X-Git-Url: https://pintos-os.org/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=src%2Flanguage%2Fstats%2Frank.q;h=c225370e459269cc572430f87e533f7a591662c4;hb=2cf38ce51a9f34961d68a75e0b312a591b5c9abf;hp=13facbdbbf59f85e43bd350578d4633cb07b880e;hpb=c3ac5a8af9c449072c7e872ca70a78c1755ae309;p=pspp-builds.git diff --git a/src/language/stats/rank.q b/src/language/stats/rank.q index 13facbdb..c225370e 100644 --- a/src/language/stats/rank.q +++ b/src/language/stats/rank.q @@ -1,5 +1,5 @@ /* PSPP - a program for statistical analysis. - Copyright (C) 2005, 2006, 2007 Free Software Foundation, Inc. + Copyright (C) 2005, 2006, 2007, 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 @@ -19,7 +19,6 @@ #include #include -#include #include #include #include @@ -29,6 +28,7 @@ #include #include #include +#include #include #include #include @@ -152,7 +152,7 @@ static enum mv_class exclude_values; static struct rank_spec *rank_specs; static size_t n_rank_specs; -static struct case_ordering *sc; +static struct subcase sc; static const struct variable **group_vars; static size_t n_group_vars; @@ -233,14 +233,14 @@ create_var_label (struct variable *dest_var, static bool -rank_cmd (struct dataset *ds, const struct case_ordering *sc, +rank_cmd (struct dataset *ds, const struct subcase *sc, const struct rank_spec *rank_specs, int n_rank_specs) { struct dictionary *d = dataset_dict (ds); bool ok = true; int i; - for (i = 0 ; i < case_ordering_get_var_cnt (sc) ; ++i ) + for (i = 0 ; i < subcase_get_n_fields (sc) ; ++i ) { /* Rank variable at index I in SC. */ struct casegrouper *split_grouper; @@ -249,25 +249,22 @@ rank_cmd (struct dataset *ds, const struct case_ordering *sc, proc_discard_output (ds); split_grouper = casegrouper_create_splits (proc_open (ds), d); - output = autopaging_writer_create (dict_get_next_value_idx (d)); + output = autopaging_writer_create (dict_get_proto (d)); while (casegrouper_get_next_group (split_grouper, &split_group)) { - struct case_ordering *ordering; + struct subcase ordering; struct casereader *ordered; struct casegrouper *by_grouper; struct casereader *by_group; - int j; /* Sort this split group by the BY variables as primary keys and the rank variable as secondary key. */ - ordering = case_ordering_create (); - for (j = 0; j < n_group_vars; j++) - case_ordering_add_var (ordering, group_vars[j], SRT_ASCEND); - case_ordering_add_var (ordering, - case_ordering_get_var (sc, i), - case_ordering_get_direction (sc, i)); - ordered = sort_execute (split_group, ordering); + subcase_init_vars (&ordering, group_vars, n_group_vars); + subcase_add_var (&ordering, src_vars[i], + subcase_get_direction (sc, i)); + ordered = sort_execute (split_group, &ordering); + subcase_destroy (&ordering); /* Rank the rank variable within this split group. */ by_grouper = casegrouper_create_vars (ordered, @@ -479,7 +476,7 @@ rank_sorted_file (struct casereader *input, { struct casereader *pass1, *pass2, *pass2_1; struct casegrouper *tie_grouper; - struct ccase c; + struct ccase *c; double w = 0.0; double cc = 0.0; int tie_group = 1; @@ -492,8 +489,8 @@ rank_sorted_file (struct casereader *input, casereader_split (input, &pass1, &pass2); /* Pass 1: Get total group weight. */ - for (; casereader_read (pass1, &c); case_destroy (&c)) - w += dict_get_case_weight (dict, &c, NULL); + for (; (c = casereader_read (pass1)) != NULL; case_unref (c)) + w += dict_get_case_weight (dict, c, NULL); casereader_destroy (pass1); /* Pass 2: Do ranking. */ @@ -510,21 +507,22 @@ rank_sorted_file (struct casereader *input, casewriter_get_taint (output)); /* Pass 2.1: Sum up weight for tied cases. */ - for (; casereader_read (pass2_1, &c); case_destroy (&c)) - tw += dict_get_case_weight (dict, &c, NULL); + for (; (c = casereader_read (pass2_1)) != NULL; case_unref (c)) + tw += dict_get_case_weight (dict, c, NULL); cc += tw; casereader_destroy (pass2_1); /* Pass 2.2: Rank tied cases. */ - while (casereader_read (pass2_2, &c)) + while ((c = casereader_read (pass2_2)) != NULL) { + c = case_unshare (c); for (i = 0; i < n_rank_specs; ++i) { const struct variable *dst_var = rs[i].destvars[dest_idx]; - double *dst_value = &case_data_rw (&c, dst_var)->f; + double *dst_value = &case_data_rw (c, dst_var)->f; *dst_value = rank_func[rs[i].rfunc] (tw, cc, cc_1, tie_group, w); } - casewriter_write (output, &c); + casewriter_write (output, c); } casereader_destroy (pass2_2); @@ -535,11 +533,12 @@ rank_sorted_file (struct casereader *input, /* Transformation function to enumerate all the cases */ static int -create_resort_key (void *key_var_, struct ccase *cc, casenumber case_num) +create_resort_key (void *key_var_, struct ccase **cc, casenumber case_num) { struct variable *key_var = key_var_; - case_data_rw(cc, key_var)->f = case_num; + *cc = case_unshare (*cc); + case_data_rw (*cc, key_var)->f = case_num; return TRNS_CONTINUE; } @@ -625,8 +624,7 @@ rank_cleanup(void) rank_specs = NULL; n_rank_specs = 0; - case_ordering_destroy (sc); - sc = NULL; + subcase_destroy (&sc); free (src_vars); src_vars = NULL; @@ -641,6 +639,7 @@ cmd_rank (struct lexer *lexer, struct dataset *ds) size_t i; n_rank_specs = 0; + subcase_init_empty (&sc); if ( !parse_rank (lexer, ds, &cmd, NULL) ) { rank_cleanup (); @@ -660,12 +659,12 @@ cmd_rank (struct lexer *lexer, struct dataset *ds) rank_specs = xmalloc (sizeof (*rank_specs)); rank_specs[0].rfunc = RANK; rank_specs[0].destvars = - xcalloc (case_ordering_get_var_cnt (sc), sizeof (struct variable *)); + xcalloc (subcase_get_n_fields (&sc), sizeof (struct variable *)); n_rank_specs = 1; } - assert ( case_ordering_get_var_cnt (sc) == n_src_vars); + assert ( subcase_get_n_fields (&sc) == n_src_vars); /* Create variables for all rank destinations which haven't already been created with INTO. @@ -715,48 +714,44 @@ cmd_rank (struct lexer *lexer, struct dataset *ds) if ( rank_specs[i].rfunc == NORMAL || rank_specs[i].rfunc == PROPORTION ) - tab_output_text (TAT_PRINTF, - _("%s into %s(%s of %s using %s BY %s)"), - var_get_name (src_vars[v]), - var_get_name (rank_specs[i].destvars[v]), - function_name[rank_specs[i].rfunc], - var_get_name (src_vars[v]), - fraction_name(), - ds_cstr (&varlist) - ); + tab_output_text_format (0, + _("%s into %s(%s of %s using %s BY %s)"), + var_get_name (src_vars[v]), + var_get_name (rank_specs[i].destvars[v]), + function_name[rank_specs[i].rfunc], + var_get_name (src_vars[v]), + fraction_name(), + ds_cstr (&varlist)); else - tab_output_text (TAT_PRINTF, - _("%s into %s(%s of %s BY %s)"), - var_get_name (src_vars[v]), - var_get_name (rank_specs[i].destvars[v]), - function_name[rank_specs[i].rfunc], - var_get_name (src_vars[v]), - ds_cstr (&varlist) - ); + tab_output_text_format (0, + _("%s into %s(%s of %s BY %s)"), + var_get_name (src_vars[v]), + var_get_name (rank_specs[i].destvars[v]), + function_name[rank_specs[i].rfunc], + var_get_name (src_vars[v]), + ds_cstr (&varlist)); ds_destroy (&varlist); } else { if ( rank_specs[i].rfunc == NORMAL || rank_specs[i].rfunc == PROPORTION ) - tab_output_text (TAT_PRINTF, - _("%s into %s(%s of %s using %s)"), - var_get_name (src_vars[v]), - var_get_name (rank_specs[i].destvars[v]), - function_name[rank_specs[i].rfunc], - var_get_name (src_vars[v]), - fraction_name() - ); + tab_output_text_format (0, + _("%s into %s(%s of %s using %s)"), + var_get_name (src_vars[v]), + var_get_name (rank_specs[i].destvars[v]), + function_name[rank_specs[i].rfunc], + var_get_name (src_vars[v]), + fraction_name()); else - tab_output_text (TAT_PRINTF, - _("%s into %s(%s of %s)"), - var_get_name (src_vars[v]), - var_get_name (rank_specs[i].destvars[v]), - function_name[rank_specs[i].rfunc], - var_get_name (src_vars[v]) - ); + tab_output_text_format (0, + _("%s into %s(%s of %s)"), + var_get_name (src_vars[v]), + var_get_name (rank_specs[i].destvars[v]), + function_name[rank_specs[i].rfunc], + var_get_name (src_vars[v])); } } } @@ -773,17 +768,17 @@ cmd_rank (struct lexer *lexer, struct dataset *ds) add_transformation (ds, create_resort_key, 0, order); /* Do the ranking */ - result = rank_cmd (ds, sc, rank_specs, n_rank_specs); + result = rank_cmd (ds, &sc, rank_specs, n_rank_specs); /* Put the active file back in its original order. Delete our sort key, which we don't need anymore. */ { - struct case_ordering *ordering = case_ordering_create (); struct casereader *sorted; - case_ordering_add_var (ordering, order, SRT_ASCEND); + /* FIXME: loses error conditions. */ + proc_discard_output (ds); - sorted = sort_execute (proc_open (ds), ordering); + sorted = sort_execute_1var (proc_open (ds), order); result = proc_commit (ds) && result; dict_delete_var (dataset_dict (ds), order); @@ -808,10 +803,9 @@ rank_custom_variables (struct lexer *lexer, struct dataset *ds, struct cmd_rank && lex_token (lexer) != T_ALL) return 2; - sc = parse_case_ordering (lexer, dataset_dict (ds), NULL); - if (sc == NULL) + if (!parse_sort_criteria (lexer, dataset_dict (ds), &sc, &src_vars, NULL)) return 0; - case_ordering_get_vars (sc, &src_vars, &n_src_vars); + n_src_vars = subcase_get_n_fields (&sc); if ( lex_match (lexer, T_BY) ) { @@ -845,8 +839,7 @@ parse_rank_function (struct lexer *lexer, struct dictionary *dict, struct cmd_ra rank_specs[n_rank_specs - 1].destvars = NULL; rank_specs[n_rank_specs - 1].destvars = - xcalloc (case_ordering_get_var_cnt (sc), - sizeof (struct variable *)); + xcalloc (subcase_get_n_fields (&sc), sizeof (struct variable *)); if (lex_match_id (lexer, "INTO")) { @@ -860,7 +853,7 @@ parse_rank_function (struct lexer *lexer, struct dictionary *dict, struct cmd_ra msg(SE, _("Variable %s already exists."), lex_tokid (lexer)); return 0; } - if ( var_count >= case_ordering_get_var_cnt (sc) ) + if ( var_count >= subcase_get_n_fields (&sc) ) { msg(SE, _("Too many variables in INTO clause.")); return 0;