a new function in json2mcf to get the id of the first token of a sentence

maca_tools/src/json2mcf.c

@@ -251,12 +251,8 @@ void update_segment(word_buffer *wb, int start, int end, char *label, char *stat
     else
       word_set_pos(w, label_code);
   }
-
 }
 
-
-
-
 void process_segment(json_attr_val *avl, word_buffer *wb)
 {
   int start, end;
@@ -305,20 +301,43 @@ void process_link(json_attr_val *avl, word_buffer *wb)
   //  printf("link : orig = %d dest = %d label = %s status_link = %s status_lab = %s\n", orig, dest, label, status_link, status_lab);
 }
 
-void process_links(json_struct *segments, word_buffer *wb)
+void process_links(json_struct *links, word_buffer *wb)
 {
   json_struct *link;
   //  printf("process_links\n");
-  for(link = segments->u.first; link != NULL; link = link->next){
+  for(link = links->u.first; link != NULL; link = link->next){
     process_link(link->u.attr_val_list, wb);
   }
 
 }
 
+int get_id_of_first_token_in_document(json_struct *document)
+{
+  json_attr_val *avl = NULL;
+  json_struct *tokens, *token;
+  json_attr_val *avl2 = NULL;
+  for(avl = document->u.attr_val_list; avl != NULL; avl = avl->next){
+    if(!strcmp(avl->attr, (char *)"tokens")){
+      tokens = avl->val;
+      if(tokens){
+	token = tokens->u.first;
+	if(token){
+	  for(avl2 = token->u.attr_val_list; avl2 != NULL; avl2 = avl2->next){
+	    if(!strcmp(avl2->attr, (char *)"id"))
+	      return (int)avl2->val->u.number;
+	  }
+	}
+      }
+    }
+  }
+  return -1;
+}
+
 void process_document(json_struct *document, word_buffer *wb)
 {
   json_attr_val *avl = NULL;
-  //  printf("process_document\n");
+  int offset = get_id_of_first_token_in_document(document);
+  printf("process_document, offset = %d\n", offset);
   for(avl = document->u.attr_val_list; avl != NULL; avl = avl->next){
     //    if(!strcmp(avl->attr, (char *)"id")) printf("id = %s\n", avl->val->u.string);
     if(!strcmp(avl->attr, (char *)"segments")) process_segments(avl->val, wb);

maca_tools/src/mcf2json.c

@@ -184,8 +184,6 @@ void print_link(FILE *output_file, word_buffer *wb, int index_first_word, int in
   word *w = word_buffer_get_word_n(wb, index);
   
   fprintf(output_file, "{");
-  
-  //  fprintf(output_file, "\"orig\": %d, ", word_get_offset(w));
   fprintf(output_file, "\"orig\": %d, ", index - index_first_word);
   fprintf(output_file, "\"dest\":");
   if(gov_col){
@@ -193,7 +191,6 @@ void print_link(FILE *output_file, word_buffer *wb, int index_first_word, int in
       fprintf(output_file, "-1");
     else{
       word *gov = word_buffer_get_word_n(wb, word_get_gov(w) + index);
-//      fprintf(output_file, "%d", word_get_offset(gov));
       fprintf(output_file, "%d", word_get_gov(w) + index - index_first_word);
     }
   }
@@ -240,9 +237,7 @@ void print_segment(FILE *output_file, word_buffer *wb, int index_first_word, int
   word *w = word_buffer_get_word_n(wb, index);
 
   fprintf(output_file, "{ ");
-  /* fprintf(output_file, "\"start\": %d, ", word_get_offset(w)); */
   fprintf(output_file, "\"start\": %d, ", index - index_first_word);
-  /* fprintf(output_file, "\"end\": %d, ", word_get_offset(w) + word_get_length(w) - 1); */
   fprintf(output_file, "\"end\": %d, ", index - index_first_word);
   fprintf(output_file, "\"label\": \"");
   
@@ -278,16 +273,12 @@ void print_segments(FILE *output_file, word_buffer *wb, int index_first_word, in
 void print_token(FILE *output_file, word_buffer *wb, int index)
 {
   int form_col =  mcd_get_form_col(word_buffer_get_mcd(wb));
-  int offset_col =  mcd_get_offset_col(word_buffer_get_mcd(wb));
   int length_col =  mcd_get_length_col(word_buffer_get_mcd(wb));
   word *w = word_buffer_get_word_n(wb, index);
   char token[5000];
   int length_token, i;
 
   fprintf(output_file, "{ ");
-  if(word_get_offset(w) != -1)
-    fprintf(output_file, "\"id\": %d, ", word_get_offset(w));
-  else
   fprintf(output_file, "\"id\": %d, ", word_get_index(w));
   fprintf(output_file, "\"word\": \"");
   if(form_col != -1){