *);
           match = match_search_pattern(pattern, author, date, message,
-                                       changed_paths, iterpool);
+                                       changed_paths, buf, iterpool);
           if (!match)
             break;
         }
@@ -331,7 +350,7 @@ svn_cl__log_entry_receiver(void *baton,
 
   if (lb->search_patterns &&
       ! match_search_patterns(lb->search_patterns, author, date, message,
-                              log_entry->changed_paths2, pool))
+                              log_entry->changed_paths2, &lb->buffer, pool))
     {
       if (log_entry->has_children)
         {
@@ -535,7 +554,7 @@ svn_cl__log_entry_receiver_xml(void *bat
   /* Match search pattern before XML-escaping. */
   if (lb->search_patterns &&
       ! match_search_patterns(lb->search_patterns, author, date, message,
-                              log_entry->changed_paths2, pool))
+                              log_entry->changed_paths2, &lb->buffer, pool))
     {
       if (log_entry->has_children)
         {
@@ -795,6 +814,7 @@ svn_cl__log(apr_getopt_t *os,
   lb.diff_extensions = opt_state->extensions;
   lb.merge_stack = NULL;
   lb.search_patterns = opt_state->search_patterns;
+  svn_membuf__create(&lb.buffer, 0, pool);
   lb.pool = pool;
 
   if (opt_state->xml)

Modified: subversion/trunk/subversion/svn/svn.c
URL: http://svn.apache.org/viewvc/subversion/trunk/subversion/svn/svn.c?rev=1731300&r1=1731299&r2=1731300&view=diff
==============================================================================
--- subversion/trunk/subversion/svn/svn.c (original)
+++ subversion/trunk/subversion/svn/svn.c Fri Feb 19 22:11:11 2016
@@ -56,6 +56,7 @@
 #include "private/svn_opt_private.h"
 #include "private/svn_cmdline_private.h"
 #include "private/svn_subr_private.h"
+#include "private/svn_utf_private.h"
 
 #include "svn_private_config.h"
 
@@ -1868,6 +1869,7 @@ sub_main(int *exit_code, int argc, const
   svn_boolean_t reading_file_from_stdin = FALSE;
   apr_hash_t *changelists;
   apr_hash_t *cfg_hash;
+  svn_membuf_t buf;
 
   received_opts = apr_array_make(pool, SVN_OPT_MAX_OPTIONS, sizeof(int));
 
@@ -1888,6 +1890,9 @@ sub_main(int *exit_code, int argc, const
   /* Init our changelists hash. */
   changelists = apr_hash_make(pool);
 
+  /* Init the temporary buffer. */
+  svn_membuf__create(&buf, 0, pool);
+
   /* Begin processing arguments. */
   opt_state.start_revision.kind = svn_opt_revision_unspecified;
   opt_state.end_revision.kind = svn_opt_revision_unspecified;
@@ -2392,11 +2397,19 @@ sub_main(int *exit_code, int argc, const
         break;
       case opt_search:
         SVN_ERR(svn_utf_cstring_to_utf8(&utf8_opt_arg, opt_arg, pool));
-        add_search_pattern_group(&opt_state, utf8_opt_arg, pool);
+        SVN_ERR(svn_utf__normalize(&utf8_opt_arg, utf8_opt_arg,
+                                   strlen(utf8_opt_arg), TRUE, &buf));
+        add_search_pattern_group(&opt_state,
+                                 apr_pstrdup(pool, utf8_opt_arg),
+                                 pool);
         break;
       case opt_search_and:
         SVN_ERR(svn_utf_cstring_to_utf8(&utf8_opt_arg, opt_arg, pool));
-        add_search_pattern_to_latest_group(&opt_state, utf8_opt_arg, pool);
+        SVN_ERR(svn_utf__normalize(&utf8_opt_arg, utf8_opt_arg,
+                                   strlen(utf8_opt_arg), TRUE, &buf));
+        add_search_pattern_to_latest_group(&opt_state,
+                                           apr_pstrdup(pool, utf8_opt_arg),
+                                           pool);
       case opt_remove_unversioned:
         opt_state.remove_unversioned = TRUE;
         break;

Modified: subversion/trunk/subversion/tests/cmdline/log_tests.py
URL: http://svn.apache.org/viewvc/subversion/trunk/subversion/tests/cmdline/log_tests.py?rev=1731300&r1=1731299&r2=1731300&view=diff
==============================================================================
--- subversion/trunk/subversion/tests/cmdline/log_tests.py (original)
+++ subversion/trunk/subversion/tests/cmdline/log_tests.py Fri Feb 19 22:11:11 2016
@@ -2288,13 +2288,13 @@ def log_search(sbox):
   log_chain = parse_log_output(output)
   check_log_chain(log_chain, [7, 6, 3])
 
-  # search is case-sensitive
+  # search is case-insensitive
   exit_code, output, err = svntest.actions.run_and_verify_svn(
                              None, [], 'log', '--search',
                              'FOR REVISION [367]')
 
   log_chain = parse_log_output(output)
-  check_log_chain(log_chain, [])
+  check_log_chain(log_chain, [7, 6, 3])
 
   # multi-pattern search
   exit_code, output, err = svntest.actions.run_and_verify_svn(

Modified: subversion/trunk/subversion/tests/libsvn_subr/utf-test.c
URL: http://svn.apache.org/viewvc/subversion/trunk/subversion/tests/libsvn_subr/utf-test.c?rev=1731300&r1=1731299&r2=1731300&view=diff
==============================================================================
--- subversion/trunk/subversion/tests/libsvn_subr/utf-test.c (original)
+++ subversion/trunk/subversion/tests/libsvn_subr/utf-test.c Fri Feb 19 22:11:11 2016
@@ -823,6 +823,97 @@ test_utf_conversions(apr_pool_t *pool)
 }
 
 
+static svn_error_t *
+test_utf_normalize(apr_pool_t *pool)
+{
+  /* Normalized: NFC */
+  static const char nfc[] =
+    "\xe1\xb9\xa8"              /* S with dot above and below */
+    "\xc5\xaf"                  /* u with ring */
+    "\xe1\xb8\x87"              /* b with macron below */
+    "\xe1\xb9\xbd"              /* v with tilde */
+    "\xe1\xb8\x9d"              /* e with breve and cedilla */
+    "\xc8\x91"                  /* r with double grave */
+    "\xc5\xa1"                  /* s with caron */
+    "\xe1\xb8\xaf"              /* i with diaeresis and acute */
+    "\xe1\xbb\x9d"              /* o with grave and hook */
+    "\xe1\xb9\x8b";             /* n with circumflex below */
+
+  /* Normalized: NFC, case folded */
+  static const char nfc_casefold[] =
+    "\xe1\xb9\xa9"              /* s with dot above and below */
+    "\xc5\xaf"                  /* u with ring */
+    "\xe1\xb8\x87"              /* b with macron below */
+    "\xe1\xb9\xbd"              /* v with tilde */
+    "\xe1\xb8\x9d"              /* e with breve and cedilla */
+    "\xc8\x91"                  /* r with double grave */
+    "\xc5\xa1"                  /* s with caron */
+    "\xe1\xb8\xaf"              /* i with diaeresis and acute */
+    "\xe1\xbb\x9d"              /* o with grave and hook */
+    "\xe1\xb9\x8b";             /* n with circumflex below */
+
+  /* Normalized: NFD */
+  static const char nfd[] =
+    "S\xcc\xa3\xcc\x87"         /* S with dot above and below */
+    "u\xcc\x8a"                 /* u with ring */
+    "b\xcc\xb1"                 /* b with macron below */
+    "v\xcc\x83"                 /* v with tilde */
+    "e\xcc\xa7\xcc\x86"         /* e with breve and cedilla */
+    "r\xcc\x8f"                 /* r with double grave */
+    "s\xcc\x8c"                 /* s with caron */
+    "i\xcc\x88\xcc\x81"         /* i with diaeresis and acute */
+    "o\xcc\x9b\xcc\x80"         /* o with grave and hook */
+    "n\xcc\xad";                /* n with circumflex below */
+
+  /* Mixed, denormalized */
+  static const char mixup[] =
+    "S\xcc\x87\xcc\xa3"         /* S with dot above and below */
+    "\xc5\xaf"                  /* u with ring */
+    "b\xcc\xb1"                 /* b with macron below */
+    "\xe1\xb9\xbd"              /* v with tilde */
+    "e\xcc\xa7\xcc\x86"         /* e with breve and cedilla */
+    "\xc8\x91"                  /* r with double grave */
+    "s\xcc\x8c"                 /* s with caron */
+    "\xe1\xb8\xaf"              /* i with diaeresis and acute */
+    "o\xcc\x80\xcc\x9b"         /* o with grave and hook */
+    "\xe1\xb9\x8b";             /* n with circumflex below */
+
+  /* Invalid UTF-8 */
+  static const char invalid[] =
+    "\xe1\xb9\xa8"              /* S with dot above and below */
+    "\xc5\xaf"                  /* u with ring */
+    "\xe1\xb8\x87"              /* b with macron below */
+    "\xe1\xb9\xbd"              /* v with tilde */
+    "\xe1\xb8\x9d"              /* e with breve and cedilla */
+    "\xc8\x91"                  /* r with double grave */
+    "\xc5\xa1"                  /* s with caron */
+    "\xe1\xb8\xaf"              /* i with diaeresis and acute */
+    "\xe6"                      /* Invalid byte */
+    "\xe1\xb9\x8b";             /* n with circumflex below */
+
+  const char *result;
+  svn_membuf_t buf;
+
+  svn_membuf__create(&buf, 0, pool);
+  SVN_ERR(svn_utf__normalize(&result, nfd, strlen(nfd), FALSE, &buf));
+  SVN_TEST_STRING_ASSERT(result, nfc);
+  SVN_ERR(svn_utf__normalize(&result, nfd, strlen(nfd), TRUE, &buf));
+  SVN_TEST_STRING_ASSERT(result, nfc_casefold);
+  SVN_ERR(svn_utf__normalize(&result, mixup, strlen(mixup), FALSE, &buf));
+  SVN_TEST_STRING_ASSERT(result, nfc);
+  SVN_ERR(svn_utf__normalize(&result, mixup, strlen(mixup), TRUE, &buf));
+  SVN_TEST_STRING_ASSERT(result, nfc_casefold);
+
+  SVN_TEST_ASSERT_ERROR(svn_utf__normalize(&result, invalid, strlen(invalid),
+                                           FALSE, &buf),
+                        SVN_ERR_UTF8PROC_ERROR);
+  SVN_TEST_ASSERT_ERROR(svn_utf__normalize(&result, invalid, strlen(invalid),
+                                           TRUE, &buf),
+                        SVN_ERR_UTF8PROC_ERROR);
+
+  return SVN_NO_ERROR;
+}
+
 

 /* The test table.  */
 
@@ -849,6 +940,8 @@ static struct svn_test_descriptor_t test
                    "test svn_utf__is_normalized"),
     SVN_TEST_PASS2(test_utf_conversions,
                    "test svn_utf__utf{16,32}_to_utf8"),
+    SVN_TEST_PASS2(test_utf_normalize,
+                   "test svn_utf__normalize"),
     SVN_TEST_NULL
   };