Add the /BLANK subcommand to AUTORECODE

diff --git a/Smake b/Smake
index 4cbe44ece9a3060f9549b2fe7e345591b5f48623..383104dd9d4c485058241a4cc89963fe99a6e47a 100644 (file)
--- a/Smake
+++ b/Smake
@@ -43,6 +43,7 @@ GNULIB_MODULES = \
        inttostr \
        localcharset \
         mbchar \
+        mbiter \
        memcasecmp \
        memchr \
        mempcpy \

diff --git a/doc/transformation.texi b/doc/transformation.texi
index f1d1cbf2617b649c5393b10428325329c696447d..883584b91f794bb4ccd8091b5cbaf6ff17fdc8af 100644 (file)
--- a/doc/transformation.texi
+++ b/doc/transformation.texi
@@ -223,6 +223,7 @@ AUTORECODE VARIABLES=src_vars INTO dest_vars
         [ /DESCENDING ]
         [ /PRINT ]
         [ /GROUP ]
+        [ /BLANK = @{VALID, MISSING@} ]
 @end display
 
 The @cmd{AUTORECODE} procedure considers the @var{n} values that a variable
@@ -247,6 +248,11 @@ The GROUP subcommand is relevant only if more than one variable is to be
 recoded.   It causes a single mapping between source and target values to
 be used, instead of one map per variable.
 
+If /BLANK=MISSING is given, then string variables which contain only 
+whitespace are recoded as SYSMIS.  If /BLANK=VALID is given then they
+will be allocated a value like any other.  /BLANK is not relevant
+to numeric values. /BLANK=VALID is the default.
+
 @cmd{AUTORECODE} is a procedure.  It causes the data to be read.
 
 @node COMPUTE

diff --git a/src/language/stats/autorecode.c b/src/language/stats/autorecode.c
index a4d34ccf608e7c784334d6a23546da546df1f095..33cab462c4d6ab226480722fe31cae218bdda6ff 100644 (file)
--- a/src/language/stats/autorecode.c
+++ b/src/language/stats/autorecode.c
@@ -38,6 +38,7 @@
 
 #include "gl/xalloc.h"
 #include "gl/vasnprintf.h"
+#include "gl/mbiter.h"
 
 #include "gettext.h"
 #define _(msgid) gettext (msgid)
@@ -77,6 +78,9 @@ struct autorecode_pgm
 
   /* Hash table of "struct arc_item"s. */
   struct hmap *global_items;
+
+  bool blank_valid;
+  const struct dictionary *dict;
 };
 
 static trns_proc_func autorecode_trns_proc;
@@ -87,13 +91,35 @@ static void arc_free (struct autorecode_pgm *);
 static struct arc_item *find_arc_item (const struct arc_spec *, const union value *,
                                        size_t hash);
 
+static bool
+value_is_blank (const union value *val, int width, const struct dictionary *dict)
+{
+  mbi_iterator_t iter;
+  const char *str = CHAR_CAST_BUG (const char*, value_str (val, width));
+  char *text = recode_string (UTF8, dict_get_encoding (dict), str, width);
+
+  for (mbi_init (iter, text, width); mbi_avail (iter); mbi_advance (iter))
+    {
+      mbchar_t c = mbi_cur (iter);
+
+      if ( ! mb_isblank (c))
+       {
+         free (text);
+         return false;
+       }
+    }
+
+  free (text);
+  return true;
+}
+
 /* Performs the AUTORECODE procedure. */
 int
 cmd_autorecode (struct lexer *lexer, struct dataset *ds)
 {
   struct autorecode_pgm *arc = NULL;
-  struct dictionary *dict = dataset_dict (ds);
 
+  struct dictionary *dict = dataset_dict (ds);
   const struct variable **src_vars = NULL;
   char **dst_names = NULL;
   size_t n_srcs = 0;
@@ -110,17 +136,19 @@ cmd_autorecode (struct lexer *lexer, struct dataset *ds)
 
   /* Create procedure. */
   arc = xzalloc (sizeof *arc);
+  arc->blank_valid = true;
+  arc->dict = dataset_dict (ds);
 
   /* Parse variable lists. */
   lex_match_id (lexer, "VARIABLES");
   lex_match (lexer, T_EQUALS);
-  if (!parse_variables_const (lexer, dict, &src_vars, &n_srcs,
+  if (!parse_variables_const (lexer, arc->dict, &src_vars, &n_srcs,
                               PV_NO_DUPLICATE))
     goto error;
   if (!lex_force_match_id (lexer, "INTO"))
     goto error;
   lex_match (lexer, T_EQUALS);
-  if (!parse_DATA_LIST_vars (lexer, dict, &dst_names, &n_dsts,
+  if (!parse_DATA_LIST_vars (lexer, arc->dict, &dst_names, &n_dsts,
                              PV_NO_DUPLICATE))
     goto error;
   if (n_dsts != n_srcs)
@@ -135,7 +163,7 @@ cmd_autorecode (struct lexer *lexer, struct dataset *ds)
     {
       const char *name = dst_names[i];
 
-      if (dict_lookup_var (dict, name) != NULL)
+      if (dict_lookup_var (arc->dict, name) != NULL)
         {
           msg (SE, _("Target variable %s duplicates existing variable %s."),
                name, name);
@@ -155,6 +183,22 @@ cmd_autorecode (struct lexer *lexer, struct dataset *ds)
          arc->global_items = xmalloc (sizeof (*arc->global_items));
          hmap_init (arc->global_items);
        }
+      else if (lex_match_id (lexer, "BLANK"))
+       {
+         lex_match (lexer, T_EQUALS);
+         if (lex_match_id (lexer, "VALID"))
+           {
+             arc->blank_valid = true;
+           }
+         else if (lex_match_id (lexer, "MISSING"))
+           {
+             arc->blank_valid = false;
+           }
+         else
+           goto error;
+       }
+      else
+       goto error;
     }
 
   if (lex_token (lexer) != T_ENDCMD)
@@ -166,6 +210,7 @@ cmd_autorecode (struct lexer *lexer, struct dataset *ds)
   arc->specs = xmalloc (n_dsts * sizeof *arc->specs);
   arc->n_specs = n_dsts;
 
+
   for (i = 0; i < n_dsts; i++)
     {
       struct arc_spec *spec = &arc->specs[i];
@@ -195,7 +240,10 @@ cmd_autorecode (struct lexer *lexer, struct dataset *ds)
         struct arc_item *item;
 
         item = find_arc_item (spec, value, hash);
-        if (item == NULL)
+        if ( (item == NULL)
+            &&  
+            ( arc->blank_valid || var_is_numeric (spec->src) || ! value_is_blank (value, width, arc->dict))
+            )
           {
             item = xmalloc (sizeof *item);
            item->width = width;
@@ -251,7 +299,7 @@ cmd_autorecode (struct lexer *lexer, struct dataset *ds)
            {
              const char *str = CHAR_CAST_BUG (const char*, value_str (from, src_width));
 
-             recoded_value = recode_string (UTF8, dict_get_encoding (dict), str, src_width);
+             recoded_value = recode_string (UTF8, dict_get_encoding (arc->dict), str, src_width);
            }
          else
            recoded_value = asnprintf (NULL, &len, "%g", from->f);
@@ -374,9 +422,8 @@ autorecode_trns_proc (void *arc_, struct ccase **c,
       const struct arc_spec *spec = &arc->specs[i];
       int width = var_get_width (spec->src);
       const union value *value = case_data (*c, spec->src);
-      struct arc_item *item;
+      const struct arc_item *item = find_arc_item (spec, value, value_hash (value, width, 0));
 
-      item = find_arc_item (spec, value, value_hash (value, width, 0));
       case_data_rw (*c, spec->dst)->f = item ? item->to : SYSMIS;
     }
 

diff --git a/tests/language/stats/autorecode.at b/tests/language/stats/autorecode.at
index 7f5e6eb376aec3e421bc4f4490a791bb56083d14..c3d8d3cc7745e4ca2396059a46640eec58bb0a3b 100644 (file)
--- a/tests/language/stats/autorecode.at
+++ b/tests/language/stats/autorecode.at
@@ -206,3 +206,34 @@ four        ,nought,2.00,4.00
 ])
 
 AT_CLEANUP
+
+
+
+AT_SETUP([AUTORECODE /blank])
+
+AT_DATA([auto-blank.sps],  [dnl
+data list notable list /x (a8) y * z (a16).
+begin data.
+one   2  fred
+two   4  ""
+""    4  fred
+""    2  charliebrown
+three 2  charliebrown
+end data.
+
+autorecode variables x y z into a b c  /blank=missing.
+
+list a b c y.
+])
+
+AT_CHECK([pspp -O format=csv auto-blank.sps], [0], [dnl
+Table: Data List
+a,b,c,y
+1.00,1.00,2.00,2.00
+3.00,2.00,.  ,4.00
+.  ,2.00,2.00,4.00
+.  ,1.00,1.00,2.00
+2.00,1.00,1.00,2.00
+])
+
+AT_CLEANUP