Find a substring, replace a substring according the case - c#

What's the easiest and fastest way to find a sub-string(template) in a string and replace it with something else following the template's letter case (if all lower case - replace with lowercase, if all upper case - replace with uppercase, if begins with uppercase and so on...)
so if the substring is in curly braces
"{template}" becomes "replaced content"
"{TEMPLATE}" becomes "REPLACED CONTENT" and
"{Template}" becomes "Replaced content" but
"{tEMPLATE}" becomes "rEPLACED CONTENT"

Well, you could use regular expressions and a match evaluator callback like this:
regex = new Regex(#"\{(?<value>.*?)\}",
RegexOptions.CultureInvariant | RegexOptions.ExplicitCapture);
string replacedText = regex.Replace(<text>,
new MatchEvaluator(this.EvaluateMatchCallback));
And your evaluator callback would do something like this:
private string EvaluateMatchCallback(Match match) {
string templateInsert = match.Groups["value"].Value;
// or whatever
string replacedText = GetReplacementTextBasedOnTemplateValue(templateInsert);
return replacedText;
}
Once you get the regex match value you can just do a case-sensitive comparison and return the correct replacement value.
EDIT I sort of assumed you were trying to find the placeholders in a block of text rather than worry about the casing per se, if your pattern is valid all the time then you can just check the first two characters of the placeholder itself and that will tell you the casing you need to use in the replacement expression:
string foo = "teMPLATE";
if (char.IsLower(foo[0])) {
if (char.IsLower(foo[1])) {
// first lower and second lower
}
else {
// first lower and second upper
}
}
else {
if (char.IsLower(foo[1])) {
// first upper and second lower
}
else {
// first upper and second upper
}
}
I would still use a regular expression to match the replacement placeholder, but that's just me.

You can check the case of the first two letters of the placeholder and choose one of the four case transforming strategies for the inserted text.
public static string Convert(string input, bool firstIsUpper, bool restIsUpper)
{
string firstLetter = input.Substring(0, 1);
firstLetter = firstIsUpper ? firstLetter.ToUpper() : firstLetter.ToLower();
string rest = input.Substring(1);
rest = restIsUpper ? rest.ToUpper() : rest.ToLower();
return firstLetter + rest;
}
public static string Replace(string input, Dictionary<string, string> valueMap)
{
var ms = Regex.Matches(input, "{(\\w+?)}");
int i = 0;
var sb = new StringBuilder();
for (int j = 0; j < ms.Count; j++)
{
string pattern = ms[j].Groups[1].Value;
string key = pattern.ToLower();
bool firstIsUpper = char.IsUpper(pattern[0]);
bool restIsUpper = char.IsUpper(pattern[1]);
sb.Append(input.Substring(i, ms[j].Index - i));
sb.Append(Convert(valueMap[key], firstIsUpper, restIsUpper));
i = ms[j].Index + ms[j].Length;
}
return sb.ToString();
}
public static void DoStuff()
{
Console.WriteLine(Replace("--- {aAA} --- {AAA} --- {Aaa}", new Dictionary<string,string> {{"aaa", "replacement"}}));
}

Ended up doing that:
public static string ReplaceWithTemplate(this string original, string pattern, string replacement)
{
var template = Regex.Match(original, pattern, RegexOptions.IgnoreCase).Value.Remove(0, 1);
template = template.Remove(template.Length - 1);
var chars = new List<char>();
var isLetter = false;
for (int i = 0; i < replacement.Length; i++)
{
if (i < (template.Length)) isLetter = Char.IsUpper(template[i]);
chars.Add(Convert.ToChar(
isLetter ? Char.ToUpper(replacement[i])
: Char.ToLower(replacement[i])));
}
return new string(chars.ToArray());
}

Related

Method that takes a message and index, creates a substring using the index

Problem: I want to write a method that takes a message/index pair like this:
("Hello, I am *Name1, how are you doing *Name2?", 2)
The index refers to the asterisk delimited name in the message. So if the index is 1, it should refer to *Name1, if it's 2 it should refer to *Name2.
The method should return just the name with the asterisk (*Name2).
I have attempted to play around with substrings, taking the first delimited * and ending when we reach a character that isn't a letter, number, underscore or hyphen, but the logic just isn't setting in.
I know this is similar to a few problems on SO but I can't find anything this specific. Any help is appreciated.
This is what's left of my very vague attempt so far. Based on this thread:
public string GetIndexedNames(string message, int index)
{
int strStart = message.IndexOf("#") + "#".Length;
int strEnd = message.LastIndexOf(" ");
String result = message.Substring(strStart, strEnd - strStart);
}
If you want to do it the old school way, then something like:
public static void Main(string[] args)
{
string message = "Hello, I am *Name1, how are you doing *Name2?";
string name1 = GetIndexedNames(message, "*", 1);
string name2 = GetIndexedNames(message, "*", 2);
Console.WriteLine(message);
Console.WriteLine(name1);
Console.WriteLine(name2);
Console.ReadLine();
}
public static string GetIndexedNames(string message, string singleCharDelimiter, int index)
{
string valid = "abcdefghijlmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789_-";
string[] parts = message.Split(singleCharDelimiter.ToArray());
if (parts.Length >= index)
{
StringBuilder sb = new StringBuilder();
for(int i = 0; i < parts[index].Length; i++)
{
string character = parts[index].Substring(i, 1);
if (valid.Contains(character))
{
sb.Append(character);
}
else
{
return sb.ToString();
}
}
return sb.ToString();
}
return "";
}
You can try using regular expressions to match the names. Assuming that name is a sequence of word characters (letters or digits):
using System.Linq;
using System.Text.RegularExpressions;
...
// Either name with asterisk *Name or null
// index is 1-based
private static ObtainName(string source, int index) => Regex
.Matches(source, #"\*\w+")
.Cast<Match>()
.Select(match => match.Value)
.Distinct() // in case the same name repeats several times
.ElementAtOrDefault(index - 1);
Demo:
string name = ObtainName(
"Hello, I am *Name1, how are you doing *Name2?", 2);
Console.Write(name);
Outcome:
*Name2
Perhaps not the most elegant solution, but if you want to use IndexOf, use a loop:
public static string GetIndexedNames(string message, int index, char marker='*')
{
int lastFound = 0;
for (int i = 0; i < index; i++) {
lastFound = message.IndexOf(marker, lastFound+1);
if (lastFound == -1) return null;
}
var space = message.IndexOf(' ', lastFound);
return space == -1 ? message.Substring(lastFound) : message.Substring(lastFound, space - lastFound);
}

How can I add a character to a string based on another string in C#?

I need to create a function that receives two strings, representing the word to be completed and the reference word, as well as a character corresponding to the proposed letter, and returns a string corresponding to the word to be completed in which all occurrences of the proposed letter have been added, relative to the reference word.
Example: CompleterMot (".. IM ..", "ANIMAL", 'A') should return "A.IMA.".
I don't understand how I can add all occurrences of the letter in the word that will be completed.
static string CompleterMot(string motincomplet, string motoriginal, char lettrepropos)
{
string output = " ";
for (int i = 0; i < motoriginal.Length; i++)
{
if((motoriginal[i] == lettrepropos))
{
output = motincomplet;
if(output[i] != lettrepropos)
output += (char)(lettrepropos);
}
}
return output;
}
In final I had ..IM..A and I don't know how to fix my code.
In your loop, you are doing this : output = motincomplet; this override the previous result. Then you append the expected letter to the output that gives "..IM.." + 'A' as result.
You can use a StringBuilder for string manipulation, that's pretty nice and allow you to directly change a character at a given index :
static string CompleterMot(string motincomplet, string motoriginal, char lettrepropos)
{
var sb = new System.Text.StringBuilder(motincomplet);
for (int i = 0; i < motoriginal.Length; i++)
{
if (motoriginal[i] == lettrepropos)
{
sb[i] = lettrepropos;
}
}
return sb.ToString();
}

Getting text from string between two characters

I am trying to pull apart a string and get the text between two "#"'s. I have found ways to do it in Java and php and I am assuming they are similar in C#, i just jeep failing and pretty sure its PEBKAC. So I though I would ask.
Example- I want to programatically pull out "filenameid" and "Name" from this string:
'#filenameid#30day#Name#.xls'
Try splitting:
String source = "#filenameid#30day#Name#.xls";
String[] chunks = source
.Split(new Char[] { '#' }, StringSplitOptions.RemoveEmptyEntries);
then take appropriate chunks:
String id = chunks[0];
String period = chunks[1];
String name = chunks[2];
Use capturing groups.
#"#([^#]*)#"
Get the string you want from group index 1. Note that lookarounds won't work here.
If your string has same format always you can do the following:
string a = "#filenameid#30day#Name#.xls";
string[]split=a.Split('#');
string fileID = split[1];
string name = split[3];
You can achieve it using Regex.
void Main()
{
const string Expression = #"#([^#]*)*#";
const string TestSample = #"'#filenameid#30day#Name#.xls'";
Regex regex = new Regex(Expression);
regex.Matches(TestSample)
.Cast<Match>()
.Select(match => match.Captures[0].Value.Replace("#", ""))
.ToList()
.ForEach(Console.WriteLine);
}
Here is a low-level solution for the problem:
static void Main(string[] args) {
string text = "#filenameid#30day#Name#.xls";
int frameStart = 0;
int match = 0;
// loop on characters
for(int i = 0; i < text.Length; i++) {
char c = text[i];
switch(c) {
case '#':
// evaluate frame (text between meshes)
switch(match) {
// match at index 1
case 1:
Console.Write("filenameid=");
Console.WriteLine(text.Substring(frameStart, i - frameStart));
break;
// match at index 3
case 3:
Console.Write("name=");
Console.WriteLine(text.Substring(frameStart, i - frameStart));
break;
}
// move to next frame
frameStart = i + 1;
match++;
break;
}
}
// count of matches is match + 1
Console.ReadKey();
}

Convert Dash-Separated String to camelCase via C#

I have a large XML file that contain tag names that implement the dash-separated naming convention. How can I use C# to convert the tag names to the camel case naming convention?
The rules are:
1. Convert all characters to lower case
2. Capitalize the first character after each dash
3. Remove all dashes
Example
Before Conversion
<foo-bar>
<a-b-c></a-b-c>
</foo-bar>
After Conversion
<fooBar>
<aBC></aBC>
</fooBar>
Here's a code example that works, but it's slow to process - I'm thinking that there is a better way to accomplish my goal.
string ConvertDashToCamelCase(string input)
{
input = input.ToLower();
char[] ca = input.ToCharArray();
StringBuilder sb = new StringBuilder();
for(int i = 0; i < ca.Length; i++)
{
if(ca[i] == '-')
{
string t = ca[i + 1].ToString().toUpper();
sb.Append(t);
i++;
}
else
{
sb.Append(ca[i].ToString());
}
}
return sb.ToString();
}
The reason your original code was slow is because you're calling ToString all over the place unnecessarily. There's no need for that. There's also no need for the intermediate array of char. The following should be much faster, and faster than the version that uses String.Split, too.
string ConvertDashToCamelCase(string input)
{
StringBuilder sb = new StringBuilder();
bool caseFlag = false;
for (int i = 0; i < input.Length; ++i)
{
char c = input[i];
if (c == '-')
{
caseFlag = true;
}
else if (caseFlag)
{
sb.Append(char.ToUpper(c));
caseFlag = false;
}
else
{
sb.Append(char.ToLower(c));
}
}
return sb.ToString();
}
I'm not going to claim that the above is the fastest possible. In fact, there are several obvious optimizations that could save some time. But the above is clean and clear: easy to understand.
The key is the caseFlag, which you use to indicate that the next character copied should be set to upper case. Also note that I don't automatically convert the entire string to lower case. There's no reason to, since you'll be looking at every character anyway and can do the appropriate conversion at that time.
The idea here is that the code doesn't do any more work than it absolutely has to.
For completeness, here's also a regular expression one-liner (inspred by this JavaScript answer):
string ConvertDashToCamelCase(string input) =>
Regex.Replace(input, "-.", m => m.Value.ToUpper().Substring(1));
It replaces all occurrences of -x with x converted to upper case.
Special cases:
If you want lower-case all other characters, replace input with input.ToLower() inside the expression:
string ConvertDashToCamelCase(string input) =>
Regex.Replace(input.ToLower(), "-.", m => m.Value.ToUpper().Substring(1));
If you want to support multiple dashes between words (dash--case) and have all of the dashes removed (dashCase), replace - with -+ in the regular expression (to greedily match all sequences of dashes) and keep only the final character:
string ConvertDashToCamelCase(string input) =>
Regex.Replace(input, "-+.", m => m.Value.ToUpper().Substring(m.Value.Length - 1));
If you want to support multiple dashes between words (dash--case) and remove only the final one (dash-Case), change the regular expression to match only a dash followed by a non-dash (rather than a dash followed by any character):
string ConvertDashToCamelCase(string input) =>
Regex.Replace(input, "-[^-]", m => m.Value.ToUpper().Substring(1));
string ConvertDashToCamelCase(string input)
{
string[] words = input.Split('-');
words = words.Select(element => wordToCamelCase(element));
return string.Join("", words);
}
string wordToCamelCase(string input)
{
return input.First().ToString().ToUpper() + input.Substring(1).ToLower();
}
Here is an updated version of #Jim Mischel's answer that will ignore the content - i.e. it will only camelCase tag names.
string ConvertDashToCamelCase(string input)
{
StringBuilder sb = new StringBuilder();
bool caseFlag = false;
bool tagFlag = false;
for(int i = 0; i < input.Length; i++)
{
char c = input[i];
if(tagFlag)
{
if (c == '-')
{
caseFlag = true;
}
else if (caseFlag)
{
sb.Append(char.ToUpper(c));
caseFlag = false;
}
else
{
sb.Append(char.ToLower(c));
}
}
else
{
sb.Append(c);
}
// Reset tag flag if necessary
if(c == '>' || c == '<')
{
tagFlag = (c == '<');
}
}
return sb.ToString();
}
using System;
using System.Text;
public class MyString
{
public static string ToCamelCase(string str)
{
char[] s = str.ToCharArray();
StringBuilder sb = new StringBuilder();
for(int i = 0; i < s.Length; i++)
{
if (s[i] == '-' || s[i] == '_')
sb.Append(Char.ToUpper(s[++i]));
else
sb.Append(s[i]);
}
return sb.ToString();
}
}

Trim to first number

Is there a way to trim a string to the first numeric digit from left AND right using standard .NET tools? Or I need to write my own function (not difficult, but I'd rather use standard methods). I need the following outputs for the provided inputs:
Input Output
-----------------------
abc123def 123
;'-2s;35(r 2s;35
abc12de3f4g 12de3f4
You'll need to use regular expressions
string TrimToDigits(string text)
{
var pattern = #"\d.*\d";
var regex = new Regex(pattern);
Match m = regex.Match(text); // m is the first match
if (m.Success)
{
return m.Value;
}
return String.Empty;
}
If you want to call this like you normally would the String.Trim() method, you can create it as an extension method.
static class StringExtensions
{
static string TrimToDigits(this string text)
{
// ...
}
}
And then you can call it like this:
var trimmedString = otherString.TrimToDigits();
No, there is no built in way. You will have to write your own method to do this.
No, I don't think there is. Method though:
for (int i = 0; i < str.Length; i++)
{
if (char.IsDigit(str[i]))
{
break;
}
str = string.Substring(1);
}
for (int i = str.Length - 1; i > 0; i--)
{
if (char.IsDigit(str[i]))
{
break;
}
str = string.Substring(0, str.Length - 1);
}
I think this'll work.

Categories

Resources