Download OO Design for Course Scheduling: Class Diagram and Design Specification - Prof. Yuanfang C and more Study Guides, Projects, Research Software Engineering in PDF only on Docsity!
Design Document for OO Design for Case Study 1 (Course
Scheduling)
1 Overview
The goal of this project is to develop a system for scheduling the courses in a computer sci-
ence department, based on the input about classrooms, lecture times, and time preferences
of the different instructors. Different conditions have to be satisfied by the final schedule
and which have been specified in the SRS.
This document specifies the final system design for the system. It also gives some
explanations on how the design evolved and why some design decisions were taken.
2 Class Diagram
We have discussed how the design of this case study evolved in the book. Here we just
mention some of the key observations and design decisions that led to the final class diagram.
• From the problem specification, we identify the following classes: TimeTable, Course,
Room, LectureSlot, CToBeSched (course to be scheduled), InputFile 1, and InputFile 2.
• TimeTable is an important class. It is an aggregation of many TimeTableEntry, each
of which is a collection of a Course, a Room where the course is scheduled, and a
LectureSlot in which the course is scheduled.
• From file 1, we get: RoomDB, CourseDB, and SlotDB, each of which is an aggregation
of many members of Room, Course, and Slot, respectively.
• From file 2, we get a TableOfCToBeSched, which is an aggregation of many CToBeSched.
• For scheduling, courses were divided into four four different types: PGwithPref,
UGwithPref, PGwithoutPref, and UGwithoutPref.
• We have a class ConflictTable into which different conflicts for the different time
slots of different courses are stored, and from where they are later printed. This is a
list of ConflictTableEntry.
• Implementation concerns also added a template class List, an internal class PGReserve,
The final class diagram after the design is shown in Figure 1.
In this design, mapping between the modules in the design (which essentially is the
module view) and the component view given in the architecture document is not straight-
forward. Most of the classes are used by all the three components. However, the processing
of the system is as per the architecture. Let us look at the main() function in this design:
main (int argc, char *argv[]) {
InputFile_1 in1;
InputFile_2 in2;
TableOfCtoBeSched tbl;
CourseDB cDB;
RoomDB rDB;
SlotDB sDB;
TimeTable *timeTable;
ConflictTable *conflictTable;
// From main, first the databases are built by in1.build_CRS_DBs()
// Then table from file 2 is built by in2.buildCtoBeSched( )
// Rooms are then sorted, and scheduling is done by tbl.scheduleAll( )
// Timetable and conflict table are then printed.
As this shows, the processing is still done in three stages, each stage passing the infor-
mation to the next stage. The pipes are really the parameters and objects being passed
between the various objects created when the system executes.
3 Design Specification
Here we specify the design shown in Figure 1 precisely. We use C++ structures to specify
the design.
// This is the specification for the Object-Oriented Design for the Case Study. // It gives all the major classes and most of the major data members and // operations for these classes. The major parameters of the various operations // are also given, which shows how objects are being made visible. All the // major object declarations are also given. Comments are added only where // they are needed over and above the discussions above.
class Course { private: char *courseID; public: *tellCourseID (); isAValidID (); // To test whether courseID is invalid bool isAPGCourse (); void setAttr (char *); };
class Room { private: int roomNo; int capacity; public: Room (int num, int cap); bool tellAttrib (int &num, int &cap); bool areAttribValid (); // Atrribute validation };
Figure 1: Final class diagram for the case study (cont.)
class LectureSlot { private: char *slotID; public: char *tellSlotID (); bool isAValidSlot (); void setAttr (char *id); };
// Generic List Abstraction template class List { private: Element_t listElem[ MAX_ELEMENTS ]; int currentPtr; int stateMarker; public: bool isListEmpty (); bool insertElement (Element_t element);
// Next two operations needed to traverse the list void reset (); // Reset the CurrentPtr, if you want to scan from start bool getNextElement (Element_t &element); // In tables we will represent a room, course, etc. as an index in its DB. // Operations for random access in the list are therefore needed. bool getIthElement (int, Element_t &); bool setIthElement (int, Element_t); };
// Abstraction for room database class RoomDB { private: List<Room *> roomDB; public: bool insert (Room *); void sortRooms (); // Sort on capacity - makes looking for a room easier bool getSuitableRoom (enrol); // get smallest room that fits };
// Abstraction for course database class CourseDB { private: List<Course *> courseDB; public: bool insert (Course *); bool lookUp (int &, Course &); // Search the DB and return the index };
// Abstraction for class database class SlotDB { private: List<LectureSlot *> slotDB; public: bool insert (LectureSlot *); bool lookUp (int &, LectureSlot &); // Return index of the lecture slot };
// Object for File 1 class InputFile_1 { private: ifstream inFile; char *fileName; // Parsing functions bool semicolonExists (char *); // TRUE if semicolon is present in the line bool buildRoomDB (RoomDB &); // Parse file until roomDB is built bool build_CS_DB (CourseDB &, SlotDB &); // Build course and slot DBs char skipOverWhiteSpaces (); void getNextRoomEntry (int &, int &, int &); void getNextToken (int &, char *&); public: build_CRS_DBs (char *fName,CourseDB &cDB, RoomDB &rDB, SlotDB &sDB); };
// A course that has to be scheduled - from file 2. Forms the base class class CtoBeSched {
List<CtoBeSched *> table [NUM_OF_CATEGORIES]; PGReserve *pgReserve; // Reservation chart for PGsansP public: bool scheduleAll (TimeTable *&, SlotDB *,RoomDB *, ConflictTable *&); // Sends getScheduled message to all courses bool insert (int, CtoBeSched *); };
class TimeTableEntry { // One entry in the timetable private: int coursePtr; // Index into Course DB bool booked; // Some other course booked it? public: void setAttr (int cptr, bool mark); bool isThisBooked (); int tellWho (); void printEntry (CourseDB *cDB); };
class TimeTable { private: int PGOccupied[ MAX_SLOTS ]; TimeTableEntry table[ MAX_ROOMS ][ MAX_SLOTS ]; public: bool bookASlot (int, int, int); // Set table[i][j] = ptr ; bool isColumnEmpty (int colIndex); bool setPGOccupied (int slotIndex,int coursePtr); int PGwho (int coursePtr); int who (int roomPtr, int slotPtr); void printTimeTable (CourseDB *, RoomDB *, SlotDB *); bool isAlreadyOccupied (int roomPtr, int slotPtr); huntForSuitableRoom (int, int, RoomDB *, int &); };
class ConflictTable_Entry { private: int errorCode; int coursePtr; int conflictCourse; int preference; int roomPtr; public: void setAttr (int E, int c,int cc,int p,int r); void printAttr (CourseDB *,SlotDB *,RoomDB *); };
class ConflictTable { private: List< ConflictTable_Entry *> table; public: bool insertEntry (int, int, int, int, int); void printTable (CourseDB *,SlotDB *,RoomDB *); };
class InputFile_2 {
private: char *fileName; // Parsing functions char skipOverWhiteSpaces (); char *getNextToken (); char *getNextPref (); bool prefGiven (); public: bool buildCtoBeSched (char *fName, TableOfCtoBeSched &, CourseDB *, SlotDB *); };
main (int argc, char *argv[]) { InputFile_1 in1; InputFile_2 in2; TableOfCtoBeSched tbl; CourseDB cDB; RoomDB rDB; SlotDB sDB; TimeTable *timeTable; ConflictTable *conflictTable;
// From main, first the databases are built by in1.build_CRS_DBs() // Then table from file 2 is built by in2.buildCtoBeSched( ) // Rooms are then sorted, and scheduling is done by tbl.scheduleAll( ) // Timetable and conflict table are then printed. }
