{*******************************************************}
{                                                       }
{       Tocsg.PCRE                                      }
{                                                       }
{       Copyright (C) 2022 sunk                         }
{                                                       }
{*******************************************************}

unit Tocsg.PCRE;

interface

uses
  Tocsg.Obj, Classes, SysUtils,
  WinApi.Windows, System.RegularExpressions;
  
const
  PT_RX_IDNUM   = '\d{2}(0[1-9]|1[0-2])(0[1-9]|[12][0-9]|3[01])\W?[1-4]\d{6}'; // or \d{2}(0[1-9]|1[0-2])(0[1-9]|[12][0-9]|3[01])-[1-4]\d{7}
//3 : JCB나 다이너스, 4 : 비자, 5 : 마스터, 6 : 중국의 은련 카드, 9 : 세계 공통 국내 전용카드
  PT_RX_CARDNUM = '(3|4|5|6|9)\d{3}\W?\d{4}\W?\d{4}\W?\d{4}';
  PT_RX_EMAIL   = '[\w\-]+@(?:(?:[\w\-]{2,}\.)+[a-zA-Z]{2,})';
  //EMAIL = [_a-zA-Z0-9-]+([-+.][_a-zA-Z0-9]+)*[\^ &%*@]{0,2}\@[\^ &%*@]{0,2}[_a-zA-Z0-9]+([-.][_a-zA-Z0-9]+)*\.[-a-zA-Z0-9]+([-.][_a-zA-Z0-9]+)*;
//  PT_RX_URL     = '((ftp|https?)://|www)([a-zA-Z_0-9\W]*)[-\w.]+(/([\w/_.]*(\?\S+)?)?)?'; // 이건 매우좋지 않다. x 않좋은 예라 남겨놓음
//  PT_RX_URL     = '((?:https?://)?(?:www\.)?[-a-z\d]{1,9}\.[-a-z\d]{2,5}(?:\.[-a-z\d]{2,4})?)'; // 이건 . 으로 구분된건 앵간에서 다 검출한다. 오탐이 많음
  PT_RX_URL     = '((ftp|mms|https?)://([-\w\.]+)+(:\d+)?(/([\w/_\.]*(\?\S+)?)?)?)'; // 조건을 ftp, mms, http, https로 시작하는걸로 고정 13_1107 11:26:56 sunk

  PT_RX_IP      = '(((\d{1,2})|(1\d{2})|(2[0-4]\d)|(25[0-5]))\.){3}((\d{1,2})|(1\d{2})|(2[0-4]\d)|(25[0-5]))';
  PT_RX_HP      = '(\s|^)(0(1[016789]))(\)|-| )\d{3,4}(-| )\d{4}(?=([^\d-])|$)';
  PT_RX_PHONE   = '0(2|([3-6][1-5]))\W?\d{3,4}\W?\d{4}';  

type
  TTgPcre = class(TTgObject)
  public
    class function GetMatchValues(sSrcText, sMatchText: String;
                                        var sResult: String;
                                        bResultClear: Boolean = true;
                                        aRexOpt: TRegExOptions = [roIgnoreCase, roMultiLine]): Integer;
  end;

function GetCountOverlapWordsCount(sText: String): Integer;
function GetCountOverlapWords(sText: String; sDm: String = ''): String;
function RemoveOverlapWords(sText: String; sDm: String = ''): String;

implementation

uses
  Tocsg.Strings, Tocsg.Safe, Tocsg.Exception;

function GetCountOverlapWordsCount(sText: String): Integer;
var
  StrList: TStringList;
  i, nPos, nLen, nHit: Integer;
  sKwd: String;
begin
  Result := 0;
  Guard(StrList, TStringList.Create);
  if SplitString(sText, ',', StrList) = 0 then
    exit;

  for i := 0 to StrList.Count - 1 do
  begin
    sKwd := StrList[i];
    nPos := Pos('(x', sKwd);
    if nPos > 0 then
    begin
      nLen := Length(sKwd) - 1;
      SetLength(sKwd, nLen);
      Inc(Result, StrToIntDef(Copy(sKwd, nPos + 2, nLen - nPos + 2), 1));
      SetLength(sKwd, nPos - 1);
    end else
      Inc(Result);
  end;
end;

function GetCountOverlapWords(sText: String; sDm: String = ''): String;
var
  ComList,
  OverlWordList: TStringList;
  i, nLastIdx,
  nOverlCnt: Integer;
  sCheck: String;
begin
  Result := '';
  if sText = '' then
    exit;
  Guard(ComList, TStringList.Create);
  Guard(OverlWordList, TStringList.Create);
  SplitString(sText, ',', ComList);
  while ComList.Count > 0 do
  begin
    nLastIdx := ComList.Count - 1;
    sCheck := ComList[nLastIdx];
    ComList.Delete(nLastIdx);

    nOverlCnt := 0;
    for i := nLastIdx - 1 downto 0 do
    begin
      if sCheck = ComList[i] then
      begin
        Inc(nOverlCnt);
        ComList.Delete(i);
      end;
    end;

    if nOverlCnt > 0 then
      sCheck := Format('%s(x%d)', [sCheck, nOverlCnt+1]);
    OverlWordList.Add(sCheck);
  end;

  if (sDm <> '') and (sDm <> ',') then
  begin
    for i := 0 to OverlWordList.Count - 1 do
      SumString(Result, OverlWordList[i], sDm);
  end else
    Result := OverlWordList.CommaText;
end;

function RemoveOverlapWords(sText: String; sDm: String = ''): String;
var
  ComList,
  OverlWordList: TStringList;
  i, nLastIdx: Integer;
  sCheck: String;
begin
  Result := '';
  if sText = '' then
    exit;
  Guard(ComList, TStringList.Create);
  Guard(OverlWordList, TStringList.Create);
  SplitString(sText, ',', ComList);
  while ComList.Count > 0 do
  begin
    nLastIdx := ComList.Count - 1;
    sCheck := ComList[nLastIdx];
    ComList.Delete(nLastIdx);

    for i := nLastIdx - 1 downto 0 do
    begin
      if sCheck = ComList[i] then
        ComList.Delete(i);
    end;

    OverlWordList.Add(sCheck);
  end;

  if (sDm <> '') and (sDm <> ',') then
  begin
    for i := 0 to OverlWordList.Count - 1 do
      SumString(Result, OverlWordList[i], sDm);
  end else
    Result := OverlWordList.CommaText;
end;

{ TTgPcre }

class function TTgPcre.GetMatchValues(sSrcText, sMatchText: String;
                                        var sResult: String;
                                        bResultClear: Boolean = true;
                                        aRexOpt: TRegExOptions = [roIgnoreCase, roMultiLine]): Integer;
//                                        aRexOpt: TRegExOptions = [roIgnoreCase]): Integer;
var
  rx: TRegEx;
  mc: TMatchCollection;
  i: Integer;
begin
  Result := 0;
  try
    sSrcText := Trim(sSrcText);
    if sSrcText = '' then
      exit;

    if sSrcText = #13#10 then
      exit;

    if bResultClear then
      sResult := '';

    if sMatchText = '' then
      exit;

//    rx := TRegEx.Create(TRegEx.Escape(sMatchText, true), aRexOpt);
    rx := TRegEx.Create(sMatchText, aRexOpt);

    mc := rx.Matches(sSrcText);
    Result := mc.Count;
    if Result > 0 then
    begin
      for i := 0 to Result - 1 do
        if mc.Item[i].Value <> '' then
          SumString(sResult, mc.Item[i].Value, ',')
        else
          Dec(Result);
    end;
  except
    on E: Exception do
      ETgException.TraceException(E, 'Fail .. TTgPcre.GetMatchValues()');
  end;
end;

end.