X-Git-Url: https://pintos-os.org/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=src%2Fmath%2Fcategoricals.h;h=2cd55b5ec3afa1a6d8630d754bf682518c817c56;hb=76e1b0aef6b1cc3b911c4833565109eaa424f9a5;hp=09ced7d7acc2a9900fd67039697b98273aaafe0c;hpb=3025fc73778a733bfd2aeefd57da1b1c1955a675;p=pspp diff --git a/src/math/categoricals.h b/src/math/categoricals.h index 09ced7d7ac..2cd55b5ec3 100644 --- a/src/math/categoricals.h +++ b/src/math/categoricals.h @@ -1,5 +1,5 @@ /* PSPP - a program for statistical analysis. - Copyright (C) 2009 Free Software Foundation, Inc. + Copyright (C) 2009, 2010, 2011 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 @@ -19,11 +19,12 @@ #define _CATEGORICALS__ #include -#include +#include "data/missing-values.h" struct categoricals; struct variable; struct ccase; +struct interaction; union value ; @@ -36,7 +37,7 @@ typedef void update_func (void *user_data, typedef void *user_data_create_func (void *aux1, void *aux2); -struct categoricals *categoricals_create (const struct variable *const *v, size_t n_vars, +struct categoricals *categoricals_create (const struct interaction **, size_t n_int, const struct variable *wv, enum mv_class exclude, user_data_create_func *udf, update_func *update, void *aux1, void *aux2); @@ -61,11 +62,21 @@ size_t categoricals_total (const struct categoricals *cat); */ size_t categoricals_get_n_variables (const struct categoricals *cat); -void categoricals_done (struct categoricals *cat); -const struct variable * categoricals_get_variable_by_subscript (const struct categoricals *cat, int subscript); +/* + Must be called (once) before any call to the *_by_subscript or *_by_category + functions, but AFTER any calls to categoricals_update +*/ +void categoricals_done (const struct categoricals *cat); -const union value * categoricals_get_value_by_subscript (const struct categoricals *cat, int subscript); + +/* + The *_by_subscript functions use the short map. + Their intended use is by covariance matrix routines, where normally 1 less than + the total number of distinct values of each categorical variable should + be considered. + */ +const struct variable * categoricals_get_variable_by_subscript (const struct categoricals *cat, int subscript); double categoricals_get_weight_by_subscript (const struct categoricals *cat, int subscript); @@ -74,10 +85,15 @@ double categoricals_get_sum_by_subscript (const struct categoricals *cat, int su double categoricals_get_binary_by_subscript (const struct categoricals *cat, int subscript, const struct ccase *c); -void * categoricals_get_user_data_by_subscript (const struct categoricals *cat, int subscript); +/* These use the long map. Useful for descriptive statistics. */ + +/* Return the value corresponding to the N'th category */ +const union value * categoricals_get_value_by_category (const struct categoricals *cat, int n); +void * categoricals_get_user_data_by_category (const struct categoricals *cat, int category); +const struct ccase * categoricals_get_case_by_category (const struct categoricals *cat, int subscript); #endif