Bug List

Global common_dbg

XXX need to init lib's props --RAM

Global fi_update_seen_on_network (gnet_src_t srcid)

FIXME: also remove ranges when a download source is no longer available.

Global fi_gui_update_display (time_t now)

FIXME: We should remember for every node when it was last updated and only refresh every node at most once every second. This information should be kept in a struct pointed to by the row user_data and should be automatically freed when removing the row (see upload stats code).

Global st_search_callback )(gpointer ctx, struct shared_file *sf)

FIXME: The type of this callback is too specific.

Global create_directory (const gchar *dir)

FIXME: This might fail with ``fancy'' file permissions. The directories should be created from leaf to root instead of vice-versa.

Global formatted_connection_pongs (gchar *field, host_type_t htype, gint num)

XXX Refactoring note: there is a need for generic header formatting routines, and especially the dumping routing, which could be taught basic formatting and splitting so that very long lines are dumped using continuations. --RAM, 10/01/2002

Global nodes_gui_update_nodes_display (time_t now)

FIXME: We should remember for every node when it was last updated and only refresh every node at most once every second. This information should be kept in a struct pointed to by the row user_data and should be automatically freed when removing the row (see upload stats code).

Global nodes_gui_update_nodes_display (time_t now)

FIXME: We should remember for every node when it was last updated and only refresh every node at most once every second. This information should be kept in a struct pointed to by the row user_data and should be automatically freed when removing the row (see upload stats code).

Global node_added_callback (gpointer data)

FIXME: uses node_added which is a global variable in nodes.c. This should instead be contained with the argument to this call.

Global add_parent_with_sha1 (GHashTable *ht, gpointer key, GtkCTreeNode *data)

FIXME: The "key" is an atom of the record's SHA1, why don't we create that atom here, as we free it in "remove_parent_with_sha1"? Emile 02/15/2004

---