/* PSPP - a program for statistical analysis.
- Copyright (C) 1997-9, 2000, 2006, 2008, 2009, 2010 Free Software Foundation, Inc.
+ Copyright (C) 1997-9, 2000, 2006, 2008, 2009, 2010, 2011, 2012, 2014 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
#include <config.h>
+#include "language/stats/aggregate.h"
+
#include <stdlib.h>
-#include <data/any-writer.h>
-#include <data/case.h>
-#include <data/casegrouper.h>
-#include <data/casereader.h>
-#include <data/casewriter.h>
-#include <data/dictionary.h>
-#include <data/file-handle-def.h>
-#include <data/format.h>
-#include <data/procedure.h>
-#include <data/settings.h>
-#include <data/subcase.h>
-#include <data/sys-file-writer.h>
-#include <data/variable.h>
-#include <language/command.h>
-#include <language/data-io/file-handle.h>
-#include <language/lexer/lexer.h>
-#include <language/lexer/variable-parser.h>
-#include <language/stats/sort-criteria.h>
-#include <libpspp/assertion.h>
-#include <libpspp/message.h>
-#include <libpspp/misc.h>
-#include <libpspp/pool.h>
-#include <libpspp/str.h>
-#include <math/moments.h>
-#include <math/sort.h>
-#include <math/statistic.h>
-#include <math/percentiles.h>
-
-#include "aggregate.h"
-
-#include "minmax.h"
-#include "xalloc.h"
+#include "data/any-writer.h"
+#include "data/case.h"
+#include "data/casegrouper.h"
+#include "data/casereader.h"
+#include "data/casewriter.h"
+#include "data/dataset.h"
+#include "data/dictionary.h"
+#include "data/file-handle-def.h"
+#include "data/format.h"
+#include "data/settings.h"
+#include "data/subcase.h"
+#include "data/sys-file-writer.h"
+#include "data/variable.h"
+#include "language/command.h"
+#include "language/data-io/file-handle.h"
+#include "language/lexer/lexer.h"
+#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/misc.h"
+#include "libpspp/pool.h"
+#include "libpspp/str.h"
+#include "math/moments.h"
+#include "math/percentiles.h"
+#include "math/sort.h"
+#include "math/statistic.h"
+
+#include "gl/c-strcase.h"
+#include "gl/minmax.h"
+#include "gl/xalloc.h"
#include "gettext.h"
#define _(msgid) gettext (msgid)
/* Attributes of aggregation functions. */
const struct agr_func agr_func_tab[] =
{
- {"SUM", N_("Sum of values"), AGR_SV_YES, 0, -1, {FMT_F, 8, 2}},
- {"MEAN", N_("Mean average"), AGR_SV_YES, 0, -1, {FMT_F, 8, 2}},
- {"MEDIAN", N_("Median average"), AGR_SV_YES, 0, -1, {FMT_F, 8, 2}},
- {"SD", N_("Standard deviation"), AGR_SV_YES, 0, -1, {FMT_F, 8, 2}},
- {"MAX", N_("Maximum value"), AGR_SV_YES, 0, VAL_STRING, {-1, -1, -1}},
- {"MIN", N_("Minimum value"), AGR_SV_YES, 0, VAL_STRING, {-1, -1, -1}},
- {"PGT", N_("Percentage greater than"), AGR_SV_YES, 1, VAL_NUMERIC, {FMT_F, 5, 1}},
- {"PLT", N_("Percentage less than"), AGR_SV_YES, 1, VAL_NUMERIC, {FMT_F, 5, 1}},
- {"PIN", N_("Percentage included in range"), AGR_SV_YES, 2, VAL_NUMERIC, {FMT_F, 5, 1}},
- {"POUT", N_("Percentage excluded from range"), AGR_SV_YES, 2, VAL_NUMERIC, {FMT_F, 5, 1}},
- {"FGT", N_("Fraction greater than"), AGR_SV_YES, 1, VAL_NUMERIC, {FMT_F, 5, 3}},
- {"FLT", N_("Fraction less than"), AGR_SV_YES, 1, VAL_NUMERIC, {FMT_F, 5, 3}},
- {"FIN", N_("Fraction included in range"), AGR_SV_YES, 2, VAL_NUMERIC, {FMT_F, 5, 3}},
- {"FOUT", N_("Fraction excluded from range"), AGR_SV_YES, 2, VAL_NUMERIC, {FMT_F, 5, 3}},
- {"N", N_("Number of cases"), AGR_SV_NO, 0, VAL_NUMERIC, {FMT_F, 7, 0}},
- {"NU", N_("Number of cases (unweighted)"), AGR_SV_OPT, 0, VAL_NUMERIC, {FMT_F, 7, 0}},
- {"NMISS", N_("Number of missing values"), AGR_SV_YES, 0, VAL_NUMERIC, {FMT_F, 7, 0}},
+ {"SUM", N_("Sum of values"), AGR_SV_YES, 0, -1, {FMT_F, 8, 2}},
+ {"MEAN", N_("Mean average"), AGR_SV_YES, 0, -1, {FMT_F, 8, 2}},
+ {"MEDIAN", N_("Median average"), AGR_SV_YES, 0, -1, {FMT_F, 8, 2}},
+ {"SD", N_("Standard deviation"), AGR_SV_YES, 0, -1, {FMT_F, 8, 2}},
+ {"MAX", N_("Maximum value"), AGR_SV_YES, 0, VAL_STRING, {-1, -1, -1}},
+ {"MIN", N_("Minimum value"), AGR_SV_YES, 0, VAL_STRING, {-1, -1, -1}},
+ {"PGT", N_("Percentage greater than"), AGR_SV_YES, 1, VAL_NUMERIC, {FMT_F, 5, 1}},
+ {"PLT", N_("Percentage less than"), AGR_SV_YES, 1, VAL_NUMERIC, {FMT_F, 5, 1}},
+ {"PIN", N_("Percentage included in range"), AGR_SV_YES, 2, VAL_NUMERIC, {FMT_F, 5, 1}},
+ {"POUT", N_("Percentage excluded from range"), AGR_SV_YES, 2, VAL_NUMERIC, {FMT_F, 5, 1}},
+ {"FGT", N_("Fraction greater than"), AGR_SV_YES, 1, VAL_NUMERIC, {FMT_F, 5, 3}},
+ {"FLT", N_("Fraction less than"), AGR_SV_YES, 1, VAL_NUMERIC, {FMT_F, 5, 3}},
+ {"FIN", N_("Fraction included in range"), AGR_SV_YES, 2, VAL_NUMERIC, {FMT_F, 5, 3}},
+ {"FOUT", N_("Fraction excluded from range"), AGR_SV_YES, 2, VAL_NUMERIC, {FMT_F, 5, 3}},
+ {"N", N_("Number of cases"), AGR_SV_NO, 0, VAL_NUMERIC, {FMT_F, 7, 0}},
+ {"NU", N_("Number of cases (unweighted)"), AGR_SV_OPT, 0, VAL_NUMERIC, {FMT_F, 7, 0}},
+ {"NMISS", N_("Number of missing values"), AGR_SV_YES, 0, VAL_NUMERIC, {FMT_F, 7, 0}},
{"NUMISS", N_("Number of missing values (unweighted)"), AGR_SV_YES, 0, VAL_NUMERIC, {FMT_F, 7, 0}},
- {"FIRST", N_("First non-missing value"), AGR_SV_YES, 0, VAL_STRING, {-1, -1, -1}},
- {"LAST", N_("Last non-missing value"), AGR_SV_YES, 0, VAL_STRING, {-1, -1, -1}},
- {NULL, NULL, AGR_SV_NO, 0, -1, {-1, -1, -1}},
+ {"FIRST", N_("First non-missing value"), AGR_SV_YES, 0, VAL_STRING, {-1, -1, -1}},
+ {"LAST", N_("Last non-missing value"), AGR_SV_YES, 0, VAL_STRING, {-1, -1, -1}},
+ {NULL, NULL, AGR_SV_NO, 0, -1, {-1, -1, -1}},
};
/* Missing value types. */
lex_match (lexer, T_EQUALS);
if (!lex_match (lexer, T_ASTERISK))
{
- out_file = fh_parse (lexer, FH_REF_FILE | FH_REF_SCRATCH);
+ out_file = fh_parse (lexer, FH_REF_FILE, dataset_session (ds));
if (out_file == NULL)
goto error;
}
if ( agr.add_variables )
agr.dict = dict_clone (dict);
else
- agr.dict = dict_create ();
+ agr.dict = dict_create (dict_get_encoding (dict));
dict_set_label (agr.dict, dict_get_label (dict));
dict_set_documents (agr.dict, dict_get_documents (dict));
lex_match (lexer, T_EQUALS);
if (!lex_match_id (lexer, "COLUMNWISE"))
{
- lex_error (lexer, _("expecting %s"), "COLUMNWISE");
+ lex_error_expecting (lexer, "COLUMNWISE");
goto error;
}
agr.missing = COLUMNWISE;
if (out_file == NULL)
{
- /* The active file will be replaced by the aggregated data,
+ /* The active dataset will be replaced by the aggregated data,
so TEMPORARY is moot. */
proc_cancel_temporary_transformations (ds);
proc_discard_output (ds);
else
{
dump_aggregate_info (&agr, output, c);
- case_unref (c);
}
+ case_unref (c);
}
if (!casegrouper_destroy (grouper))
goto error;
if (next_input == NULL)
goto error;
- proc_set_active_file (ds, next_input, agr.dict);
+ dataset_set_dict (ds, agr.dict);
+ dataset_set_source (ds, next_input);
agr.dict = NULL;
}
else
{
size_t n_dest_prev = n_dest;
- if (!parse_DATA_LIST_vars (lexer, &dest, &n_dest,
+ if (!parse_DATA_LIST_vars (lexer, dict, &dest, &n_dest,
(PV_APPEND | PV_SINGLE | PV_NO_SCRATCH
| PV_NO_DUPLICATE)))
goto error;
if (lex_is_string (lexer))
{
- struct string label;
- ds_init_string (&label, lex_tokstr (lexer));
-
- ds_truncate (&label, 255);
- dest_label[n_dest - 1] = ds_xstrdup (&label);
+ dest_label[n_dest - 1] = xstrdup (lex_tokcstr (lexer));
lex_get (lexer);
- ds_destroy (&label);
}
}
goto error;
}
- ds_assign_string (&function_name, lex_tokstr (lexer));
- exclude = ds_chomp (&function_name, '.') ? MV_SYSTEM : MV_ANY;
+ ds_assign_substring (&function_name, lex_tokss (lexer));
+ exclude = ds_chomp_byte (&function_name, '.') ? MV_SYSTEM : MV_ANY;
for (function = agr_func_tab; function->name; function++)
- if (!strcasecmp (function->name, ds_cstr (&function_name)))
+ if (!c_strcasecmp (function->name, ds_cstr (&function_name)))
break;
if (NULL == function->name)
{
{
if (function->src_vars == AGR_SV_YES)
{
- lex_force_match (lexer, T_LPAREN);
goto error;
}
}
lex_match (lexer, T_COMMA);
if (lex_is_string (lexer))
{
- arg[i].c = ds_xstrdup (lex_tokstr (lexer));
+ arg[i].c = recode_string (dict_get_encoding (agr->dict),
+ "UTF-8", lex_tokcstr (lexer),
+ -1);
type = VAL_STRING;
}
else if (lex_is_number (lexer))
free (iter);
}
if (agr->dict != NULL)
- dict_destroy (agr->dict);
+ dict_unref (agr->dict);
}
\f
/* Execution. */
iter->int1 = 1;
break;
case MAX | FSTRING:
- if (memcmp (iter->string, value_str (v, src_width), src_width) < 0)
- memcpy (iter->string, value_str (v, src_width), src_width);
+ /* Need to do some kind of Unicode collation thingy here */
+ if (memcmp (iter->string, v->s, src_width) < 0)
+ memcpy (iter->string, v->s, src_width);
iter->int1 = 1;
break;
case MIN:
iter->int1 = 1;
break;
case MIN | FSTRING:
- if (memcmp (iter->string, value_str (v, src_width), src_width) > 0)
- memcpy (iter->string, value_str (v, src_width), src_width);
+ if (memcmp (iter->string, v->s, src_width) > 0)
+ memcpy (iter->string, v->s, src_width);
iter->int1 = 1;
break;
case FGT:
break;
case FGT | FSTRING:
case PGT | FSTRING:
- if (memcmp (iter->arg[0].c,
- value_str (v, src_width), src_width) < 0)
+ if (memcmp (iter->arg[0].c, v->s, src_width) < 0)
iter->dbl[0] += weight;
iter->dbl[1] += weight;
break;
break;
case FLT | FSTRING:
case PLT | FSTRING:
- if (memcmp (iter->arg[0].c,
- value_str (v, src_width), src_width) > 0)
+ if (memcmp (iter->arg[0].c, v->s, src_width) > 0)
iter->dbl[0] += weight;
iter->dbl[1] += weight;
break;
break;
case FIN | FSTRING:
case PIN | FSTRING:
- if (memcmp (iter->arg[0].c,
- value_str (v, src_width), src_width) <= 0
- && memcmp (iter->arg[1].c,
- value_str (v, src_width), src_width) >= 0)
+ if (memcmp (iter->arg[0].c, v->s, src_width) <= 0
+ && memcmp (iter->arg[1].c, v->s, src_width) >= 0)
iter->dbl[0] += weight;
iter->dbl[1] += weight;
break;
break;
case FOUT | FSTRING:
case POUT | FSTRING:
- if (memcmp (iter->arg[0].c,
- value_str (v, src_width), src_width) > 0
- || memcmp (iter->arg[1].c,
- value_str (v, src_width), src_width) < 0)
+ if (memcmp (iter->arg[0].c, v->s, src_width) > 0
+ || memcmp (iter->arg[1].c, v->s, src_width) < 0)
iter->dbl[0] += weight;
iter->dbl[1] += weight;
break;
case FIRST | FSTRING:
if (iter->int1 == 0)
{
- memcpy (iter->string, value_str (v, src_width), src_width);
+ memcpy (iter->string, v->s, src_width);
iter->int1 = 1;
}
break;
iter->int1 = 1;
break;
case LAST | FSTRING:
- memcpy (iter->string, value_str (v, src_width), src_width);
+ memcpy (iter->string, v->s, src_width);
iter->int1 = 1;
break;
case NMISS:
case MAX | FSTRING:
case MIN | FSTRING:
if (i->int1)
- memcpy (value_str_rw (v, width), i->string, width);
+ memcpy (v->s, i->string, width);
else
value_set_missing (v, width);
break;
case FIRST | FSTRING:
case LAST | FSTRING:
if (i->int1)
- memcpy (value_str_rw (v, width), i->string, width);
+ memcpy (v->s, i->string, width);
else
value_set_missing (v, width);
break;