The Objective

The objective of this assignment is to give you practice with project and function design. It will also give you an opportunity to work with reading information from a file, using structures, an array of structures, passing by reference, string comparison, sorting, and some formatted printing.

The Background

There are many companies that will pay people to find out which websites are popular so that they may better select where to place their adverstisements. We at CMSC201 Inc. are interested in where our students spend most of their time while using the Internet so that we may make our website more appealing to the student population.

I have installed a data gathering program on a set of computers which have been distributed throughout the campus. This program monitors and logs every website any student visits while logged into one of these computers. Yes, all the places you have visited. The logs generated have been gathered and formed into one extremely large data file we will be working with for this project named History.dat.

Data mining is a field in computer science where researchers look at creating new ways of analyzing large amounts of data. Their goal is to summarize any relationships that exist, and/or find relationships that are not as easily visible. This is one field that is currently growing in the UMBC Computer Science Department.

The Task

Design and code a project that will allow you to read in information from a data file, store it in arrays of structures so that the file can be closed, and then use the information in the arrays to answer some statistical questions. We are interested in knowing:

• the total number of unique websites found in the file
• total number of each of the domain name endings (.net, .gov... etc).
  We have chosen to limit this to: .net, .gov, .edu, and .com
• total number of websites that were visited more than once
• total time logged
• the statistical breakdown by type of websites visited
• the best content to add to our website ( based on number of hits )

The user will need to enter the following information:

• the name of the data file to be used (a string of size 30 or less)

The data file contains the following information per line:

• web address
• site type (EDUCATION, FUN, NEWS, and SEARCH are the only type of sites)
• logged time in seconds

• www - is the access code
• umbc - is the domain name
• edu - is the domain name ending

The Specifications

• You must use the following structure definitions and typedefs:
  typedef struct address { char accessCode[MAX_ACCESS_CODE_SIZE]; char domainName[MAX_DOMAIN_SIZE]; char domainEnding[MAX_DOMAIN_ENDING_SIZE]; }ADDRESS; typedef struct website { ADDRESS webAddress; char siteType[MAX_SITE_TYPE_SIZE]; int hits; int seconds; }WEBSITE; where MAX_ACCESS_CODE_SIZE is defined to be 15, MAX_DOMAIN_SIZE is 30, MAX_DOMAIN_ENDING_SIZE is defined to be 4, and MAX_SITE_TYPE_SIZE is 10.
  ADDRESS & these 4 #defines HAVE ALREADY BEEN WRITTEN AND EXIST IN proj3helper.h
  You'll need to define WEBSITE in your dataMine.h file.

• You will store the information that you get from the file into arrays, where each element is a website. Using 4 arrays will make your task much simplier. These arrays will contain websites and should be capable of maintaining up to 100 websites each. You MUST have arrays named: netSites, govSites, eduSites, and comSites that contain websites of corresponding domain name endings.

• You must ask the user to enter the name of the data file to be used and you must allow that filename to be as much as 30 characters long. Any data file we use for testing will have exactly the same format as History.dat.

• You must have a function named ComputeTotals() and it must have the following prototype: void ComputeTotals(WEBSITE array[], int *pTotalUniqueSites, int *pTotalNonUniqueSites, int *pTotalTimeLogged, int *pTotalFunSites, int *pTotalSearchSites, int *pTotalEducationSites, int *pTotalNewsSites); where :
  pTotalUniqueSites, pTotalTimeLogged are pointers to variables totalUniqueSites, totalLoggedTime.
  pTotalNonUniqueSites is a pointer to a variable which counts the total number of websites with more than one hit.
  The pointers pTotalFunSites, pTotalNewsSites, pTotalSearchSites, and pTotalEducationSites all point to variables which hold the number of sites based on the content types of websites.

• You must have a function named ArrayLength() and must have the following prototype: int ArrayLength(WEBSITE array[]); which takes an array of websites, which might not be full, and returns the total number of websites in the array.
  HINT: Filling the arrays initially with a constant DUMMY website as a place holder is necessary to determine how many websites have been entered into the array.
  A DUMMY wesbite might have the form:
  accessCode = "WWW"
  domainName = "DUMMY"
  domainEnding = "COM"
  hits = -1
  seconds = -1
  type = ""
  

• You must have a function named IsPresent() and it must have the following prototype: int IsPresent(WEBSITE array[], WEBSITE newSite); which takes an array of websites and an individual website, checks for a duplicate, and returns one of two possible values. One possible value is the index to the duplicate. If no duplicate is found, the value is the index of the first available open location in the array.
  HINT: In order for a website to be considered a duplicate, it must share the same accessCode, domainName, and domainEnding. You may even want to make this a function.
  In the event there this is a duplicate website, the following must be done:
  
  • increase the hits of the first copy by 1
  • add the time spent on the duplicate to the original's logged time.
  
  
• Converting a string into an ADDRESS is a difficult task, so we have provided some code which can be attained by executing this command: cp /afs/umbc.edu/users/r/b/rberge1/pub/proj3/proj3helper.* . In this file there is a function: ADDRESS CreateAddress(char webaddress[]); This function breaks a web address down and fills the members of the ADDRESS structure based on a '.' token.

• YOU MUST Sort each array. Sorting the arrays must result in the arrays being in descending order based on the number of hits.
  HINT: You may use code that has been provided for you in the lecture notes or that you have written in lab. However, you must modify it to handle an array of websites, and that it sorts the array in descending order. Also, please make sure you account for the array not being entirely full when it is to be sorted.

• Printing Statistics: will involve calculating the percentage of hits for each site type (FUN, EDUCATION, SEARCH & NEWS). You can find this value by dividing the total number of hits for each type by the total number of unique websites.

• Forming the Suggestion: will involve comparing the total number of hits for each site type. Print out the suggested content type to be added based on these comparisons.

• If you would like to have a function called OpenFile(), then it should return a FILE* to the calling function. You are NOT required to have a function called OpenFile(). You may open the file in main() and then pass the FILE* to the function that reads from the file if you prefer. Don't forget to close the file as soon as you have finished reading from it.

• Do NOT use dynamic memory allocation for the arrays of websites, just declare them (in main). If there are more websites in the file than can be stored in any array, which are of size 100, then you should print an error message to the user and exit the program.

EXTRA CREDIT

• The file has the time spent on each website in seconds. For extra credit convert this value to HH:MM:SS any time it is printed to the screen. This will be worth 5 points.

Sample Run

Submitting the Program

submitls cs201 Proj3