csv-file-writer: Write all line breaks in system-specific format.

[pspp] / src / language / dictionary / missing-values.c

diff --git a/src/language/dictionary/missing-values.c b/src/language/dictionary/missing-values.c
index bbc2c84d4d89e225cb8540e08662a3ed2b044378..a369c2ea6999a802f00b9d8b723f244d2f4a2ffe 100644 (file)
--- a/src/language/dictionary/missing-values.c
+++ b/src/language/dictionary/missing-values.c
@@ -1,5 +1,5 @@
 /* PSPP - a program for statistical analysis.
-   Copyright (C) 1997-9, 2000, 2006, 2009, 2010 Free Software Foundation, Inc.
+   Copyright (C) 1997-9, 2000, 2006, 2009, 2010, 2011, 2013 Free Software Foundation, Inc.
 
    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
@@ -18,18 +18,20 @@
 
 #include <stdlib.h>
 
-#include <data/data-in.h>
-#include <data/missing-values.h>
-#include <data/procedure.h>
-#include <data/value.h>
-#include <data/variable.h>
-#include <data/format.h>
-#include <language/command.h>
-#include <language/lexer/lexer.h>
-#include <language/lexer/value-parser.h>
-#include <language/lexer/variable-parser.h>
-#include <libpspp/message.h>
-#include <libpspp/str.h>
+#include "data/data-in.h"
+#include "data/dictionary.h"
+#include "data/dataset.h"
+#include "data/format.h"
+#include "data/missing-values.h"
+#include "data/value.h"
+#include "data/variable.h"
+#include "language/command.h"
+#include "language/lexer/lexer.h"
+#include "language/lexer/value-parser.h"
+#include "language/lexer/variable-parser.h"
+#include "libpspp/i18n.h"
+#include "libpspp/message.h"
+#include "libpspp/str.h"
 
 #include "gettext.h"
 #define _(msgid) gettext (msgid)
@@ -37,21 +39,21 @@
 int
 cmd_missing_values (struct lexer *lexer, struct dataset *ds)
 {
+  struct dictionary *dict = dataset_dict (ds);
   struct variable **v = NULL;
   size_t nv;
 
-  int retval = CMD_FAILURE;
-  bool deferred_errors = false;
+  bool ok = true;
 
   while (lex_token (lexer) != T_ENDCMD)
     {
       size_t i;
 
-      if (!parse_variables (lexer, dataset_dict (ds), &v, &nv, PV_NONE))
-        goto done;
+      if (!parse_variables (lexer, dict, &v, &nv, PV_NONE))
+        goto error;
 
       if (!lex_force_match (lexer, T_LPAREN))
-        goto done;
+        goto error;
 
       for (i = 0; i < nv; i++)
         var_clear_missing_values (v[i]);
@@ -68,7 +70,7 @@ cmd_missing_values (struct lexer *lexer, struct dataset *ds)
                 msg (SE, _("Cannot mix numeric variables (e.g. %s) and "
                            "string variables (e.g. %s) within a single list."),
                      var_get_name (n), var_get_name (s));
-                goto done;
+                goto error;
               }
 
           if (var_is_numeric (v[0]))
@@ -81,44 +83,54 @@ cmd_missing_values (struct lexer *lexer, struct dataset *ds)
                   bool ok;
 
                   if (!parse_num_range (lexer, &x, &y, &type))
-                    goto done;
+                    goto error;
 
                   ok = (x == y
                         ? mv_add_num (&mv, x)
                         : mv_add_range (&mv, x, y));
                   if (!ok)
-                    deferred_errors = true;
+                    ok = false;
 
                   lex_match (lexer, T_COMMA);
                 }
             }
           else
             {
+              const char *encoding = dict_get_encoding (dict);
+
               mv_init (&mv, MV_MAX_STRING);
               while (!lex_match (lexer, T_RPAREN))
                 {
-                  uint8_t value[MV_MAX_STRING];
-                  size_t length;
+                  const char *utf8_s;
+                  size_t utf8_trunc_len;
+                  size_t utf8_len;
+
+                  char *raw_s;
 
                   if (!lex_force_string (lexer))
                     {
-                      deferred_errors = true;
+                      ok = false;
                       break;
                     }
 
-                  length = ds_length (lex_tokstr (lexer));
-                  if (length > MV_MAX_STRING)
-                    {
-                      msg (SE, _("Truncating missing value to maximum "
-                                 "acceptable length (%d bytes)."),
-                           MV_MAX_STRING);
-                      length = MV_MAX_STRING;
-                    }
-                  memset (value, ' ', MV_MAX_STRING);
-                  memcpy (value, ds_data (lex_tokstr (lexer)), length);
-
-                  if (!mv_add_str (&mv, value))
-                    deferred_errors = true;
+                  /* Truncate the string to fit in 8 bytes in the dictionary
+                     encoding. */
+                  utf8_s = lex_tokcstr (lexer);
+                  utf8_len = ss_length (lex_tokss (lexer));
+                  utf8_trunc_len = utf8_encoding_trunc_len (utf8_s, encoding,
+                                                            MV_MAX_STRING);
+                  if (utf8_trunc_len < utf8_len)
+                    msg (SE, _("Truncating missing value to maximum "
+                               "acceptable length (%d bytes)."),
+                         MV_MAX_STRING);
+
+                  /* Recode to dictionary encoding and add. */
+                  raw_s = recode_string (encoding, "UTF-8",
+                                         utf8_s, utf8_trunc_len);
+                  if (!mv_add_str (&mv, CHAR_CAST (const uint8_t *, raw_s),
+                                   strlen (raw_s)))
+                    ok = false;
+                  free (raw_s);
 
                   lex_get (lexer);
                   lex_match (lexer, T_COMMA);
@@ -134,7 +146,7 @@ cmd_missing_values (struct lexer *lexer, struct dataset *ds)
                   msg (SE, _("Missing values provided are too long to assign "
                              "to variable of width %d."),
                        var_get_width (v[i]));
-                  deferred_errors = true;
+                  ok = false;
                 }
             }
 
@@ -145,12 +157,12 @@ cmd_missing_values (struct lexer *lexer, struct dataset *ds)
       free (v);
       v = NULL;
     }
-  retval = lex_end_of_command (lexer);
 
- done:
   free (v);
-  if (deferred_errors)
-    retval = CMD_FAILURE;
-  return retval;
+  return ok ? CMD_SUCCESS : CMD_FAILURE;
+
+error:
+  free (v);
+  return CMD_FAILURE;
 }