clish_xmlnode_t *clish_xmlnode_next_child(clish_xmlnode_t *parent,
clish_xmlnode_t *curchild)
{
node_t *roxc;
if (!parent)
return NULL;
roxc = xmlnode_to_node(curchild);
if (roxc) {
return node_to_xmlnode(roxml_get_next_sibling(roxc));
} else {
node_t *roxp = xmlnode_to_node(parent);
node_t *child = NULL;
int count;
count = roxml_get_chld_nb(roxp);
if (count)
child = roxml_get_chld(roxp, NULL, 0);
return node_to_xmlnode(child);
}
return NULL;
}
static inline void ots_get_xml_node(node_t** node, const char* name) {
#if DEBUG > 1
fprintf(stdout, "%d: %s\n", __LINE__, roxml_get_name(*node, NULL, 0));
#endif /* DEBUG > 1 */
while ((*node != NULL) &&
g_strcmp0(roxml_get_name(*node, NULL, 0), (const char*)name)) {
*node = roxml_get_next_sibling(*node);
#if DEBUG > 2
fprintf(stdout, "%s\n", roxml_get_name(*node, NULL, 0));
#endif /* DEBUG > 2 */
}
#if DEBUG > 0
fprintf(stdout, "%d: %s\n", __LINE__, roxml_get_name(*node, NULL, 0));
#endif /* DEBUG > 0 */
}
static inline void ots_append_xml_node_contents(GList** list, node_t** node) {
#if DEBUG > 1
fprintf(stdout, "%d: %s\n", __LINE__, roxml_get_name(*node, NULL, 0));
fprintf(stdout, "%d: %s\n", __LINE__, roxml_get_content(*node, NULL, 0, NULL));
#endif /* DEBUG > 1 */
while (*node != NULL) {
if (!g_strcmp0(roxml_get_name(*node, NULL, 0), (const char*)"rule")) {
*list = g_list_append(*list,
g_strdup(roxml_get_content(*node, NULL, 0, NULL)));
#if DEBUG > 2
fprintf(stdout, "%s\n", roxml_get_content(*node, NULL, 0, NULL));
fprintf(stdout, "%d: %s\n", __LINE__, roxml_get_name(*node, NULL, 0));
#endif /* DEBUG > 2 */
}
*node = roxml_get_next_sibling((*node));
}
}
gboolean ots_load_xml_dictionary_buf(OtsArticle* Doc, const char* buffer) {
char* head_name = NULL;
node_t* stem = NULL;
node_t* pre = NULL;
node_t* post = NULL;
node_t* syno = NULL; /* synonyms */
node_t* manual = NULL; /* manual */
node_t* step1_pre = NULL; /* step1 */
node_t* step1_post = NULL; /* step1 */
node_t* parse = NULL; /* parser rules */
node_t* pbreak = NULL;
node_t* pdbreak = NULL;
node_t* tc_words = NULL; /* term count dictionary */
node_t* tf_words = NULL; /* term frequency dictionary */
OtsStemRule* rule = Doc->stem;
node_t* head = roxml_load_buf((char*)buffer);
if (head == NULL) {
fprintf(stderr, "empty document\n");
roxml_release(RELEASE_ALL);
return (FALSE);
}
#if DEBUG > 1
fprintf(stdout, "%d: %s\n", __LINE__, roxml_get_name(head, NULL, 0));
#endif /* DEBUG > 1 */
if (g_strcmp0(roxml_get_name(head, head_name, 0), (const char*)"dictionary")) {
fprintf(stderr, "%s\n", head_name);
roxml_release(RELEASE_ALL);
roxml_close(head);
return (FALSE);
}
stem = roxml_get_chld(head, NULL, 0);
if(stem == NULL) {
roxml_release(RELEASE_ALL);
roxml_close(head);
return (FALSE);
}
ots_get_xml_node(&stem, (const char*)"stemmer");
parse = roxml_get_chld(head, NULL, 0);
if(parse == NULL) {
roxml_release(RELEASE_ALL);
roxml_close(head);
return (FALSE);
}
ots_get_xml_node(&parse, (const char*)"parser");
tc_words = roxml_get_chld(head, NULL, 0);
if(tc_words == NULL) {
roxml_release(RELEASE_ALL);
roxml_close(head);
return (FALSE);
}
ots_get_xml_node(&tc_words, (const char*)"grader-tc");
tf_words = roxml_get_chld(head, NULL, 0);
if(tf_words == NULL) {
roxml_release(RELEASE_ALL);
roxml_close(head);
return (FALSE);
}
ots_get_xml_node(&tf_words, (const char*)"grader-tf");
pre = roxml_get_chld(stem, NULL, 0);
if(pre == NULL) {
roxml_release(RELEASE_ALL);
roxml_close(head);
return (FALSE);
}
ots_get_xml_node(&pre, (const char*)"pre");
post = roxml_get_chld(stem, NULL, 0);
if(post == NULL) {
roxml_release(RELEASE_ALL);
roxml_close(head);
return (FALSE);
}
ots_get_xml_node(&post, (const char*)"post");
syno = roxml_get_chld(stem, NULL, 0);
if(syno == NULL) {
roxml_release(RELEASE_ALL);
roxml_close(head);
return (FALSE);
}
ots_get_xml_node(&syno, (const char*)"synonyms");
manual = roxml_get_chld(stem, NULL, 0);
if(manual == NULL) {
roxml_release(RELEASE_ALL);
roxml_close(head);
return (FALSE);
}
ots_get_xml_node(&manual, (const char*)"manual");
step1_pre = roxml_get_chld(stem, NULL, 0);
if(step1_pre == NULL) {
roxml_release(RELEASE_ALL);
roxml_close(head);
return (FALSE);
}
ots_get_xml_node(&step1_pre, (const char*)"step1_pre");
step1_post = roxml_get_chld(stem, NULL, 0);
if(step1_post == NULL) {
roxml_release(RELEASE_ALL);
roxml_close(head);
return (FALSE);
}
ots_get_xml_node(&step1_post, (const char*)"step1_post");
pre = roxml_get_chld(pre, NULL, 0); /* point to first word */
if(pre == NULL) {
roxml_release(RELEASE_ALL);
roxml_close(head);
return (FALSE);
}
ots_append_xml_node_contents(&rule->RemovePre, &pre);
post = roxml_get_chld(post, NULL, 0);
if(post == NULL) {
roxml_release(RELEASE_ALL);
roxml_close(head);
return (FALSE);
}
ots_append_xml_node_contents(&rule->RemovePost, &post);
syno = roxml_get_chld(syno, NULL, 0);
if(syno == NULL) {
roxml_release(RELEASE_ALL);
roxml_close(head);
return (FALSE);
}
ots_append_xml_node_contents(&rule->synonyms, &syno);
manual = roxml_get_chld(manual, NULL, 0);
if(manual == NULL) {
roxml_release(RELEASE_ALL);
roxml_close(head);
return (FALSE);
}
ots_append_xml_node_contents(&rule->manual, &manual);
step1_pre = roxml_get_chld(step1_pre, NULL, 0);
if(step1_pre == NULL) {
roxml_release(RELEASE_ALL);
roxml_close(head);
return (FALSE);
}
ots_append_xml_node_contents(&rule->step1_pre, &step1_pre);
step1_post = roxml_get_chld(step1_post, NULL, 0);
if(step1_post == NULL) {
roxml_release(RELEASE_ALL);
roxml_close(head);
return (FALSE);
}
ots_append_xml_node_contents(&rule->step1_post, &step1_post);
pbreak = roxml_get_chld(parse, NULL, 0);
if(pbreak == NULL) {
roxml_release(RELEASE_ALL);
roxml_close(head);
return (FALSE);
}
ots_get_xml_node(&pbreak, (const char*)"linebreak");
pdbreak = roxml_get_chld(parse, NULL, 0);
if(pdbreak == NULL) {
roxml_release(RELEASE_ALL);
roxml_close(head);
return (FALSE);
}
ots_get_xml_node(&pdbreak, (const char*)"linedontbreak");
/*Parser break*/
pbreak = roxml_get_chld(pbreak, NULL, 0);
if(pbreak == NULL) {
roxml_release(RELEASE_ALL);
roxml_close(head);
return (FALSE);
}
ots_append_xml_node_contents(&rule->ParserBreak, &pbreak);
/*Parser Don't break*/
pdbreak = roxml_get_chld(pdbreak, NULL, 0);
if(pdbreak == NULL) {
roxml_release(RELEASE_ALL);
roxml_close(head);
return (FALSE);
}
ots_append_xml_node_contents(&rule->ParserDontBreak, &pdbreak);
/* Term Count load dict */
tc_words = roxml_get_chld(tc_words, NULL, 0);
if(tc_words == NULL) {
roxml_release(RELEASE_ALL);
roxml_close(head);
return (FALSE);
}
#if DEBUG > 1
fprintf(stdout, "%d: %s\n", __LINE__, roxml_get_name(tc_words, NULL, 0));
fprintf(stdout, "%d: %s\n", __LINE__, roxml_get_content(tc_words, NULL, 0, NULL));
#endif /* DEBUG > 1 */
while (tc_words != NULL) {
if (!g_strcmp0(roxml_get_name(tc_words, NULL, 0), (const char*)"word")) {
Doc->dict = g_list_prepend(Doc->dict,
(gpointer)ots_new_wordEntery(
(unsigned const char*)roxml_get_content(tc_words, NULL, 0, NULL)));
}
tc_words = roxml_get_next_sibling(tc_words);
#if DEBUG > 2
fprintf(stdout, "%s\n", roxml_get_content(tc_words, NULL, 0, NULL));
#endif /* DEBUG > 2 */
}
/*Term Frequency load dict*/
tf_words = roxml_get_chld(tf_words, NULL, 0);
if(tf_words == NULL) {
roxml_release(RELEASE_ALL);
roxml_close(head);
return (FALSE);
}
#if DEBUG > 1
fprintf(stdout, "%d: %s\n", __LINE__, roxml_get_name(tf_words, NULL, 0));
fprintf(stdout, "%d: %s\n", __LINE__, roxml_get_content(tf_words, NULL, 0, NULL));
#endif /* DEBUG > 1 */
while (tf_words != NULL) {
if (!g_strcmp0(roxml_get_name(tf_words, NULL, 0), (const char*)"word")) {
const gchar* key = g_strdup(roxml_get_content(tf_words, NULL, 0, NULL));
node_t* idf_attr = roxml_get_attr(tf_words, "idf", 0);
if (idf_attr != NULL) {
const char* idf_val = roxml_get_content(idf_attr, NULL, 0, NULL);
if (idf_val != NULL) {
#if DEBUG > 3
fprintf(stdout, "%s\n", idf_val);
#endif /* DEBUG > 3 */
Doc->tf_terms = g_list_append(Doc->tf_terms,
ots_new_OtsWordTF((const unsigned char*)key, atof(idf_val)));
}
}
g_free((gpointer)key);
}
tf_words = roxml_get_next_sibling(tf_words);
#if DEBUG > 2
fprintf(stdout, "%s\n", roxml_get_content(tf_words, NULL, 0, NULL));
#endif /* DEBUG > 2 */
}
roxml_release(RELEASE_ALL);
roxml_close(head);
return (TRUE);
}
