We Keep Coding

fastest starts with search algorithm

I need to implement a search algorithm which only searches from the start of the string rather than anywhere within the string.
I am new to algorithms but from what I can see it seems as though they go through the string and find any occurrence.
I have a collection of strings (over 1 million) which need to be searched everytime the user types a keystroke.
EDIT:
This will be an incremental search. I currently have it implemented with the following code and my searches are coming back ranging between 300-700ms from over 1 million possible strings. The collection isnt ordered but there is no reason it couldnt be.
private ICollection<string> SearchCities(string searchString) {
return _cityDataSource.AsParallel().Where(x => x.ToLower().StartsWith(searchString)).ToArray();
}

I've adapted the code from this article from Visual Studio Magazine that implements a Trie.
The following program demonstrates how to use a Trie to do fast prefix searching.
In order to run this program, you will need a text file called "words.txt" with a large list of words. You can download one from Github here.
After you compile the program, copy the "words.txt" file into the same folder as the executable.
When you run the program, type a prefix (such as prefix ;)) and press return, and it will list all the words beginning with that prefix.
This should be a very fast lookup - see the Visual Studio Magazine article for more details!
using System;
using System.Collections.Generic;
using System.IO;
using System.Linq;
using System.Text;
namespace ConsoleApp1
{
class Program
{
static void Main()
{
var trie = new Trie();
trie.InsertRange(File.ReadLines("words.txt"));
Console.WriteLine("Type a prefix and press return.");
while (true)
{
string prefix = Console.ReadLine();
if (string.IsNullOrEmpty(prefix))
continue;
var node = trie.Prefix(prefix);
if (node.Depth == prefix.Length)
{
foreach (var suffix in suffixes(node))
Console.WriteLine(prefix + suffix);
}
else
{
Console.WriteLine("Prefix not found.");
}
Console.WriteLine();
}
}
static IEnumerable<string> suffixes(Node parent)
{
var sb = new StringBuilder();
return suffixes(parent, sb).Select(suffix => suffix.TrimEnd('$'));
}
static IEnumerable<string> suffixes(Node parent, StringBuilder current)
{
if (parent.IsLeaf())
{
yield return current.ToString();
}
else
{
foreach (var child in parent.Children)
{
current.Append(child.Value);
foreach (var value in suffixes(child, current))
yield return value;
--current.Length;
}
}
}
}
public class Node
{
public char Value { get; set; }
public List<Node> Children { get; set; }
public Node Parent { get; set; }
public int Depth { get; set; }
public Node(char value, int depth, Node parent)
{
Value = value;
Children = new List<Node>();
Depth = depth;
Parent = parent;
}
public bool IsLeaf()
{
return Children.Count == 0;
}
public Node FindChildNode(char c)
{
return Children.FirstOrDefault(child => child.Value == c);
}
public void DeleteChildNode(char c)
{
for (var i = 0; i < Children.Count; i++)
if (Children[i].Value == c)
Children.RemoveAt(i);
}
}
public class Trie
{
readonly Node _root;
public Trie()
{
_root = new Node('^', 0, null);
}
public Node Prefix(string s)
{
var currentNode = _root;
var result = currentNode;
foreach (var c in s)
{
currentNode = currentNode.FindChildNode(c);
if (currentNode == null)
break;
result = currentNode;
}
return result;
}
public bool Search(string s)
{
var prefix = Prefix(s);
return prefix.Depth == s.Length && prefix.FindChildNode('$') != null;
}
public void InsertRange(IEnumerable<string> items)
{
foreach (string item in items)
Insert(item);
}
public void Insert(string s)
{
var commonPrefix = Prefix(s);
var current = commonPrefix;
for (var i = current.Depth; i < s.Length; i++)
{
var newNode = new Node(s[i], current.Depth + 1, current);
current.Children.Add(newNode);
current = newNode;
}
current.Children.Add(new Node('$', current.Depth + 1, current));
}
public void Delete(string s)
{
if (!Search(s))
return;
var node = Prefix(s).FindChildNode('$');
while (node.IsLeaf())
{
var parent = node.Parent;
parent.DeleteChildNode(node.Value);
node = parent;
}
}
}
}

A couple of thoughts:
First, your million strings need to be ordered, so that you can "seek" to the first matching string and return strings until you no longer have a match...in order (seek via C# List<string>.BinarySearch, perhaps). That's how you touch the least number of strings possible.
Second, you should probably not try to hit the string list until there's a pause in input of at least 500 ms (give or take).
Third, your queries into the vastness should be async and cancelable, because it's certainly going to be the case that one effort will be superseded by the next keystroke.
Finally, any subsequent query should first check that the new search string is an append of the most recent search string...so that you can begin your subsequent seek from the last seek (saving lots of time).

I suggest using linq.
string x = "searchterm";
List<string> y = new List<string>();
List<string> Matches = y.Where(xo => xo.StartsWith(x)).ToList();
Where x is your keystroke search text term, y is your collection of strings to search, and Matches is the matches from your collection.
I tested this with the first 1 million prime numbers, here is the code adapted from above:
Stopwatch SW = new Stopwatch();
SW.Start();
string x = "2";
List<string> y = System.IO.File.ReadAllText("primes1.txt").Split(' ').ToList();
y.RemoveAll(xo => xo == " " || xo == "" || xo == "\r\r\n");
List <string> Matches = y.Where(xo => xo.StartsWith(x)).ToList();
SW.Stop();
Console.WriteLine("matches: " + Matches.Count);
Console.WriteLine("time taken: " + SW.Elapsed.TotalSeconds);
Console.Read();
Result is:
matches: 77025
time taken: 0.4240604
Of course this is testing against numbers and I don't know whether linq converts the values before, or if numbers make any difference.

Creating lists with loops by an user in C#

So I want to make a list of names. I want this list to go on until user inputs 0 to quit, after user types 0 I want all names to be displayed. You probably see what I'm trying to do from the code below...that "typedName" is there just so you see what I'm trying to do.
namespace ConsoleApplication1
{
class Program
{
static void Main(string[] args)
{
List<string> names = new List<string>();
Console.WriteLine("Type in 0 to end.");
bool over = false;
while (over != true)
{
names.Add(Console.ReadLine());
if(typedName == "0")
{
over = true;
}
}
Console.WriteLine("Entered names : ");
names.ForEach(Console.WriteLine);
Console.ReadLine();
}
}
}

First you need the typedName to be captured and then check if it is equal to 0.
if it is not add it to the list
List<string> names = new List<string>();
Console.WriteLine("Type in 0 to end.");
while (true)
{
var typedName = Console.ReadLine();
if (typedName.Equals("0"))
{
break;
}
names.Add(typedName);
}
Console.WriteLine("Entered names : ");
foreach(var name in names)
{
Console.WriteLine(name);
}
Console.ReadLine();

if(typedName == "0")
Well, what is typedName? Or what should it be? I suspect it should be the input entered by the user, something like this:
var typedName = Console.ReadLine();
You can then add it to the list by using that variable:
names.Add(typedName);
And compare it with "0" as you already do, etc.

your code is not complete that is why is not working...
you are missing the most important part:
populate the list if and only if typedName != "0"
while (!over)
{
var typedName =Console.ReadLine();
if(typedName == "0")
{
over = true;
}else
{
Console.WriteLine("Enter a name... ");
names.Add(Console.ReadLine());
}
...
}

Validating String in C#

Hi so im trying to validate my string here so that it does not allow any input that starts with: "911" so if you type: "9 11", "91 1", "9 1 1" it should go through my if statement. It works with "911" but not the others, here's my code:
using System;
using System.Collections.Generic;
namespace Phone_List
{
class Program
{
static void Main(string[] args)
{
var phoneList = new List<string>();
string input;
Console.WriteLine("Input: ");
while ((input = Console.ReadLine()) != "")
{
phoneList.Add(input);
for (int i = 0; i < phoneList.Count; i++)
{
if (phoneList[i].Substring(0, 3) == "911")
{
input.StartsWith("9 11");
input.StartsWith("9 1 1");
input.StartsWith("91 1");
Console.WriteLine("NO");
Console.ReadLine();
return;
}
else
{
Console.WriteLine("YES");
Console.ReadLine();
return;
}
}
}
}
}
}
As you can see I am trying to use "input.StartsWith("9 11");" but it does not work...

You could use the Replace method of String; the condition you describe can be formulated as follows.
input.Replace(" ", "").StartsWith("911")

Use regular expressions for checks like this.
For example:
Regex.IsMatch(input, "^\\s*9\\s*1\\s*1");
This regex matches all strings that include whitespaces in front of and between "911".

Use the following to check if the string starts with "911":
First create a copy from the input string but without any white spaces:
string input_without_white_spaces =
new string(input.ToCharArray().Where(x => !char.IsWhiteSpace(x)).ToArray());
Then you can check if the string starts with 911 like this:
if (input_without_white_spaces.StartsWith("911"))
{
...
}

bool valid = s.StartsWith("911") ||
!string.Join("",s.Split()).StartsWith("911");

search text file using c# and display the line number and the complete line that contains the search keyword

I require help to search a text file (log file) using c# and display the line number and the complete line that contains the search keyword.

This is a slight modification from: http://msdn.microsoft.com/en-us/library/aa287535%28VS.71%29.aspx
int counter = 0;
string line;
// Read the file and display it line by line.
System.IO.StreamReader file = new System.IO.StreamReader("c:\\test.txt");
while((line = file.ReadLine()) != null)
{
if ( line.Contains("word") )
{
Console.WriteLine (counter.ToString() + ": " + line);
}
counter++;
}
file.Close();

Bit late to the game on this one, but happened across this post and thought I'd add an alternative answer.
foreach (var match in File.ReadLines(#"c:\LogFile.txt")
.Select((text, index) => new { text, lineNumber = index+ 1 })
.Where(x => x.text.Contains("SEARCHWORD")))
{
Console.WriteLine("{0}: {1}", match.lineNumber, match.text);
}
This uses:
File.ReadLines, which eliminates the need for a StreamReader, and it also plays nicely with LINQ's Where clause to return a filtered set of lines from a file.
The overload of Enumerable.Select that returns each element's index, which you can then add 1 to, to get the line number for the matching line.
Sample Input:
just a sample line
another sample line
first matching SEARCHWORD line
not a match
...here's aSEARCHWORDmatch
SEARCHWORD123
asdfasdfasdf
Output:
3: first matching SEARCHWORD line
5: ...here's aSEARCHWORDmatch
6: SEARCHWORD123

To export do Excel you can use the CSV file format, like the Pessimist wrote. If you are uncertain about what to write, try entering some data in MS Excel and click on "Save As" option in the Menu and choose CSV as file type.
Take care when writing a CSV file format as in some languages the default for separating values is not the comma. In brazilian portuguese, for example, the default is comma as decimal separator, dot as thousands separator and semicolon for separating values. Mind the culture when writing that.
The other alternative is using horizontal tabs as separators. Experiment to write a string, press the TAB key and then another string and paste it into Microsoft Excel. It is the default separator in that program.
If you're using an ad-hoc solution to your specific problem, either alternatives can be used without much thinking. If you are programming something to be used by other persons (or in other environments), mind the culture specific differences.
Oh, I've just remembered now: you can write a Spreadsheet using XML, you can do that with only the .NET package. I've done that years ago with C# .NET 2.0

I had a requirement where I needed to search through a list of directories looking for particular file types, containing a specific search terms but excluding other terms.
For example let's say you wanted to look through C:\DEV and only find .cs files that have terms "WriteLine" and "Readline" but not the term "hello".
I decided to write a small c# utility to do just this:
This is how you call it:
class Program
{
//Syntax:
//FileSearch <Directory> EXT <ext1> <ext2> LIKE <TERM1> <TERM2> NOT <TERM3> <TERM4>
//Example:
//Search for all files recursively in C:\Dev with an extension of cs that contain either "WriteLine" or "Readline" but not "hello"
//FileSearch C:\DEV EXT .cs LIKE "WriteLine" "ReadLine" NOT "hello"
static void Main(string[] args)
{
if (args.Length == 0)
{
Console.WriteLine("FileSearch <Directory> EXT <EXT1> LIKE <TERM1> <TERM2> NOT <TERM3> <TERM4>");
return;
}
Search s = new Search(args);
s.DoSearch();
}
}
This is the implementation:
using System;
using System.Collections.Generic;
using System.IO;
using System.Text;
class Hit
{
public string File { get; set; }
public int LineNo { get; set; }
public int Pos { get; set; }
public string Line { get; set; }
public string SearchTerm { get; set; }
public void Print()
{
Console.WriteLine(File);
Console.Write("(" + LineNo + "," + Pos + ") ");
Console.WriteLine(Line);
}
}
class Search
{
string rootDir;
List<string> likeTerms;
List<string> notTerms;
List<string> extensions;
List<Hit> hitList = new List<Hit>();
//FileSearch <Directory> EXT .CS LIKE "TERM1" "TERM2" NOT "TERM3" "TERM4"
public Search(string[] args)
{
this.rootDir = args[0];
this.extensions = ParseTerms("EXT", "LIKE", args);
this.likeTerms = ParseTerms("LIKE", "NOT", args);
this.notTerms = ParseTerms("NOT", "", args);
Print();
}
public void Print()
{
Console.WriteLine("Search Dir:" + rootDir);
Console.WriteLine("Extensions:");
foreach (string s in extensions)
Console.WriteLine(s);
Console.WriteLine("Like Terms:");
foreach (string s in likeTerms)
Console.WriteLine(s);
Console.WriteLine("Not Terms:");
foreach (string s in notTerms)
Console.WriteLine(s);
}
private List<string> ParseTerms(string keyword, string stopword, string[] args)
{
List<string> list = new List<string>();
bool collect = false;
foreach (string arg in args)
{
string argu = arg.ToUpper();
if (argu == stopword)
break;
if (argu == keyword)
{
collect = true;
continue;
}
if(collect)
list.Add(arg);
}
return list;
}
private void SearchDir(string dir)
{
foreach (string file in Directory.GetFiles(dir, "*.*"))
{
string extension = Path.GetExtension(file);
if (extension != null && extensions.Contains(extension))
SearchFile(file);
}
foreach (string subdir in Directory.GetDirectories(dir))
SearchDir(subdir);
}
private void SearchFile(string file)
{
using (StreamReader sr = new StreamReader(file))
{
int lineNo = 0;
while (!sr.EndOfStream)
{
int pos = 0;
string term = "";
string line = sr.ReadLine();
lineNo++;
//Look through each likeTerm
foreach(string likeTerm in likeTerms)
{
pos = line.IndexOf(likeTerm, StringComparison.OrdinalIgnoreCase);
if (pos >= 0)
{
term = likeTerm;
break;
}
}
//If found make sure not in the not term
if (pos >= 0)
{
bool notTermFound = false;
//Look through each not Term
foreach (string notTerm in notTerms)
{
if (line.IndexOf(notTerm, StringComparison.OrdinalIgnoreCase) >= 0)
{
notTermFound = true;
break;
}
}
//If not term not found finally add to hitList
if (!notTermFound)
{
Hit hit = new Hit();
hit.File = file;
hit.LineNo = lineNo;
hit.Pos = pos;
hit.Line = line;
hit.SearchTerm = term;
hitList.Add(hit);
}
}
}
}
}
public void DoSearch()
{
SearchDir(rootDir);
foreach (Hit hit in hitList)
{
hit.Print();
}
}
}

Please critique my class

I've taken a few school classes along time ago on and to be honest i never really understood the concept of classes. I recently "got back on the horse" and have been trying to find some real world application for creating a class.
you may have seen that I'm trying to parse a lot of family tree data that is in an very old and antiquated format called gedcom
I created a Gedcom Reader class to read in the file , process it and make it available as two lists that contain the data that i found necessary to use
More importantly to me is i created a class to do it so I would very much like to get the experts here to tell me what i did right and what i could have done better ( I wont say wrong because the thing works and that's good enough for me)
Class:
using System;
using System.Collections.Generic;
using System.Text;
using System.IO;
namespace GedcomReader
{
class Gedcom
{
private string GedcomText = "";
public struct INDI
{
public string ID;
public string Name;
public string Sex;
public string BDay;
public bool Dead;
}
public struct FAM
{
public string FamID;
public string Type;
public string IndiID;
}
public List<INDI> Individuals = new List<INDI>();
public List<FAM> Families = new List<FAM>();
public Gedcom(string fileName)
{
using (StreamReader SR = new StreamReader(fileName))
{
GedcomText = SR.ReadToEnd();
}
ReadGedcom();
}
private void ReadGedcom()
{
string[] Nodes = GedcomText.Replace("0 #", "\u0646").Split('\u0646');
foreach (string Node in Nodes)
{
string[] SubNode = Node.Replace("\r\n", "\r").Split('\r');
if (SubNode[0].Contains("INDI"))
{
Individuals.Add(ExtractINDI(SubNode));
}
else if (SubNode[0].Contains("FAM"))
{
Families.Add(ExtractFAM(SubNode));
}
}
}
private FAM ExtractFAM(string[] Node)
{
string sFID = Node[0].Replace("# FAM", "");
string sID = "";
string sType = "";
foreach (string Line in Node)
{
// If node is HUSB
if (Line.Contains("1 HUSB "))
{
sType = "PAR";
sID = Line.Replace("1 HUSB ", "").Replace("#", "").Trim();
}
//If node for Wife
else if (Line.Contains("1 WIFE "))
{
sType = "PAR";
sID = Line.Replace("1 WIFE ", "").Replace("#", "").Trim();
}
//if node for multi children
else if (Line.Contains("1 CHIL "))
{
sType = "CHIL";
sID = Line.Replace("1 CHIL ", "").Replace("#", "");
}
}
FAM Fam = new FAM();
Fam.FamID = sFID;
Fam.Type = sType;
Fam.IndiID = sID;
return Fam;
}
private INDI ExtractINDI(string[] Node)
{
//If a individual is found
INDI I = new INDI();
if (Node[0].Contains("INDI"))
{
//Create new Structure
//Add the ID number and remove extra formating
I.ID = Node[0].Replace("#", "").Replace(" INDI", "").Trim();
//Find the name remove extra formating for last name
I.Name = Node[FindIndexinArray(Node, "NAME")].Replace("1 NAME", "").Replace("/", "").Trim();
//Find Sex and remove extra formating
I.Sex = Node[FindIndexinArray(Node, "SEX")].Replace("1 SEX ", "").Trim();
//Deterine if there is a brithday -1 means no
if (FindIndexinArray(Node, "1 BIRT ") != -1)
{
// add birthday to Struct
I.BDay = Node[FindIndexinArray(Node, "1 BIRT ") + 1].Replace("2 DATE ", "").Trim();
}
// deterimin if there is a death tag will return -1 if not found
if (FindIndexinArray(Node, "1 DEAT ") != -1)
{
//convert Y or N to true or false ( defaults to False so no need to change unless Y is found.
if (Node[FindIndexinArray(Node, "1 DEAT ")].Replace("1 DEAT ", "").Trim() == "Y")
{
//set death
I.Dead = true;
}
}
}
return I;
}
private int FindIndexinArray(string[] Arr, string search)
{
int Val = -1;
for (int i = 0; i < Arr.Length; i++)
{
if (Arr[i].Contains(search))
{
Val = i;
}
}
return Val;
}
}
}
Implementation:
using System;
using System.Windows.Forms;
using GedcomReader;
namespace WindowsFormsApplication1
{
public partial class Form1 : Form
{
public Form1()
{
InitializeComponent();
}
private void button1_Click(object sender, EventArgs e)
{
string path = #"C:\mostrecent.ged";
string outpath = #"C:\gedcom.txt";
Gedcom GD = new Gedcom(path);
GraphvizWriter GVW = new GraphvizWriter("Family Tree");
foreach(Gedcom.INDI I in GD.Individuals)
{
string color = "pink";
if (I.Sex == "M")
{
color = "blue";
}
GVW.ListNode(I.ID, I.Name, "filled", color, "circle");
if (I.ID == "ind23800")
{MessageBox.Show("stop");}
//"ind23800" [ label="Sarah Mandley",shape="circle",style="filled",color="pink" ];
}
foreach (Gedcom.FAM F in GD.Families)
{
if (F.Type == "par")
{
GVW.ConnNode(F.FamID, F.IndiID);
}
else if (F.Type =="chil")
{
GVW.ConnNode(F.IndiID, F.FamID);
}
}
string x = GVW.SB.ToString();
GVW.SaveFile(outpath);
MessageBox.Show("done");
}
}
I am particularly interested in if anything could be done about the structures i don't know if how i use them in the implementation is the greatest but again it works
Thanks alot

Quick thoughts:
Nested types should not be visible.
ValueTypes (structs) should be immutable.
Fields (class variables) should not be public. Expose them via properties instead.
Check passed arguments for invalid values, like null.

It might be more readable. It's hard to read and understand.
You may study SOLID principles (http://butunclebob.com/ArticleS.UncleBob.PrinciplesOfOod)
Robert C. Martin gave good presentation on Oredev 2008 about clean code (http://www.oredev.org/topmenu/video/agile/robertcmartincleancodeiiifunctions.4.5a2d30d411ee6ffd2888000779.html)
Some recomended books to read about code readability:
Kent Beck "Implemetation patterns"
Robert C Martin "Clean Code" Robert C
Martin "Agile Principles, Patterns
and Practices in C#"

I suggest you check this place out: http://refactormycode.com/.
For some quick things, your naming is the biggest thing I would start to change.
No need to use ALL-CAPS or abbreviated terms.
Also, FxCop will help with a lot of suggested changes. For example, FindIndexinArray would be named FindIndexInArray.
EDIT:
I don't know if this is a bug in your code or by-design, but in FindIndexinArray, you don't break from your loop once you find a match. Do you want the first (break) or last (no break) match in the array?