Implement DATASET commands.

[pspp-builds.git] / src / language / data-io / combine-files.c

diff --git a/src/language/data-io/combine-files.c b/src/language/data-io/combine-files.c
index 82c36945a2fd803de74a0ea3239c3918440c4ed3..e09b36ecad9dde80b16e521ef21cfe22c9bbc01a 100644 (file)
--- a/src/language/data-io/combine-files.c
+++ b/src/language/data-io/combine-files.c
@@ -23,9 +23,9 @@
 #include "data/case.h"
 #include "data/casereader.h"
 #include "data/casewriter.h"
+#include "data/dataset.h"
 #include "data/dictionary.h"
 #include "data/format.h"
-#include "data/procedure.h"
 #include "data/subcase.h"
 #include "data/variable.h"
 #include "language/command.h"
@@ -35,7 +35,9 @@
 #include "language/lexer/variable-parser.h"
 #include "language/stats/sort-criteria.h"
 #include "libpspp/assertion.h"
+#include "libpspp/i18n.h"
 #include "libpspp/message.h"
+#include "libpspp/string-array.h"
 #include "libpspp/taint.h"
 #include "math/sort.h"
 
@@ -159,7 +161,7 @@ combine_files (enum comb_command_type command,
 
   proc.files = NULL;
   proc.n_files = 0;
-  proc.dict = dict_create ();
+  proc.dict = dict_create (get_default_encoding ());
   proc.output = NULL;
   proc.matcher = NULL;
   subcase_init_empty (&proc.by_vars);
@@ -205,23 +207,23 @@ combine_files (enum comb_command_type command,
 
       if (lex_match (lexer, T_ASTERISK))
         {
-          if (!proc_has_active_file (ds))
+          if (!dataset_has_source (ds))
             {
-              msg (SE, _("Cannot specify the active file since no active "
-                         "file has been defined."));
+              msg (SE, _("Cannot specify the active dataset since none "
+                         "has been defined."));
               goto error;
             }
 
           if (proc_make_temporary_transformations_permanent (ds))
             msg (SE, _("This command may not be used after TEMPORARY when "
-                       "the active file is an input source.  "
+                       "the active dataset is an input source.  "
                        "Temporary transformations will be made permanent."));
 
           file->dict = dict_clone (dataset_dict (ds));
         }
       else
         {
-          file->handle = fh_parse (lexer, FH_REF_FILE | FH_REF_SCRATCH);
+          file->handle = fh_parse (lexer, FH_REF_FILE, dataset_session (ds));
           if (file->handle == NULL)
             goto error;
 
@@ -302,7 +304,8 @@ combine_files (enum comb_command_type command,
                         msg (SE, _("File %s lacks BY variable %s."),
                              fh_get_name (file->handle), name);
                       else
-                        msg (SE, _("Active file lacks BY variable %s."), name);
+                        msg (SE, _("Active dataset lacks BY variable %s."),
+                             name);
                       ok = false;
                     }
                 }
@@ -465,7 +468,8 @@ combine_files (enum comb_command_type command,
   if (active_file != NULL)
     proc_commit (ds);
 
-  proc_set_active_file (ds, casewriter_make_reader (proc.output), proc.dict);
+  dataset_set_dict (ds, proc.dict);
+  dataset_set_source (ds, casewriter_make_reader (proc.output));
   proc.dict = NULL;
   proc.output = NULL;
 
@@ -491,9 +495,8 @@ static bool
 merge_dictionary (struct dictionary *const m, struct comb_file *f)
 {
   struct dictionary *d = f->dict;
-  const char *d_docs, *m_docs;
+  const struct string_array *d_docs, *m_docs;
   int i;
-  const char *file_encoding;
 
   if (dict_get_label (m) == NULL)
     dict_set_label (m, dict_get_label (d));
@@ -507,17 +510,9 @@ merge_dictionary (struct dictionary *const m, struct comb_file *f)
      The correct thing to do would be to convert to an encoding
      which can cope with all the input files (eg UTF-8).
    */
-  file_encoding = dict_get_encoding (f->dict);
-  if ( file_encoding != NULL)
-    {
-      if ( dict_get_encoding (m) == NULL)
-       dict_set_encoding (m, file_encoding);
-      else if ( 0 != strcmp (file_encoding, dict_get_encoding (m)))
-       {
-         msg (MW,
-              _("Combining files with incompatible encodings. String data may not be represented correctly."));
-       }
-    }
+  if ( 0 != strcmp (dict_get_encoding (f->dict), dict_get_encoding (m)))
+    msg (MW, _("Combining files with incompatible encodings. String data may "
+               "not be represented correctly."));
 
   if (d_docs != NULL)
     {
@@ -525,9 +520,19 @@ merge_dictionary (struct dictionary *const m, struct comb_file *f)
         dict_set_documents (m, d_docs);
       else
         {
-          char *new_docs = xasprintf ("%s%s", m_docs, d_docs);
-          dict_set_documents (m, new_docs);
-          free (new_docs);
+          struct string_array new_docs;
+          size_t i;
+
+          new_docs.n = m_docs->n + d_docs->n;
+          new_docs.strings = xmalloc (new_docs.n * sizeof *new_docs.strings);
+          for (i = 0; i < m_docs->n; i++)
+            new_docs.strings[i] = m_docs->strings[i];
+          for (i = 0; i < d_docs->n; i++)
+            new_docs.strings[m_docs->n + i] = d_docs->strings[i];
+
+          dict_set_documents (m, &new_docs);
+
+          free (new_docs.strings);
         }
     }
 
@@ -577,7 +582,7 @@ merge_dictionary (struct dictionary *const m, struct comb_file *f)
           if (var_has_missing_values (dv) && !var_has_missing_values (mv))
             var_set_missing_values (mv, var_get_missing_values (dv));
           if (var_get_label (dv) && !var_get_label (mv))
-            var_set_label (mv, var_get_label (dv));
+            var_set_label (mv, var_get_label (dv), false);
         }
       else
         mv = dict_clone_var_assert (m, dv);