Regex to separate nationality from text

Question

I have list of Nationalities of people against their entries, most of entries are properly given, but some of the entries are given as below; Proper ones are like below;

German
Iranian
Qatar

Improper are like below;

Possibly Ethiopian

• (text before nationality)

Lebanon citizenship

• (text after nationality)

DRC and Belgian nationalities

• (nationalities joined with and)

(1) Germany (b) Algeria

• (with 1 and b as separator)

(a) Russian (b) Georgia

• (with a and b separator)

a) French, b) Tunisian

• ( with a and , and b separator)

Indonesian (as at December 2003)

• (a different text after nationality)

Iranian (Iranian citizenship)

• (another text after nationality)

Sudanese by birth

• (another text after nationality)

(1) Russian (2) USSR (until 1991)

• (text after 2nd entry)

Bahrain (citizenship revoked in January 2015)

• (different text again)

United States of America. Also believed to hold Syrian nationality

• (descriptive text containing nationality)

Tunisian (dual nationality)

• (different text)

(1) German (2) Moroccan

• ( 1 and 2 with () double parentheses)

1) Saudi Arabia 2) Qatar

• (1 and 2 with ) parentheses)

a) Central African Republic b) South Sudan

• (with single parentheses)

Iranian national and US national/citizen

• (nationality inside text and joined with and)

Kuwaiti citizenship withdrawn in 2002

• (a different text)

I need to take out only bold text (Nationalities) from given text. Nationality can be of any country, these are just samples of some countries.

How would I apply regex or any type conditions which give results as expected. I have tried to check if text contains such characters then split them. for which I need to create more that 20 conditions and which also not good approach to do this.

List<string> listOfNationalities = listOfNationalities;

List<string> multiple new List<string>();
foreach (var nationality in listOfNationalities)
{
    if(nationality.Contains("(1)"))
    {
        string[] nat = nationality.Split(')'); 
        foreach (var item in nat)
        {
            multiple.Add(item);
        }
    }
}

Answer 1

If the nationalities is provided by a fixed list of available options. You can do the following:

List<string> listOfNationalities = listOfNationalities;

List<string> validNationalities = new List<string>();
validNationalities.Add("Brazilian");
validNationalities.Add("Japanese");
validNationalities.Add("...");

List<string> multiple = listOfNationalities.Where(n => validNationalities.Contains(n));

or even simpler:

string listOfNationalities = string.Join("|",listOfNationalities);

List<string> validNationalities = new List<string>();
validNationalities.Add("Brazilian");
validNationalities.Add("Japanese");
validNationalities.Add("...");

List<string> multiple = validNationalities.Where(n => listOfNationalities.Contains(n));

In this way, you will get the two nationalities given.

Answer 2

If you already have a list of valid nationalities, and if the nationalities don't include special characters, you can use something like the following to create the regex pattern at runtime:

public string NationalitiesPattern;

public string GetNationalitiesPattern()
{
    List<string> listOfNationalities = // All valid nationalities.
    string joinedNationalities = string.Join("|", listOfNationalities);
    return $@"\b(?:{joinedNationalities})\b";       // "\b(?:German|Iranian|Qatar|etc)\b"
}

And then you can use it like this:

if (string.IsNullOrEmpty(NationalitiesPattern))
    NationalitiesPattern = GetNationalitiesPattern();

MatchCollection matches = Regex.Matches(inputString, NationalitiesPattern);
foreach (Match m in matches)
    Console.WriteLine(m.Value);