How to extract all email addresses from a string?

Small improvement by checking the validity of the emails as the duplication of the characters '@' & '.' like "aa@@aa@a.a.com" will be accepted as valid email unless you have function to check the duplication and also you can check for the special characters as well (#$%^&*()!+- ... etc).


JG

Yes, much more validation could have taken place. The template is in place though, and it is simple.

Here's a thought...parse the string into space-delimited strings using TStringlist.DelimitedText()

Then you can iterate through the parsed TStringList items, applying a much simpler pattern.

@aikimark: Kind of what my post did?

@ThievingSix

Yes.  I was commenting on the original article's RegEx link reference and other TRegEx comments in this thread.

However, there is a problem with your pattern matching code.  There is no assurance that the period comes between the position of the @-position +2 and the Length(TStringList item) -1

================
Being a bit intrigued by this question, I looked at the spec on email addresses and it is much more complicated than I thought.  Here's a good article on both the spec and validating email addresses adhering to the spec:
http://haacked.com/archive/2007/08/21/i-knew-how-to-validate-an-email-address-until-i.aspx

It might be simpler to submit email candidates that pass the first level test to a web site that looks for valid MX records associated with the candidate's domain.

aikimark,
 Many sites are no longer offering MX lookup because they can be used to validate dictionary attacks.

Like my post said, there are several ways to extract the addresses, He needs a separate regex to validate the email as being the valid FORMAT, if he wants to validate them as being valid email addresses, then that's another post altogether because he only asked how to extract them from a string.

Hi guys,

I tried EddieShipman solution, this is not working:

procedure TFormMain.Button1Click(Sender: TObject);
var Src: string;
    HTTP: TIdHTTP;
begin
    Memo1.Text := '';
    HTTP := TidHTTP.Create(Application);
    Src := HTTP.Get(Edit1.Text);
    Memo1.Text := ExtractEmails(src);
    HTTP.Free;
end;

Please help.

SHow more of your code, please. There has to be something else wrong, What didn't work and what errors or problems did you see?

What was the URL you used, too?

Hello,

The code is attached.

Thanks.

unit UnitMain;
 
interface
 
uses
  Windows, Messages, SysUtils, Variants, Classes, Graphics, Controls, Forms,
  Dialogs, IdHTTP, StdCtrls, RegExpr;
 
type
  TFormMain = class(TForm)
    Button1: TButton;
    Memo1: TMemo;
    Edit1: TEdit;
    procedure Button1Click(Sender: TObject);
  private
    { Private declarations }
  public
    { Public declarations }
  end;
 
var
  FormMain: TFormMain;
 
implementation
 
{$R *.dfm}
 
// This simple function extracts all emails from input string
// and places list of this emails into result string as CSV (comma separated values)
function ExtractEmails (const AInputString : string) : string;
 
// Note: compilation of r.e. performed during Expression assignment
// take some time, so if You will use this function many times
// it will be useless overhead.
// You can significant optimize this if You will create TRegExpr
// and precompile expression during program initialization.
 
const
  // Both these expressions work, the one commented out is the one included
  // in the demo. I placed the other one in for testing.
  EmailRE = '[\w]+@[\w]+.[\w]{2,3}';
  //        '[_a-zA-Z\d\-\.]+@[_a-zA-Z\d\-]+(\.[_a-zA-Z\d\-]+)+';
var
  r : TRegExpr;
begin
  Result := '';
 
  r := TRegExpr.Create;
  // Create the object instance.
  // Please, don't forget that - 10% of all 'bug-reports' to me caused
  // by attempts to use object without creation of it!
 
  try // ensure memory clean-up
    r.Expression := EmailRE;
    // Assign r.e. source code. It will be compiled when nessesary
    // (for example when Exec called). If there are errors in r.e.
    // run-time execption will be raised during r.e. compilation
    if r.Exec (AInputString) then
    repeat
       Result := Result + r.Match [0] + ',';
    until not r.ExecNext;
  finally
    r.Free;
  end;
end;
 
procedure TFormMain.Button1Click(Sender: TObject);
var Src: string;
    HTTP: TIdHTTP;
begin
    Memo1.Text := '';
    HTTP := TidHTTP.Create(Application);
    Src := HTTP.Get(Edit1.Text);
    Memo1.Text := ExtractEmails(src);
    HTTP.Free;
end;
 
end.

                                              

1:
2:
3:
4:
5:
6:
7:
8:
9:
10:
11:
12:
13:
14:
15:
16:
17:
18:
19:
20:
21:
22:
23:
24:
25:
26:
27:
28:
29:
30:
31:
32:
33:
34:
35:
36:
37:
38:
39:
40:
41:
42:
43:
44:
45:
46:
47:
48:
49:
50:
51:
52:
53:
54:
55:
56:
57:
58:
59:
60:
61:
62:
63:
64:
65:
66:
67:
68:
69:
70:
71:
72:
73:
74:
75:
76:
77:
78:

What is the URL you are trying to parse?

Note:  the questioner is using this code:
http://vvv.truls.org/pascal/Units.Delphi/regexpr/Demos/TRegExprClass/TRegExprClassMain.pas

The help file is here:
http://vvv.truls.org/pascal/Units.Delphi/regexpr/Help/

@robotwink
Have you read the help file?

aikimark, I'm not asking about the regex stuff, I'm asking what the URL he is passing the the http control, He says he is having problems with it.

Ah, I can't believe I missed this one. My original code(first post) works exactly like the link he provided.