Iterating backwards through an char array after finding a known word

Question

I've got a project I'm working on in C#. I've got two char array's. One is a sentence and one is a word. I've got to iterate through the sentence array until I find a word that matches the word that was turned into an word array what I'm wondering is once I find the word how do I iterate backwards through the sentence array at the point I found the word back through the same length as the word array?

Code :

String wordString = "(Four)";
String sentenceString = "Maybe the fowl of Uruguay repeaters (Four) will be found";
char[] wordArray = wordString.ToCharArray();
List<String> words = sentenceString.Split(' ').ToList<string>();
//This would be the part where I iterate through sentence
foreach (string sentence in sentArray)
{
     //Here I would need to find where the string of (Four) and trim it and see if it equals the wordString. 
     if (sentence.Contains(wordString)
     {
         //At this point I would need to go back the length of wordString which happens to be four places but I'm not sure how to do this.  And for each word I go back in the sentence I need to capture that in another string array.
      }

I don't know if I'm being clear enough on this but if I'm not please feel free to ask.. Thank you in advanced. Also what this should return is "fowl of Uruguay repeaters". So basically the use case is for the number of letters in the parenthesis the logic should return the same number of words before the word in parenthesis.

Answer 1

int i ;
string outputString = (i=sentenceString.IndexOf(wordString))<0 ?     
                          sentenceString : sentenceString.Substring(0,i) ;

Answer 2

var wordString = "(Four)";
            int wordStringInt = 4; // Just do switch case to convert your string to int
            var sentenceString = "Maybe the fowl of Uruguay repeaters (Four) will be found";
            var sentenceStringArray = sentenceString.Split(' ').ToList();
            int wordStringIndexInArray = sentenceStringArray.IndexOf(wordString) - 1;
            var stringOutPut = "";
            if (wordStringIndexInArray > 0 && wordStringIndexInArray > wordStringInt)
            {
                stringOutPut = "";
                while (wordStringInt > 0)
                {
                    stringOutPut = sentenceStringArray[wordStringInt] + " " + stringOutPut;
                    wordStringInt--;
                }

            }

Answer 3

our are you. I have few question concerned to this exercise. If the Word (four) was in beginning it should not return? or return all string? As the length of four is equal to 4 imagine if that word appear as the second word on sentence what it should return just the first word or return 4 words even including the (four) word.?

My solution is the laziest one I just see your question and decide to help.

• My solution assumes it return all the word before the (four) if the length is bigger than the word before the (four) word.
• My solution return empty string if the word (four) is in the beginning.
• My solution return Length of (four) (4) words before (four) word.

ONCE AGAIN IT IS NOT MY BEST APPROACH.

I see the code bellow:

        string wordString = "(Four)";
        string sentenceString = "Maybe the fowl of Uruguay repeaters (Four) will be found";
        //Additionally you can add splitoption to remove the empty word on split function bellow
        //Just in case there are more space in sentence.
        string[] splitedword = sentenceString.Split(' ');
        int tempBackposition = 0;
        int finalposition = 0;
        for (int i = 0; i < splitedword.Length; i++)
        {
            if (splitedword[i].Contains(wordString))
            {
                finalposition = i;
                break;
            }
        }
        tempBackposition = finalposition - wordString.Replace("(","").Replace(")","").Length;
        string output = "";
        tempBackposition= tempBackposition<0?0:tempBackposition;
        for (int i = tempBackposition; i < finalposition; i++)
        {
            output += splitedword[i] + " ";
        }
        Console.WriteLine(output);
        Console.ReadLine();

If it's not what you want can you answer my questions on top? or help me to understand were it's wrong

Answer 4

What you are matching is kind of complex, so for a more general solution you could use regular expressions.

First we declare what we are searching for:

string word = "(Four)";
string sentence = "Maybe the fowl of Uruguay repeaters (Four) will be found";

We will then search for the words in this string using regular expressions. Since we don't want to match whitespace, and we need to know where each match actually starts and we need to know the word inside the parenthesis we tell it that we optionally want opening and ending parenthesis, but we also want the contents of those as a match:

var words = Regex.Matches(sentence, @"[\p{Ps}]*(?<Content>[\w]+)[\p{Pe}]*").Cast<Match>().ToList();

[\p{Ps}] means we want opening punctuation ([{ etc. while the * indicates zero or more.

Followed is a sub-capture called Content (specified by ?<Content>) with one or more word characters. At the end we specify that we want zero or more ending punctuation.

We then need to find the word in the list of matches:

var item = words.Single(x => x.Value == word);

Then we need to find this item's index:

int index = words.IndexOf(item);

At this point we just need to know the length of the contents:

var length = item.Groups["Content"].Length;

This length we use to go back in the string 4 words

var start = words.Skip(index - length).First();

And now we have everything we need:

var result = sentence.Substring(start.Index, item.Index - start.Index);

Result should contain fowl of Uruguay repeaters.

edit: It may be a lot simpler to just figure out the count from the word rather than from the content. In that case the complete code should be the following:

string word = "(Four)";
string sentence = "Maybe the fowl of Uruguay repeaters (Four) will be found";

var wordMatch = Regex.Match(word, @"[\p{Ps}]*(?<Content>[\w]+)[\p{Pe}]*");
var length = wordMatch.Groups["Content"].Length;

var words = Regex.Matches(sentence, @"\S+").Cast<Match>().ToList();
var item = words.Single(x => x.Value == word);
int index = words.IndexOf(item);


var start = words.Skip(index - length).First();
var result = sentence.Substring(start.Index, item.Index - start.Index);

\S+ in this case means "match one or more non-whitespace character".

Answer 5

You should try something like the following, which uses Array.Copy after it finds the number word. You will still have to implement the ConvertToNum function correctly (it is hardcoded in for now), but this should be a quick and easy solution.

string[] GetWords()
{
    string sentenceString = "Maybe the fowl of Uruguay repeaters (Four) will be found";
    string[] words = sentenceString.Split();
    int num = 0;
    int i; // scope of i should remain outside the for loop
    for (i = 0; i < words.Length; i++)
    {
        string word = words[i];
        if (word.StartsWith("(") && word.EndsWith(")"))
        {
            num = ConvertToNum(word.Substring(1, word.Length - 1));
            // converted the number word we found, so we break
            break;
        }
    }
    if (num == 0)
    {
        // no number word was found in the string - return empty array
        return new string[0];
    }
        // do some extra checking if number word exceeds number of previous words
        int startIndex = i - num;
        // if it does - just start from index 0
        startIndex = startIndex < 0 ? 0 : startIndex;
        int length = i - startIndex;
        string[] output = new string[length];
        Array.Copy(words, startIndex, output, 0, length);
        return output;
}


// Convert the number word to an integer
int ConvertToNum(string numberStr)
{
    return 4; // you should implement this method correctly
}

See - Convert words (string) to Int, for help implementing the ConvertToNum solution. Obviously it could be simplified depending on the range of numbers you expect to deal with.

Answer 6

Here is my solution, im not using regex at all, for the sake of easy understanding:

        static void Main() {
            var wordString = "(Four)";
            int wordStringLength = wordString.Replace("(","").Replace(")","").Length; 
            //4, because i'm assuming '(' and ')' doesn't count. 

            var sentenceString = "Maybe the fowl of Uruguay repeaters (Four) will be found";

            //Transform into a list of words, ToList() to future use of IndexOf Method
            var sentenceStringWords = sentenceString.Split(' ').ToList();

            //Find the index of the word in the list of words
            int wordIndex = sentenceStringWords.IndexOf(wordString);

            //Get a subrange from the original list of words, going back x Times the legnth of word (in this case 4),

            var wordsToConcat = sentenceStringWords.GetRange(wordIndex-wordStringLength, wordStringLength);

            //Finally concat the output;
            var outPut = string.Join(" ", wordsToConcat);

            //Output: fowl of Uruguay repeaters

        }

Answer 7

I have a solution for you:

string wordToMatch = "(Four)";
            string sentence = "Maybe the fowl of Uruguay repeaters (Four) will be found";

            if (sentence.Contains(wordToMatch))
            {
                int length = wordToMatch.Trim(new[] { '(', ')' }).Length;
                int indexOfMatchedWord = sentence.IndexOf(wordToMatch);
                string subString1 = sentence.Substring(0, indexOfMatchedWord);
                string[] words = subString1.Split(new[] { ' ' }, StringSplitOptions.RemoveEmptyEntries);
                var reversed = words.Reverse().Take(length);

                string result = string.Join(" ", reversed.Reverse());
                Console.WriteLine(result);
                Console.ReadLine();
            }

Likely this can be improved regardng performance, but I have a feeling you don't care about that. Make sure you are using 'System.Linq'

Answer 8

I assumed empty returns when input is incomplete, feel free to correct me on that. Wasn't 100% clear in your post how this should be handled.

    private string getPartialSentence(string sentence, string word)
    {
        if (string.IsNullOrEmpty(sentence) || string.IsNullOrEmpty(word))
            return string.Empty;

        int locationInSentence = sentence.IndexOf(word, StringComparison.Ordinal);

        if (locationInSentence == -1)
            return string.Empty;

        string partialSentence = sentence.Substring(0, locationInSentence);
        string[] words = partialSentence.Split(new[] {' '}, StringSplitOptions.RemoveEmptyEntries);

        int nbWordsRequired = word.Replace("(", "").Replace(")", "").Length;

        if (words.Count() >= nbWordsRequired)
            return String.Join(" ", words.Skip(words.Count() - nbWordsRequired));

        return String.Join(" ", words);
    }

Answer 9

I went with using an enumeration and associated dictionary to pair the "(Four)" type strings with their integer values. You could just as easily (and maybe easier) go with a switch statement using

case "(Four)": { currentNumber = 4; };

I feel like the enum allows a little more flexibility though.

public enum NumberVerb
    {
        one = 1,
        two = 2,
        three = 3,
        four = 4,
        five = 5,
        six = 6,
        seven = 7,
        eight = 8,
        nine = 9,
        ten = 10,
    };

    public static Dictionary<string, NumberVerb> m_Dictionary
    {
        get
        {
            Dictionary<string, NumberVerb> temp = new Dictionary<string, NumberVerb>();

            temp.Add("(one)", NumberVerb.one);
            temp.Add("(two)", NumberVerb.two);
            temp.Add("(three)", NumberVerb.three);
            temp.Add("(four)", NumberVerb.four);
            temp.Add("(five)", NumberVerb.five);
            temp.Add("(six)", NumberVerb.six);
            temp.Add("(seven)", NumberVerb.seven);
            temp.Add("(eight)", NumberVerb.eight);
            temp.Add("(nine)", NumberVerb.nine);
            temp.Add("(ten)", NumberVerb.ten);

            return temp;
        }
    }

    static void Main(string[] args)
    {
        string resultPhrase = "";

        // Get the sentance that will be searched.
        Console.WriteLine("Please enter the starting sentance:");
        Console.WriteLine("(don't forget your keyword: ie '(Four)')");
        string sentance = Console.ReadLine();

        // Get the search word.
        Console.WriteLine("Please enter the search keyword:");
        string keyword = Console.ReadLine();

        // Set the associated number of words to backwards-iterate.
        int currentNumber = -1;
        try 
        {
            currentNumber = (int)m_Dictionary[keyword.ToLower()];
        }
        catch(KeyNotFoundException ex)
        {
            Console.WriteLine("The provided keyword was not found in the dictionary.");
        }

        // Search the sentance string for the keyword, and get the starting index.
        Console.WriteLine("Searching for phrase...");

        string[] words = sentance.Split(' ');
        int searchResultIndex = -1;

        for (int i = 0; (searchResultIndex == -1 && i < words.Length); i++)
        {
            if (words[i].Equals(keyword))
            {
                searchResultIndex = i;
            }
        }

        // Handle the search results.
        if (searchResultIndex == -1)
        {
            resultPhrase = "The keyword was not found.";
        }
        else if (searchResultIndex < currentNumber)
        {
            // Check the array boundaries with the given indexes.
            resultPhrase = "Error: Out of bounds!";
        }
        else
        {
            // Get the preceding words.
            for (int i = 0; i < currentNumber; i++)
            {
                resultPhrase = string.Format(" {0}{1}", words[searchResultIndex - 1 - i], resultPhrase);
            }
        }

        // Display the preceding words.
        Console.WriteLine(resultPhrase.Trim());

        // Exit.
        Console.ReadLine();
    }