From 97fed7964411b7bed4d8ad5bc895966de7e6b2b3 Mon Sep 17 00:00:00 2001
From: Ben Pfaff <blp@cs.stanford.edu>
Date: Sat, 23 Apr 2011 11:59:49 -0700
Subject: [PATCH] short-names: Consider character encoding when making short
 names.

Variable names and short names are always in UTF-8, but the length of
short names needs to be limited to 8 bytes in the dictionary encoding,
not in UTF-8.  This commit fixes that problem.
---
 src/data/short-names.c | 58 ++++++++++++++++--------------------------
 src/data/variable.c    | 12 +++++----
 2 files changed, 29 insertions(+), 41 deletions(-)

diff --git a/src/data/short-names.c b/src/data/short-names.c
index fbf813a3..aa157bae 100644
--- a/src/data/short-names.c
+++ b/src/data/short-names.c
@@ -23,6 +23,7 @@
 #include "data/variable.h"
 #include "libpspp/assertion.h"
 #include "libpspp/compiler.h"
+#include "libpspp/i18n.h"
 #include "libpspp/message.h"
 #include "libpspp/str.h"
 #include "libpspp/stringi-set.h"
@@ -30,37 +31,6 @@
 #include "gettext.h"
 #define _(msgid) gettext (msgid)
 
-/* Sets V's short name to BASE, followed by a suffix of the form
-   _A, _B, _C, ..., _AA, _AB, etc. according to the value of
-   SUFFIX_NUMBER.  Truncates BASE as necessary to fit. */
-static void
-set_var_short_name_suffix (struct variable *v, size_t i,
-                           const char *base, int suffix_number)
-{
-  char suffix[SHORT_NAME_LEN + 1];
-  char short_name[SHORT_NAME_LEN + 1];
-  int len, ofs;
-
-  assert (suffix_number >= 0);
-
-  /* Compose suffix. */
-  suffix[0] = '_';
-  if (!str_format_26adic (suffix_number, &suffix[1], sizeof suffix - 1))
-    msg (SE, _("Variable suffix too large."));
-  len = strlen (suffix);
-
-  /* Append suffix to V's short name. */
-  str_copy_trunc (short_name, sizeof short_name, base);
-  if (strlen (short_name) + len > SHORT_NAME_LEN)
-    ofs = SHORT_NAME_LEN - len;
-  else
-    ofs = strlen (short_name);
-  strcpy (short_name + ofs, suffix);
-
-  /* Set name. */
-  var_set_short_name (v, i, short_name);
-}
-
 static void
 claim_short_name (struct variable *v, size_t i,
                   struct stringi_set *short_names)
@@ -83,13 +53,28 @@ assign_short_name (struct variable *v, size_t i,
 
   for (trial = 0; ; trial++)
     {
+      char suffix[SHORT_NAME_LEN + 1];
+      char *short_name;
+
+      /* Compose suffix. */
       if (trial == 0)
-        var_set_short_name (v, i, var_get_name (v));
+        suffix[0] = '\0';
       else
-        set_var_short_name_suffix (v, i, var_get_name (v), trial);
+        {
+          suffix[0] = '_';
+          str_format_26adic (trial, &suffix[1], sizeof suffix - 1);
+        }
 
-      if (stringi_set_insert (short_names, var_get_short_name (v, i)))
-        break;
+      /* Set name. */
+      short_name = utf8_encoding_concat (var_get_name (v), suffix,
+                                         var_get_encoding (v), SHORT_NAME_LEN);
+      if (stringi_set_insert (short_names, short_name))
+        {
+          var_set_short_name (v, i, short_name);
+          free (short_name);
+          return;
+        }
+      free (short_name);
     }
 }
 
@@ -133,7 +118,8 @@ short_names_assign (struct dictionary *d)
     {
       struct variable *v = dict_get_var (d, i);
       const char *name = var_get_name (v);
-      if (strlen (name) <= SHORT_NAME_LEN)
+      int len = recode_string_len (var_get_encoding (v), "UTF-8", name, -1);
+      if (len <= SHORT_NAME_LEN)
         var_set_short_name (v, 0, name);
     }
 
diff --git a/src/data/variable.c b/src/data/variable.c
index c83c31ad..17a1542c 100644
--- a/src/data/variable.c
+++ b/src/data/variable.c
@@ -782,10 +782,12 @@ var_get_short_name (const struct variable *var, size_t idx)
   return idx < var->short_name_cnt ? var->short_names[idx] : NULL;
 }
 
-/* Sets VAR's short name with the given IDX to SHORT_NAME,
-   truncating it to SHORT_NAME_LEN characters and converting it
-   to uppercase in the process.  Specifying a null pointer for
-   SHORT_NAME clears the specified short name. */
+/* Sets VAR's short name with the given IDX to the UTF-8 string SHORT_NAME.
+   The caller must already have checked that, in the dictionary encoding,
+   SHORT_NAME is no more than SHORT_NAME_LEN bytes long.  The new short name
+   will be converted to uppercase.
+
+   Specifying a null pointer for SHORT_NAME clears the specified short name. */
 void
 var_set_short_name (struct variable *var, size_t idx, const char *short_name)
 {
@@ -811,7 +813,7 @@ var_set_short_name (struct variable *var, size_t idx, const char *short_name)
           for (i = old_cnt; i < var->short_name_cnt; i++)
             var->short_names[i] = NULL;
         }
-      var->short_names[idx] = xstrndup (short_name, MAX_SHORT_STRING);
+      var->short_names[idx] = xstrdup (short_name);
       str_uppercase (var->short_names[idx]);
     }
 
-- 
2.30.2