DESCRIPTION

The locon is used to describe a connector of the logical view  
of a figure. Connectors are describing the cell interface, and are the link between its physical and logical views, so the names must be the same.

The declarations needed to work on locon are available in the header file "/labo/include/mlo404.h", where '404' is the actual mbk version.

The following C structure supports the description of the logical connector :

typedef struct locon {
   struct locon     *NEXT;
   char             *NAME;
   struct losig     *SIG;
   void             *ROOT;
   char             DIRECTION;
   char             TYPE;
   struct ptype     *USER;
} locon_list;

NEXT

Pointer to the next locon of the list.

NAME

Name of the connector. It identifies the connector, so it should unique in a given logical figure. A logical connector with a given name correspond to the physical connector or connectors of this name.

SIG

Pointer to the signal linked to the connector. This signal is, of course, unique. See losig(3) for details.

ROOT

Pointer to the object it belongs to. This object may be either a lofig, or a loins, depending on the type of the connector. See lofig(3) and loins for details.

DIRECTION

This indicates the way of the information going through the connector. Six directions are legal :

IN

input

OUT

output

INOUT

input and output

UNKNOWN

no information is available

TRISTATE

tristate output

TRANSCV

tristate output and input

TYPE

Character indicating if the connector belongs to a model or an instance. It can take two values :

EXTERNAL

for figure connectors

INTERNAL

for instance connectors

The TYPE is needed to appropriatly cast the ROOT field, and must be approriatly filled by the parsers. A misuse of it may cause strange behaviours.

USER

Pointer to a ptype list, see ptype for details, that is a general purpose pointer used to share informations on the connector.