TweetFollow Us on Twitter

Jan 97 Challenge

Volume Number: 13 (1997)
Issue Number: 1
Column Tag: Programmer's Challenge

Programmer's Challenge

By Bob Boonstra, Westford, MA

BeSort

The theme of this issue of the magazine is the Be Operating System, so it seemed appropriate to focus the Challenge on the BeOS, and give you a chance to use the BeOS for Macintosh CD-ROM bundled in this issue. Since the BeOS will be new to most of you, this Challenge will be a simple one. The problem is to write a window class that will display and sort a list of strings by one of three specified methods: a bubble sort, an exchange sort, and an algorithm of your choosing. The prototype for the class you should write is

typedef enum SortType {
 kBubbleSort = 1,
 kExchangeSort = 2,
 kMySort = 3
} SortType;

class SortWindow : public BWindow {

public:
 SortWindow(BRect frame); 
virtual void   DoSort(
 char *thingsToSort[],  /* list of strings to sort, also returns sorted list */
 int numberOfThings, /* number of thingsToSort */
 SortType sortMethod);  /* sort method to use */
};

My test code will open three instances of your window class and ask each one to sort a copy of the same list of strings: one by the kBubbleSort method, one by kExchangeSort, and one by kMySort. Your SortWindow constructor should create a BListView and attach it to your SortWindow. When the DoSort method is invoked, you should display the thingsToSort and sort them into ascending ASCII order by the sortMethod algorithm. Each time two thingsToSort are exchanged, the BListView display should be updated. When the sort is complete, DoSort should post a B_QUIT_REQUESTED message to the application. The list should be sorted in place and returned in thingsToSort.

This will be a native PowerPC Challenge, using the latest Macintosh CodeWarrior environment, targeted for the BeOS. Solutions must be coded in C++. The code will be tested on my 8500 using the BeOS. (In the event I cannot get the BeOS to run on my Mac, I will run the tests on a 2x133MHz BeBox with one processor disabled.) The winner will be the solution that completes all three sorts correctly in the minimum time.

Three Months Ago Winner

Congratulations to Andy Antoniewicz (Mountain View, CA) for narrowly beating second and third place finishers Greg Cooper and Ludovic Nicolle in the October DNA Match Challenge. Of the fifteen entries I received for this Challenge, ten worked completely correctly, two were partially correct, and the remaining three crashed my machine.

Recall that the DNA Match Challenge was essentially a string matching problem, where the strings were allowed to differ in a specified number of positions (or fewer). The object was to return the number of near matches of a fragment string found in a reference database string.

My intent had been to test the solutions submitted using very long database strings. The run times of the solutions imposed a practical limit of about 2 MB on the size of the database string in an individual test case. The fragments to be matched were all significantly shorter than the database string, as indicated in the problem statement. The tests ranged from requiring very accurate matches, with a small value for the number of differences allowed, to approximate matches that could differ in up to half of the characters in the reference string.

The problem statement allowed for a timed initialization routine that would be executed once prior to testing matches against multiple fragments for a given database string. None of the top-ranked solutions made any significant use of this option, although a number of people used it to initialize small translation tables. Several people commented that it was difficult to find a use for this initialization routine, when the scratch storage provided was smaller than the maximum size database string.

Andy's winning entry parses the fragment string to create, for each character in the DNA "alphabet", a list of offsets where that character is located in the fragment. He then walks the database string and increments a match counter for each possible alignment of the fragment that matches the database at that character position. Andy uses a circular buffer twice the size of the fragment to store the match counts, which allows him to perform bounds checking on that buffer only once per database character, rather than within the innermost character matching loop. I had to read the code several times and run through a few cases manually before the light went on and I understood the algorithm, after which I found it quite clever.

While A.C.C. Murphy's solution did not place in the top five, one of his algorithms used a refinement worthy of note. He kept a running total of the counts of characters in a fragment-sized segment of the database, only checking for a specific match if the frequency counts were close enough to those of the fragment. In test cases where the character frequencies of the fragment were significantly different than much of the database, this technique might have done very well.

The table below summarizes the results for each correct or partially correct entry, including total execution time for all of the test cases and code size. Numbers in parenthesis after a person's name indicate that person's cumulative point total for all previous Challenges, not including this one. An asterisk indicates a result that was partially correct and therefore not eligible to win.

NameLanguageTotal TimeCode Size
Andy Antoniewicz (4)C123823728
Greg Cooper (17)C126233872
Ludovic Nicolle (14)C126646528
Michael Panchenko (6)C148243800
Bjorn Davidsson (4)C149171424
A.C.C. Murphy (10)C1510861800
Ernst Munter (224)C++176521752
Peter Lewis (32)C195312240
Mark DayC197385848
Larry Landry (29)C2671181376
Alan Hart (*)C210701848
Xin Xu (*)C2230151576

TOP 20 CONTESTANTS

Here are the Top 20 Contestants for the Programmer's Challenge. The numbers below include points awarded over the 24 most recent contests, including points earned by this month's entrants.

RankNamePoints
1.Munter, Ernst193
2.Gregg, Xan114
3.Larsson, Gustav87
4.Lengyel, Eric40
5.[Name deleted]40
6.Lewis, Peter32
7.Boring, Randy27
8.Cooper, Greg27
9.Antoniewicz, Andy24
10.Beith, Gary24
11.Kasparian, Raffi22
12.Cutts, Kevin21
13.Nicolle, Ludovic21
14.Picao, Miguel Cruz21
15.Brown, Jorg20
16.Gundrum, Eric20
17.Karsh, Bill19
18.Stenger, Allen19
19.Mallett, Jeff17
20.Nevard, John17

There are three ways to earn points: (1) scoring in the top 5 of any Challenge, (2) being the first person to find a bug in a published winning solution or, (3) being the first person to suggest a Challenge that I use. The points you can win are:

1st place20 points
2nd place10 points
3rd place7 points
4th place4 points
5th place2 points
finding bug2 points
suggesting Challenge2 points

Here is Andy's winning solution:

DNA_Match.c

© 1996 Andy Antoniewicz

/* 
Problem:
DNA string match with wildcard & constant distance

Notes:
No setup calculations on the database are performed. It seemed kind of pointless to attempt the equivalent 
of a 50 to 1 loss, less compression of the database ( it was 1000 to 1 for the first problem statement ).

This is a simple and quick single byte per pass index and count type algorithm. It uses three tables: a byte 
to code index table "aTable", a fragment index list array "alphaList",
and a hit count circular array "matchQueue". A precursor list is built and used once to build the alphaList 
array, and I do not clean up after (plenty of allocated storage).
Once built, the aTable contains an index into the alphaList for each given alphabet letter in the fragment. 
The alphaList contains a sequential list of all occurrences of that character in the fragment string.

The algorithm then increments the match count for all possible alignments of the fragment for each database 
character. Since the maximum misalignment is less than the fragment size, only the previous fragmentSize 
database characters need to be considered for matches at any particular time. Hence the circular queue 

The database search execution time is order( pN ) where N = number of database characters
p = average fragment entries per alphabet char 
( for example: A=3,C=3,G=2,T=2 -
> p = 2.5 )
The storage used is almost totally dependent on the fragment size. Storage used in bytes
    = 20     ( storage struct )
    + 256         ( 16bit aTable )
    + 2 * fragment chars( 16bit precursor )
    + 2 * fragment chars( 16bit alphaList )
    + 2 * alphabet chars (  more alphaList )
    + 4 * fragment chars( 2x16bit matchQueue )

 ********* ********* ********* ********* ********* ********/
 
#define kAlphaTableSize   128
#define kEndOfAlphaList   -1
 
typedef struct {
 long storageSize; // total storage available
 long usedStorage; // total storage used
 long fragmentSize;// size in chars of fragment
 short  *alphaList;// start of fragment index list
 short  *matchQueue; // start of match count queue
 
 short  alphaTable[ kAlphaTableSize ];
 short  precursor[]; // start of match count queue
} DNAStore;

/********* ********* ********* ********* ********* ********
 Function Prototypes
 ********* ********* ********* ********* ********* ********/

void InitMatch(
 char *alphabet, // legal characters in database
 char *database, // the reference database
 void *storage,  // pre-allocated storage
 long storageSize// size of storage in bytes
);

long FindMatch(  // return number of matches
 char *alphabet, // legal characters in database
 char *database, // the reference database
 void *storage,  // pre-allocated storage
 char *fragment, // the fragment to find
 long diffsAllowed,// num of diffs allowed between
    //  a "match" and the database
 long matchPosition[]// match return array
);

void BuildAlphaList(
 char   *alphabet, // legal characters in database
 char   *fragment, // the fragment to find; 0 term
 DNAStore *storage // my storage area
);

/********* ********* ********* ********* ********* ********
 * InitMatch
 * This routine does nothing except to store the
 *  size of memory allocated by the calling program.
 * All of the structures used are based on the fragment
 * to be searched for.
 ********* ********* ********* ********* ********* ********/

InitMatch
void InitMatch(
 char *alphabet, // legal characters in database
 char *database, // the reference database
 void *storage,  // pre-allocated storage
 long storageSize// size of storage in bytes
)
{
 ((DNAStore*)storage)->storageSize = storageSize;
 
}// end of InitMatch

/********* ********* ********* ********* ********* ********
 * BuildAlphaList
 * 
 * This routine has three parts:
 * Build the AlphaTable
 * The AlphaTable is an index table used to find
 * the start location inside the AlphaList
 * that corresponds to the given alphabet
 * character.
 * Build the AlphaList precursor
 * This is a linked list of the indexes to each
 * character in the fragment. It is used once
 * to build the AlphaList and never re-used.
 * Build the AlphaList
 * This is a sorted list of indexes for each
 * alphabet character. The head of each list is
 * stored in the AlphaTable, and each list is
 * ended by a -1.
 * 
 ********* ********* ********* ********* ********* ********/

BuildAlphaList
void BuildAlphaList(
 char   *alphabet, // legal characters in database
 char   *fragment, // the fragment to find; 0 term
 DNAStore *storage // my storage area
)
{
 short  *aTable; // ptr in the AlphaTable
 short  *precursor;// the AlphaList precursor
 short  *alphaList;// the AlphaList

 char *aString;  // ptr to a character string
 long aChar;// a character from a string
 
 long count;
 short  index;

  /*********
     * Initialize alphabet entries to kEndOfAlphaList
     *     Proper function for characters not in the alphabet
     *     requires that the other entries be preset to zero.
     *     This is automatic if the storage is cleared by
     *     the calling program and if the alphabet does not
     *     change between calls to InitMatch.
     */
 aTable = storage->alphaTable;
 aString = alphabet - 1;
 while( (aChar = (long)(*(++aString)) ) > 0x00 )
 {
 *(aTable+aChar) = kEndOfAlphaList;
 }
 
  /*********
     * Build precursor linked index list
     *     This list is only used to produce the 
     *     AlphaList below. The precursor and
     *     the AlphaList are rebuilt for each
     *     new fragment that will be searched for.
     */
 precursor= storage->precursor;
 aString= fragment - 1;
 index  = 0;
 count  = 0;
 while( (aChar = (long)(*(++aString))) > 0x00 )
 {
 index = *(aTable+aChar); // get prev head
 *(aTable+aChar) = count; // put new head
 *(precursor+count) = index;// store prev head
 count++;
 }
 storage->fragmentSize = count;
 
  /*********
     * Build AlphaList
     *     by walking each alphabet character's precursor 
     *     list and writing the index list into the 
     *     AlphaList the algorithm gets a sorted list of
     *     indexes into the fragment for each letter in the
     *     alphabet. The aTable will point to the first entry
     *     in the AlphaList and the last entry of each 
     *     character list is equal to the constant 
     *     kEndOfAlphaList ( = -1 ). Note that AlphaList
     *     location 0 is used to ignore database characters
     *     which are not in the given alphabet (it was free).
     */
 alphaList = precursor + count;
 *(alphaList) = kEndOfAlphaList;
 aString = alphabet - 1;
 count = 1;
 while( (aChar = (long)(*(++aString)) ) > 0x00 )
 {
 index = *(aTable+aChar);
 *(aTable+aChar) = count;
 while( index != kEndOfAlphaList )
 {
 *(alphaList+count) = index;
 index = *(precursor+index);
 count++;
 }
 *(alphaList+count) = kEndOfAlphaList;
 count++;
 }
 storage->alphaList= alphaList;
 storage->matchQueue = alphaList + count + 1;
 
}// end of BuildAlphaList


/********* ********* ********* ********* ********* ********
 * FindMatch
 *
 * For each character in the database increment every
 * match count that has the same character ( a hit ).
 * If after all possible alignments have been tallied,
 * the match count is greater than or equal to the value
 * of the threshold, then a match has been found.
 * Add the matching entry to the return array and continue
 * until all database characters have been tested.
 * 
 ********* ********* ********* ********* ********* ********/

FindMatch
long FindMatch(  // return number of matches
 char *alphabet, // legal characters in database
 char *database, // the reference database
 void *storage,  // pre-allocated storage
 char *fragment, // the fragment to find
 long diffsAllowed,// num of diffs allowed between
    //  a "match" and the fragment
 long matchPosition[]// match return array
)
{
 DNAStore *theStore; // typed storage
 short  *matchTop; // top of match array
 short  *matchCur; // current match array entry
 short  *aTable; // alpha to aList index table
 short  *aList;  // array of hit offsets
 short  *hitOffset;// current hit offset entry
 char   *dbString; // current database char entry
 long   dbChar;  // current database character
 long   count;   // current database location
 long   numMatch;// number of matches found
 long   fragSize;// fragment size in bytes
 long   loop;    // loop counter
 long   hitPos;  // current hit offset position
 long   threshold; // the threshold for matching
 
 theStore = (DNAStore*)storage;
 BuildAlphaList( alphabet,fragment,theStore);
 fragSize = theStore->fragmentSize;
 matchTop = theStore->matchQueue;
 matchCur = matchTop + fragSize;
 
 theStore->usedStorage
 = (long)(matchCur+fragSize)-(long)theStore;
    /*********
     * Clear the Match Count Array
     *     clear out the match counts for the entire array
     */
 for( count=0; count<fragSize; count++ )
 {
 *(matchCur+fragSize) = 0;
 *matchCur- = 0;
 }

    /*********
     * Walk the Database
     *
     *    The match count array is a double size circular
     *    queue of match counts. It is double size so I do
     *    not need to check the array bounds in the inner
     *    match count increment loop.
     */
  
 dbString = database - 1;
 aTable = theStore->alphaTable;
 aList = theStore->alphaList;
 threshold = fragSize-diffsAllowed;

 numMatch = 0;
 count = -fragSize;// count = current database loc.
 
 while( (dbChar = (long)(*(++dbString))) > 0x00 )
 {
    // circular queue reset to center
 if( matchCur == matchTop )
 {
 matchCur += fragSize;
 }
 
    // check for a match to the fragment
 if((*matchCur+*(matchCur+fragSize)) >= threshold )
 {
 matchPosition[ numMatch ] = count;
 numMatch++;
 }
 
    // clear both old counts
 *matchCur = 0;
 *(matchCur+fragSize) = 0;

    // increment match counts for all possible
    //     fragment alignments
 hitOffset = aList + *(aTable + dbChar) - 1;
 while( (hitPos = (long)(*(++hitOffset))) >= 0 )
 {
 *(matchCur + hitPos ) += 1;
 }
 
    // The match count queue is walked in reverse
    //     order through memory because the alphaList
    //     indexes are positive.
 matchCur-;
 count++;
 }
 
    /*********
     * Check the remaining match count entries for
     *    any fragments that extend beyond the end of
     *    the database.
     */
  
 for( loop=0; loop < diffsAllowed; loop++)
 {
 if( matchCur == matchTop )
 {
 matchCur += fragSize;
 }
 
 if((*matchCur+*(matchCur+fragSize)) >= threshold )
 {
 matchPosition[ numMatch ] = count;
 numMatch++;
 }
 matchCur-;
 count++; 
 }
 
 return(numMatch);
 }
 
    // end of DNA_Match.c

 
AAPL
$102.50
Apple Inc.
+0.25
MSFT
$45.43
Microsoft Corpora
+0.55
GOOG
$571.60
Google Inc.
+2.40

MacTech Search:
Community Search:

Software Updates via MacUpdate

Path Finder 6.5.5 - Powerful, award-winn...
Path Finder is a file browser that combines the familiar Finder interface with the powerful utilities and innovative features. Just a small selection of the Path Finder 6 feature set: Dual pane... Read more
QuarkXPress 10.2.1 - Desktop publishing...
With QuarkXPress, you can communicate in all the ways you need to -- and always look professional -- in print and digital media, all in a single tool. Features include: Easy to Use -- QuarkXPress is... Read more
Skype 6.19.0.450 - Voice-over-internet p...
Skype allows you to talk to friends, family and co-workers across the Internet without the inconvenience of long distance telephone charges. Using peer-to-peer data transmission technology, Skype... Read more
VueScan 9.4.41 - Scanner software with a...
VueScan is a scanning program that works with most high-quality flatbed and film scanners to produce scans that have excellent color fidelity and color balance. VueScan is easy to use, and has... Read more
Cloud 3.0.0 - File sharing from your men...
Cloud is simple file sharing for the Mac. Drag a file from your Mac to the CloudApp icon in the menubar and we take care of the rest. A link to the file will automatically be copied to your clipboard... Read more
LibreOffice 4.3.1.2 - Free Open Source o...
LibreOffice is an office suite (word processor, spreadsheet, presentations, drawing tool) compatible with other major office suites. The Document Foundation is coordinating development and... Read more
SlingPlayer Plugin 3.3.20.505 - Browser...
SlingPlayer is the screen interface software that works hand-in-hand with the hardware inside the Slingbox to make your TV viewing experience just like that at home. It features an array of... Read more
Get Lyrical 3.8 - Auto-magically adds ly...
Get Lyrical auto-magically add lyrics to songs in iTunes. You can choose either a selection of tracks, or the current track. Or turn on "Active Tagging" to get lyrics for songs as you play them.... Read more
Viber 4.2.2 - Send messages and make cal...
Viber lets you send free messages and make free calls to other Viber users, on any device and network, in any country! Viber syncs your contacts, messages and call history with your mobile device,... Read more
Cocktail 7.6 - General maintenance and o...
Cocktail is a general purpose utility for OS X that lets you clean, repair and optimize your Mac. It is a powerful digital toolset that helps hundreds of thousands of Mac users around the world get... Read more

Latest Forum Discussions

See All

Rhonna Designs Magic (Photography)
Rhonna Designs Magic 1.0 Device: iOS Universal Category: Photography Price: $1.99, Version: 1.0 (iTunes) Description: Want to sprinkle *magic* on your photos? With RD Magic, you can add colors, filters, light leaks, bokeh, edges,... | Read more »
This Week at 148Apps: August 25-29, 2014
Shiny Happy App Reviews   | Read more »
Qube Kingdom – Tips, Tricks, Strategies,...
Qube Kingdom is a tower defense game from DeNA. You rally your troops – magicians, archers, knights, barbarians, and others – and fight against an evil menace looking to dominate your kingdom of tiny squares. Planning a war isn’t easy, so here are a... | Read more »
Qube Kingdom Review
Qube Kingdom Review By Nadia Oxford on August 29th, 2014 Our Rating: :: KIND OF A SQUARE KINGDOMUniversal App - Designed for iPhone and iPad Qube Kingdom has cute visuals, but it’s a pretty basic tower defense game at heart.   | Read more »
Fire in the Hole Review
Fire in the Hole Review By Rob Thomas on August 29th, 2014 Our Rating: :: WALK THE PLANKUniversal App - Designed for iPhone and iPad Seafoam’s Fire in the Hole looks like a bright, 8-bit throwback, but there’s not enough booty to... | Read more »
Alien Creeps TD is Now Available Worldwi...
Alien Creeps TD is Now Available Worldwide Posted by Ellis Spice on August 29th, 2014 [ permalink ] Universal App - Designed for iPhone and iPad | Read more »
Dodo Master Review
Dodo Master Review By Jordan Minor on August 29th, 2014 Our Rating: :: NEST EGGiPad Only App - Designed for the iPad Dodo Master is tough but fair, and that’s what makes it a joy to play.   | Read more »
Motorsport Manager Review
Motorsport Manager Review By Lee Hamlet on August 29th, 2014 Our Rating: :: MARVELOUS MANAGEMENTUniversal App - Designed for iPhone and iPad Despite its depth and sense of tactical freedom, Motorsport Manager is one of the most... | Read more »
Motorsport Manager – Beginner Tips, Tric...
The world of Motorsport management can be an unforgiving and merciless one, so to help with some of the stress that comes with running a successful race team, here are a few hints and tips to leave your opponents in the dust. | Read more »
CalPal Update Brings the App to 2.0, Add...
CalPal Update Brings the App to 2.0, Adds Lots of New Stuff Posted by Ellis Spice on August 29th, 2014 [ permalink ] | Read more »

Price Scanner via MacPrices.net

Apple now offering refurbished 21-inch 1.4GHz...
The Apple Store is now offering Apple Certified Refurbished 21″ 1.4GHz iMacs for $929 including free shipping plus Apple’s standard one-year warranty. Their price is $170 off the cost of new models,... Read more
Save $50 on the 2.5GHz Mac mini, on sale for...
B&H Photo has the 2.5GHz Mac mini on sale for $549.99 including free shipping. That’s $50 off MSRP, and B&H will also include a free copy of Parallels Desktop software. NY sales tax only. Read more
Save up to $300 on an iMac with Apple refurbi...
The Apple Store has Apple Certified Refurbished iMacs available for up to $300 off the cost of new models. Apple’s one-year warranty is standard, and shipping is free. These are the best prices on... Read more
The Rise of Phablets
Carlisle & Gallagher Consulting Group, a businesses and technology consulting firm focused solely on the financial services industry, has released an infographic depicting the convergence of... Read more
Bad Driver Database App Allows Good Drivers t...
Bad Driver Database 1.4 by Facile Group is a new iOS and Android app that lets users instantly input and see how many times a careless, reckless or just plain stupid driver has been added to the... Read more
Eddy – Cloud Music Player for iPhone/iPad Fre...
Ukraine based CapableBits announces the release of Eddy, its tiny, but smart and powerful cloud music player for iPhone and iPad that allows users to stream or download music directly from cloud... Read more
A&D Medical Launches Its WellnessConnecte...
For consumers and the healthcare providers and loved ones who care for them, A&D Medical, a leader in connected health and biometric measurement devices and services, has launched its... Read more
Anand Lal Shimpi Retires From AnandTech
Anand Lal Shimpi, whose AnandTech Website is famous for its meticulously detailed and thoroughgoing reviews and analysis, is packing it in. Lal Shimpi, who founded the tech site at age 14 in 1997,... Read more
2.5GHz Mac mini, Apple refurbished, in stock...
The Apple Store has Apple Certified Refurbished 2.5GHz Mac minis available for $509, $90 off MSRP. Apple’s one-year warranty is included, and shipping is free. Read more
13-inch 2.5GHz MacBook Pro on sale for $999,...
B&H Photo has the 13″ 2.5GHz MacBook Pro on sale for $999.99 including free shipping plus NY sales tax only. Their price is $100 off MSRP. Read more

Jobs Board

*Apple* Retail - Multiple Positions (US) - A...
Sales Specialist - Retail Customer Service and Sales Transform Apple Store visitors into loyal Apple customers. When customers enter the store, you're also the Read more
*Apple* Retail - Multiple Positions (US) - A...
Sales Specialist - Retail Customer Service and Sales Transform Apple Store visitors into loyal Apple customers. When customers enter the store, you're also the Read more
*Apple* Retail - Multiple Positions (US) - A...
Sales Specialist - Retail Customer Service and Sales Transform Apple Store visitors into loyal Apple customers. When customers enter the store, you're also the Read more
*Apple* Retail - Multiple Positions (US) - A...
Sales Specialist - Retail Customer Service and Sales Transform Apple Store visitors into loyal Apple customers. When customers enter the store, you're also the Read more
Senior Event Manager, *Apple* Retail Market...
…This senior level position is responsible for leading and imagining the Apple Retail Team's global event strategy. Delivering an overarching brand story; in-store, Read more
All contents are Copyright 1984-2011 by Xplain Corporation. All rights reserved. Theme designed by Icreon.