This section will document any updates applied to the project since original release:

• revision #: <description> (DATESTAMP)

In this project, we take our first opportunity to undergo a complete code re-write of node functionality, and we will also introduce the necessary functionality for doubly-linked node operations.

As this is a rewrite, dln0 is not based on any of the code you have written up to this point. As such, the transition process is slightly different. Two different approaches are outlined below:

lab46:~/src/data/sll4$ make get-dln0
...

The “get-” functionality is distinct from the “upgrade-” you have been using to transition between the sll* projects. When you upgrade, your existing code is copied over, because the next project builds upon what you did previously.

But when you “get” dln0, you are getting an entirely new project skeleton- NONE of your existing code is copied over (the structure has changed enough where copying your own code would have been rather problematic).

Once you run “make get-dln0” you should have a dln0 directory that you can access and commence working on just as you have with the other project directories.

Just as we did with the first project in the series, sln1, we can also use grabit to obtain this project:

lab46:~/src/data$ grabit data
ERROR: must specify class and project!
example:
    grabit discrete matrixadd

Projects available for data:
    * dln0
    * sln1

lab46:~/src/data$ 

So, we can just go ahead and do:

lab46:~/src/data$ grabit data dln0
...

And voila! Freshly grabbed dln0 in our ~/src/data/ directory.

For this project, we're going to be re-implementing MOST of the previous node and list functions. There have been a few changes, namely:

1

#ifndef _NODE_H
#define _NODE_H
 
//////////////////////////////////////////////////////////////////////
//
// Additional useful information in data.h
//
#include "data.h"
 
//////////////////////////////////////////////////////////////////////
//
// node struct definition
//
struct node {
    union  info      payload;
    struct node     *there;
    struct node     *back;
};
 
//////////////////////////////////////////////////////////////////////
//
// function prototypes
//
code_t mknode(Node **, char);    // allocate new node containing value
code_t cpnode(Node  *, Node **); // duplicate node
code_t rmnode(Node **);          // deallocate node
 
#endif

There is an addition of a “back” node pointer, to allow connections to our previous neighbors.

The node info element has been changed as well… instead of a singular value, it is now a union by the name of payload, which contains a value entry (a char entry), a data entry (Node pointer), and an other entry (a void pointer).

You'll notice that node.h includes a file called data.h; This header will contain predominantly useful #define statements, typedefs, and support function prototypes to make our lives easier. It will see additional content added with future projects.

1

#ifndef _DATA_H
#define _DATA_H
 
//////////////////////////////////////////////////////////////////////
//
// We make use of NULL, so we need stdlib
//
#include <stdlib.h>
 
//////////////////////////////////////////////////////////////////////
//
// Set up union for node payload (multipurpose use)
//
union info {
    char         value;
    struct node *data;
    void        *other;
};
 
//////////////////////////////////////////////////////////////////////
//
// node struct helper defines
//
#define  VALUE   payload.value
#define  DATA    payload.data
#define  OTHER   payload.other
 
//////////////////////////////////////////////////////////////////////
//
// create some peers to NULL for our endeavors: UNDEFINED
//
#if !defined(UNDEFINED)
    #define UNDEFINED ((void*)1)
#endif
 
//////////////////////////////////////////////////////////////////////
//
// custom types (mostly for shortening typing)
//
typedef struct node            Node;   // because we deserve nice things
typedef unsigned long long int code_t; // status code data type
typedef unsigned long long int ulli;
typedef   signed long long int slli;
 
//////////////////////////////////////////////////////////////////////
//
// Status codes for the doubly linked node implementation
//
#define  DLN_SUCCESS         0x0000000000000100
#define  DLN_MALLOC_FAIL     0x0000000000000200
#define  DLN_ALREADY_ALLOC   0x0000000000000400
#define  DLN_NULL            0x0000000000000800
#define  DLN_ERROR           0x0000000000001000
#define  DLN_INVALID         0x0000000000002000
#define  DLN_DEFAULT_FAIL    0x0000000000004000
#define  DLN_RESERVED_CODE   0x0000000000008000
 
#endif

You'll notice the presence of a set of #define's in the data.h header file. These are intended to be used to report on various states of node status after performing various operations.

They are not exclusive- in some cases, multiple states can be applied. The intent is that you will OR together all pertinent states and return that from the function.

• DLN_SUCCESS - everything went according to plan, no errors encountered, average case
• DLN_MALLOC_FAIL - memory allocation failed (considered in error)
• DLN_ALREADY_ALLOC - memory has already been allocated (considered in error)
• DLN_NULL - result is NULL (probably in error)
• DLN_DEFAULT_FAIL - default state of unimplemented functions (default error)
• DLN_ERROR - some error occurred
• DLN_INVALID - invalid use (NULL pointer)
• DLN_RESERVED_CODE - reserved for future use (not used at present time)

For example, in the case of “DLN_MALLOC_FAIL”, there are actually a total of three states raised:

• DLN_ERROR (a problem has occurred)
• DLN_MALLOC_FAIL (a problem has occurred when using malloc())
• DLN_NULL (no memory allocated, so node cannot be anything but NULL)

ALL THREE states must be returned from the function in question should such an occurrence take place.

You'll notice these #defines map to numeric values, and particular ones at that. This is to our supreme advantage: if you understand how numbers work, you should have an easy time of working with these status codes.

Finally we have support.h… this header will contain some information on helper functions utilized in the various unit tests. You really don't need to bother with this… in fact, do not use any of these functions in your implementation.

In src/node/, you will find skeletons of what was previously there, ready for you to re-implement.

Figure out what is going on, the connections, and make sure you understand it.

Be sure to focus on implementing the functionality from scratch (the more you do this from scratch, vs. referencing old code, the more it will help you).

To assist you in verifying a correct implementation, a fully working implementation of the node and list libraries should resemble the following (when running the respective verify script):

Here is what you should get for node:

lab46:~/src/data/dln0$ make check
======================================================
=    Verifying Doubly-Linked  Node Functionality     =
======================================================
    [mknode] Total:  12, Matches:  12, Mismatches:   0
    [cpnode] Total:  17, Matches:  17, Mismatches:   0
    [rmnode] Total:   4, Matches:   4, Mismatches:   0
======================================================
   [RESULTS] Total:  33, Matches:  33, Mismatches:   0
======================================================
lab46:~/src/data/dln0$