/* PSPP - a program for statistical analysis.
- Copyright (C) 1997-9, 2000 Free Software Foundation, Inc.
+ Copyright (C) 1997-9, 2000, 2009, 2010 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 <language/dictionary/split-file.h>
#include <language/lexer/lexer.h>
#include <language/lexer/variable-parser.h>
-#include <libpspp/alloc.h>
#include <libpspp/array.h>
#include <libpspp/compiler.h>
-#include <libpspp/magic.h>
#include <libpspp/message.h>
#include <libpspp/assertion.h>
#include <math/moments.h>
-#include <output/manager.h>
-#include <output/table.h>
+#include <output/tab.h>
+
+#include "xalloc.h"
#include "gettext.h"
#define _(msgid) gettext (msgid)
static const struct dsc_statistic_info dsc_info[DSC_N_STATS] =
{
{"MEAN", N_("Mean"), MOMENT_MEAN},
- {"SEMEAN", N_("S E Mean"), MOMENT_VARIANCE},
+ {"SEMEAN", N_("S.E. Mean"), MOMENT_VARIANCE},
{"STDDEV", N_("Std Dev"), MOMENT_VARIANCE},
{"VARIANCE", N_("Variance"), MOMENT_VARIANCE},
{"KURTOSIS", N_("Kurtosis"), MOMENT_KURTOSIS},
- {"SEKURTOSIS", N_("S E Kurt"), MOMENT_NONE},
+ {"SEKURTOSIS", N_("S.E. Kurt"), MOMENT_NONE},
{"SKEWNESS", N_("Skewness"), MOMENT_SKEWNESS},
- {"SESKEWNESS", N_("S E Skew"), MOMENT_NONE},
+ {"SESKEWNESS", N_("S.E. Skew"), MOMENT_NONE},
{"RANGE", N_("Range"), MOMENT_NONE},
{"MINIMUM", N_("Minimum"), MOMENT_NONE},
{"MAXIMUM", N_("Maximum"), MOMENT_NONE},
struct dsc_var
{
const struct variable *v; /* Variable to calculate on. */
- char z_name[LONG_NAME_LEN + 1]; /* Name for z-score variable. */
+ char z_name[VAR_NAME_LEN + 1]; /* Name for z-score variable. */
double valid, missing; /* Valid, missing counts. */
struct moments *moments; /* Moments. */
double min, max; /* Maximum and mimimum values. */
dsc->show_stats = dsc->calc_stats = DEFAULT_STATS;
/* Parse DESCRIPTIVES. */
- while (lex_token (lexer) != '.')
+ while (lex_token (lexer) != T_ENDCMD)
{
if (lex_match_id (lexer, "MISSING"))
{
- lex_match (lexer, '=');
- while (lex_token (lexer) != '.' && lex_token (lexer) != '/')
+ lex_match (lexer, T_EQUALS);
+ while (lex_token (lexer) != T_ENDCMD && lex_token (lexer) != T_SLASH)
{
if (lex_match_id (lexer, "VARIABLE"))
dsc->missing_type = DSC_VARIABLE;
lex_error (lexer, NULL);
goto error;
}
- lex_match (lexer, ',');
+ lex_match (lexer, T_COMMA);
}
}
else if (lex_match_id (lexer, "SAVE"))
save_z_scores = 1;
else if (lex_match_id (lexer, "FORMAT"))
{
- lex_match (lexer, '=');
- while (lex_token (lexer) != '.' && lex_token (lexer) != '/')
+ lex_match (lexer, T_EQUALS);
+ while (lex_token (lexer) != T_ENDCMD && lex_token (lexer) != T_SLASH)
{
if (lex_match_id (lexer, "LABELS"))
dsc->show_var_labels = 1;
lex_error (lexer, NULL);
goto error;
}
- lex_match (lexer, ',');
+ lex_match (lexer, T_COMMA);
}
}
else if (lex_match_id (lexer, "STATISTICS"))
{
- lex_match (lexer, '=');
+ lex_match (lexer, T_EQUALS);
dsc->show_stats = 0;
- while (lex_token (lexer) != '.' && lex_token (lexer) != '/')
+ while (lex_token (lexer) != T_ENDCMD && lex_token (lexer) != T_SLASH)
{
if (lex_match (lexer, T_ALL))
dsc->show_stats |= (1ul << DSC_N_STATS) - 1;
dsc->show_stats |= DEFAULT_STATS;
else
dsc->show_stats |= 1ul << (match_statistic (lexer));
- lex_match (lexer, ',');
+ lex_match (lexer, T_COMMA);
}
if (dsc->show_stats == 0)
dsc->show_stats = DEFAULT_STATS;
}
else if (lex_match_id (lexer, "SORT"))
{
- lex_match (lexer, '=');
+ lex_match (lexer, T_EQUALS);
if (lex_match_id (lexer, "NAME"))
dsc->sort_by_stat = DSC_NAME;
else
if (dsc->sort_by_stat == DSC_NONE )
dsc->sort_by_stat = DSC_MEAN;
}
- if (lex_match (lexer, '('))
+ if (lex_match (lexer, T_LPAREN))
{
if (lex_match_id (lexer, "A"))
dsc->sort_ascending = 1;
dsc->sort_ascending = 0;
else
lex_error (lexer, NULL);
- lex_force_match (lexer, ')');
+ lex_force_match (lexer, T_RPAREN);
}
}
else if (var_cnt == 0)
{
- if (lex_look_ahead (lexer) == '=')
+ if (lex_look_ahead (lexer) == T_EQUALS)
{
lex_match_id (lexer, "VARIABLES");
- lex_match (lexer, '=');
+ lex_match (lexer, T_EQUALS);
}
- while (lex_token (lexer) != '.' && lex_token (lexer) != '/')
+ while (lex_token (lexer) != T_ENDCMD && lex_token (lexer) != T_SLASH)
{
int i;
}
dsc->var_cnt = var_cnt;
- if (lex_match (lexer, '('))
+ if (lex_match (lexer, T_LPAREN))
{
if (lex_token (lexer) != T_ID)
{
lex_error (lexer, NULL);
goto error;
}
- if (try_name (dict, dsc, lex_tokid (lexer)))
+ if (try_name (dict, dsc, lex_tokcstr (lexer)))
{
- strcpy (dsc->vars[dsc->var_cnt - 1].z_name, lex_tokid (lexer));
+ strcpy (dsc->vars[dsc->var_cnt - 1].z_name, lex_tokcstr (lexer));
z_cnt++;
}
else
msg (SE, _("Z-score variable name %s would be"
- " a duplicate variable name."), lex_tokid (lexer));
+ " a duplicate variable name."), lex_tokcstr (lexer));
lex_get (lexer);
- if (!lex_force_match (lexer, ')'))
+ if (!lex_force_match (lexer, T_RPAREN))
goto error;
}
}
goto error;
}
- lex_match (lexer, '/');
+ lex_match (lexer, T_SLASH);
}
if (var_cnt == 0)
{
generate_z_varname (const struct dictionary *dict, struct dsc_proc *dsc, char *z_name,
const char *var_name, int *z_cnt)
{
- char name[LONG_NAME_LEN + 1];
+ char name[VAR_NAME_LEN + 1];
/* Try a name based on the original variable name. */
name[0] = 'Z';
cnt++;
}
- t = tab_create (2, cnt + 1, 0);
+ t = tab_create (2, cnt + 1);
tab_title (t, _("Mapping of variables to corresponding Z-scores."));
- tab_columns (t, SOM_COL_DOWN, 1);
tab_headers (t, 0, 0, 1, 0);
tab_box (t, TAL_1, TAL_1, TAL_0, TAL_1, 0, 0, 1, cnt);
tab_hline (t, TAL_2, 0, 1, 1);
tab_text (t, 0, 0, TAB_CENTER | TAT_TITLE, _("Source"));
tab_text (t, 1, 0, TAB_CENTER | TAT_TITLE, _("Target"));
- tab_dim (t, tab_natural_dimensions);
{
size_t i, y;
(either system or user-missing values that weren't included).
*/
static int
-descriptives_trns_proc (void *trns_, struct ccase * c,
+descriptives_trns_proc (void *trns_, struct ccase **c,
casenumber case_idx UNUSED)
{
struct dsc_trns *t = trns_;
assert(t->vars);
for (vars = t->vars; vars < t->vars + t->var_cnt; vars++)
{
- double score = case_num (c, *vars);
+ double score = case_num (*c, *vars);
if (var_is_num_missing (*vars, score, t->exclude))
{
all_sysmis = 1;
}
}
+ *c = case_unshare (*c);
for (z = t->z_scores; z < t->z_scores + t->z_score_cnt; z++)
{
- double input = case_num (c, z->src_var);
- double *output = &case_data_rw (c, z->z_var)->f;
+ double input = case_num (*c, z->src_var);
+ double *output = &case_data_rw (*c, z->z_var)->f;
if (z->mean == SYSMIS || z->std_dev == SYSMIS || all_sysmis
|| var_is_num_missing (z->src_var, input, t->exclude))
struct dataset *ds)
{
struct casereader *pass1, *pass2;
- struct ccase c;
+ struct ccase *c;
size_t i;
- if (!casereader_peek (group, 0, &c))
+ c = casereader_peek (group, 0);
+ if (c == NULL)
{
casereader_destroy (group);
return;
}
- output_split_file_values (ds, &c);
- case_destroy (&c);
+ output_split_file_values (ds, c);
+ case_unref (c);
group = casereader_create_filter_weight (group, dataset_dict (ds),
NULL, NULL);
dsc->valid = 0.;
/* First pass to handle most of the work. */
- for (; casereader_read (pass1, &c); case_destroy (&c))
+ for (; (c = casereader_read (pass1)) != NULL; case_unref (c))
{
- double weight = dict_get_case_weight (dataset_dict (ds), &c, NULL);
+ double weight = dict_get_case_weight (dataset_dict (ds), c, NULL);
/* Check for missing values. */
- if (listwise_missing (dsc, &c))
+ if (listwise_missing (dsc, c))
{
dsc->missing_listwise += weight;
if (dsc->missing_type == DSC_LISTWISE)
for (i = 0; i < dsc->var_cnt; i++)
{
struct dsc_var *dv = &dsc->vars[i];
- double x = case_num (&c, dv->v);
+ double x = case_num (c, dv->v);
if (var_is_num_missing (dv->v, x, dsc->exclude))
{
/* Second pass for higher-order moments. */
if (dsc->max_moment > MOMENT_MEAN)
{
- for (; casereader_read (pass2, &c); case_destroy (&c))
+ for (; (c = casereader_read (pass2)) != NULL; case_unref (c))
{
- double weight = dict_get_case_weight (dataset_dict (ds), &c, NULL);
+ double weight = dict_get_case_weight (dataset_dict (ds), c, NULL);
/* Check for missing values. */
- if (dsc->missing_type == DSC_LISTWISE && listwise_missing (dsc, &c))
+ if (dsc->missing_type == DSC_LISTWISE && listwise_missing (dsc, c))
continue;
for (i = 0; i < dsc->var_cnt; i++)
{
struct dsc_var *dv = &dsc->vars[i];
- double x = case_num (&c, dv->v);
+ double x = case_num (c, dv->v);
if (var_is_num_missing (dv->v, x, dsc->exclude))
continue;
sort (dsc->vars, dsc->var_cnt, sizeof *dsc->vars,
descriptives_compare_dsc_vars, dsc);
- t = tab_create (nc, dsc->var_cnt + 1, 0);
+ t = tab_create (nc, dsc->var_cnt + 1);
tab_headers (t, 1, 0, 1, 0);
tab_box (t, TAL_1, TAL_1, -1, -1, 0, 0, nc - 1, dsc->var_cnt);
tab_box (t, -1, -1, -1, TAL_1, 1, 0, nc - 1, dsc->var_cnt);
tab_hline (t, TAL_2, 0, nc - 1, 1);
tab_vline (t, TAL_2, 1, 0, dsc->var_cnt);
- tab_dim (t, tab_natural_dimensions);
nc = 0;
tab_text (t, nc++, 0, TAB_LEFT | TAT_TITLE, _("Variable"));
nc = 0;
tab_text (t, nc++, i + 1, TAB_LEFT, var_get_name (dv->v));
- tab_text (t, nc++, i + 1, TAT_PRINTF, "%g", dv->valid);
+ tab_text_format (t, nc++, i + 1, 0, "%g", dv->valid);
if (dsc->format == DSC_SERIAL)
- tab_text (t, nc++, i + 1, TAT_PRINTF, "%g", dv->missing);
+ tab_text_format (t, nc++, i + 1, 0, "%g", dv->missing);
+
for (j = 0; j < DSC_N_STATS; j++)
if (dsc->show_stats & (1ul << j))
- tab_float (t, nc++, i + 1, TAB_NONE, dv->stats[j], 10, 3);
+ tab_double (t, nc++, i + 1, TAB_NONE, dv->stats[j], NULL);
}
tab_title (t, _("Valid cases = %g; cases with missing value(s) = %g."),