#include <setjmp.h>
#include <stdlib.h>
+#include <libpspp/i18n.h>
#include <libpspp/assertion.h>
#include <libpspp/message.h>
#include <libpspp/compiler.h>
#include <libpspp/hash.h>
#include <libpspp/array.h>
+#include <data/attributes.h>
#include <data/case.h>
#include <data/casereader-provider.h>
#include <data/casereader.h>
struct fh_lock *lock; /* Mutual exclusion for file handle. */
FILE *file; /* File stream. */
bool error; /* I/O or corruption error? */
- size_t value_cnt; /* Number of "union value"s in struct case. */
+ struct caseproto *proto; /* Format of output cases. */
/* File format. */
enum integer_format integer_format; /* On-disk integer format. */
struct variable **,
int value_idx);
+static void sys_msg (struct sfm_reader *r, int class,
+ const char *format, va_list args)
+ PRINTF_FORMAT (3, 0);
static void sys_warn (struct sfm_reader *, const char *, ...)
PRINTF_FORMAT (2, 3);
-
static void sys_error (struct sfm_reader *, const char *, ...)
PRINTF_FORMAT (2, 3)
NO_RETURN;
static void read_string (struct sfm_reader *, char *, size_t);
static void skip_bytes (struct sfm_reader *, size_t);
-static struct variable_to_value_map *open_variable_to_value_map (
- struct sfm_reader *, size_t size);
-static void close_variable_to_value_map (struct sfm_reader *r,
- struct variable_to_value_map *);
-static bool read_variable_to_value_map (struct sfm_reader *,
- struct dictionary *,
- struct variable_to_value_map *,
- struct variable **var, char **value,
- int *warning_cnt);
+static struct text_record *open_text_record (struct sfm_reader *, size_t size);
+static void close_text_record (struct sfm_reader *r,
+ struct text_record *);
+static bool read_variable_to_value_pair (struct sfm_reader *,
+ struct dictionary *,
+ struct text_record *,
+ struct variable **var, char **value);
+static void text_warn (struct sfm_reader *r, struct text_record *text,
+ const char *format, ...)
+ PRINTF_FORMAT (3, 4);
+static char *text_get_token (struct text_record *,
+ struct substring delimiters);
+static bool text_match (struct text_record *, char c);
+static bool text_read_short_name (struct sfm_reader *, struct dictionary *,
+ struct text_record *,
+ struct substring delimiters,
+ struct variable **);
static bool close_reader (struct sfm_reader *r);
\f
struct sfm_read_info *);
static void read_machine_integer_info (struct sfm_reader *,
size_t size, size_t count,
- struct sfm_read_info *);
+ struct sfm_read_info *,
+ struct dictionary *
+ );
static void read_machine_float_info (struct sfm_reader *,
size_t size, size_t count);
static void read_display_parameters (struct sfm_reader *,
static void read_long_string_map (struct sfm_reader *,
size_t size, size_t count,
struct dictionary *);
+static void read_data_file_attributes (struct sfm_reader *,
+ size_t size, size_t count,
+ struct dictionary *);
+static void read_variable_attributes (struct sfm_reader *,
+ size_t size, size_t count,
+ struct dictionary *);
+static void read_long_string_value_labels (struct sfm_reader *,
+ size_t size, size_t count,
+ struct dictionary *);
+
+/* Convert all the strings in DICT from the dict encoding to UTF8 */
+static void
+recode_strings (struct dictionary *dict)
+{
+ int i;
+
+ const char *enc = dict_get_encoding (dict);
+
+ if ( NULL == enc)
+ enc = get_default_encoding ();
+
+ for (i = 0 ; i < dict_get_var_cnt (dict); ++i)
+ {
+ /* Convert the long variable name */
+ struct variable *var = dict_get_var (dict, i);
+ const char *native_name = var_get_name (var);
+ char *utf8_name = recode_string (UTF8, enc, native_name, -1);
+ if ( 0 != strcmp (utf8_name, native_name))
+ {
+ if ( NULL == dict_lookup_var (dict, utf8_name))
+ dict_rename_var (dict, var, utf8_name);
+ else
+ msg (MW,
+ _("Recoded variable name duplicates an existing `%s' within system file."), utf8_name);
+ }
+
+ free (utf8_name);
+
+ /* Convert the variable label */
+ if (var_has_label (var))
+ {
+ char *utf8_label = recode_string (UTF8, enc, var_get_label (var), -1);
+ var_set_label (var, utf8_label);
+ free (utf8_label);
+ }
+
+ if (var_has_value_labels (var))
+ {
+ const struct val_lab *vl = NULL;
+ const struct val_labs *vlabs = var_get_value_labels (var);
+
+ for (vl = val_labs_first (vlabs); vl != NULL; vl = val_labs_next (vlabs, vl))
+ {
+ const union value *val = val_lab_get_value (vl);
+ const char *label = val_lab_get_label (vl);
+ char *new_label = NULL;
+ new_label = recode_string (UTF8, enc, label, -1);
+
+ var_replace_value_label (var, val, new_label);
+ free (new_label);
+ }
+ }
+ }
+}
/* Opens the system file designated by file handle FH for
reading. Reads the system file's dictionary into *DICT.
r->has_long_var_names = true;
}
+ recode_strings (*dict);
+
/* Read record 999 data, which is just filler. */
read_int (r);
dictionary and may destroy or modify its variables. */
sfm_dictionary_to_sfm_vars (*dict, &r->sfm_vars, &r->sfm_var_cnt);
pool_register (r->pool, free, r->sfm_vars);
+ r->proto = caseproto_ref_pool (dict_get_proto (*dict), r->pool);
pool_free (r->pool, var_by_value_idx);
- r->value_cnt = dict_get_next_value_idx (*dict);
return casereader_create_sequential
- (NULL, r->value_cnt,
+ (NULL, r->proto,
r->case_cnt == -1 ? CASENUMBER_MAX: r->case_cnt,
&sys_file_casereader_class, r);
read_bytes (r, raw_bias, sizeof raw_bias);
if (float_identify (100.0, raw_bias, sizeof raw_bias, &r->float_format) == 0)
{
- sys_warn (r, _("Compression bias is not the usual "
- "value of 100, or system file uses unrecognized "
- "floating-point format."));
+ uint8_t zero_bias[8] = { 0, 0, 0, 0, 0, 0, 0, 0 };
+
+ if (memcmp (raw_bias, zero_bias, 8))
+ sys_warn (r, _("Compression bias is not the usual "
+ "value of 100, or system file uses unrecognized "
+ "floating-point format."));
+ else
+ {
+ /* Some software is known to write all-zeros to this
+ field. Such software also writes floating-point
+ numbers in the format that we expect by default
+ (it seems that all software most likely does, in
+ reality), so don't warn in this case. */
+ }
+
if (r->integer_format == INTEGER_MSB_FIRST)
r->float_format = FLOAT_IEEE_DOUBLE_BE;
else
/* Create variable. */
if (width < 0 || width > 255)
- sys_error (r, _("Bad variable width %d."), width);
+ sys_error (r, _("Bad width %d for variable %s."), width, name);
var = dict_create_var (dict, name, width);
if (var == NULL)
sys_error (r,
struct missing_values mv;
int i;
- mv_init (&mv, var_get_width (var));
+ mv_init_pool (r->pool, &mv, var_get_width (var));
if (var_is_numeric (var))
{
if (missing_value_code < -3 || missing_value_code > 3
}
else
{
+ int mv_width = MAX (width, 8);
+ union value value;
+
if (missing_value_code < 1 || missing_value_code > 3)
sys_error (r, _("String missing value indicator field is not "
"0, 1, 2, or 3."));
- if (var_is_long_string (var))
- sys_warn (r, _("Ignoring missing values on long string variable "
- "%s, which PSPP does not yet support."), name);
+
+ value_init (&value, mv_width);
+ value_set_missing (&value, mv_width);
for (i = 0; i < missing_value_code; i++)
{
- char string[9];
- read_string (r, string, sizeof string);
- mv_add_str (&mv, string);
+ uint8_t *s = value_str_rw (&value, mv_width);
+ read_bytes (r, s, 8);
+ mv_add_str (&mv, s);
}
+ value_destroy (&value, mv_width);
}
- if (!var_is_long_string (var))
- var_set_missing_values (var, &mv);
+ var_set_missing_values (var, &mv);
}
/* Set formats. */
switch (subtype)
{
case 3:
- read_machine_integer_info (r, size, count, info);
+ read_machine_integer_info (r, size, count, info, dict);
return;
case 4:
break;
case 17:
- /* Text field that defines variable attributes. New in
- SPSS 14. */
- break;
+ read_data_file_attributes (r, size, count, dict);
+ return;
+
+ case 18:
+ read_variable_attributes (r, size, count, dict);
+ return;
case 20:
/* New in SPSS 16. Contains a single string that describes
the character encoding, e.g. "windows-1252". */
- break;
+ {
+ char *encoding = pool_calloc (r->pool, size, count + 1);
+ read_string (r, encoding, count + 1);
+ dict_set_encoding (dict, encoding);
+ return;
+ }
case 21:
/* New in SPSS 16. Encodes value labels for long string
variables. */
- sys_warn (r, _("Ignoring value labels for long string variables, "
- "which PSPP does not yet support."));
- break;
+ read_long_string_value_labels (r, size, count, dict);
+ return;
default:
sys_warn (r, _("Unrecognized record type 7, subtype %d. Please send a copy of this file, and the syntax which created it to %s"),
/* Read record type 7, subtype 3. */
static void
read_machine_integer_info (struct sfm_reader *r, size_t size, size_t count,
- struct sfm_read_info *info)
+ struct sfm_read_info *info,
+ struct dictionary *dict)
{
int version_major = read_int (r);
int version_minor = read_int (r);
int float_representation = read_int (r);
int compression_code UNUSED = read_int (r);
int integer_representation = read_int (r);
- int character_code UNUSED = read_int (r);
+ int character_code = read_int (r);
int expected_float_format;
int expected_integer_format;
gettext (endian[integer_representation == 1]),
gettext (endian[expected_integer_format == 1]));
}
+
+
+ /*
+ Record 7 (20) provides a much more reliable way of
+ setting the encoding.
+ The character_code is used as a fallback only.
+ */
+ if ( NULL == dict_get_encoding (dict))
+ {
+ switch (character_code)
+ {
+ case 1:
+ dict_set_encoding (dict, "EBCDIC-US");
+ break;
+ case 2:
+ case 3:
+ /* These ostensibly mean "7-bit ASCII" and "8-bit ASCII"[sic]
+ respectively. However, there are known to be many files
+ in the wild with character code 2, yet have data which are
+ clearly not ascii.
+ Therefore we ignore these values.
+ */
+ return;
+ case 4:
+ dict_set_encoding (dict, "MS_KANJI");
+ break;
+ case 65000:
+ dict_set_encoding (dict, "UTF-7");
+ break;
+ case 65001:
+ dict_set_encoding (dict, "UTF-8");
+ break;
+ default:
+ {
+ char enc[100];
+ snprintf (enc, 100, "CP%d", character_code);
+ dict_set_encoding (dict, enc);
+ }
+ break;
+ };
+ }
}
/* Read record type 7, subtype 4. */
size, count);
if (sysmis != SYSMIS)
- sys_warn (r, _("File specifies unexpected value %g as SYSMIS."), sysmis);
+ sys_warn (r, _("File specifies unexpected value %g as %s."),
+ sysmis, "SYSMIS");
+
if (highest != HIGHEST)
- sys_warn (r, _("File specifies unexpected value %g as HIGHEST."), highest);
+ sys_warn (r, _("File specifies unexpected value %g as %s."),
+ highest, "HIGHEST");
+
if (lowest != LOWEST)
- sys_warn (r, _("File specifies unexpected value %g as LOWEST."), lowest);
+ sys_warn (r, _("File specifies unexpected value %g as %s."),
+ lowest, "LOWEST");
}
/* Read record type 7, subtype 11, which specifies how variables
read_long_var_name_map (struct sfm_reader *r, size_t size, size_t count,
struct dictionary *dict)
{
- struct variable_to_value_map *map;
+ struct text_record *text;
struct variable *var;
char *long_name;
- int warning_cnt = 0;
- map = open_variable_to_value_map (r, size * count);
- while (read_variable_to_value_map (r, dict, map, &var, &long_name,
- &warning_cnt))
+ text = open_text_record (r, size * count);
+ while (read_variable_to_value_pair (r, dict, text, &var, &long_name))
{
char **short_names;
size_t short_name_cnt;
}
free (short_names);
}
- close_variable_to_value_map (r, map);
+ close_text_record (r, text);
r->has_long_var_names = true;
}
read_long_string_map (struct sfm_reader *r, size_t size, size_t count,
struct dictionary *dict)
{
- struct variable_to_value_map *map;
+ struct text_record *text;
struct variable *var;
char *length_s;
- int warning_cnt = 0;
- map = open_variable_to_value_map (r, size * count);
- while (read_variable_to_value_map (r, dict, map, &var, &length_s,
- &warning_cnt))
+ text = open_text_record (r, size * count);
+ while (read_variable_to_value_pair (r, dict, text, &var, &length_s))
{
size_t idx = var_get_dict_index (var);
long int length;
dict_delete_consecutive_vars (dict, idx + 1, segment_cnt - 1);
var_set_width (var, length);
}
- close_variable_to_value_map (r, map);
+ close_text_record (r, text);
dict_compact_values (dict);
}
struct label
{
- char raw_value[8]; /* Value as uninterpreted bytes. */
+ uint8_t raw_value[8]; /* Value as uninterpreted bytes. */
union value value; /* Value. */
char *label; /* Null-terminated label string. */
};
struct variable **var = NULL; /* Associated variables. */
int var_cnt; /* Number of associated variables. */
+ int max_width; /* Maximum width of string variables. */
int i;
/* Read the list of variables. */
var = pool_nalloc (subpool, var_cnt, sizeof *var);
+ max_width = 0;
for (i = 0; i < var_cnt; i++)
{
var[i] = lookup_var_by_value_idx (r, var_by_value_idx, read_int (r));
- if (var_is_long_string (var[i]))
- sys_error (r, _("Value labels are not allowed on long string "
- "variables (%s)."), var_get_name (var[i]));
+ if (var_get_width (var[i]) > 8)
+ sys_error (r, _("Value labels may not be added to long string "
+ "variables (e.g. %s) using records types 3 and 4."),
+ var_get_name (var[i]));
+ max_width = MAX (max_width, var_get_width (var[i]));
}
/* Type check the variables. */
{
struct label *label = labels + i;
+ value_init_pool (subpool, &label->value, max_width);
if (var_is_alpha (var[0]))
- buf_copy_rpad (label->value.s, sizeof label->value.s,
- label->raw_value, sizeof label->raw_value);
+ u8_buf_copy_rpad (value_str_rw (&label->value, max_width), max_width,
+ label->raw_value, sizeof label->raw_value, ' ');
else
label->value.f = float_get_double (r->float_format, label->raw_value);
}
label->value.f, var_get_name (v));
else
sys_warn (r, _("Duplicate value label for \"%.*s\" on %s."),
- var_get_width (v), label->value.s,
+ max_width, value_str (&label->value, max_width),
var_get_name (v));
}
}
pool_destroy (subpool);
}
+
+/* Reads a set of custom attributes from TEXT into ATTRS.
+ ATTRS may be a null pointer, in which case the attributes are
+ read but discarded. */
+static void
+read_attributes (struct sfm_reader *r, struct text_record *text,
+ struct attrset *attrs)
+{
+ do
+ {
+ struct attribute *attr;
+ char *key;
+ int index;
+
+ /* Parse the key. */
+ key = text_get_token (text, ss_cstr ("("));
+ if (key == NULL)
+ return;
+
+ attr = attribute_create (key);
+ for (index = 1; ; index++)
+ {
+ /* Parse the value. */
+ char *value;
+ size_t length;
+
+ value = text_get_token (text, ss_cstr ("\n"));
+ if (value == NULL)
+ {
+ text_warn (r, text, _("Error parsing attribute value %s[%d]"),
+ key, index);
+ break;
+ }
+
+ length = strlen (value);
+ if (length >= 2 && value[0] == '\'' && value[length - 1] == '\'')
+ {
+ value[length - 1] = '\0';
+ attribute_add_value (attr, value + 1);
+ }
+ else
+ {
+ text_warn (r, text,
+ _("Attribute value %s[%d] is not quoted: %s"),
+ key, index, value);
+ attribute_add_value (attr, value);
+ }
+
+ /* Was this the last value for this attribute? */
+ if (text_match (text, ')'))
+ break;
+ }
+ if (attrs != NULL)
+ attrset_add (attrs, attr);
+ else
+ attribute_destroy (attr);
+ }
+ while (!text_match (text, '/'));
+}
+
+/* Reads record type 7, subtype 17, which lists custom
+ attributes on the data file. */
+static void
+read_data_file_attributes (struct sfm_reader *r,
+ size_t size, size_t count,
+ struct dictionary *dict)
+{
+ struct text_record *text = open_text_record (r, size * count);
+ read_attributes (r, text, dict_get_attributes (dict));
+ close_text_record (r, text);
+}
+
+static void
+skip_long_string_value_labels (struct sfm_reader *r, size_t n_labels)
+{
+ size_t i;
+
+ for (i = 0; i < n_labels; i++)
+ {
+ size_t value_length, label_length;
+
+ value_length = read_int (r);
+ skip_bytes (r, value_length);
+ label_length = read_int (r);
+ skip_bytes (r, label_length);
+ }
+}
+
+static void
+read_long_string_value_labels (struct sfm_reader *r,
+ size_t size, size_t count,
+ struct dictionary *d)
+{
+ const off_t start = ftello (r->file);
+ while (ftello (r->file) - start < size * count)
+ {
+ char var_name[VAR_NAME_LEN + 1];
+ size_t n_labels, i;
+ struct variable *v;
+ union value value;
+ int var_name_len;
+ int width;
+
+ /* Read header. */
+ var_name_len = read_int (r);
+ if (var_name_len > VAR_NAME_LEN)
+ sys_error (r, _("Variable name length in long string value label "
+ "record (%d) exceeds %d-byte limit."),
+ var_name_len, VAR_NAME_LEN);
+ read_string (r, var_name, var_name_len + 1);
+ width = read_int (r);
+ n_labels = read_int (r);
+
+ v = dict_lookup_var (d, var_name);
+ if (v == NULL)
+ {
+ sys_warn (r, _("Ignoring long string value record for "
+ "unknown variable %s."), var_name);
+ skip_long_string_value_labels (r, n_labels);
+ continue;
+ }
+ if (var_is_numeric (v))
+ {
+ sys_warn (r, _("Ignoring long string value record for "
+ "numeric variable %s."), var_name);
+ skip_long_string_value_labels (r, n_labels);
+ continue;
+ }
+ if (width != var_get_width (v))
+ {
+ sys_warn (r, _("Ignoring long string value record for variable %s "
+ "because the record's width (%d) does not match the "
+ "variable's width (%d)"),
+ var_name, width, var_get_width (v));
+ skip_long_string_value_labels (r, n_labels);
+ continue;
+ }
+
+ /* Read values. */
+ value_init_pool (r->pool, &value, width);
+ for (i = 0; i < n_labels; i++)
+ {
+ size_t value_length, label_length;
+ char label[256];
+ bool skip = false;
+
+ /* Read value. */
+ value_length = read_int (r);
+ if (value_length == width)
+ read_bytes (r, value_str_rw (&value, width), width);
+ else
+ {
+ sys_warn (r, _("Ignoring long string value %zu for variable %s, "
+ "with width %d, that has bad value width %zu."),
+ i, var_get_name (v), width, value_length);
+ skip_bytes (r, value_length);
+ skip = true;
+ }
+
+ /* Read label. */
+ label_length = read_int (r);
+ read_string (r, label, MIN (sizeof label, label_length + 1));
+ if (label_length >= sizeof label)
+ {
+ /* Skip and silently ignore label text after the
+ first 255 bytes. The maximum documented length
+ of a label is 120 bytes so this is more than
+ generous. */
+ skip_bytes (r, sizeof label - (label_length + 1));
+ }
+
+ if (!skip && !var_add_value_label (v, &value, label))
+ sys_warn (r, _("Duplicate value label for \"%.*s\" on %s."),
+ width, value_str (&value, width), var_get_name (v));
+ }
+ }
+}
+
+
+/* Reads record type 7, subtype 18, which lists custom
+ attributes on individual variables. */
+static void
+read_variable_attributes (struct sfm_reader *r,
+ size_t size, size_t count,
+ struct dictionary *dict)
+{
+ struct text_record *text = open_text_record (r, size * count);
+ for (;;)
+ {
+ struct variable *var;
+ if (!text_read_short_name (r, dict, text, ss_cstr (":"), &var))
+ break;
+ read_attributes (r, text, var != NULL ? var_get_attributes (var) : NULL);
+ }
+ close_text_record (r, text);
+}
+
\f
/* Case reader. */
static void read_error (struct casereader *, const struct sfm_reader *);
static bool read_case_number (struct sfm_reader *, double *);
-static bool read_case_string (struct sfm_reader *, char *, size_t);
+static bool read_case_string (struct sfm_reader *, uint8_t *, size_t);
static int read_opcode (struct sfm_reader *);
static bool read_compressed_number (struct sfm_reader *, double *);
-static bool read_compressed_string (struct sfm_reader *, char *);
-static bool read_whole_strings (struct sfm_reader *, char *, size_t);
+static bool read_compressed_string (struct sfm_reader *, uint8_t *);
+static bool read_whole_strings (struct sfm_reader *, uint8_t *, size_t);
static bool skip_whole_strings (struct sfm_reader *, size_t);
-/* Reads one case from READER's file into C. Returns true only
- if successful. */
-static bool
-sys_file_casereader_read (struct casereader *reader, void *r_,
- struct ccase *c)
+/* Reads and returns one case from READER's file. Returns a null
+ pointer if not successful. */
+static struct ccase *
+sys_file_casereader_read (struct casereader *reader, void *r_)
{
struct sfm_reader *r = r_;
+ struct ccase *volatile c;
int i;
if (r->error)
- return false;
+ return NULL;
- case_create (c, r->value_cnt);
+ c = case_create (r->proto);
if (setjmp (r->bail_out))
{
casereader_force_error (reader);
- case_destroy (c);
- return false;
+ case_unref (c);
+ return NULL;
}
for (i = 0; i < r->sfm_var_cnt; i++)
struct sfm_var *sv = &r->sfm_vars[i];
union value *v = case_data_rw_idx (c, sv->case_index);
- if (sv->width == 0)
+ if (sv->var_width == 0)
{
if (!read_case_number (r, &v->f))
goto eof;
}
else
{
- if (!read_case_string (r, v->s + sv->offset, sv->width))
+ uint8_t *s = value_str_rw (v, sv->var_width);
+ if (!read_case_string (r, s + sv->offset, sv->segment_width))
goto eof;
if (!skip_whole_strings (r, ROUND_DOWN (sv->padding, 8)))
partial_record (r);
}
}
- return true;
+ return c;
eof:
- case_destroy (c);
+ case_unref (c);
if (i != 0)
partial_record (r);
if (r->case_cnt != -1)
read_error (reader, r);
- return false;
+ return NULL;
}
/* Issues an error that R ends in a partial record. */
Returns true if successful, false if end of file is
reached immediately. */
static bool
-read_case_string (struct sfm_reader *r, char *s, size_t length)
+read_case_string (struct sfm_reader *r, uint8_t *s, size_t length)
{
size_t whole = ROUND_DOWN (length, 8);
size_t partial = length % 8;
if (partial)
{
- char bounce[8];
+ uint8_t bounce[8];
if (!read_whole_strings (r, bounce, sizeof bounce))
{
if (whole)
Returns true if successful, false if end of file is
reached immediately. */
static bool
-read_compressed_string (struct sfm_reader *r, char *dst)
+read_compressed_string (struct sfm_reader *r, uint8_t *dst)
{
switch (read_opcode (r))
{
Returns true if successful, false if end of file is
reached immediately. */
static bool
-read_whole_strings (struct sfm_reader *r, char *s, size_t length)
+read_whole_strings (struct sfm_reader *r, uint8_t *s, size_t length)
{
assert (length % 8 == 0);
if (!r->compressed)
static bool
skip_whole_strings (struct sfm_reader *r, size_t length)
{
- char buffer[1024];
+ uint8_t buffer[1024];
assert (length < sizeof buffer);
return read_whole_strings (r, buffer, length);
}
return NULL;
}
\f
-/* Helpers for reading records that contain "variable=value"
- pairs. */
+/* Helpers for reading records that contain structured text
+ strings. */
+
+/* Maximum number of warnings to issue for a single text
+ record. */
+#define MAX_TEXT_WARNINGS 5
/* State. */
-struct variable_to_value_map
+struct text_record
{
struct substring buffer; /* Record contents. */
size_t pos; /* Current position in buffer. */
+ int n_warnings; /* Number of warnings issued or suppressed. */
};
-/* Reads SIZE bytes into a "variable=value" map for R,
- and returns the map. */
-static struct variable_to_value_map *
-open_variable_to_value_map (struct sfm_reader *r, size_t size)
+/* Reads SIZE bytes into a text record for R,
+ and returns the new text record. */
+static struct text_record *
+open_text_record (struct sfm_reader *r, size_t size)
{
- struct variable_to_value_map *map = pool_alloc (r->pool, sizeof *map);
+ struct text_record *text = pool_alloc (r->pool, sizeof *text);
char *buffer = pool_malloc (r->pool, size + 1);
read_bytes (r, buffer, size);
- map->buffer = ss_buffer (buffer, size);
- map->pos = 0;
- return map;
+ text->buffer = ss_buffer (buffer, size);
+ text->pos = 0;
+ text->n_warnings = 0;
+ return text;
}
-/* Closes MAP and frees its storage.
- Not really needed, because the pool will free the map anyway,
- but can be used to free it earlier. */
+/* Closes TEXT, frees its storage, and issues a final warning
+ about suppressed warnings if necesary. */
static void
-close_variable_to_value_map (struct sfm_reader *r,
- struct variable_to_value_map *map)
+close_text_record (struct sfm_reader *r, struct text_record *text)
{
- pool_free (r->pool, ss_data (map->buffer));
+ if (text->n_warnings > MAX_TEXT_WARNINGS)
+ sys_warn (r, _("Suppressed %d additional related warnings."),
+ text->n_warnings - MAX_TEXT_WARNINGS);
+ pool_free (r->pool, ss_data (text->buffer));
}
-/* Reads the next variable=value pair from MAP.
+/* Reads a variable=value pair from TEXT.
Looks up the variable in DICT and stores it into *VAR.
Stores a null-terminated value into *VALUE. */
static bool
-read_variable_to_value_map (struct sfm_reader *r, struct dictionary *dict,
- struct variable_to_value_map *map,
- struct variable **var, char **value,
- int *warning_cnt)
+read_variable_to_value_pair (struct sfm_reader *r, struct dictionary *dict,
+ struct text_record *text,
+ struct variable **var, char **value)
{
- int max_warnings = 5;
-
for (;;)
{
- struct substring short_name_ss, value_ss;
+ if (!text_read_short_name (r, dict, text, ss_cstr ("="), var))
+ return false;
+
+ *value = text_get_token (text, ss_buffer ("\t\0", 2));
+ if (*value == NULL)
+ return false;
- if (!ss_tokenize (map->buffer, ss_cstr ("="), &map->pos, &short_name_ss)
- || !ss_tokenize (map->buffer, ss_buffer ("\t\0", 2), &map->pos,
- &value_ss))
- {
- if (*warning_cnt > max_warnings)
- sys_warn (r, _("Suppressed %d additional variable map warnings."),
- *warning_cnt - max_warnings);
- return false;
- }
+ text->pos += ss_span (ss_substr (text->buffer, text->pos, SIZE_MAX),
+ ss_buffer ("\t\0", 2));
- map->pos += ss_span (ss_substr (map->buffer, map->pos, SIZE_MAX),
- ss_buffer ("\t\0", 2));
+ if (*var != NULL)
+ return true;
+ }
+}
- ss_data (short_name_ss)[ss_length (short_name_ss)] = '\0';
- *var = lookup_var_by_short_name (dict, ss_data (short_name_ss));
- if (*var == NULL)
- {
- if (++*warning_cnt <= max_warnings)
- sys_warn (r, _("Variable map refers to unknown variable %s."),
- ss_data (short_name_ss));
- continue;
- }
+static bool
+text_read_short_name (struct sfm_reader *r, struct dictionary *dict,
+ struct text_record *text, struct substring delimiters,
+ struct variable **var)
+{
+ char *short_name = text_get_token (text, delimiters);
+ if (short_name == NULL)
+ return false;
- ss_data (value_ss)[ss_length (value_ss)] = '\0';
- *value = ss_data (value_ss);
+ *var = lookup_var_by_short_name (dict, short_name);
+ if (*var == NULL)
+ text_warn (r, text, _("Variable map refers to unknown variable %s."),
+ short_name);
+ return true;
+}
+
+/* Displays a warning for the current file position, limiting the
+ number to MAX_TEXT_WARNINGS for TEXT. */
+static void
+text_warn (struct sfm_reader *r, struct text_record *text,
+ const char *format, ...)
+{
+ if (text->n_warnings++ < MAX_TEXT_WARNINGS)
+ {
+ va_list args;
+ va_start (args, format);
+ sys_msg (r, MW, format, args);
+ va_end (args);
+ }
+}
+
+static char *
+text_get_token (struct text_record *text, struct substring delimiters)
+{
+ struct substring token;
+
+ if (!ss_tokenize (text->buffer, delimiters, &text->pos, &token))
+ return NULL;
+ ss_data (token)[ss_length (token)] = '\0';
+ return ss_data (token);
+}
+
+static bool
+text_match (struct text_record *text, char c)
+{
+ if (text->buffer.string[text->pos] == c)
+ {
+ text->pos++;
return true;
}
+ else
+ return false;
}
\f
/* Messages. */
/* PSPP - a program for statistical analysis.
- Copyright (C) 1997-9, 2000 Free Software Foundation, Inc.
+ Copyright (C) 1997-9, 2000, 2009 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/command.h>
#include <language/dictionary/split-file.h>
#include <language/lexer/lexer.h>
+#include <language/lexer/value-parser.h>
+#include <libpspp/array.h>
#include <libpspp/assertion.h>
#include <libpspp/compiler.h>
#include <libpspp/hash.h>
#include <data/format.h>
#include "xalloc.h"
+#include "xmemdup0.h"
#include "gettext.h"
#define _(msgid) gettext (msgid)
/* (specification)
"T-TEST" (tts_):
- +groups=custom;
- testval=double;
- +variables=varlist("PV_NO_SCRATCH | PV_NUMERIC");
- +pairs=custom;
- missing=miss:!analysis/listwise,
- incl:include/!exclude;
- +format=fmt:!labels/nolabels;
- criteria=:cin(d:criteria,"%s > 0. && %s < 1.").
+ +groups=custom;
+ testval=double;
+ +variables=varlist("PV_NO_SCRATCH | PV_NUMERIC");
+ +pairs=custom;
+ missing=miss:!analysis/listwise,
+ incl:include/!exclude;
+ +format=fmt:!labels/nolabels;
+ criteria=:cin(d:criteria,"%s > 0. && %s < 1.").
*/
/* (declarations) */
/* (functions) */
-
-/* Variable for the GROUPS subcommand, if given. */
-static struct variable *indep_var;
-
enum comparison
{
- CMP_LE = -2,
- CMP_EQ = 0,
+ CMP_LE,
+ CMP_EQ,
};
-struct group_properties
-{
- /* The comparison criterion */
- enum comparison criterion;
-
- /* The width of the independent variable */
- int indep_width ;
-
- union {
- /* The value of the independent variable at which groups are determined to
- belong to one group or the other */
- double critical_value;
-
-
- /* The values of the independent variable for each group */
- union value g_value[2];
- } v ;
-
-};
-
-
-static struct group_properties gp ;
-
-
-
-/* PAIRS: Number of pairs to be compared ; each pair. */
-static int n_pairs = 0 ;
+/* A pair of variables to be compared. */
struct pair
-{
- /* The variables comprising the pair */
- const struct variable *v[2];
-
- /* The number of valid variable pairs */
- double n;
-
- /* The sum of the members */
- double sum[2];
-
- /* sum of squares of the members */
- double ssq[2];
-
- /* Std deviation of the members */
- double std_dev[2];
-
-
- /* Sample Std deviation of the members */
- double s_std_dev[2];
-
- /* The means of the members */
- double mean[2];
-
- /* The correlation coefficient between the variables */
- double correlation;
-
- /* The sum of the differences */
- double sum_of_diffs;
-
- /* The sum of the products */
- double sum_of_prod;
-
- /* The mean of the differences */
- double mean_diff;
-
- /* The sum of the squares of the differences */
- double ssq_diffs;
+ {
+ const struct variable *v[2]; /* The paired variables. */
+ double n; /* The number of valid variable pairs */
+ double sum[2]; /* The sum of the members */
+ double ssq[2]; /* sum of squares of the members */
+ double std_dev[2]; /* Std deviation of the members */
+ double s_std_dev[2]; /* Sample Std deviation of the members */
+ double mean[2]; /* The means of the members */
+ double correlation; /* Correlation coefficient between the variables. */
+ double sum_of_diffs; /* The sum of the differences */
+ double sum_of_prod; /* The sum of the products */
+ double mean_diff; /* The mean of the differences */
+ double ssq_diffs; /* The sum of the squares of the differences */
+ double std_dev_diff; /* The std deviation of the differences */
+ };
- /* The std deviation of the differences */
- double std_dev_diff;
+/* Which mode was T-TEST invoked */
+enum t_test_mode {
+ T_1_SAMPLE, /* One-sample tests. */
+ T_IND_SAMPLES, /* Independent-sample tests. */
+ T_PAIRED /* Paired-sample tests. */
};
-static struct pair *pairs=0;
-
-static int parse_value (struct lexer *lexer, union value * v, enum val_type);
-
-/* Structures and Functions for the Statistics Summary Box */
-struct ssbox;
-typedef void populate_ssbox_func (struct ssbox *ssb,
- const struct dictionary *,
- struct cmd_t_test *cmd);
-typedef void finalize_ssbox_func (struct ssbox *ssb);
+/* Total state of a T-TEST procedure. */
+struct t_test_proc
+ {
+ enum t_test_mode mode; /* Mode that T-TEST was invoked in. */
+ double criteria; /* Confidence interval in (0, 1). */
+ enum mv_class exclude; /* Classes of missing values to exclude. */
+ bool listwise_missing; /* Drop whole case if one missing var? */
+ struct fmt_spec weight_format; /* Format of weight variable. */
+
+ /* Dependent variables. */
+ const struct variable **vars;
+ size_t n_vars;
+
+ /* For mode == T_1_SAMPLE. */
+ double testval;
+
+ /* For mode == T_PAIRED only. */
+ struct pair *pairs;
+ size_t n_pairs;
+
+ /* For mode == T_IND_SAMPLES only. */
+ struct variable *indep_var; /* Independent variable. */
+ enum comparison criterion; /* Type of comparison. */
+ double critical_value; /* CMP_LE only: Grouping threshold value. */
+ union value g_value[2]; /* CMP_EQ only: Per-group indep var values. */
+ };
+/* Statistics Summary Box */
struct ssbox
-{
- struct tab_table *t;
-
- populate_ssbox_func *populate;
- finalize_ssbox_func *finalize;
-
-};
-
-/* Create a ssbox */
-void ssbox_create (struct ssbox *ssb, struct cmd_t_test *cmd, int mode);
-
-/* Populate a ssbox according to cmd */
-void ssbox_populate (struct ssbox *ssb, const struct dictionary *dict,
- struct cmd_t_test *cmd);
-
-/* Submit and destroy a ssbox */
-void ssbox_finalize (struct ssbox *ssb);
-
-/* A function to create, populate and submit the Paired Samples Correlation
- box */
-static void pscbox (const struct dictionary *);
+ {
+ struct tab_table *t;
+ void (*populate) (struct ssbox *, struct t_test_proc *);
+ void (*finalize) (struct ssbox *);
+ };
+static void ssbox_create (struct ssbox *, struct t_test_proc *);
+static void ssbox_populate (struct ssbox *, struct t_test_proc *);
+static void ssbox_finalize (struct ssbox *);
-/* Structures and Functions for the Test Results Box */
-struct trbox;
+/* Paired Samples Correlation box */
+static void pscbox (struct t_test_proc *);
-typedef void populate_trbox_func (struct trbox *trb,
- const struct dictionary *dict,
- struct cmd_t_test *cmd);
-typedef void finalize_trbox_func (struct trbox *trb);
+/* Test Results Box. */
struct trbox {
struct tab_table *t;
- populate_trbox_func *populate;
- finalize_trbox_func *finalize;
-};
-
-/* Create a trbox */
-void trbox_create (struct trbox *trb, struct cmd_t_test *cmd, int mode);
-
-/* Populate a ssbox according to cmd */
-static void trbox_populate (struct trbox *trb, const struct dictionary *dict,
- struct cmd_t_test *cmd);
-
-/* Submit and destroy a ssbox */
-void trbox_finalize (struct trbox *trb);
-
-/* Which mode was T-TEST invoked */
-enum {
- T_1_SAMPLE = 0 ,
- T_IND_SAMPLES,
- T_PAIRED
-};
-
-
-static int common_calc (const struct dictionary *dict,
- const struct ccase *, void *,
- enum mv_class);
-static void common_precalc (struct cmd_t_test *);
-static void common_postcalc (struct cmd_t_test *);
-
-static int one_sample_calc (const struct dictionary *dict, const struct ccase *, void *, enum mv_class);
-static void one_sample_precalc (struct cmd_t_test *);
-static void one_sample_postcalc (struct cmd_t_test *);
-
-static int paired_calc (const struct dictionary *dict, const struct ccase *,
- struct cmd_t_test*, enum mv_class);
-static void paired_precalc (struct cmd_t_test *);
-static void paired_postcalc (struct cmd_t_test *);
-
-static void group_precalc (struct cmd_t_test *);
-static int group_calc (const struct dictionary *dict, const struct ccase *,
- struct cmd_t_test *, enum mv_class);
-static void group_postcalc (struct cmd_t_test *);
-
-
-static void calculate (struct cmd_t_test *,
- struct casereader *,
- const struct dataset *);
-
-static int mode;
-
-static struct cmd_t_test cmd;
+ void (*populate) (struct trbox *, struct t_test_proc *);
+ void (*finalize) (struct trbox *);
+ };
-static bool bad_weight_warn = false;
+static void trbox_create (struct trbox *, struct t_test_proc *);
+static void trbox_populate (struct trbox *, struct t_test_proc *);
+static void trbox_finalize (struct trbox *);
+static void calculate (struct t_test_proc *, struct casereader *,
+ const struct dataset *);
static int compare_group_binary (const struct group_statistics *a,
- const struct group_statistics *b,
- const struct group_properties *p);
-
-
-static unsigned hash_group_binary (const struct group_statistics *g,
- const struct group_properties *p);
-
-
+ const struct group_statistics *b,
+ const struct t_test_proc *);
+static unsigned hash_group_binary (const struct group_statistics *g,
+ const struct t_test_proc *p);
int
cmd_t_test (struct lexer *lexer, struct dataset *ds)
{
+ struct cmd_t_test cmd;
+ struct t_test_proc proc;
struct casegrouper *grouper;
struct casereader *group;
- bool ok;
+ struct variable *wv;
+ bool ok = false;
- if ( !parse_t_test (lexer, ds, &cmd, NULL) )
- return CMD_FAILURE;
+ proc.pairs = NULL;
+ proc.n_pairs = 0;
+ proc.vars = NULL;
+ proc.indep_var = NULL;
+ if (!parse_t_test (lexer, ds, &cmd, &proc))
+ goto parse_failed;
- if (! cmd.sbc_criteria)
- cmd.criteria=0.95;
+ wv = dict_get_weight (dataset_dict (ds));
+ proc.weight_format = wv ? *var_get_print_format (wv) : F_8_0;
- {
- int m=0;
- if (cmd.sbc_testval) ++m;
- if (cmd.sbc_groups) ++m;
- if (cmd.sbc_pairs) ++m;
-
- if ( m != 1)
- {
- msg (SE,
- _ ("TESTVAL, GROUPS and PAIRS subcommands are mutually exclusive.")
- );
- free_t_test (&cmd);
- return CMD_FAILURE;
- }
- }
-
- if (cmd.sbc_testval)
- mode=T_1_SAMPLE;
- else if (cmd.sbc_groups)
- mode=T_IND_SAMPLES;
- else
- mode=T_PAIRED;
-
- if ( mode == T_PAIRED)
+ if ((cmd.sbc_testval != 0) + (cmd.sbc_groups != 0) + (cmd.sbc_pairs != 0)
+ != 1)
{
- if (cmd.sbc_variables)
- {
- msg (SE, _ ("VARIABLES subcommand is not appropriate with PAIRS"));
- free_t_test (&cmd);
- return CMD_FAILURE;
- }
- else
- {
- /* Iterate through the pairs and put each variable that is a
- member of a pair into cmd.v_variables */
+ msg (SE, _("Exactly one of TESTVAL, GROUPS and PAIRS subcommands "
+ "must be specified."));
+ goto done;
+ }
- int i;
- struct hsh_iterator hi;
- struct const_hsh_table *hash;
- const struct variable *v;
+ proc.mode = (cmd.sbc_testval ? T_1_SAMPLE
+ : cmd.sbc_groups ? T_IND_SAMPLES
+ : T_PAIRED);
+ proc.criteria = cmd.sbc_criteria ? cmd.criteria : 0.95;
+ proc.exclude = cmd.incl != TTS_INCLUDE ? MV_ANY : MV_SYSTEM;
+ proc.listwise_missing = cmd.miss == TTS_LISTWISE;
- hash = const_hsh_create (n_pairs, compare_vars_by_name, hash_var_by_name,
- 0, 0);
+ if (proc.mode == T_1_SAMPLE)
+ proc.testval = cmd.n_testval[0];
- for (i=0; i < n_pairs; ++i)
- {
- const_hsh_insert (hash, pairs[i].v[0]);
- const_hsh_insert (hash, pairs[i].v[1]);
- }
+ if (proc.mode == T_PAIRED)
+ {
+ size_t i, j;
- assert (cmd.n_variables == 0);
- cmd.n_variables = const_hsh_count (hash);
-
- cmd.v_variables = xnrealloc (cmd.v_variables, cmd.n_variables,
- sizeof *cmd.v_variables);
- /* Iterate through the hash */
- for (i=0,v = const_hsh_first (hash, &hi);
- v != 0;
- v = const_hsh_next (hash, &hi) )
- cmd.v_variables[i++]=v;
- const_hsh_destroy (hash);
+ if (cmd.sbc_variables)
+ {
+ msg (SE, _("VARIABLES subcommand may not be used with PAIRS."));
+ goto done;
}
+
+ /* Fill proc.vars with the unique variables from pairs. */
+ proc.n_vars = proc.n_pairs * 2;
+ proc.vars = xmalloc (sizeof *proc.vars * proc.n_vars);
+ for (i = j = 0; i < proc.n_pairs; i++)
+ {
+ proc.vars[j++] = proc.pairs[i].v[0];
+ proc.vars[j++] = proc.pairs[i].v[1];
+ }
+ proc.n_vars = sort_unique (proc.vars, proc.n_vars, sizeof *proc.vars,
+ compare_var_ptrs_by_name, NULL);
}
- else if ( !cmd.sbc_variables)
+ else
{
- msg (SE, _ ("One or more VARIABLES must be specified."));
- free_t_test (&cmd);
- return CMD_FAILURE;
+ if (!cmd.n_variables)
+ {
+ msg (SE, _("One or more VARIABLES must be specified."));
+ goto done;
+ }
+ proc.n_vars = cmd.n_variables;
+ proc.vars = cmd.v_variables;
+ cmd.v_variables = NULL;
}
- bad_weight_warn = true;
-
/* Data pass. */
grouper = casegrouper_create_splits (proc_open (ds), dataset_dict (ds));
while (casegrouper_get_next_group (grouper, &group))
- calculate (&cmd, group, ds);
+ calculate (&proc, group, ds);
ok = casegrouper_destroy (grouper);
ok = proc_commit (ds) && ok;
- n_pairs=0;
- free (pairs);
- pairs=0;
-
- if ( mode == T_IND_SAMPLES)
+ if (proc.mode == T_IND_SAMPLES)
{
int v;
/* Destroy any group statistics we created */
- for (v = 0 ; v < cmd.n_variables ; ++v )
+ for (v = 0; v < proc.n_vars; v++)
{
- struct group_proc *grpp = group_proc_get (cmd.v_variables[v]);
+ struct group_proc *grpp = group_proc_get (proc.vars[v]);
hsh_destroy (grpp->group_hash);
}
}
+done:
free_t_test (&cmd);
- return ok ? CMD_SUCCESS : CMD_CASCADING_FAILURE;
+parse_failed:
+ if (proc.indep_var != NULL)
+ {
+ int width = var_get_width (proc.indep_var);
+ value_destroy (&proc.g_value[0], width);
+ value_destroy (&proc.g_value[1], width);
+ }
+ free (proc.vars);
+ free (proc.pairs);
+ return ok ? CMD_SUCCESS : CMD_FAILURE;
}
static int
-tts_custom_groups (struct lexer *lexer, struct dataset *ds, struct cmd_t_test *cmd UNUSED,
- void *aux UNUSED)
+tts_custom_groups (struct lexer *lexer, struct dataset *ds,
+ struct cmd_t_test *cmd UNUSED, void *proc_)
{
- int n_group_values=0;
+ struct t_test_proc *proc = proc_;
+ int n_values;
+ int width;
lex_match (lexer, '=');
- indep_var = parse_variable (lexer, dataset_dict (ds));
- if (!indep_var)
+ proc->indep_var = parse_variable (lexer, dataset_dict (ds));
+ if (proc->indep_var == NULL)
{
lex_error (lexer, "expecting variable name in GROUPS subcommand");
return 0;
}
-
- if (var_is_long_string (indep_var))
- {
- msg (SE, _ ("Long string variable %s is not valid here."),
- var_get_name (indep_var));
- return 0;
- }
+ width = var_get_width (proc->indep_var);
+ value_init (&proc->g_value[0], width);
+ value_init (&proc->g_value[1], width);
if (!lex_match (lexer, '('))
+ n_values = 0;
+ else
{
- if (var_is_numeric (indep_var))
- {
- gp.v.g_value[0].f = 1;
- gp.v.g_value[1].f = 2;
-
- gp.criterion = CMP_EQ;
-
- n_group_values = 2;
-
- return 1;
- }
+ if (!parse_value (lexer, &proc->g_value[0], width))
+ return 0;
+ lex_match (lexer, ',');
+ if (lex_match (lexer, ')'))
+ n_values = 1;
else
- {
- msg (SE, _ ("When applying GROUPS to a string variable, two "
- "values must be specified."));
- return 0;
- }
+ {
+ if (!parse_value (lexer, &proc->g_value[1], width)
+ || !lex_force_match (lexer, ')'))
+ return 0;
+ n_values = 2;
+ }
}
- if (!parse_value (lexer, &gp.v.g_value[0], var_get_type (indep_var)))
- return 0;
-
- lex_match (lexer, ',');
- if (lex_match (lexer, ')'))
+ if (var_is_numeric (proc->indep_var))
{
- if (var_is_alpha (indep_var))
+ proc->criterion = n_values == 1 ? CMP_LE : CMP_EQ;
+ if (n_values == 1)
+ proc->critical_value = proc->g_value[0].f;
+ else if (n_values == 0)
{
- msg (SE, _ ("When applying GROUPS to a string variable, two "
- "values must be specified."));
- return 0;
+ proc->g_value[0].f = 1;
+ proc->g_value[1].f = 2;
}
- gp.criterion = CMP_LE;
- gp.v.critical_value = gp.v.g_value[0].f;
-
- n_group_values = 1;
- return 1;
}
-
- if (!parse_value (lexer, &gp.v.g_value[1], var_get_type (indep_var)))
- return 0;
-
- n_group_values = 2;
- if (!lex_force_match (lexer, ')'))
- return 0;
-
- if ( n_group_values == 2 )
- gp.criterion = CMP_EQ ;
else
- gp.criterion = CMP_LE ;
-
-
- if ( var_is_alpha (indep_var))
{
- buf_copy_rpad (gp.v.g_value [0].s, var_get_width (indep_var),
- gp.v.g_value [0].s, strlen (gp.v.g_value[0].s));
-
- buf_copy_rpad (gp.v.g_value [1].s, var_get_width (indep_var),
- gp.v.g_value [1].s, strlen (gp.v.g_value[1].s));
+ proc->criterion = CMP_EQ;
+ if (n_values != 2)
+ {
+ msg (SE, _("When applying GROUPS to a string variable, two "
+ "values must be specified."));
+ return 0;
+ }
}
-
return 1;
}
+static void
+add_pair (struct t_test_proc *proc,
+ const struct variable *v0, const struct variable *v1)
+{
+ struct pair *p = &proc->pairs[proc->n_pairs++];
+ p->v[0] = v0;
+ p->v[1] = v1;
+}
static int
-tts_custom_pairs (struct lexer *lexer, struct dataset *ds, struct cmd_t_test *cmd UNUSED, void *aux UNUSED)
+tts_custom_pairs (struct lexer *lexer, struct dataset *ds,
+ struct cmd_t_test *cmd UNUSED, void *proc_)
{
- const struct variable **vars;
- size_t n_vars;
- size_t n_pairs_local;
-
- size_t n_before_WITH;
- size_t n_after_WITH = SIZE_MAX;
- int paired ; /* Was the PAIRED keyword given ? */
-
- lex_match (lexer, '=');
+ struct t_test_proc *proc = proc_;
- n_vars=0;
- if (!parse_variables_const (lexer, dataset_dict (ds), &vars, &n_vars,
- PV_DUPLICATE | PV_NUMERIC | PV_NO_SCRATCH))
- {
- free (vars);
- return 0;
- }
- assert (n_vars);
-
- n_before_WITH = 0;
- if (lex_match (lexer, T_WITH))
- {
- n_before_WITH = n_vars;
- if (!parse_variables_const (lexer, dataset_dict (ds), &vars, &n_vars,
- PV_DUPLICATE | PV_APPEND
- | PV_NUMERIC | PV_NO_SCRATCH))
- {
- free (vars);
- return 0;
- }
- n_after_WITH = n_vars - n_before_WITH;
- }
+ const struct variable **vars1 = NULL;
+ size_t n_vars1 = 0;
- paired = (lex_match (lexer, '(') && lex_match_id (lexer, "PAIRED") && lex_match (lexer, ')'));
+ const struct variable **vars2 = NULL;
+ size_t n_vars2 = 0;
- /* Determine the number of pairs needed */
- if (paired)
- {
- if (n_before_WITH != n_after_WITH)
- {
- free (vars);
- msg (SE, _ ("PAIRED was specified but the number of variables "
- "preceding WITH (%zu) did not match the number "
- "following (%zu)."),
- n_before_WITH, n_after_WITH);
- return 0;
- }
- n_pairs_local = n_before_WITH;
- }
- else if (n_before_WITH > 0) /* WITH keyword given, but not PAIRED keyword */
- {
- n_pairs_local = n_before_WITH * n_after_WITH ;
- }
- else /* Neither WITH nor PAIRED keyword given */
- {
- if (n_vars < 2)
- {
- free (vars);
- msg (SE, _ ("At least two variables must be specified "
- "on PAIRS."));
- return 0;
- }
-
- /* how many ways can you pick 2 from n_vars ? */
- n_pairs_local = n_vars * (n_vars - 1) / 2;
- }
+ bool paired = false;
+ size_t n_total_pairs;
+ size_t i, j;
- /* Allocate storage for the pairs */
- pairs = xnrealloc (pairs, n_pairs + n_pairs_local, sizeof *pairs);
+ lex_match (lexer, '=');
- /* Populate the pairs with the appropriate variables */
- if ( paired )
- {
- int i;
+ if (!parse_variables_const (lexer, dataset_dict (ds), &vars1, &n_vars1,
+ PV_DUPLICATE | PV_NUMERIC | PV_NO_SCRATCH))
+ return 0;
- assert (n_pairs_local == n_vars / 2);
- for (i = 0; i < n_pairs_local; ++i)
- {
- pairs[i].v[n_pairs] = vars[i];
- pairs[i].v[n_pairs + 1] = vars[i + n_pairs_local];
- }
- }
- else if (n_before_WITH > 0) /* WITH keyword given, but not PAIRED keyword */
+ if (lex_match (lexer, T_WITH))
{
- int i,j;
- size_t p = n_pairs;
-
- for (i=0 ; i < n_before_WITH ; ++i )
- {
- for (j=0 ; j < n_after_WITH ; ++j)
- {
- pairs[p].v[0] = vars[i];
- pairs[p].v[1] = vars[j+n_before_WITH];
- ++p;
- }
- }
+ if (!parse_variables_const (lexer, dataset_dict (ds), &vars2, &n_vars2,
+ PV_DUPLICATE | PV_NUMERIC | PV_NO_SCRATCH))
+ {
+ free (vars1);
+ return 0;
+ }
+
+ if (lex_match (lexer, '(')
+ && lex_match_id (lexer, "PAIRED")
+ && lex_match (lexer, ')'))
+ {
+ paired = true;
+ if (n_vars1 != n_vars2)
+ {
+ msg (SE, _("PAIRED was specified but the number of variables "
+ "preceding WITH (%zu) did not match the number "
+ "following (%zu)."),
+ n_vars1, n_vars2);
+ free (vars1);
+ free (vars2);
+ return 0;
+ }
+ }
}
- else /* Neither WITH nor PAIRED given */
+ else
{
- size_t i,j;
- size_t p=n_pairs;
-
- for (i=0 ; i < n_vars ; ++i )
+ if (n_vars1 < 2)
{
- for (j=i+1 ; j < n_vars ; ++j)
- {
- pairs[p].v[0] = vars[i];
- pairs[p].v[1] = vars[j];
- ++p;
- }
+ free (vars1);
+ msg (SE, _("At least two variables must be specified on PAIRS."));
+ return 0;
}
}
- n_pairs+=n_pairs_local;
+ /* Allocate storage for the new pairs. */
+ n_total_pairs = proc->n_pairs + (paired ? n_vars1
+ : n_vars2 > 0 ? n_vars1 * n_vars2
+ : n_vars1 * (n_vars1 - 1) / 2);
+ proc->pairs = xnrealloc (proc->pairs, n_total_pairs, sizeof *proc->pairs);
- free (vars);
- return 1;
-}
-
-/* Parses the current token (numeric or string, depending on type)
- value v and returns success. */
-static int
-parse_value (struct lexer *lexer, union value * v, enum val_type type)
-{
- if (type == VAL_NUMERIC)
- {
- if (!lex_force_num (lexer))
- return 0;
- v->f = lex_tokval (lexer);
- }
+ /* Populate the pairs with the appropriate variables. */
+ if (paired)
+ for (i = 0; i < n_vars1; i++)
+ add_pair (proc, vars1[i], vars2[i]);
+ else if (n_vars2 > 0)
+ for (i = 0; i < n_vars1; i++)
+ for (j = 0; j < n_vars2; j++)
+ add_pair (proc, vars1[i], vars2[j]);
else
- {
- if (!lex_force_string (lexer))
- return 0;
- memset (v->s, ' ', MAX_SHORT_STRING);
- strncpy (v->s, ds_cstr (lex_tokstr (lexer)), ds_length (lex_tokstr (lexer)));
- }
-
- lex_get (lexer);
+ for (i = 0; i < n_vars1; i++)
+ for (j = i + 1; j < n_vars1; j++)
+ add_pair (proc, vars1[i], vars1[j]);
+ assert (proc->n_pairs == n_total_pairs);
+ free (vars1);
+ free (vars2);
return 1;
}
+\f
+/* Implementation of the SSBOX object. */
+static void ssbox_base_init (struct ssbox *, int cols, int rows);
+static void ssbox_base_finalize (struct ssbox *);
+static void ssbox_one_sample_init (struct ssbox *, struct t_test_proc *);
+static void ssbox_independent_samples_init (struct ssbox *, struct t_test_proc *);
+static void ssbox_paired_init (struct ssbox *, struct t_test_proc *);
-/* Implementation of the SSBOX object */
-
-void ssbox_base_init (struct ssbox *this, int cols,int rows);
-
-void ssbox_base_finalize (struct ssbox *ssb);
-
-void ssbox_one_sample_init (struct ssbox *this,
- struct cmd_t_test *cmd );
-
-void ssbox_independent_samples_init (struct ssbox *this,
- struct cmd_t_test *cmd);
-
-void ssbox_paired_init (struct ssbox *this,
- struct cmd_t_test *cmd);
-
-
-/* Factory to create an ssbox */
-void
-ssbox_create (struct ssbox *ssb, struct cmd_t_test *cmd, int mode)
+/* Factory to create an ssbox. */
+static void
+ssbox_create (struct ssbox *ssb, struct t_test_proc *proc)
{
- switch (mode)
- {
- case T_1_SAMPLE:
- ssbox_one_sample_init (ssb,cmd);
- break;
- case T_IND_SAMPLES:
- ssbox_independent_samples_init (ssb,cmd);
- break;
- case T_PAIRED:
- ssbox_paired_init (ssb,cmd);
- break;
- default:
- NOT_REACHED ();
- }
+ switch (proc->mode)
+ {
+ case T_1_SAMPLE:
+ ssbox_one_sample_init (ssb, proc);
+ break;
+ case T_IND_SAMPLES:
+ ssbox_independent_samples_init (ssb, proc);
+ break;
+ case T_PAIRED:
+ ssbox_paired_init (ssb, proc);
+ break;
+ default:
+ NOT_REACHED ();
+ }
}
-
-
/* Despatcher for the populate method */
-void
-ssbox_populate (struct ssbox *ssb, const struct dictionary *dict,
- struct cmd_t_test *cmd)
+static void
+ssbox_populate (struct ssbox *ssb, struct t_test_proc *proc)
{
- ssb->populate (ssb, dict, cmd);
+ ssb->populate (ssb, proc);
}
-
/* Despatcher for finalize */
-void
+static void
ssbox_finalize (struct ssbox *ssb)
{
ssb->finalize (ssb);
}
-
/* Submit the box and clear up */
-void
+static void
ssbox_base_finalize (struct ssbox *ssb)
{
tab_submit (ssb->t);
}
-
-
/* Initialize a ssbox struct */
-void
-ssbox_base_init (struct ssbox *this, int cols,int rows)
+static void
+ssbox_base_init (struct ssbox *this, int cols, int rows)
{
this->finalize = ssbox_base_finalize;
this->t = tab_create (cols, rows, 0);
tab_columns (this->t, SOM_COL_DOWN, 1);
- tab_headers (this->t,0,0,1,0);
- tab_box (this->t, TAL_2, TAL_2, TAL_0, TAL_1, 0, 0, cols -1, rows -1 );
- tab_hline (this->t, TAL_2,0,cols-1,1);
- tab_dim (this->t, tab_natural_dimensions);
+ tab_headers (this->t, 0, 0, 1, 0);
+ tab_box (this->t, TAL_2, TAL_2, TAL_0, TAL_1, 0, 0, cols - 1, rows - 1);
+ tab_hline (this->t, TAL_2, 0, cols- 1, 1);
+ tab_dim (this->t, tab_natural_dimensions, NULL);
}
+\f
+/* ssbox implementations. */
-void ssbox_one_sample_populate (struct ssbox *ssb,
- const struct dictionary *,
- struct cmd_t_test *cmd);
+static void ssbox_one_sample_populate (struct ssbox *, struct t_test_proc *);
+static void ssbox_independent_samples_populate (struct ssbox *,
+ struct t_test_proc *);
+static void ssbox_paired_populate (struct ssbox *, struct t_test_proc *);
/* Initialize the one_sample ssbox */
-void
-ssbox_one_sample_init (struct ssbox *this,
- struct cmd_t_test *cmd )
+static void
+ssbox_one_sample_init (struct ssbox *this, struct t_test_proc *proc)
{
- const int hsize=5;
- const int vsize=cmd->n_variables+1;
+ const int hsize = 5;
+ const int vsize = proc->n_vars + 1;
this->populate = ssbox_one_sample_populate;
- ssbox_base_init (this, hsize,vsize);
- tab_title (this->t, _ ("One-Sample Statistics"));
- tab_vline (this->t, TAL_2, 1,0,vsize - 1);
- tab_text (this->t, 1, 0, TAB_CENTER | TAT_TITLE, _ ("N"));
- tab_text (this->t, 2, 0, TAB_CENTER | TAT_TITLE, _ ("Mean"));
- tab_text (this->t, 3, 0, TAB_CENTER | TAT_TITLE, _ ("Std. Deviation"));
- tab_text (this->t, 4, 0, TAB_CENTER | TAT_TITLE, _ ("SE. Mean"));
+ ssbox_base_init (this, hsize, vsize);
+ tab_title (this->t, _("One-Sample Statistics"));
+ tab_vline (this->t, TAL_2, 1, 0, vsize - 1);
+ tab_text (this->t, 1, 0, TAB_CENTER | TAT_TITLE, _("N"));
+ tab_text (this->t, 2, 0, TAB_CENTER | TAT_TITLE, _("Mean"));
+ tab_text (this->t, 3, 0, TAB_CENTER | TAT_TITLE, _("Std. Deviation"));
+ tab_text (this->t, 4, 0, TAB_CENTER | TAT_TITLE, _("SE. Mean"));
}
-static void ssbox_independent_samples_populate (struct ssbox *ssb,
- const struct dictionary *,
- struct cmd_t_test *cmd);
-
/* Initialize the independent samples ssbox */
-void
-ssbox_independent_samples_init (struct ssbox *this,
- struct cmd_t_test *cmd)
+static void
+ssbox_independent_samples_init (struct ssbox *this, struct t_test_proc *proc)
{
int hsize=6;
- int vsize = cmd->n_variables*2 +1;
+ int vsize = proc->n_vars * 2 + 1;
this->populate = ssbox_independent_samples_populate;
- ssbox_base_init (this, hsize,vsize);
- tab_vline (this->t, TAL_GAP, 1, 0,vsize - 1);
- tab_title (this->t, _ ("Group Statistics"));
- tab_text (this->t, 1, 0, TAB_CENTER | TAT_TITLE, var_get_name (indep_var));
- tab_text (this->t, 2, 0, TAB_CENTER | TAT_TITLE, _ ("N"));
- tab_text (this->t, 3, 0, TAB_CENTER | TAT_TITLE, _ ("Mean"));
- tab_text (this->t, 4, 0, TAB_CENTER | TAT_TITLE, _ ("Std. Deviation"));
- tab_text (this->t, 5, 0, TAB_CENTER | TAT_TITLE, _ ("SE. Mean"));
+ ssbox_base_init (this, hsize, vsize);
+ tab_vline (this->t, TAL_GAP, 1, 0, vsize - 1);
+ tab_title (this->t, _("Group Statistics"));
+ tab_text (this->t, 1, 0, TAB_CENTER | TAT_TITLE,
+ var_get_name (proc->indep_var));
+ tab_text (this->t, 2, 0, TAB_CENTER | TAT_TITLE, _("N"));
+ tab_text (this->t, 3, 0, TAB_CENTER | TAT_TITLE, _("Mean"));
+ tab_text (this->t, 4, 0, TAB_CENTER | TAT_TITLE, _("Std. Deviation"));
+ tab_text (this->t, 5, 0, TAB_CENTER | TAT_TITLE, _("SE. Mean"));
}
-
/* Populate the ssbox for independent samples */
static void
ssbox_independent_samples_populate (struct ssbox *ssb,
- const struct dictionary *dict,
- struct cmd_t_test *cmd)
+ struct t_test_proc *proc)
{
int i;
- const struct variable *wv = dict_get_weight (dict);
- const struct fmt_spec *wfmt = wv ? var_get_print_format (wv) : &F_8_0;
-
- char *val_lab[2] = {NULL, NULL};
+ char *val_lab[2];
double indep_value[2];
- char prefix[2][3]={"",""};
+ char prefix[2][3];
- if ( var_is_numeric (indep_var) )
+ for (i = 0; i < 2; i++)
{
- const char *s;
+ union value *value = &proc->g_value[i];
+ int width = var_get_width (proc->indep_var);
- s = var_lookup_value_label (indep_var, &gp.v.g_value[0]);
- val_lab[0] = s ? strdup (s) : NULL;
+ indep_value[i] = (proc->criterion == CMP_LE ? proc->critical_value
+ : value->f);
- s = var_lookup_value_label (indep_var, &gp.v.g_value[1]);
- val_lab[1] = s ? strdup (s) : NULL;
- }
- else
- {
- val_lab[0] = calloc (sizeof (char), MAX_SHORT_STRING + 1);
- val_lab[1] = calloc (sizeof (char), MAX_SHORT_STRING + 1);
- memcpy (val_lab[0], gp.v.g_value[0].s, MAX_SHORT_STRING);
- memcpy (val_lab[1], gp.v.g_value[1].s, MAX_SHORT_STRING);
+ if (val_type_from_width (width) == VAL_NUMERIC)
+ {
+ const char *s = var_lookup_value_label (proc->indep_var, value);
+ val_lab[i] = s ? xstrdup (s) : xasprintf ("%g", indep_value[i]);
+ }
+ else
+ val_lab[i] = xmemdup0 (value_str (value, width), width);
}
- if (gp.criterion == CMP_LE )
+ if (proc->criterion == CMP_LE)
{
- strcpy (prefix[0],">=");
- strcpy (prefix[1],"<");
- indep_value[0] = gp.v.critical_value;
- indep_value[1] = gp.v.critical_value;
+ strcpy (prefix[0], ">=");
+ strcpy (prefix[1], "<");
}
else
{
- indep_value[0] = gp.v.g_value[0].f;
- indep_value[1] = gp.v.g_value[1].f;
+ strcpy (prefix[0], "");
+ strcpy (prefix[1], "");
}
- assert (ssb->t);
-
- for (i=0; i < cmd->n_variables; ++i)
+ for (i = 0; i < proc->n_vars; i++)
{
- const struct variable *var = cmd->v_variables[i];
+ const struct variable *var = proc->vars[i];
struct hsh_table *grp_hash = group_proc_get (var)->group_hash;
int count=0;
- tab_text (ssb->t, 0, i*2+1, TAB_LEFT,
- var_get_name (cmd->v_variables[i]));
-
- if (val_lab[0])
- tab_text (ssb->t, 1, i*2+1, TAB_LEFT | TAT_PRINTF,
- "%s%s", prefix[0], val_lab[0]);
- else
- tab_text (ssb->t, 1, i*2+1, TAB_LEFT | TAT_PRINTF,
- "%s%g", prefix[0], indep_value[0]);
-
-
- if (val_lab[1])
- tab_text (ssb->t, 1, i*2+1+1, TAB_LEFT | TAT_PRINTF,
- "%s%s", prefix[1], val_lab[1]);
- else
- tab_text (ssb->t, 1, i*2+1+1, TAB_LEFT | TAT_PRINTF,
- "%s%g", prefix[1], indep_value[1]);
-
+ tab_text (ssb->t, 0, i * 2 + 1, TAB_LEFT,
+ var_get_name (proc->vars[i]));
+ tab_text_format (ssb->t, 1, i * 2 + 1, TAB_LEFT,
+ "%s%s", prefix[0], val_lab[0]);
+ tab_text_format (ssb->t, 1, i * 2 + 1+ 1, TAB_LEFT,
+ "%s%s", prefix[1], val_lab[1]);
/* Fill in the group statistics */
- for ( count = 0 ; count < 2 ; ++count )
+ for (count = 0; count < 2; count++)
{
union value search_val;
-
struct group_statistics *gs;
- if ( gp.criterion == CMP_LE )
- {
- if ( count == 0 )
- {
- /* >= case */
- search_val.f = gp.v.critical_value + 1.0;
- }
- else
- {
- /* less than ( < ) case */
- search_val.f = gp.v.critical_value - 1.0;
- }
- }
+ if (proc->criterion == CMP_LE)
+ search_val.f = proc->critical_value + (count == 0 ? 1.0 : -1.0);
else
- {
- search_val = gp.v.g_value[count];
- }
+ search_val = proc->g_value[count];
- gs = hsh_find (grp_hash, (void *) &search_val);
+ gs = hsh_find (grp_hash, &search_val);
assert (gs);
- tab_double (ssb->t, 2, i*2+count+1, TAB_RIGHT, gs->n, wfmt);
- tab_double (ssb->t, 3, i*2+count+1, TAB_RIGHT, gs->mean, NULL);
- tab_double (ssb->t, 4, i*2+count+1, TAB_RIGHT, gs->std_dev, NULL);
- tab_double (ssb->t, 5, i*2+count+1, TAB_RIGHT, gs->se_mean, NULL);
+ tab_double (ssb->t, 2, i * 2 + count+ 1, TAB_RIGHT, gs->n,
+ &proc->weight_format);
+ tab_double (ssb->t, 3, i * 2 + count+ 1, TAB_RIGHT, gs->mean, NULL);
+ tab_double (ssb->t, 4, i * 2 + count+ 1, TAB_RIGHT, gs->std_dev,
+ NULL);
+ tab_double (ssb->t, 5, i * 2 + count+ 1, TAB_RIGHT, gs->se_mean,
+ NULL);
}
}
free (val_lab[0]);
free (val_lab[1]);
}
-
-static void ssbox_paired_populate (struct ssbox *ssb,
- const struct dictionary *dict,
- struct cmd_t_test *cmd);
-
/* Initialize the paired values ssbox */
-void
-ssbox_paired_init (struct ssbox *this, struct cmd_t_test *cmd UNUSED)
+static void
+ssbox_paired_init (struct ssbox *this, struct t_test_proc *proc)
{
- int hsize=6;
-
- int vsize = n_pairs*2+1;
+ int hsize = 6;
+ int vsize = proc->n_pairs * 2 + 1;
this->populate = ssbox_paired_populate;
- ssbox_base_init (this, hsize,vsize);
- tab_title (this->t, _ ("Paired Sample Statistics"));
- tab_vline (this->t,TAL_GAP,1,0,vsize-1);
- tab_vline (this->t,TAL_2,2,0,vsize-1);
- tab_text (this->t, 2, 0, TAB_CENTER | TAT_TITLE, _ ("Mean"));
- tab_text (this->t, 3, 0, TAB_CENTER | TAT_TITLE, _ ("N"));
- tab_text (this->t, 4, 0, TAB_CENTER | TAT_TITLE, _ ("Std. Deviation"));
- tab_text (this->t, 5, 0, TAB_CENTER | TAT_TITLE, _ ("SE. Mean"));
+ ssbox_base_init (this, hsize, vsize);
+ tab_title (this->t, _("Paired Sample Statistics"));
+ tab_vline (this->t, TAL_GAP, 1, 0, vsize - 1);
+ tab_vline (this->t, TAL_2, 2, 0, vsize - 1);
+ tab_text (this->t, 2, 0, TAB_CENTER | TAT_TITLE, _("Mean"));
+ tab_text (this->t, 3, 0, TAB_CENTER | TAT_TITLE, _("N"));
+ tab_text (this->t, 4, 0, TAB_CENTER | TAT_TITLE, _("Std. Deviation"));
+ tab_text (this->t, 5, 0, TAB_CENTER | TAT_TITLE, _("SE. Mean"));
}
-
/* Populate the ssbox for paired values */
-void
-ssbox_paired_populate (struct ssbox *ssb, const struct dictionary *dict,
- struct cmd_t_test *cmd UNUSED)
+static void
+ssbox_paired_populate (struct ssbox *ssb, struct t_test_proc *proc)
{
int i;
- const struct variable *wv = dict_get_weight (dict);
- const struct fmt_spec *wfmt = wv ? var_get_print_format (wv) : &F_8_0;
-
- assert (ssb->t);
-
- for (i=0; i < n_pairs; ++i)
+ for (i = 0; i < proc->n_pairs; i++)
{
+ struct pair *p = &proc->pairs[i];
int j;
- tab_text (ssb->t, 0, i*2+1, TAB_LEFT | TAT_PRINTF , _ ("Pair %d"),i);
-
- for (j=0 ; j < 2 ; ++j)
+ tab_text_format (ssb->t, 0, i * 2 + 1, TAB_LEFT, _("Pair %d"), i);
+ for (j=0; j < 2; j++)
{
- struct group_statistics *gs;
-
- gs = &group_proc_get (pairs[i].v[j])->ugs;
-
/* Titles */
-
- tab_text (ssb->t, 1, i*2+j+1, TAB_LEFT,
- var_get_name (pairs[i].v[j]));
+ tab_text (ssb->t, 1, i * 2 + j + 1, TAB_LEFT,
+ var_get_name (p->v[j]));
/* Values */
- tab_double (ssb->t,2, i*2+j+1, TAB_RIGHT, pairs[i].mean[j], NULL);
- tab_double (ssb->t,3, i*2+j+1, TAB_RIGHT, pairs[i].n, wfmt);
- tab_double (ssb->t,4, i*2+j+1, TAB_RIGHT, pairs[i].std_dev[j], NULL);
- tab_double (ssb->t,5, i*2+j+1, TAB_RIGHT,
- pairs[i].std_dev[j]/sqrt (pairs[i].n), NULL);
-
+ tab_double (ssb->t, 2, i * 2 + j + 1, TAB_RIGHT, p->mean[j], NULL);
+ tab_double (ssb->t, 3, i * 2 + j + 1, TAB_RIGHT, p->n,
+ &proc->weight_format);
+ tab_double (ssb->t, 4, i * 2 + j + 1, TAB_RIGHT, p->std_dev[j],
+ NULL);
+ tab_double (ssb->t, 5, i * 2 + j + 1, TAB_RIGHT,
+ p->std_dev[j] /sqrt (p->n), NULL);
}
}
}
/* Populate the one sample ssbox */
-void
-ssbox_one_sample_populate (struct ssbox *ssb, const struct dictionary *dict,
- struct cmd_t_test *cmd)
+static void
+ssbox_one_sample_populate (struct ssbox *ssb, struct t_test_proc *proc)
{
int i;
- const struct variable *wv = dict_get_weight (dict);
- const struct fmt_spec *wfmt = wv ? var_get_print_format (wv) : &F_8_0;
-
- assert (ssb->t);
-
- for (i=0; i < cmd->n_variables; ++i)
+ for (i = 0; i < proc->n_vars; i++)
{
- struct group_statistics *gs = &group_proc_get (cmd->v_variables[i])->ugs;
+ struct group_statistics *gs = &group_proc_get (proc->vars[i])->ugs;
- tab_text (ssb->t, 0, i+1, TAB_LEFT, var_get_name (cmd->v_variables[i]));
- tab_double (ssb->t,1, i+1, TAB_RIGHT, gs->n, wfmt);
- tab_double (ssb->t,2, i+1, TAB_RIGHT, gs->mean, NULL);
- tab_double (ssb->t,3, i+1, TAB_RIGHT, gs->std_dev, NULL);
- tab_double (ssb->t,4, i+1, TAB_RIGHT, gs->se_mean, NULL);
+ tab_text (ssb->t, 0, i + 1, TAB_LEFT, var_get_name (proc->vars[i]));
+ tab_double (ssb->t, 1, i + 1, TAB_RIGHT, gs->n, &proc->weight_format);
+ tab_double (ssb->t, 2, i + 1, TAB_RIGHT, gs->mean, NULL);
+ tab_double (ssb->t, 3, i + 1, TAB_RIGHT, gs->std_dev, NULL);
+ tab_double (ssb->t, 4, i + 1, TAB_RIGHT, gs->se_mean, NULL);
}
}
-
-
-
+\f
/* Implementation of the Test Results box struct */
-void trbox_base_init (struct trbox *self,size_t n_vars, int cols);
-void trbox_base_finalize (struct trbox *trb);
-
-void trbox_independent_samples_init (struct trbox *trb,
- struct cmd_t_test *cmd );
-
-static void trbox_independent_samples_populate (struct trbox *trb,
- const struct dictionary *dict,
- struct cmd_t_test *cmd);
-
-void trbox_one_sample_init (struct trbox *self,
- struct cmd_t_test *cmd );
-
-static void trbox_one_sample_populate (struct trbox *trb,
- const struct dictionary *,
- struct cmd_t_test *cmd);
-
-void trbox_paired_init (struct trbox *self,
- struct cmd_t_test *cmd );
-
-static void trbox_paired_populate (struct trbox *trb,
- const struct dictionary *,
- struct cmd_t_test *cmd);
-
-
+static void trbox_base_init (struct trbox *, size_t n_vars, int cols);
+static void trbox_base_finalize (struct trbox *);
+static void trbox_independent_samples_init (struct trbox *,
+ struct t_test_proc *);
+static void trbox_independent_samples_populate (struct trbox *,
+ struct t_test_proc *);
+static void trbox_one_sample_init (struct trbox *, struct t_test_proc *);
+static void trbox_one_sample_populate (struct trbox *, struct t_test_proc *);
+static void trbox_paired_init (struct trbox *, struct t_test_proc *);
+static void trbox_paired_populate (struct trbox *, struct t_test_proc *);
/* Create a trbox according to mode*/
-void
-trbox_create (struct trbox *trb,
- struct cmd_t_test *cmd, int mode)
+static void
+trbox_create (struct trbox *trb, struct t_test_proc *proc)
{
- switch (mode)
- {
- case T_1_SAMPLE:
- trbox_one_sample_init (trb,cmd);
- break;
- case T_IND_SAMPLES:
- trbox_independent_samples_init (trb,cmd);
- break;
- case T_PAIRED:
- trbox_paired_init (trb,cmd);
- break;
- default:
- NOT_REACHED ();
- }
+ switch (proc->mode)
+ {
+ case T_1_SAMPLE:
+ trbox_one_sample_init (trb, proc);
+ break;
+ case T_IND_SAMPLES:
+ trbox_independent_samples_init (trb, proc);
+ break;
+ case T_PAIRED:
+ trbox_paired_init (trb, proc);
+ break;
+ default:
+ NOT_REACHED ();
+ }
}
-/* Populate a trbox according to cmd */
+/* Populate a trbox according to proc */
static void
-trbox_populate (struct trbox *trb, const struct dictionary *dict,
- struct cmd_t_test *cmd)
+trbox_populate (struct trbox *trb, struct t_test_proc *proc)
{
- trb->populate (trb, dict, cmd);
+ trb->populate (trb, proc);
}
/* Submit and destroy a trbox */
-void
+static void
trbox_finalize (struct trbox *trb)
{
trb->finalize (trb);
}
/* Initialize the independent samples trbox */
-void
+static void
trbox_independent_samples_init (struct trbox *self,
- struct cmd_t_test *cmd UNUSED)
+ struct t_test_proc *proc)
{
- const int hsize=11;
- const int vsize=cmd->n_variables*2+3;
+ const int hsize = 11;
+ const int vsize = proc->n_vars * 2 + 3;
assert (self);
self->populate = trbox_independent_samples_populate;
- trbox_base_init (self,cmd->n_variables*2,hsize);
- tab_title (self->t,_ ("Independent Samples Test"));
- tab_hline (self->t,TAL_1,2,hsize-1,1);
- tab_vline (self->t,TAL_2,2,0,vsize-1);
- tab_vline (self->t,TAL_1,4,0,vsize-1);
- tab_box (self->t,-1,-1,-1,TAL_1, 2,1,hsize-2,vsize-1);
- tab_hline (self->t,TAL_1, hsize-2,hsize-1,2);
- tab_box (self->t,-1,-1,-1,TAL_1, hsize-2,2,hsize-1,vsize-1);
+ trbox_base_init (self, proc->n_vars * 2, hsize);
+ tab_title (self->t, _("Independent Samples Test"));
+ tab_hline (self->t, TAL_1, 2, hsize - 1, 1);
+ tab_vline (self->t, TAL_2, 2, 0, vsize - 1);
+ tab_vline (self->t, TAL_1, 4, 0, vsize - 1);
+ tab_box (self->t, -1, -1, -1, TAL_1, 2, 1, hsize - 2, vsize - 1);
+ tab_hline (self->t, TAL_1, hsize - 2, hsize - 1, 2);
+ tab_box (self->t, -1, -1, -1, TAL_1, hsize - 2, 2, hsize - 1, vsize - 1);
tab_joint_text (self->t, 2, 0, 3, 0,
- TAB_CENTER,_ ("Levene's Test for Equality of Variances"));
- tab_joint_text (self->t, 4,0,hsize-1,0,
- TAB_CENTER,_ ("t-test for Equality of Means"));
-
- tab_text (self->t,2,2, TAB_CENTER | TAT_TITLE,_ ("F"));
- tab_text (self->t,3,2, TAB_CENTER | TAT_TITLE,_ ("Sig."));
- tab_text (self->t,4,2, TAB_CENTER | TAT_TITLE,_ ("t"));
- tab_text (self->t,5,2, TAB_CENTER | TAT_TITLE,_ ("df"));
- tab_text (self->t,6,2, TAB_CENTER | TAT_TITLE,_ ("Sig. (2-tailed)"));
- tab_text (self->t,7,2, TAB_CENTER | TAT_TITLE,_ ("Mean Difference"));
- tab_text (self->t,8,2, TAB_CENTER | TAT_TITLE,_ ("Std. Error Difference"));
- tab_text (self->t,9,2, TAB_CENTER | TAT_TITLE,_ ("Lower"));
- tab_text (self->t,10,2, TAB_CENTER | TAT_TITLE,_ ("Upper"));
-
- tab_joint_text (self->t, 9, 1, 10, 1, TAB_CENTER | TAT_PRINTF,
- _ ("%g%% Confidence Interval of the Difference"),
- cmd->criteria*100.0);
-
+ TAB_CENTER, _("Levene's Test for Equality of Variances"));
+ tab_joint_text (self->t, 4, 0, hsize- 1, 0,
+ TAB_CENTER, _("t-test for Equality of Means"));
+
+ tab_text (self->t, 2, 2, TAB_CENTER | TAT_TITLE, _("F"));
+ tab_text (self->t, 3, 2, TAB_CENTER | TAT_TITLE, _("Sig."));
+ tab_text (self->t, 4, 2, TAB_CENTER | TAT_TITLE, _("t"));
+ tab_text (self->t, 5, 2, TAB_CENTER | TAT_TITLE, _("df"));
+ tab_text (self->t, 6, 2, TAB_CENTER | TAT_TITLE, _("Sig. (2-tailed)"));
+ tab_text (self->t, 7, 2, TAB_CENTER | TAT_TITLE, _("Mean Difference"));
+ tab_text (self->t, 8, 2, TAB_CENTER | TAT_TITLE, _("Std. Error Difference"));
+ tab_text (self->t, 9, 2, TAB_CENTER | TAT_TITLE, _("Lower"));
+ tab_text (self->t, 10, 2, TAB_CENTER | TAT_TITLE, _("Upper"));
+
+ tab_joint_text_format (self->t, 9, 1, 10, 1, TAB_CENTER,
+ _("%g%% Confidence Interval of the Difference"),
+ proc->criteria * 100.0);
}
/* Populate the independent samples trbox */
static void
trbox_independent_samples_populate (struct trbox *self,
- const struct dictionary *dict UNUSED,
- struct cmd_t_test *cmd)
+ struct t_test_proc *proc)
{
int i;
- assert (self);
- for (i=0; i < cmd->n_variables; ++i)
+ for (i = 0; i < proc->n_vars; i++)
{
- double p,q;
+ double p, q;
double t;
double df;
double std_err_diff;
double mean_diff;
- const struct variable *var = cmd->v_variables[i];
+ double se2;
+
+ const struct variable *var = proc->vars[i];
struct group_proc *grp_data = group_proc_get (var);
struct hsh_table *grp_hash = grp_data->group_hash;
- struct group_statistics *gs0 ;
- struct group_statistics *gs1 ;
+ struct group_statistics *gs0;
+ struct group_statistics *gs1;
union value search_val;
- if ( gp.criterion == CMP_LE )
- search_val.f = gp.v.critical_value - 1.0;
+ if (proc->criterion == CMP_LE)
+ search_val.f = proc->critical_value - 1.0;
else
- search_val = gp.v.g_value[0];
+ search_val = proc->g_value[0];
- gs0 = hsh_find (grp_hash, (void *) &search_val);
+ gs0 = hsh_find (grp_hash, &search_val);
assert (gs0);
- if ( gp.criterion == CMP_LE )
- search_val.f = gp.v.critical_value + 1.0;
+ if (proc->criterion == CMP_LE)
+ search_val.f = proc->critical_value + 1.0;
else
- search_val = gp.v.g_value[1];
+ search_val = proc->g_value[1];
- gs1 = hsh_find (grp_hash, (void *) &search_val);
+ gs1 = hsh_find (grp_hash, &search_val);
assert (gs1);
- tab_text (self->t, 0, i*2+3, TAB_LEFT, var_get_name (cmd->v_variables[i]));
-
- tab_text (self->t, 1, i*2+3, TAB_LEFT, _ ("Equal variances assumed"));
-
-
- tab_double (self->t, 2, i*2+3, TAB_CENTER, grp_data->levene, NULL);
+ tab_text (self->t, 0, i * 2 + 3, TAB_LEFT, var_get_name (proc->vars[i]));
+ tab_text (self->t, 1, i * 2 + 3, TAB_LEFT, _("Equal variances assumed"));
+ tab_double (self->t, 2, i * 2 + 3, TAB_CENTER, grp_data->levene, NULL);
/* Now work out the significance of the Levene test */
- df1 = 1; df2 = grp_data->ugs.n - 2;
+ df1 = 1;
+ df2 = grp_data->ugs.n - 2;
q = gsl_cdf_fdist_Q (grp_data->levene, df1, df2);
+ tab_double (self->t, 3, i * 2 + 3, TAB_CENTER, q, NULL);
- tab_double (self->t, 3, i*2+3, TAB_CENTER, q, NULL);
+ df = gs0->n + gs1->n - 2.0;
+ tab_double (self->t, 5, i * 2 + 3, TAB_RIGHT, df, NULL);
- df = gs0->n + gs1->n - 2.0 ;
- tab_double (self->t, 5, i*2+3, TAB_RIGHT, df, NULL);
+ pooled_variance = (gs0->n * pow2 (gs0->s_std_dev)
+ + gs1->n *pow2 (gs1->s_std_dev)) / df ;
- pooled_variance = ( (gs0->n )*pow2 (gs0->s_std_dev)
- +
- (gs1->n )*pow2 (gs1->s_std_dev)
- ) / df ;
+ t = (gs0->mean - gs1->mean) / sqrt (pooled_variance);
+ t /= sqrt ((gs0->n + gs1->n) / (gs0->n * gs1->n));
- t = (gs0->mean - gs1->mean) / sqrt (pooled_variance) ;
- t /= sqrt ((gs0->n + gs1->n)/ (gs0->n*gs1->n));
-
- tab_double (self->t, 4, i*2+3, TAB_RIGHT, t, NULL);
+ tab_double (self->t, 4, i * 2 + 3, TAB_RIGHT, t, NULL);
p = gsl_cdf_tdist_P (t, df);
q = gsl_cdf_tdist_Q (t, df);
- tab_double (self->t, 6, i*2+3, TAB_RIGHT, 2.0* (t>0?q:p), NULL);
+ tab_double (self->t, 6, i * 2 + 3, TAB_RIGHT, 2.0 * (t > 0 ? q : p),
+ NULL);
mean_diff = gs0->mean - gs1->mean;
- tab_double (self->t, 7, i*2+3, TAB_RIGHT, mean_diff, NULL);
-
+ tab_double (self->t, 7, i * 2 + 3, TAB_RIGHT, mean_diff, NULL);
- std_err_diff = sqrt ( pow2 (gs0->se_mean) + pow2 (gs1->se_mean));
- tab_double (self->t, 8, i*2+3, TAB_RIGHT, std_err_diff, NULL);
+ std_err_diff = sqrt (pow2 (gs0->se_mean) + pow2 (gs1->se_mean));
+ tab_double (self->t, 8, i * 2 + 3, TAB_RIGHT, std_err_diff, NULL);
/* Now work out the confidence interval */
- q = (1 - cmd->criteria)/2.0; /* 2-tailed test */
+ q = (1 - proc->criteria)/2.0; /* 2-tailed test */
- t = gsl_cdf_tdist_Qinv (q,df);
- tab_double (self->t, 9, i*2+3, TAB_RIGHT,
- mean_diff - t * std_err_diff, NULL);
+ t = gsl_cdf_tdist_Qinv (q, df);
+ tab_double (self->t, 9, i * 2 + 3, TAB_RIGHT,
+ mean_diff - t * std_err_diff, NULL);
- tab_double (self->t, 10, i*2+3, TAB_RIGHT,
- mean_diff + t * std_err_diff, NULL);
+ tab_double (self->t, 10, i * 2 + 3, TAB_RIGHT,
+ mean_diff + t * std_err_diff, NULL);
- {
- double se2;
/* Now for the \sigma_1 != \sigma_2 case */
- tab_text (self->t, 1, i*2+3+1,
- TAB_LEFT, _ ("Equal variances not assumed"));
-
+ tab_text (self->t, 1, i * 2 + 3 + 1,
+ TAB_LEFT, _("Equal variances not assumed"));
- se2 = (pow2 (gs0->s_std_dev)/ (gs0->n -1) ) +
- (pow2 (gs1->s_std_dev)/ (gs1->n -1) );
+ se2 = ((pow2 (gs0->s_std_dev) / (gs0->n - 1)) +
+ (pow2 (gs1->s_std_dev) / (gs1->n - 1)));
- t = mean_diff / sqrt (se2) ;
- tab_double (self->t, 4, i*2+3+1, TAB_RIGHT, t, NULL);
+ t = mean_diff / sqrt (se2);
+ tab_double (self->t, 4, i * 2 + 3 + 1, TAB_RIGHT, t, NULL);
- df = pow2 (se2) / (
- (pow2 (pow2 (gs0->s_std_dev)/ (gs0->n - 1 ))
- / (gs0->n -1 )
- )
- +
- (pow2 (pow2 (gs1->s_std_dev)/ (gs1->n - 1 ))
- / (gs1->n -1 )
- )
- ) ;
-
- tab_double (self->t, 5, i*2+3+1, TAB_RIGHT, df, NULL);
+ df = pow2 (se2) / ((pow2 (pow2 (gs0->s_std_dev) / (gs0->n - 1))
+ / (gs0->n - 1))
+ + (pow2 (pow2 (gs1->s_std_dev) / (gs1->n - 1))
+ / (gs1->n - 1)));
+ tab_double (self->t, 5, i * 2 + 3 + 1, TAB_RIGHT, df, NULL);
p = gsl_cdf_tdist_P (t, df);
q = gsl_cdf_tdist_Q (t, df);
- tab_double (self->t, 6, i*2+3+1, TAB_RIGHT, 2.0* (t>0?q:p), NULL);
+ tab_double (self->t, 6, i * 2 + 3 + 1, TAB_RIGHT, 2.0 * (t > 0 ? q : p),
+ NULL);
/* Now work out the confidence interval */
- q = (1 - cmd->criteria)/2.0; /* 2-tailed test */
+ q = (1 - proc->criteria) / 2.0; /* 2-tailed test */
t = gsl_cdf_tdist_Qinv (q, df);
- tab_double (self->t, 7, i*2+3+1, TAB_RIGHT, mean_diff, NULL);
-
-
- tab_double (self->t, 8, i*2+3+1, TAB_RIGHT, std_err_diff, NULL);
-
-
- tab_double (self->t, 9, i*2+3+1, TAB_RIGHT,
- mean_diff - t * std_err_diff, NULL);
-
- tab_double (self->t, 10, i*2+3+1, TAB_RIGHT,
- mean_diff + t * std_err_diff, NULL);
- }
+ tab_double (self->t, 7, i * 2 + 3 + 1, TAB_RIGHT, mean_diff, NULL);
+ tab_double (self->t, 8, i * 2 + 3 + 1, TAB_RIGHT, std_err_diff, NULL);
+ tab_double (self->t, 9, i * 2 + 3 + 1, TAB_RIGHT,
+ mean_diff - t * std_err_diff, NULL);
+ tab_double (self->t, 10, i * 2 + 3 + 1, TAB_RIGHT,
+ mean_diff + t * std_err_diff, NULL);
}
}
/* Initialize the paired samples trbox */
-void
-trbox_paired_init (struct trbox *self,
- struct cmd_t_test *cmd UNUSED)
+static void
+trbox_paired_init (struct trbox *self, struct t_test_proc *proc)
{
-
const int hsize=10;
- const int vsize=n_pairs+3;
+ const int vsize=proc->n_pairs+ 3;
self->populate = trbox_paired_populate;
- trbox_base_init (self,n_pairs,hsize);
- tab_title (self->t, _ ("Paired Samples Test"));
- tab_hline (self->t,TAL_1,2,6,1);
- tab_vline (self->t,TAL_2,2,0,vsize - 1);
- tab_joint_text (self->t,2,0,6,0,TAB_CENTER,_ ("Paired Differences"));
- tab_box (self->t,-1,-1,-1,TAL_1, 2,1,6,vsize-1);
- tab_box (self->t,-1,-1,-1,TAL_1, 6,0,hsize-1,vsize-1);
- tab_hline (self->t,TAL_1,5,6, 2);
- tab_vline (self->t,TAL_GAP,6,0,1);
-
- tab_joint_text (self->t, 5, 1, 6, 1, TAB_CENTER | TAT_PRINTF,
- _ ("%g%% Confidence Interval of the Difference"),
- cmd->criteria*100.0);
-
- tab_text (self->t, 2, 2, TAB_CENTER | TAT_TITLE, _ ("Mean"));
- tab_text (self->t, 3, 2, TAB_CENTER | TAT_TITLE, _ ("Std. Deviation"));
- tab_text (self->t, 4, 2, TAB_CENTER | TAT_TITLE, _ ("Std. Error Mean"));
- tab_text (self->t, 5, 2, TAB_CENTER | TAT_TITLE, _ ("Lower"));
- tab_text (self->t, 6, 2, TAB_CENTER | TAT_TITLE, _ ("Upper"));
- tab_text (self->t, 7, 2, TAB_CENTER | TAT_TITLE, _ ("t"));
- tab_text (self->t, 8, 2, TAB_CENTER | TAT_TITLE, _ ("df"));
- tab_text (self->t, 9, 2, TAB_CENTER | TAT_TITLE, _ ("Sig. (2-tailed)"));
+ trbox_base_init (self, proc->n_pairs, hsize);
+ tab_title (self->t, _("Paired Samples Test"));
+ tab_hline (self->t, TAL_1, 2, 6, 1);
+ tab_vline (self->t, TAL_2, 2, 0, vsize - 1);
+ tab_joint_text (self->t, 2, 0, 6, 0, TAB_CENTER, _("Paired Differences"));
+ tab_box (self->t, -1, -1, -1, TAL_1, 2, 1, 6, vsize - 1);
+ tab_box (self->t, -1, -1, -1, TAL_1, 6, 0, hsize - 1, vsize - 1);
+ tab_hline (self->t, TAL_1, 5, 6, 2);
+ tab_vline (self->t, TAL_GAP, 6, 0, 1);
+
+ tab_joint_text_format (self->t, 5, 1, 6, 1, TAB_CENTER,
+ _("%g%% Confidence Interval of the Difference"),
+ proc->criteria*100.0);
+
+ tab_text (self->t, 2, 2, TAB_CENTER | TAT_TITLE, _("Mean"));
+ tab_text (self->t, 3, 2, TAB_CENTER | TAT_TITLE, _("Std. Deviation"));
+ tab_text (self->t, 4, 2, TAB_CENTER | TAT_TITLE, _("Std. Error Mean"));
+ tab_text (self->t, 5, 2, TAB_CENTER | TAT_TITLE, _("Lower"));
+ tab_text (self->t, 6, 2, TAB_CENTER | TAT_TITLE, _("Upper"));
+ tab_text (self->t, 7, 2, TAB_CENTER | TAT_TITLE, _("t"));
+ tab_text (self->t, 8, 2, TAB_CENTER | TAT_TITLE, _("df"));
+ tab_text (self->t, 9, 2, TAB_CENTER | TAT_TITLE, _("Sig. (2-tailed)"));
}
/* Populate the paired samples trbox */
static void
trbox_paired_populate (struct trbox *trb,
- const struct dictionary *dict,
- struct cmd_t_test *cmd UNUSED)
+ struct t_test_proc *proc)
{
int i;
- const struct variable *wv = dict_get_weight (dict);
- const struct fmt_spec *wfmt = wv ? var_get_print_format (wv) : &F_8_0;
-
- for (i=0; i < n_pairs; ++i)
+ for (i = 0; i < proc->n_pairs; i++)
{
- double p,q;
+ struct pair *pair = &proc->pairs[i];
+ double p, q;
double se_mean;
- double n = pairs[i].n;
+ double n = pair->n;
double t;
double df = n - 1;
- tab_text (trb->t, 0, i+3, TAB_LEFT | TAT_PRINTF, _ ("Pair %d"),i);
-
- tab_text (trb->t, 1, i+3, TAB_LEFT | TAT_PRINTF, "%s - %s",
- var_get_name (pairs[i].v[0]),
- var_get_name (pairs[i].v[1]));
-
- tab_double (trb->t, 2, i+3, TAB_RIGHT, pairs[i].mean_diff, NULL);
-
- tab_double (trb->t, 3, i+3, TAB_RIGHT, pairs[i].std_dev_diff, NULL);
+ tab_text_format (trb->t, 0, i + 3, TAB_LEFT, _("Pair %d"), i);
+ tab_text_format (trb->t, 1, i + 3, TAB_LEFT, "%s - %s",
+ var_get_name (pair->v[0]),
+ var_get_name (pair->v[1]));
+ tab_double (trb->t, 2, i + 3, TAB_RIGHT, pair->mean_diff, NULL);
+ tab_double (trb->t, 3, i + 3, TAB_RIGHT, pair->std_dev_diff, NULL);
/* SE Mean */
- se_mean = pairs[i].std_dev_diff / sqrt (n) ;
- tab_double (trb->t, 4, i+3, TAB_RIGHT, se_mean, NULL);
+ se_mean = pair->std_dev_diff / sqrt (n);
+ tab_double (trb->t, 4, i + 3, TAB_RIGHT, se_mean, NULL);
/* Now work out the confidence interval */
- q = (1 - cmd->criteria)/2.0; /* 2-tailed test */
+ q = (1 - proc->criteria) / 2.0; /* 2-tailed test */
t = gsl_cdf_tdist_Qinv (q, df);
- tab_double (trb->t, 5, i+3, TAB_RIGHT,
- pairs[i].mean_diff - t * se_mean , NULL);
+ tab_double (trb->t, 5, i + 3, TAB_RIGHT,
+ pair->mean_diff - t * se_mean, NULL);
+ tab_double (trb->t, 6, i + 3, TAB_RIGHT,
+ pair->mean_diff + t * se_mean, NULL);
- tab_double (trb->t, 6, i+3, TAB_RIGHT,
- pairs[i].mean_diff + t * se_mean , NULL);
+ t = ((pair->mean[0] - pair->mean[1])
+ / sqrt ((pow2 (pair->s_std_dev[0]) + pow2 (pair->s_std_dev[1])
+ - (2 * pair->correlation
+ * pair->s_std_dev[0] * pair->s_std_dev[1]))
+ / (n - 1)));
- t = (pairs[i].mean[0] - pairs[i].mean[1])
- / sqrt (
- ( pow2 (pairs[i].s_std_dev[0]) + pow2 (pairs[i].s_std_dev[1]) -
- 2 * pairs[i].correlation *
- pairs[i].s_std_dev[0] * pairs[i].s_std_dev[1] )
- / (n - 1)
- );
-
- tab_double (trb->t, 7, i+3, TAB_RIGHT, t, NULL);
+ tab_double (trb->t, 7, i + 3, TAB_RIGHT, t, NULL);
/* Degrees of freedom */
- tab_double (trb->t, 8, i+3, TAB_RIGHT, df, wfmt);
+ tab_double (trb->t, 8, i + 3, TAB_RIGHT, df, &proc->weight_format);
- p = gsl_cdf_tdist_P (t, df);
- q = gsl_cdf_tdist_P (t, df);
+ p = gsl_cdf_tdist_P (t,df);
+ q = gsl_cdf_tdist_Q (t,df);
- tab_double (trb->t, 9, i+3, TAB_RIGHT, 2.0* (t>0?q:p), NULL);
-
+ tab_double (trb->t, 9, i + 3, TAB_RIGHT, 2.0 * (t > 0 ? q : p), NULL);
}
}
/* Initialize the one sample trbox */
-void
-trbox_one_sample_init (struct trbox *self, struct cmd_t_test *cmd )
+static void
+trbox_one_sample_init (struct trbox *self, struct t_test_proc *proc)
{
- const int hsize=7;
- const int vsize=cmd->n_variables+3;
+ const int hsize = 7;
+ const int vsize = proc->n_vars + 3;
self->populate = trbox_one_sample_populate;
- trbox_base_init (self, cmd->n_variables,hsize);
- tab_title (self->t, _ ("One-Sample Test"));
+ trbox_base_init (self, proc->n_vars, hsize);
+ tab_title (self->t, _("One-Sample Test"));
tab_hline (self->t, TAL_1, 1, hsize - 1, 1);
tab_vline (self->t, TAL_2, 1, 0, vsize - 1);
- tab_joint_text (self->t, 1, 0, hsize-1,0, TAB_CENTER | TAT_PRINTF,
- _ ("Test Value = %f"), cmd->n_testval[0]);
-
- tab_box (self->t, -1, -1, -1, TAL_1, 1,1,hsize-1,vsize-1);
+ tab_joint_text_format (self->t, 1, 0, hsize - 1, 0, TAB_CENTER,
+ _("Test Value = %f"), proc->testval);
+ tab_box (self->t, -1, -1, -1, TAL_1, 1, 1, hsize - 1, vsize - 1);
- tab_joint_text (self->t,5,1,6,1,TAB_CENTER | TAT_PRINTF,
- _ ("%g%% Confidence Interval of the Difference"),
- cmd->criteria*100.0);
- tab_vline (self->t,TAL_GAP,6,1,1);
- tab_hline (self->t,TAL_1,5,6,2);
- tab_text (self->t, 1, 2, TAB_CENTER | TAT_TITLE, _ ("t"));
- tab_text (self->t, 2, 2, TAB_CENTER | TAT_TITLE, _ ("df"));
- tab_text (self->t, 3, 2, TAB_CENTER | TAT_TITLE, _ ("Sig. (2-tailed)"));
- tab_text (self->t, 4, 2, TAB_CENTER | TAT_TITLE, _ ("Mean Difference"));
- tab_text (self->t, 5, 2, TAB_CENTER | TAT_TITLE, _ ("Lower"));
- tab_text (self->t, 6, 2, TAB_CENTER | TAT_TITLE, _ ("Upper"));
+ tab_joint_text_format (self->t, 5, 1, 6, 1, TAB_CENTER,
+ _("%g%% Confidence Interval of the Difference"),
+ proc->criteria * 100.0);
+ tab_vline (self->t, TAL_GAP, 6, 1, 1);
+ tab_hline (self->t, TAL_1, 5, 6, 2);
+ tab_text (self->t, 1, 2, TAB_CENTER | TAT_TITLE, _("t"));
+ tab_text (self->t, 2, 2, TAB_CENTER | TAT_TITLE, _("df"));
+ tab_text (self->t, 3, 2, TAB_CENTER | TAT_TITLE, _("Sig. (2-tailed)"));
+ tab_text (self->t, 4, 2, TAB_CENTER | TAT_TITLE, _("Mean Difference"));
+ tab_text (self->t, 5, 2, TAB_CENTER | TAT_TITLE, _("Lower"));
+ tab_text (self->t, 6, 2, TAB_CENTER | TAT_TITLE, _("Upper"));
}
-
/* Populate the one sample trbox */
static void
-trbox_one_sample_populate (struct trbox *trb,
- const struct dictionary *dict,
- struct cmd_t_test *cmd)
+trbox_one_sample_populate (struct trbox *trb, struct t_test_proc *proc)
{
int i;
- const struct variable *wv = dict_get_weight (dict);
- const struct fmt_spec *wfmt = wv ? var_get_print_format (wv) : &F_8_0;
-
assert (trb->t);
- for (i=0; i < cmd->n_variables; ++i)
+ for (i = 0; i < proc->n_vars; i++)
{
double t;
- double p,q;
+ double p, q;
double df;
- struct group_statistics *gs = &group_proc_get (cmd->v_variables[i])->ugs;
-
+ struct group_statistics *gs = &group_proc_get (proc->vars[i])->ugs;
- tab_text (trb->t, 0, i+3, TAB_LEFT, var_get_name (cmd->v_variables[i]));
+ tab_text (trb->t, 0, i + 3, TAB_LEFT, var_get_name (proc->vars[i]));
- t = (gs->mean - cmd->n_testval[0] ) * sqrt (gs->n) / gs->std_dev ;
+ t = (gs->mean - proc->testval) * sqrt (gs->n) / gs->std_dev;
- tab_double (trb->t, 1, i+3, TAB_RIGHT, t, NULL);
+ tab_double (trb->t, 1, i + 3, TAB_RIGHT, t, NULL);
/* degrees of freedom */
df = gs->n - 1;
- tab_double (trb->t, 2, i+3, TAB_RIGHT, df, wfmt);
+ tab_double (trb->t, 2, i + 3, TAB_RIGHT, df, &proc->weight_format);
p = gsl_cdf_tdist_P (t, df);
q = gsl_cdf_tdist_Q (t, df);
/* Multiply by 2 to get 2-tailed significance, makeing sure we've got
the correct tail*/
- tab_double (trb->t, 3, i+3, TAB_RIGHT, 2.0* (t>0?q:p), NULL);
+ tab_double (trb->t, 3, i + 3, TAB_RIGHT, 2.0 * (t > 0 ? q : p), NULL);
+ tab_double (trb->t, 4, i + 3, TAB_RIGHT, gs->mean_diff, NULL);
- tab_double (trb->t, 4, i+3, TAB_RIGHT, gs->mean_diff, NULL);
-
- q = (1 - cmd->criteria)/2.0; /* 2-tailed test */
+ q = (1 - proc->criteria) / 2.0; /* 2-tailed test */
t = gsl_cdf_tdist_Qinv (q, df);
- tab_double (trb->t, 5, i+3, TAB_RIGHT,
+ tab_double (trb->t, 5, i + 3, TAB_RIGHT,
gs->mean_diff - t * gs->se_mean, NULL);
-
- tab_double (trb->t, 6, i+3, TAB_RIGHT,
+ tab_double (trb->t, 6, i + 3, TAB_RIGHT,
gs->mean_diff + t * gs->se_mean, NULL);
}
}
/* Base initializer for the generalized trbox */
-void
+static void
trbox_base_init (struct trbox *self, size_t data_rows, int cols)
{
const size_t rows = 3 + data_rows;
self->finalize = trbox_base_finalize;
self->t = tab_create (cols, rows, 0);
- tab_headers (self->t,0,0,3,0);
- tab_box (self->t, TAL_2, TAL_2, TAL_0, TAL_0, 0, 0, cols -1, rows -1);
- tab_hline (self->t, TAL_2,0,cols-1,3);
- tab_dim (self->t, tab_natural_dimensions);
+ tab_headers (self->t, 0, 0, 3, 0);
+ tab_box (self->t, TAL_2, TAL_2, TAL_0, TAL_0, 0, 0, cols - 1, rows - 1);
+ tab_hline (self->t, TAL_2, 0, cols- 1, 3);
+ tab_dim (self->t, tab_natural_dimensions, NULL);
}
-
/* Base finalizer for the trbox */
-void
+static void
trbox_base_finalize (struct trbox *trb)
{
tab_submit (trb->t);
}
-
-/* Create , populate and submit the Paired Samples Correlation box */
+/* Create, populate and submit the Paired Samples Correlation box */
static void
-pscbox (const struct dictionary *dict)
+pscbox (struct t_test_proc *proc)
{
- const struct variable *wv = dict_get_weight (dict);
- const struct fmt_spec *wfmt = wv ? var_get_print_format (wv) : &F_8_0;
-
- const int rows = 1 + n_pairs;
- const int cols = 5;
+ const int rows=1+proc->n_pairs;
+ const int cols=5;
int i;
struct tab_table *table;
- table = tab_create (cols,rows,0);
+ table = tab_create (cols, rows, 0);
tab_columns (table, SOM_COL_DOWN, 1);
- tab_headers (table,0,0,1,0);
- tab_box (table, TAL_2, TAL_2, TAL_0, TAL_1, 0, 0, cols -1, rows -1 );
+ tab_headers (table, 0, 0, 1, 0);
+ tab_box (table, TAL_2, TAL_2, TAL_0, TAL_1, 0, 0, cols - 1, rows - 1);
tab_hline (table, TAL_2, 0, cols - 1, 1);
tab_vline (table, TAL_2, 2, 0, rows - 1);
- tab_dim (table, tab_natural_dimensions);
- tab_title (table, _ ("Paired Samples Correlations"));
+ tab_dim (table, tab_natural_dimensions, NULL);
+ tab_title (table, _("Paired Samples Correlations"));
/* column headings */
- tab_text (table, 2,0, TAB_CENTER | TAT_TITLE, _ ("N"));
- tab_text (table, 3,0, TAB_CENTER | TAT_TITLE, _ ("Correlation"));
- tab_text (table, 4,0, TAB_CENTER | TAT_TITLE, _ ("Sig."));
+ tab_text (table, 2, 0, TAB_CENTER | TAT_TITLE, _("N"));
+ tab_text (table, 3, 0, TAB_CENTER | TAT_TITLE, _("Correlation"));
+ tab_text (table, 4, 0, TAB_CENTER | TAT_TITLE, _("Sig."));
- for (i=0; i < n_pairs; ++i)
+ for (i = 0; i < proc->n_pairs; i++)
{
- double p,q;
-
- double df = pairs[i].n -2;
-
- double correlation_t =
- pairs[i].correlation * sqrt (df) /
- sqrt (1 - pow2 (pairs[i].correlation));
-
+ struct pair *pair = &proc->pairs[i];
+ double p, q;
+ double df = pair->n -2;
+ double correlation_t = (pair->correlation * sqrt (df) /
+ sqrt (1 - pow2 (pair->correlation)));
/* row headings */
- tab_text (table, 0,i+1, TAB_LEFT | TAT_TITLE | TAT_PRINTF,
- _ ("Pair %d"), i);
-
- tab_text (table, 1,i+1, TAB_LEFT | TAT_TITLE | TAT_PRINTF,
- _ ("%s & %s"),
- var_get_name (pairs[i].v[0]),
- var_get_name (pairs[i].v[1]));
-
+ tab_text_format (table, 0, i + 1, TAB_LEFT | TAT_TITLE,
+ _("Pair %d"), i);
+ tab_text_format (table, 1, i + 1, TAB_LEFT | TAT_TITLE,
+ _("%s & %s"),
+ var_get_name (pair->v[0]),
+ var_get_name (pair->v[1]));
/* row data */
- tab_double (table, 2, i+1, TAB_RIGHT, pairs[i].n, wfmt);
- tab_double (table, 3, i+1, TAB_RIGHT, pairs[i].correlation, NULL);
+ tab_double (table, 2, i + 1, TAB_RIGHT, pair->n, &proc->weight_format);
+ tab_double (table, 3, i + 1, TAB_RIGHT, pair->correlation, NULL);
p = gsl_cdf_tdist_P (correlation_t, df);
q = gsl_cdf_tdist_Q (correlation_t, df);
-
- tab_double (table, 4, i+1, TAB_RIGHT, 2.0* (correlation_t>0?q:p), NULL);
+ tab_double (table, 4, i + 1, TAB_RIGHT,
+ 2.0 * (correlation_t > 0 ? q : p), NULL);
}
tab_submit (table);
}
-
-
-
-
+\f
/* Calculation Implementation */
-/* Per case calculations common to all variants of the T test */
-static int
+/* Calculations common to all variants of the T test. */
+static void
common_calc (const struct dictionary *dict,
- const struct ccase *c,
- void *_cmd,
- enum mv_class exclude)
+ struct t_test_proc *proc,
+ struct casereader *reader)
{
+ struct ccase *c;
int i;
- struct cmd_t_test *cmd = (struct cmd_t_test *)_cmd;
-
- double weight = dict_get_case_weight (dict, c, NULL);
-
-
- /* Listwise has to be implicit if the independent variable is missing ?? */
- if ( cmd->sbc_groups )
- {
- if (var_is_value_missing (indep_var, case_data (c, indep_var), exclude))
- return 0;
- }
- for (i = 0; i < cmd->n_variables ; ++i)
+ for (i = 0; i < proc->n_vars; i++)
{
- const struct variable *v = cmd->v_variables[i];
- const union value *val = case_data (c, v);
-
- if (!var_is_value_missing (v, val, exclude))
- {
- struct group_statistics *gs;
- gs = &group_proc_get (v)->ugs;
-
- gs->n += weight;
- gs->sum += weight * val->f;
- gs->ssq += weight * val->f * val->f;
- }
+ struct group_statistics *gs = &group_proc_get (proc->vars[i])->ugs;
+ gs->sum = 0;
+ gs->n = 0;
+ gs->ssq = 0;
+ gs->sum_diff = 0;
}
- return 0;
-}
-
-/* Pre calculations common to all variants of the T test */
-static void
-common_precalc ( struct cmd_t_test *cmd )
-{
- int i=0;
- for (i=0; i< cmd->n_variables ; ++i)
+ for (; (c = casereader_read (reader)) != NULL; case_unref (c))
{
- struct group_statistics *gs;
- gs= &group_proc_get (cmd->v_variables[i])->ugs;
-
- gs->sum=0;
- gs->n=0;
- gs->ssq=0;
- gs->sum_diff=0;
+ double weight = dict_get_case_weight (dict, c, NULL);
+
+ /* Listwise has to be implicit if the independent variable
+ is missing ?? */
+ if (proc->mode == T_IND_SAMPLES)
+ {
+ if (var_is_value_missing (proc->indep_var,
+ case_data (c, proc->indep_var),
+ proc->exclude))
+ continue;
+ }
+
+ for (i = 0; i < proc->n_vars; i++)
+ {
+ const struct variable *v = proc->vars[i];
+ const union value *val = case_data (c, v);
+
+ if (!var_is_value_missing (v, val, proc->exclude))
+ {
+ struct group_statistics *gs;
+ gs = &group_proc_get (v)->ugs;
+
+ gs->n += weight;
+ gs->sum += weight * val->f;
+ gs->ssq += weight * pow2 (val->f);
+ }
+ }
}
-}
+ casereader_destroy (reader);
-/* Post calculations common to all variants of the T test */
-void
-common_postcalc (struct cmd_t_test *cmd)
-{
- int i=0;
-
- for (i=0; i< cmd->n_variables ; ++i)
+ for (i = 0; i < proc->n_vars; i++)
{
- struct group_statistics *gs;
- gs= &group_proc_get (cmd->v_variables[i])->ugs;
-
- gs->mean=gs->sum / gs->n;
- gs->s_std_dev= sqrt (
- ( (gs->ssq / gs->n ) - gs->mean * gs->mean )
- ) ;
-
- gs->std_dev= sqrt (
- gs->n/ (gs->n-1) *
- ( (gs->ssq / gs->n ) - gs->mean * gs->mean )
- ) ;
+ struct group_statistics *gs = &group_proc_get (proc->vars[i])->ugs;
+ gs->mean = gs->sum / gs->n;
+ gs->s_std_dev = sqrt (((gs->ssq / gs->n) - pow2 (gs->mean)));
+ gs->std_dev = sqrt (gs->n / (gs->n- 1)
+ * ((gs->ssq / gs->n) - pow2 (gs->mean)));
gs->se_mean = gs->std_dev / sqrt (gs->n);
- gs->mean_diff= gs->sum_diff / gs->n;
+ gs->mean_diff = gs->sum_diff / gs->n;
}
}
-/* Per case calculations for one sample t test */
+/* Calculations for one sample T test. */
static int
-one_sample_calc (const struct dictionary *dict,
- const struct ccase *c, void *cmd_,
- enum mv_class exclude)
+one_sample_calc (const struct dictionary *dict, struct t_test_proc *proc,
+ struct casereader *reader)
{
+ struct ccase *c;
int i;
- struct cmd_t_test *cmd = (struct cmd_t_test *)cmd_;
-
- double weight = dict_get_case_weight (dict, c, NULL);
-
-
- for (i=0; i< cmd->n_variables ; ++i)
+ for (i = 0; i < proc->n_vars; i++)
{
- struct group_statistics *gs;
- const struct variable *v = cmd->v_variables[i];
- const union value *val = case_data (c, v);
-
- gs= &group_proc_get (cmd->v_variables[i])->ugs;
-
- if (!var_is_value_missing (v, val, exclude))
- gs->sum_diff += weight * (val->f - cmd->n_testval[0]);
+ struct group_statistics *gs = &group_proc_get (proc->vars[i])->ugs;
+ gs->sum_diff = 0;
}
- return 0;
-}
-
-/* Pre calculations for one sample t test */
-static void
-one_sample_precalc ( struct cmd_t_test *cmd )
-{
- int i=0;
-
- for (i=0; i< cmd->n_variables ; ++i)
+ for (; (c = casereader_read (reader)) != NULL; case_unref (c))
{
- struct group_statistics *gs;
- gs= &group_proc_get (cmd->v_variables[i])->ugs;
-
- gs->sum_diff=0;
+ double weight = dict_get_case_weight (dict, c, NULL);
+ for (i = 0; i < proc->n_vars; i++)
+ {
+ const struct variable *v = proc->vars[i];
+ struct group_statistics *gs = &group_proc_get (v)->ugs;
+ const union value *val = case_data (c, v);
+ if (!var_is_value_missing (v, val, proc->exclude))
+ gs->sum_diff += weight * (val->f - proc->testval);
+ }
}
-}
-/* Post calculations for one sample t test */
-static void
-one_sample_postcalc (struct cmd_t_test *cmd)
-{
- int i=0;
-
- for (i=0; i< cmd->n_variables ; ++i)
+ for (i = 0; i < proc->n_vars; i++)
{
- struct group_statistics *gs;
- gs= &group_proc_get (cmd->v_variables[i])->ugs;
-
- gs->mean_diff = gs->sum_diff / gs->n ;
+ struct group_statistics *gs = &group_proc_get (proc->vars[i])->ugs;
+ gs->mean_diff = gs->sum_diff / gs->n;
}
-}
-
+ casereader_destroy (reader);
-static void
-paired_precalc (struct cmd_t_test *cmd UNUSED)
-{
- int i;
-
- for (i=0; i < n_pairs ; ++i )
- {
- pairs[i].n = 0;
- pairs[i].sum[0] = 0; pairs[i].sum[1] = 0;
- pairs[i].ssq[0] = 0; pairs[i].ssq[1] = 0;
- pairs[i].sum_of_prod = 0;
- pairs[i].correlation = 0;
- pairs[i].sum_of_diffs = 0;
- pairs[i].ssq_diffs = 0;
- }
-
+ return 0;
}
-
static int
-paired_calc (const struct dictionary *dict, const struct ccase *c,
- struct cmd_t_test *cmd UNUSED, enum mv_class exclude)
+paired_calc (const struct dictionary *dict, struct t_test_proc *proc,
+ struct casereader *reader)
{
+ struct ccase *c;
int i;
- double weight = dict_get_case_weight (dict, c, NULL);
-
- for (i=0; i < n_pairs ; ++i )
+ for (i = 0; i < proc->n_pairs; i++)
{
- const struct variable *v0 = pairs[i].v[0];
- const struct variable *v1 = pairs[i].v[1];
-
- const union value *val0 = case_data (c, v0);
- const union value *val1 = case_data (c, v1);
-
- if (!var_is_value_missing (v0, val0, exclude) &&
- !var_is_value_missing (v1, val1, exclude))
- {
- pairs[i].n += weight;
- pairs[i].sum[0] += weight * val0->f;
- pairs[i].sum[1] += weight * val1->f;
-
- pairs[i].ssq[0] += weight * pow2 (val0->f);
- pairs[i].ssq[1] += weight * pow2 (val1->f);
-
- pairs[i].sum_of_prod += weight * val0->f * val1->f ;
-
- pairs[i].sum_of_diffs += weight * ( val0->f - val1->f ) ;
- pairs[i].ssq_diffs += weight * pow2 (val0->f - val1->f);
- }
+ struct pair *pair = &proc->pairs[i];
+ pair->n = 0;
+ pair->sum[0] = pair->sum[1] = 0;
+ pair->ssq[0] = pair->ssq[1] = 0;
+ pair->sum_of_prod = 0;
+ pair->correlation = 0;
+ pair->sum_of_diffs = 0;
+ pair->ssq_diffs = 0;
}
- return 0;
-}
-
-static void
-paired_postcalc (struct cmd_t_test *cmd UNUSED)
-{
- int i;
+ for (; (c = casereader_read (reader)) != NULL; case_unref (c))
+ {
+ double weight = dict_get_case_weight (dict, c, NULL);
+ for (i = 0; i < proc->n_pairs; i++)
+ {
+ struct pair *pair = &proc->pairs[i];
+ const struct variable *v0 = pair->v[0];
+ const struct variable *v1 = pair->v[1];
+
+ const union value *val0 = case_data (c, v0);
+ const union value *val1 = case_data (c, v1);
+
+ if (!var_is_value_missing (v0, val0, proc->exclude)
+ && !var_is_value_missing (v1, val1, proc->exclude))
+ {
+ pair->n += weight;
+ pair->sum[0] += weight * val0->f;
+ pair->sum[1] += weight * val1->f;
+ pair->ssq[0] += weight * pow2 (val0->f);
+ pair->ssq[1] += weight * pow2 (val1->f);
+ pair->sum_of_prod += weight * val0->f * val1->f;
+ pair->sum_of_diffs += weight * (val0->f - val1->f);
+ pair->ssq_diffs += weight * pow2 (val0->f - val1->f);
+ }
+ }
+ }
- for (i=0; i < n_pairs ; ++i )
+ for (i = 0; i < proc->n_pairs; i++)
{
+ struct pair *pair = &proc->pairs[i];
+ const double n = pair->n;
int j;
- const double n = pairs[i].n;
- for (j=0; j < 2 ; ++j)
+ for (j=0; j < 2; j++)
{
- pairs[i].mean[j] = pairs[i].sum[j] / n ;
- pairs[i].s_std_dev[j] = sqrt ((pairs[i].ssq[j] / n -
- pow2 (pairs[i].mean[j]))
- );
-
- pairs[i].std_dev[j] = sqrt (n/ (n-1)* (pairs[i].ssq[j] / n -
- pow2 (pairs[i].mean[j]))
- );
+ pair->mean[j] = pair->sum[j] / n;
+ pair->s_std_dev[j] = sqrt ((pair->ssq[j] / n
+ - pow2 (pair->mean[j])));
+ pair->std_dev[j] = sqrt (n / (n- 1) * (pair->ssq[j] / n
+ - pow2 (pair->mean[j])));
}
- pairs[i].correlation = pairs[i].sum_of_prod / pairs[i].n -
- pairs[i].mean[0] * pairs[i].mean[1] ;
+ pair->correlation = (pair->sum_of_prod / pair->n
+ - pair->mean[0] * pair->mean[1]);
/* correlation now actually contains the covariance */
+ pair->correlation /= pair->std_dev[0] * pair->std_dev[1];
+ pair->correlation *= pair->n / (pair->n - 1);
- pairs[i].correlation /= pairs[i].std_dev[0] * pairs[i].std_dev[1];
- pairs[i].correlation *= pairs[i].n / ( pairs[i].n - 1 );
-
- pairs[i].mean_diff = pairs[i].sum_of_diffs / n ;
-
- pairs[i].std_dev_diff = sqrt ( n / (n - 1) * (
- ( pairs[i].ssq_diffs / n )
- -
- pow2 (pairs[i].mean_diff )
- ) );
+ pair->mean_diff = pair->sum_of_diffs / n;
+ pair->std_dev_diff = sqrt (n / (n - 1) * ((pair->ssq_diffs / n)
+ - pow2 (pair->mean_diff)));
}
+
+ casereader_destroy (reader);
+ return 0;
}
-static void
-group_precalc (struct cmd_t_test *cmd )
+static int
+group_calc (const struct dictionary *dict, struct t_test_proc *proc,
+ struct casereader *reader)
{
+ struct ccase *c;
int i;
- int j;
- for (i=0; i< cmd->n_variables ; ++i)
+ for (i = 0; i < proc->n_vars; i++)
{
- struct group_proc *ttpr = group_proc_get (cmd->v_variables[i]);
+ struct group_proc *ttpr = group_proc_get (proc->vars[i]);
+ int j;
/* There's always 2 groups for a T - TEST */
ttpr->n_groups = 2;
-
- gp.indep_width = var_get_width (indep_var);
-
ttpr->group_hash = hsh_create (2,
- (hsh_compare_func *) compare_group_binary,
- (hsh_hash_func *) hash_group_binary,
- (hsh_free_func *) free_group,
- (void *) &gp );
+ (hsh_compare_func *) compare_group_binary,
+ (hsh_hash_func *) hash_group_binary,
+ (hsh_free_func *) free_group,
+ proc);
- for (j=0 ; j < 2 ; ++j)
+ for (j = 0; j < 2; j++)
{
struct group_statistics *gs = xmalloc (sizeof *gs);
-
gs->sum = 0;
gs->n = 0;
gs->ssq = 0;
-
- if ( gp.criterion == CMP_EQ )
- {
- gs->id = gp.v.g_value[j];
- }
+ if (proc->criterion == CMP_EQ)
+ gs->id = proc->g_value[j];
else
{
- if ( j == 0 )
- gs->id.f = gp.v.critical_value - 1.0 ;
+ if (j == 0)
+ gs->id.f = proc->critical_value - 1.0;
else
- gs->id.f = gp.v.critical_value + 1.0 ;
+ gs->id.f = proc->critical_value + 1.0;
}
- hsh_insert ( ttpr->group_hash, (void *) gs );
+ hsh_insert (ttpr->group_hash, gs);
}
}
-}
-
-static int
-group_calc (const struct dictionary *dict,
- const struct ccase *c, struct cmd_t_test *cmd,
- enum mv_class exclude)
-{
- int i;
-
- const double weight = dict_get_case_weight (dict, c, NULL);
-
- const union value *gv;
-
- if (var_is_value_missing (indep_var, case_data (c, indep_var), exclude))
- return 0;
-
- gv = case_data (c, indep_var);
-
- for (i=0; i< cmd->n_variables ; ++i)
+ for (; (c = casereader_read (reader)) != NULL; case_unref (c))
{
- const struct variable *var = cmd->v_variables[i];
- const union value *val = case_data (c, var);
- struct hsh_table *grp_hash = group_proc_get (var)->group_hash;
- struct group_statistics *gs;
-
- gs = hsh_find (grp_hash, (void *) gv);
-
- /* If the independent variable doesn't match either of the values
- for this case then move on to the next case */
- if ( ! gs )
- return 0;
-
- if (!var_is_value_missing (var, val, exclude))
- {
- gs->n += weight;
- gs->sum += weight * val->f;
- gs->ssq += weight * pow2 (val->f);
- }
+ const double weight = dict_get_case_weight (dict, c, NULL);
+ const union value *gv;
+
+ if (var_is_value_missing (proc->indep_var,
+ case_data (c, proc->indep_var), proc->exclude))
+ continue;
+
+ gv = case_data (c, proc->indep_var);
+ for (i = 0; i < proc->n_vars; i++)
+ {
+ const struct variable *var = proc->vars[i];
+ const union value *val = case_data (c, var);
+ struct hsh_table *grp_hash = group_proc_get (var)->group_hash;
+ struct group_statistics *gs = hsh_find (grp_hash, gv);
+
+ /* If the independent variable doesn't match either of the values
+ for this case then move on to the next case. */
+ if (gs == NULL)
+ break;
+
+ if (!var_is_value_missing (var, val, proc->exclude))
+ {
+ gs->n += weight;
+ gs->sum += weight * val->f;
+ gs->ssq += weight * pow2 (val->f);
+ }
+ }
}
- return 0;
-}
-
-
-static void
-group_postcalc ( struct cmd_t_test *cmd )
-{
- int i;
-
- for (i = 0; i < cmd->n_variables ; ++i)
+ for (i = 0; i < proc->n_vars; i++)
{
- const struct variable *var = cmd->v_variables[i];
+ const struct variable *var = proc->vars[i];
struct hsh_table *grp_hash = group_proc_get (var)->group_hash;
struct hsh_iterator g;
struct group_statistics *gs;
- int count=0;
+ int count = 0;
- for (gs = hsh_first (grp_hash,&g);
- gs != 0;
- gs = hsh_next (grp_hash,&g))
+ for (gs = hsh_first (grp_hash, &g); gs != NULL;
+ gs = hsh_next (grp_hash, &g))
{
gs->mean = gs->sum / gs->n;
-
- gs->s_std_dev= sqrt (
- ( (gs->ssq / gs->n ) - gs->mean * gs->mean )
- ) ;
-
- gs->std_dev= sqrt (
- gs->n/ (gs->n-1) *
- ( (gs->ssq / gs->n ) - gs->mean * gs->mean )
- ) ;
-
+ gs->s_std_dev = sqrt (((gs->ssq / gs->n) - pow2 (gs->mean)));
+ gs->std_dev = sqrt (gs->n / (gs->n- 1)
+ * ((gs->ssq / gs->n) - pow2 (gs->mean)));
gs->se_mean = gs->std_dev / sqrt (gs->n);
- count ++;
+ count++;
}
assert (count == 2);
}
-}
+ casereader_destroy (reader);
+ return 0;
+}
static void
-calculate (struct cmd_t_test *cmd,
- struct casereader *input, const struct dataset *ds)
+calculate (struct t_test_proc *proc,
+ struct casereader *input, const struct dataset *ds)
{
const struct dictionary *dict = dataset_dict (ds);
struct ssbox stat_summary_box;
struct trbox test_results_box;
-
- struct casereader *pass1, *pass2, *pass3;
struct taint *taint;
- struct ccase c;
-
- enum mv_class exclude = cmd->miss != TTS_INCLUDE ? MV_ANY : MV_SYSTEM;
+ struct ccase *c;
- if (!casereader_peek (input, 0, &c))
+ c = casereader_peek (input, 0);
+ if (c == NULL)
{
casereader_destroy (input);
return;
}
- output_split_file_values (ds, &c);
- case_destroy (&c);
+ output_split_file_values (ds, c);
+ case_unref (c);
- if ( cmd->miss == TTS_LISTWISE )
+ if (proc->listwise_missing)
input = casereader_create_filter_missing (input,
- cmd->v_variables,
- cmd->n_variables,
- exclude, NULL);
-
+ proc->vars,
+ proc->n_vars,
+ proc->exclude, NULL, NULL);
input = casereader_create_filter_weight (input, dict, NULL, NULL);
-
taint = taint_clone (casereader_get_taint (input));
- casereader_split (input, &pass1, &pass2);
-
- common_precalc (cmd);
- for (; casereader_read (pass1, &c); case_destroy (&c))
- common_calc (dict, &c, cmd, exclude);
- casereader_destroy (pass1);
- common_postcalc (cmd);
- switch (mode)
+ common_calc (dict, proc, casereader_clone (input));
+ switch (proc->mode)
{
case T_1_SAMPLE:
- one_sample_precalc (cmd);
- for (; casereader_read (pass2, &c); case_destroy (&c))
- one_sample_calc (dict, &c, cmd, exclude);
- one_sample_postcalc (cmd);
+ one_sample_calc (dict, proc, input);
break;
case T_PAIRED:
- paired_precalc (cmd);
- for (; casereader_read (pass2, &c); case_destroy (&c))
- paired_calc (dict, &c, cmd, exclude);
- paired_postcalc (cmd);
+ paired_calc (dict, proc, input);
break;
case T_IND_SAMPLES:
- pass3 = casereader_clone (pass2);
-
- group_precalc (cmd);
- for (; casereader_read (pass2, &c); case_destroy (&c))
- group_calc (dict, &c, cmd, exclude);
- group_postcalc (cmd);
-
- levene (dict, pass3, indep_var, cmd->n_variables, cmd->v_variables,
- exclude);
+ group_calc (dict, proc, casereader_clone (input));
+ levene (dict, input, proc->indep_var, proc->n_vars, proc->vars,
+ proc->exclude);
break;
+ default:
+ NOT_REACHED ();
}
- casereader_destroy (pass2);
if (!taint_has_tainted_successor (taint))
{
- ssbox_create (&stat_summary_box,cmd,mode);
- ssbox_populate (&stat_summary_box, dict, cmd);
+ ssbox_create (&stat_summary_box, proc);
+ ssbox_populate (&stat_summary_box, proc);
ssbox_finalize (&stat_summary_box);
- if ( mode == T_PAIRED )
- pscbox (dict);
+ if (proc->mode == T_PAIRED)
+ pscbox (proc);
- trbox_create (&test_results_box, cmd, mode);
- trbox_populate (&test_results_box, dict, cmd);
+ trbox_create (&test_results_box, proc);
+ trbox_populate (&test_results_box, proc);
trbox_finalize (&test_results_box);
}
taint_destroy (taint);
}
-short which_group (const struct group_statistics *g,
- const struct group_properties *p);
+/* return 0 if G belongs to group 0,
+ 1 if it belongs to group 1,
+ 2 if it belongs to neither group */
+static int
+which_group (const struct group_statistics *g,
+ const struct t_test_proc *proc)
+{
+ int width = var_get_width (proc->indep_var);
+
+ if (0 == value_compare_3way (&g->id, &proc->g_value[0], width))
+ return 0;
+
+ if (0 == value_compare_3way (&g->id, &proc->g_value[1], width))
+ return 1;
+
+ return 2;
+}
/* Return -1 if the id of a is less than b; +1 if greater than and
0 if equal */
static int
compare_group_binary (const struct group_statistics *a,
- const struct group_statistics *b,
- const struct group_properties *p)
+ const struct group_statistics *b,
+ const struct t_test_proc *proc)
{
- short flag_a;
- short flag_b;
+ int flag_a;
+ int flag_b;
- if ( p->criterion == CMP_LE )
+ if (proc->criterion == CMP_LE)
{
- /* less-than comparision is not meaningfull for
- alpha variables, so we shouldn't ever arrive here */
- assert (p->indep_width == 0 ) ;
-
- flag_a = ( a->id.f < p->v.critical_value ) ;
- flag_b = ( b->id.f < p->v.critical_value ) ;
+ flag_a = (a->id.f < proc->critical_value);
+ flag_b = (b->id.f < proc->critical_value);
}
else
{
- flag_a = which_group (a, p);
- flag_b = which_group (b, p);
+ flag_a = which_group (a, proc);
+ flag_b = which_group (b, proc);
}
- if (flag_a < flag_b )
- return -1;
+ if (flag_a < flag_b)
+ return - 1;
return (flag_a > flag_b);
}
static unsigned
hash_group_binary (const struct group_statistics *g,
- const struct group_properties *p)
-{
- short flag = -1;
-
- if ( p->criterion == CMP_LE )
- {
- /* Not meaningfull to do a less than compare for alpha values ? */
- assert (p->indep_width == 0 ) ;
- flag = ( g->id.f < p->v.critical_value ) ;
- }
- else if ( p->criterion == CMP_EQ)
- {
- flag = which_group (g,p);
- }
- else
- NOT_REACHED ();
-
- return flag;
-}
-
-/* return 0 if G belongs to group 0,
- 1 if it belongs to group 1,
- 2 if it belongs to neither group */
-short
-which_group (const struct group_statistics *g,
- const struct group_properties *p)
+ const struct t_test_proc *proc)
{
- if ( 0 == compare_values (&g->id, &p->v.g_value[0], p->indep_width))
- return 0;
-
- if ( 0 == compare_values (&g->id, &p->v.g_value[1], p->indep_width))
- return 1;
-
- return 2;
+ return (proc->criterion == CMP_LE
+ ? g->id.f < proc->critical_value
+ : which_group (g, proc));
}
/*
projects / pspp-builds.git / commitdiff