how to test a string if it contains special characters

[titan nyquist]

How do you test a string to see if it contains special characters? I
want to ensure that any names typed into my form has only letters (and
maybe allow a dash and an apostrophe).

I can loop RealName.Contains("..."), but there must be a more elegant
solution.

 

[Alberto Poblacion]

How do you test a string to see if it contains special characters? I
want to ensure that any names typed into my form has only letters (and
maybe allow a dash and an apostrophe).

I can loop RealName.Contains("..."), but there must be a more elegant
solution.

You can use a Regular Expression:

RegEx re = new RegEx("^[-'a-zA-Z]*$");
if (re.IsMatch(stringToTest)) ....//string is correct

 

[titan nyquist]

Thanks!

You can use a Regular Expression:

RegEx re = new RegEx("^[-'a-zA-Z]*$");
if (re.IsMatch(stringToTest)) ....//string is correct

Ok, do I have this right...

^ anchors to front
$ anchors to back
a-z = any char from a..z
A-Z = any char from A..Z
* = 0 or more of the preceeding

what does ' or -' do?

 

[titan nyquist]

Ok, do I have this right...

^ anchors to front
$ anchors to back
a-z = any char from a..z
A-Z = any char from A..Z
* = 0 or more of the preceeding

what does ' or -' do?

Oops...

^ = negates / negative

 

[titan nyquist]

what does ' or -' do?

I don't know where my head is... -' is just including ' and - to the
allowed character list. Now I get it all. Thank you!

Titan

 

[Alberto Poblacion]

RegEx re = new RegEx("^[-'a-zA-Z]*$");
if (re.IsMatch(stringToTest)) ....//string is correct

Ok, do I have this right...

^ anchors to front
$ anchors to back
a-z = any char from a..z
A-Z = any char from A..Z
* = 0 or more of the preceeding

what does ' or -' do?

You said that you might also allow dashes and apostrophes, so that's why
I added ' and -. If you only want letters, then leave just [a-zA-Z].

Oops...
^ = negates / negative

No, when the caret is used at the beginning of the expression your first
interpretation was right: Anchor at front.

 

[titan nyquist]

Thanks for clarifying!

 

[=?ISO-8859-1?Q?G=F6ran_Andersson?=]

Oops...

^ = negates / negative

Outside a set ^ matches the start of the string, inside a set (as first
character) it negates the set.

 

[Tim Sprout]

How do you test a string to see if it contains special characters?
Iwant to ensure that any names typed into my form has only letters
(and maybe allow a dash and an apostrophe).

I can loop RealName.Contains("..."), but there must be a more
elegant solution.

Another possible approach:

Create a Custom User Control TextBox and override OnKeyPress to
handle ASCII characters.


protected override void OnKeyPress(System.Windows.Forms.KeyPressEventArgs e)
{
// KeyChar property of the event Gets or Sets the
// character corresponding to the key pressed
// and returns the ASCII character that is composed.
// Handled property of the event Gets or Sets a value
// indicating whether the System.Windows.Forms.Control
// .KeyPress event was handled
// and returns true if the event is handled; otherwise,
// false

int i = (int)e.KeyChar; //cast KeyChar property to integer

// Capital letters, small case letters, dash (or minus),
// and apostrophe keys are not handled (handled returns
// false) so are passed to the textBox and displayed.
// All other KeyPresses are handled (handled returns true)
// so are not passed to the textBox and are not displayed.

if (i >= 65 && i <= 90 || i >= 96 && i <= 122 || i == 39)
{
e.Handled = false;
return;
}
e.Handled = true;
}


-Tim Sprout

 

[Peter Duniho]

if (i >= 65 && i <= 90 || i >= 96 && i <= 122 || i== 39)
{
e.Handled = false;
return;
}

If one is going to do it that way, IMHO it is better to not convert to
integers. Even if you just did "if (e.KeyChar >= 'A' && e.KeyChar <='Z'
|| e.KeyChar >= 'a' && e.KeyChar <= 'z' || e.KeyChar == '\'' || e.KeyChar
== '-')", that would be more readable and more maintainable. Even better
is that you can do "if (Char.IsLetter(e.KeyChar) || e.KeyChar == '\'' ||
e.KeyChar == '-')", which is even more readable and maintainable.

(ignoring for the moment that your code leaves out one of the possible
characters...the fact that I have to go to an ASCII table to figure out
whether you forgot the apostrophe or the hyphen is an example of the lack
of readability of your code ).

Pete

 

[Tim Sprout]

if (i >= 65 && i <= 90 || i >= 96 && i <= 122 || i == 39)
{
e.Handled = false;
return;
}

If one is going to do it that way, IMHO it is better to not convert to
integers. Even if you just did "if (e.KeyChar >= 'A' && e.KeyChar <= 'Z'
|| e.KeyChar >= 'a' && e.KeyChar <= 'z' || e.KeyChar == '\'' || e.KeyChar
== '-')", that would be more readable and more maintainable. Even better
is that you can do "if (Char.IsLetter(e.KeyChar) || e.KeyChar == '\'' ||
e.KeyChar == '-')", which is even more readable and maintainable.

(ignoring for the moment that your code leaves out one of the possible
characters...the fact that I have to go to an ASCII table to figure out
whether you forgot the apostrophe or the hyphen is an example of the lack
of readability of your code ).

Pete



Thanks for the comments, Pete!

-Tim Sprout

 

[Mihai N.]

How do you test a string to see if it contains special characters? I

want to ensure that any names typed into my form has only letters (and
maybe allow a dash and an apostrophe).

Be carefull what you check for. "letters" is language dependent.

If you test with (i >= 65 && i <= 90 || i >= 96 && i <= 122 || i == 39)
or regex ranges [a-zA-Z] or other such English-centric ideas, you will
affect other languages using characters outside these ranges.

And that is almost every language out there, with very few exceptions

 

[Alberto Poblacion]

Be carefull what you check for. "letters" is language dependent.

If you test with (i >= 65 && i <= 90 || i >= 96 && i <= 122 || i == 39)
or regex ranges [a-zA-Z] or other such English-centric ideas, you will
affect other languages using characters outside these ranges.

And that is almost every language out there, with very few exceptions

If you follow the Regular Expression route, you can test for characters
in Unicode groups and block ranges with the syntax \p{name}, where "name" is
a named character class. For instance, to test for all Alpha characters, you
use \p{L}. Another useful one is \p{IsBasicLatin} (warning: case
sensitive), which tests for the Unicode range that includes the English
alphabet.

 

[Peter Duniho]

Be carefull what you check for. "letters" is language dependent.

If you test with (i >= 65 && i <= 90 || i >= 96 && i <= 122 ||i == 39)
or regex ranges [a-zA-Z] or other such English-centric ideas, you will
affect other languages using characters outside these ranges.

Good point...one more good reason to use Char.IsLetter() instead of
explicitly checking the character code.

I also appreciate Alberto's information...I had no idea you could test for
specific Unicode character flags using regular expressions. Very cool.

Pete