readme.txt

数据结构c＋＋语言描述的源代码

LIST OF ALL C++ PROGRAMS IN CHAPTER 15

All programs were originally developed and tested using
Borland C++ version 5.01 for Windows 95.  The codes in this
directory have been altered as necessary so that they would compile and
execute properly with Microsoft's Visual C++ Version 5.0.
The primary changes result from the following:
1. The new function returns 0 on failure instead of throwing
   the xalloc exception.
2. The set new handler function has a different format.
3. Visual C++ does not recognize the fact that when an exception
   is thrown, control cannot return to the throw point.  As a result,
   the compiler gives the error ``not all control paths return a value''
   when compiling several of the original methods.  To get around
   this problem, unreachable return statements have been added.
4. Visual C++ has a different scoping rule for loop indexes
   declared within the loop header.  For example the loop index i
   declared as below
   for (int i = 0; i <= n; i++) {
       ...
       }

   is available outside the for loop block.  This causes a variable
   redefined error if we have another for loop with the header
   for (int i ...)

   To overcome this problem, the ``int'' from succeeding for loops
   has been deleted.
5. The max and min functions are not included in stdlib.h.
   Therefore, equivalent functions have been defined and placed
   in the file dosmax.h.
6. Visual C++ appears to have problems resolving template functions
   at times.  For example, some of the functions in binary.h, bbloadd.cpp
   and bbknap.cpp had to be changed from template functions to
   integer functions to get them to compile.
7. Visual C++ appears to have problems deleting arrays of type T
   when T has a default destructor.  For example, deleting
   the chain iterator array Pos (see DeactivatePos) in lnbase
   causes a crash.  A null destructor has been defined in citer1.h
   to overcome this problem.

For programs that require input to be provided, sample input is given in the
corresponding .dat file.  This sample input is NOT intended to constitute
an adequate test set.  For information on designing test data, see
Section 1.5 of the text.  To run the codes, you may need some of the .h files
from earlier chapters.


The output generated by each program is given in the corresponding .out file.

CHAPTER 15

Program 15.1 Recursive function for knapsack problem
file: rknap.cpp

Program 15.2 Computation of f and x
file: dknap.cpp

Program 15.3 Recursive computation of s, kay, and an optimal combining
file: rvbits.cpp

Program 15.4 Recursive computation avoiding recomputations
file: rcvbits.cpp

Program 15.5 Iterative computation of s and kay.
file: dvbits.cpp

Program 15.6 Recursive computation of c(i,j) and kay(i,j)
file: rmult.cpp

Program 15.7 Computing c(i,j) without recomputations
file: rcmult.cpp

Program 15.8 Iterative computation of c and kay
file: dmult.cpp

Program 15.9 Computation of c and kay
file: awd.h, allpairs.cpp, allpairs.dat

Program 15.10 Output a shortest path
file: awd.h, allpairs.cpp, allpairs.dat

Program 15.11 Finding a maximum non-crossing subset of nets
file: mns.cpp