document.c

使用具有增量学习的监控式学习方法。包括几个不同的分类算法。

/* Copyright (C) 2002  Mikael Ylikoski
 * See the accompanying file "README" for the full copyright notice */
/**
 * @file
 * Structured document interface.
 *
 * @author  Mikael Ylikoski
 * @date    2002
 */

#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include "document.h"
#include "rfc822.h"
#include "utility.h"


/**
 * Document.
 */
struct document_ {
    enum document_type dt;	/**< Document type */
    struct rfc822 *msg;		/**< Email message */
    text_part *parts;		/**< Text parts of document */
    char *charset;		/**< Character set */
    char *from;			/**< Sender */
    char *subject;		/**< Subject */
    char *language;		/**< Language */
    //time_t time;
};


/**
 * Create a new document.
 *
 * @param source  source text; must not be used or freed by caller afterwards
 * @param dt      document type
 */
document *
document_new (char *source, enum document_type dt) {
    document *doc;

    doc = my_malloc (sizeof(document));
    doc->dt = dt;
    doc->from = NULL;
    doc->subject = NULL;
    doc->language = NULL;
    doc->charset = NULL;
    doc->parts = NULL;
    if (source)
	switch (doc->dt) {
	case PLAIN:
	    document_add_text (doc, NULL, source, strlen (source));
	    break;
	case RFC822:
	    doc->msg = rfc822_new (source);
	    /*
	    if (!doc->msg) {
		free (doc);
		return NULL;
	    }
	    */
	    break;
	}

    return doc;
}


/**
 * Add a text to a document.
 *
 * @param doc      document to add to
 * @param charset  character set of text
 * @param text     text to add
 * @param len      length of text
 */
void
document_add_text (document *doc, char *charset, char *text, int len) {
    text_part *tp, *pt;

    tp = my_malloc (sizeof(text_part));
    tp->text = text;
    tp->len = len;
    tp->charset = charset;
    tp->next = NULL;

    if (!doc->parts)
	doc->parts = tp;
    else {
	for (pt = doc->parts; pt->next; pt = pt->next)
	    ;
	pt->next = tp;
    }
}


/**
 * Free memory used by a document.
 *
 * @param doc  document to free
 */
void
document_free (document *doc) {
    text_part *tp, *pt;

    switch (doc->dt) {
    case PLAIN:
	for (tp = doc->parts; tp; tp = pt) {
	    pt = tp->next;
	    if (tp->charset)
		free (tp->charset);
	    if (tp->text)
		free (tp->text);
	    free (tp);
	}
	if (doc->from)
	    free (doc->from);
	if (doc->subject)
	    free (doc->subject);
	/*if (doc->language)
	  free (doc->language);*/
	break;
    case RFC822:
	if (doc->msg)
	    rfc822_free (doc->msg);
	break;
    }
    free (doc);
}


int
document_set_from (document *doc, char *from) {
    switch (doc->dt) {
    case PLAIN:
	doc->from = from;
	break;
    case RFC822:
	return -1;
    }
    return 0;
}


int
document_set_subject (document *doc, char *subject) {
    switch (doc->dt) {
    case PLAIN:
	doc->subject = subject;
	break;
    case RFC822:
	return -1;
    }
    return 0;
}


int
document_set_language (document *doc, char *lang) {
    doc->language = lang;
    return 0;
}


int
document_set_rfc822 (document *doc, char *src) {
    switch (doc->dt) {
    case PLAIN:
	doc->dt = RFC822;
	// free
	break;
    case RFC822:
	if (doc->msg)
	    rfc822_free (doc->msg);
	break;
    }
    doc->msg = rfc822_new (src);
    if (!doc->msg)
	return -1;
    return 0;
}


/**
 * Get sender line of document.
 */
const char *
document_get_from (document *doc) {
    switch (doc->dt) {
    case PLAIN:
	return doc->from;
    case RFC822:
	if (doc->msg)
	    return rfc822_get_from (doc->msg);
    }
    return NULL;
}


/**
 * Get sender name of document.
 */
const char *
document_get_from_name (document *doc) {
    switch (doc->dt) {
    case PLAIN:
	break;
    case RFC822:
	if (doc->msg)
	    return rfc822_get_from_name (doc->msg);
    }
    return NULL;
}


/**
 * Get sender address of document.
 */
const char *
document_get_from_address (document *doc) {
    switch (doc->dt) {
    case PLAIN:
	break;
    case RFC822:
	if (doc->msg)
	    return rfc822_get_from_address (doc->msg);
    }
    return NULL;
}


/**
 * Get receiver line of document.
 */
const char *
document_get_to (document *doc) {
    switch (doc->dt) {
    case PLAIN:
	break;
    case RFC822:
	if (doc->msg)
	    return rfc822_get_to (doc->msg);
    }
    return NULL;
}


/**
 * Get subject line of document.
 */
const char *
document_get_subject (document *doc) {
    switch (doc->dt) {
    case PLAIN:
	return doc->subject;
    case RFC822:
	if (doc->msg)
	    return rfc822_get_subject (doc->msg);
    }
    return NULL;
}


/**
 * Get time of document.
 */
time_t
document_get_time (document *doc) {
    switch (doc->dt) {
    case RFC822:
	if (doc->msg)
	    return rfc822_get_date (doc->msg);
    default:
	return -1;
    }
}


/**
 * Get language of document.
 */
const char *
document_get_language (document *doc) {
    return doc->language;
}


/**
 * Get charset of first part.
 */
const char *
document_get_charset (document *doc) {
    if (doc->charset)
	return doc->charset;
    return "ISO-8859-1";
}


text_part *
document_get_parts (document *doc) {
    switch (doc->dt) {
    case PLAIN:
	return doc->parts;
    case RFC822:
	if (doc->msg)
	    return rfc822_get_parts (doc->msg);
    }
    return NULL;
}