超过 1024 个字符的字符串上的 PosEx

Question

我有一个简单的函数可以从 HTML 的 ​meta property="og:image" 中获取值：

Function HTML_GetOGImage(HTML:String;Var URL:String):Boolean;
Var
  I,A,B,C:Integer;
Begin
  Result:=False;
  A:=PosEx('og:image',HTML,1);
  If A<>0 Then B:=PosEx('content="',HTML,A+9)Else Exit;
  If B<>0 Then C:=PosEx('"',HTML,B+9)Else Exit;
  For I:=B+9 To C-1 Do URL:=URL+HTML[I];
  Result:=True;
End;

它工作得很好，但是在某些网站上，HTML 是一长串没有回车的字符，meta 的位置高于 1024，函数返回，什么都不做。 （即here）。

处理长字符串的最佳做法是什么？ PosEx 是否限制了字符串的长度？

Answer 1

这不可能是PosEx() 的错。可以通过强制一种您认为是罪魁祸首的情况来简单地对此进行反测试：

var
  s: String;
  i: Integer;
begin
  // Just a long string
  SetLength( s, 2000 );
  for i:= 1 to 2000 do s[i]:= Chr( i mod 10+ $30 );

  // The text to be found is clearly after 1024 characters
  s:= s+ 'og:image';
  i:= PosEx( 'og:image', s, 1 );

  // Should not be 0 but instead 2001
  Writeln( i );

如果此代码适用于您并且找到了搜索文本，那么您的错误就在其他地方。 PosEx()（和Pos()）对于更大的字符串（即一百万个字符）也应该没有问题。我认为您的理解是错误的：您希望您的变量 HTML 包含您想要的内容，但您是否检查过它的长度或内容（通过将其保存到文件中）？

Answer 2

PosEx() 完美运行，没有任何问题。你的问题是错误的

在你的 delphi 中运行它，我没有发现任何问题。

Unit1.dfm

object Form1: TForm1
  Left = 0
  Top = 0
  Caption = 'Form1'
  ClientHeight = 231
  ClientWidth = 505
  Color = clBtnFace
  Font.Charset = DEFAULT_CHARSET
  Font.Color = clWindowText
  Font.Height = -11
  Font.Name = 'Tahoma'
  Font.Style = []
  OldCreateOrder = False
  OnShow = FormShow
  PixelsPerInch = 96
  TextHeight = 13
  object WebBrowser1: TWebBrowser
    Left = 0
    Top = 0
    Width = 505
    Height = 231
    Align = alClient
    TabOrder = 0
    OnDownloadComplete = WebBrowser1DownloadComplete
    ExplicitLeft = 120
    ExplicitTop = 73
    ExplicitWidth = 300
    ExplicitHeight = 150
    ControlData = {
      4C00000031340000E01700000000000000000000000000000000000000000000
      000000004C000000000000000000000001000000E0D057007335CF11AE690800
      2B2E126208000000000000004C0000000114020000000000C000000000000046
      8000000000000000000000000000000000000000000000000000000000000000
      00000000000000000100000000000000000000000000000000000000}
  end
end

unit1.pas

unit Unit1;

interface

uses
  Winapi.Windows, Winapi.Messages, System.SysUtils, System.Variants, System.Classes, Vcl.Graphics,
      Vcl.Controls, Vcl.Forms, Vcl.Dialogs, Vcl.OleCtrls, SHDocVw;

const
  siteURL = 'https://www.leparisien.fr/societe/sante/covid-19-fin-de-la-periode-disolement-de-jean-castex-dont-lepouse-avait-ete-testee-positive-suivez-notre-direct-16-06-2021-BKACFU43MBAATAOLEN7BYHNNIY.php';

type
  TForm1 = class(TForm)
    WebBrowser1: TWebBrowser;
    procedure FormShow(Sender: TObject);
    procedure WebBrowser1DownloadComplete(Sender: TObject);
  private
    { Private declarations }
  public
    { Public declarations }
  end;

var
  Form1: TForm1;

implementation

{$R *.dfm}

uses System.StrUtils, ActiveX;

function HTML_GetOGImage(HTML:String;var URL:String) : Boolean;
bar
  I,A,B,C:Integer;
begin
  Result:=False;
  A := PosEx('og:image',HTML,1);
  If A<>0 Then B:=PosEx('content="',HTML,A+9)Else Exit;
  If B<>0 Then C:=PosEx('"',HTML,B+9)Else Exit;
  URL := Copy(HTML, B+9, C-B-9); //I prefer <- instead of -> For I:=B+9 To C-1 Do URL:=URL+HTML[I];

  ShowMessage('A = ' + IntToStr(A) +#13#10+ 'B = ' + IntToStr(B) +#13#10+ 'C = ' + IntToStr(C)); //remove this line after test
  Result := True;
end;

procedure TForm1.FormShow(Sender: TObject);
begin
  WebBrowser1.Navigate(siteURL);
end;

procedure TForm1.WebBrowser1DownloadComplete(Sender: TObject);
var
  LStream: TStringStream;
  Stream : IStream;
  LPersistStreamInit : IPersistStreamInit;
  URL : String;
begin
  if not Assigned(WebBrowser1.Document) then exit;
  LStream := TStringStream.Create('');
  try
    LPersistStreamInit := WebBrowser1.Document as IPersistStreamInit;
    Stream := TStreamAdapter.Create(LStream,soReference);
    LPersistStreamInit.Save(Stream,true);
    HTML_GetOGImage(LStream.DataString, URL);
  finally
    LStream.Free();
  end;
end;

end.