Delphi 通過MSHTML實現一個HTML解析類

最近經常會模擬網頁提交返回網頁源碼，然後獲得網頁中相應的元素，於是需要常常解析Html中相應的各種元素，網絡是個好東西，搜索一番，就找到了好幾個Delphi版本的HtmlParser的類庫，試着使用了幾個，發現解析起來都不完整，或多或少的回出現一些問題！於是想到了如果界面上有一個瀏覽器，我們可以通過WebBrowser的Document接口對網頁元素進行操作，很是方便！但是模擬網頁提交，界面上是不一定要出現 WebBrowser的，肯定有辦法，不通過WebBrowser就直接解析HTML的，那便是我不要WebBrowser這個外殼，只要他裏面的 Document文檔接口對象就能實現對Html的解析了，查找了一番MSDN，然後Google一下，果然可行，構建方法如下：

//創建 IHTMLDocument2接口
CoCreateInstance(CLASS_HTMLDocument, nil, CLSCTX_INPROC_SERVER, IID_IHTMLDocument2, FHtmlDoc);

接口創建好了之後就能夠對文檔元素進行解析了，很是爽快！

結合了我自己的特有操作，我對Combobox,Table,Frame等一些網頁元素做了相應的封裝，實現了一個HTMLParser，大致代碼如下：

這裏只給出聲明，代碼請在最後下載

代碼

(* **************************************************** *)
(*                 得閒工作室                           *)
(*               網頁元素操作類庫                       *)
(*                                                      *)
(*               DxHtmlElement Unit                     *)
(*     Copyright(c) 2008-2010  不得閒                   *)
(*     email:[email protected]     QQ:75492895     *)
(* **************************************************** *)
unit DxHtmlElement;

interface
uses Windows,sysUtils,Clipbrd,MSHTML,ActiveX,OleCtrls,Graphics,TypInfo;

{ Get EleMent Type }
function IsSelectElement(eleElement: IHTMLElement): Boolean;
function IsPwdElement(eleElement: IHTMLElement): Boolean;
function IsTextElement(element: IHTMLElement): boolean;
function IsTableElement(element: IHTMLElement): Boolean;
function IsElementCollection(element: IHTMLElement): Boolean;
function IsChkElement(element: IHTMLElement): boolean;
function IsRadioBtnElement(element: IHTMLElement): boolean;
function IsMemoElement(element: IHTMLElement): boolean;
function IsFormElement(element: IHTMLElement): boolean;
function IsIMGElement(element: IHTMLElement): boolean;
function IsInIMGElement(element: IHTMLElement): boolean;
function IsLabelElement(element: IHTMLElement): boolean;
function IsLinkElement(element: IHTMLElement): boolean;
function IsListElement(element: IHTMLElement): boolean;
function IsControlElement(element: IHTMLElement): boolean;
function IsObjectElement(element: IHTMLElement): boolean;
function IsFrameElement(element: IHTMLElement): boolean;
function IsInPutBtnElement(element: IHTMLElement): boolean;
function IsInHiddenElement(element: IHTMLElement): boolean;
function IsSubmitElement(element: IHTMLElement): boolean;
{ Get ImgElement Data }
function GetPicIndex(doc: IHTMLDocument2; Src: string ; Alt: string ): Integer;
function GetPicElement(doc: IHTMLDocument2;imgName: string ;src: string ;Alt: string ): IHTMLImgElement;
function GetRegCodePic(doc: IHTMLDocument2;ImgName: string ; Src: string ; Alt: string ): TPicture; overload ;
function GetRegCodePic(doc: IHTMLDocument2;Index: integer): TPicture; overload ;
function GetRegCodePic(doc: IHTMLDocument2;element: IHTMLIMGElement): TPicture; overload ;

type
  TObjectFromLResult = function (LRESULT: lResult; const IID: TIID; WPARAM: wParam; out pObject): HRESULT; stdcall ;
  TEleMentType = (ELE_UNKNOW,ELE_TEXT,ELE_PWD,ELE_SELECT,ELE_CHECKBOX,ELE_RADIOBTN,ELE_MEMO,ELE_FORM,ELE_IMAGE,
  ELE_LABEL,ELE_LINK,ELE_LIST,ELE_CONTROL,ELE_OBJECT,ELE_FRAME,ELE_INPUTBTN,ELE_INIMAGE,ELE_INHIDDEN);

function GetElementType(element: IHTMLELEMENT): TEleMentType;
function GetElementTypeName(element: IHTMLELEMENT): string ;
function GetHtmlTableCell(aTable: IHTMLTable;aRow,aCol: Integer): IHTMLElement;
function GetHtmlTable(aDoc: IHTMLDocument2; aIndex: Integer): IHTMLTable;
function GetWebBrowserHtmlTableCellText(Doc: IHTMLDocument2;
          const TableIndex, RowIndex, ColIndex: Integer; var ResValue: string ):   Boolean;
function GetHtmlTableRowHtml(aTable: IHTMLTable; aRow: Integer): IHTMLElement;

function GetWebBrowserHtmlTableCellHtml(Doc: IHTMLDocument2;
          const TableIndex,RowIndex,ColIndex: Integer; var ResValue: string ):   Boolean;
function GeHtmlTableHtml(aTable: IHTMLTable; aRow: Integer): IHTMLElement;
function GetWebBrowserHtmlTableHtml(Doc: IHTMLDocument2;
          const TableIndex,RowIndex: Integer; var ResValue: string ):   Boolean;

type
  TDxWebFrameCollection = class ;
  TDxWebElementCollection = class ;

  TLoadState = (Doc_Loading,Doc_Completed,Doc_Invalidate);

  TDxWebFrame = class
   private
    FFrame: IHTMLWINDOW2;
    FElementCollections: TDxWebElementCollection;
    FWebFrameCollections: TDxWebFrameCollection;
     function GetSrc: string ;
     function GetElementCount: integer;
     function GetWebFrameCollections: TDxWebFrameCollection;
     function GetElementCollections: TDxWebElementCollection;
     function GetDocument: IHTMLDOCUMENT2;
     function GetReadState: TLoadState;
     function GetIsLoaded: boolean;
     procedure SetFrame( const Value: IHTMLWINDOW2);
     function GetName: string ;
   public
    Constructor Create(IFrame: IHTMLWINDOW2);
    Destructor Destroy; override ;
     property Frame: IHTMLWINDOW2 read FFrame write SetFrame;
     property Src: string read GetSrc;
     property Document: IHTMLDOCUMENT2 read GetDocument;
     property Name: string read GetName;
     property Frames: TDxWebFrameCollection read GetWebFrameCollections;
     property ElementCount: integer read GetElementCount;
     property ElementCollections: TDxWebElementCollection read GetElementCollections;
     property ReadyState: TLoadState read GetReadState;
     property IsLoaded: boolean read GetIsLoaded;
   end ;

  TDxWebFrameCollection = Class
   private
    FFrameCollection: IHTMLFramesCollection2;
    Frame: TDxWebFrame;
     function GetCount: integer;
     function GetFrameInterfaceByIndex(index: integer): IHTMLWINDOW2;
     function GetFrameInterfaceByName(Name: string ): IHTMLWINDOW2;
     function GetFrameByIndex(index: integer): TDxWebFrame;
     function GetFrameByName(Name: string ): TDxWebFrame;
     procedure SetFrameCollection( const Value: IHTMLFramesCollection2);
   public
    Constructor Create(ACollection: IHTMLFramesCollection2);
    Destructor Destroy; override ;
     property FrameCollection: IHTMLFramesCollection2 read FFrameCollection write SetFrameCollection;
     property Count: integer read GetCount;
     property FrameInterfaceByIndex[index: integer]: IHTMLWINDOW2 read GetFrameInterfaceByIndex;
     property FrameInterfaceByName[Name: string ]: IHTMLWINDOW2 read GetFrameInterfaceByName;

     property FrameByIndex[index: integer]: TDxWebFrame read GetFrameByIndex;
     property FrameByName[Name: string ]: TDxWebFrame read GetFrameByName;
   end ;

  TDxWebElementCollection = class
   private
    FCollection: IHTMLElementCollection;
    FChildCollection:  TDxWebElementCollection;
     function GetCollection(index: String): TDxWebElementCollection;
     function GetCount: integer;
     function GetElement(itemName: string ; index: integer): IHTMLElement;
     function GetElementByName(itemName: string ): IHTMLELEMENT;
     function GetElementByIndex(index: integer): IHTMLELEMENT;
     procedure SetCollection( const Value: IHTMLElementCollection);
   public
    Constructor Create(ACollection: IHTMLElementCollection);
    Destructor Destroy; override ;
     property Collection: IHTMLElementCollection read FCollection write SetCollection;
     property ChildElementCollection[index: String]: TDxWebElementCollection read GetCollection;
     property ElementCount: integer read GetCount;
     property Element[itemName: string ;index: integer]: IHTMLElement read GetElement;
     property ElementByName[itemName: string ]: IHTMLELEMENT read GetElementByName;
     property ElementByIndex[index: integer]: IHTMLELEMENT read GetElementByIndex;
   end ;

  TLinkCollection = class (TDxWebElementCollection)

   end ;
  TDxWebTable = class ;

  TDxTableCollection = class
   private
    FTableCollection: IHTMLElementCollection;
    FDocument: IHTMLDOCUMENT2;
    FWebTable: TDxWebTable;
     function GetTableInterfaceByName(AName: string ): IHTMLTABLE;
     procedure SetDocument(Value: IHTMLDOCUMENT2);
     function GetTableInterfaceByIndex(index: integer): IHTMLTABLE;
     function GetCount: integer;
     function GetTableByIndex(index: integer): TDxWebTable;
     function GetTableByName(AName: string ): TDxWebTable;
   public
    Constructor Create(Doc: IHTMLDOCUMENT2);
     destructor Destroy; override ;
     property TableInterfaceByName[AName: string ]: IHTMLTABLE read GetTableInterfaceByName;
     property TableInterfaceByIndex[index: integer]: IHTMLTABLE read GetTableInterfaceByIndex;

     property TableByName[AName: string ]: TDxWebTable read GetTableByName;
     property TableByIndex[index: integer]: TDxWebTable read GetTableByIndex;

     property Document: IHTMLDOCUMENT2 read FDocument write SetDocument;
     property Count: integer read GetCount;
   end ;

  TDxWebTable = class
   private
    FTableInterface: IHTMLTABLE;
     function GetRowCount: integer;
     procedure SetTableInterface( const Value: IHTMLTABLE);
     function GetCell(ACol, ARow: integer): string ;
     function GetRowColCount(RowIndex: integer): integer;
     function GetInnerHtml: string ;
     function GetInnerText: string ;
     function GetCellElement(ACol, ARow: Integer): IHTMLTableCell;
   public
    Constructor Create(ATable: IHTMLTABLE);
     property TableInterface: IHTMLTABLE read FTableInterface write SetTableInterface;
     property RowCount: integer read GetRowCount;
     property Cell[ACol: integer;ARow: integer]: string read GetCell;
     property CellElement[ACol: Integer;ARow: Integer]: IHTMLTableCell read GetCellElement;
     property RowColCount[RowIndex: integer]: integer read GetRowColCount;
     property InnerHtml: string read GetInnerHtml;
     property InnerText: string read GetInnerText;
   end ;

  TDxWebCombobox = class
   private
    FHtmlSelect: IHTMLSelectElement;
     function GetCount: Integer;
     procedure SetItemIndex( const Value: Integer);
     function GetItemIndex: Integer;
     function GetName: string ;
     procedure SetName( const Value: string );
     function GetValue: string ;
     procedure SetValue( const Value: string );
     procedure SetCombInterface( const Value: IHTMLSelectElement);
     function GetItemByName(EleName: string ): string ;
     function GetItemByIndex(index: integer): string ;
     function GetItemAttribute(index: Integer; AttribName: string ): OleVariant;
   public
     constructor Create(AWebCombo: IHTMLSelectElement);
     procedure Add(Ele: IHTMLElement);
     procedure Insert(Ele: IHTMLElement;Index: Integer);
     procedure Remove(index: Integer);

     property CombInterface: IHTMLSelectElement read FHtmlSelect write SetCombInterface;
     property Count: Integer read GetCount;
     property ItemIndex: Integer read GetItemIndex write SetItemIndex;
     property ItemByIndex[index: integer]: string read GetItemByIndex;
     property ItemByName[EleName: string ]: string read GetItemByName;
     property ItemAttribute[index: Integer;AttribName: string ]: OleVariant read GetItemAttribute;
     property Name: string read GetName write SetName;
     property value: string read GetValue write SetValue;
   end ;

implementation
end .

HTMLParser解析類的代碼實現單元

代碼

(* **************************************************** *)
(*                 得閒工作室                           *)
(*               HTML解析單元庫                         *)
(*                                                      *)
(*               DxHtmlParser Unit                      *)
(*     Copyright(c) 2008-2010  不得閒                   *)
(*     email:[email protected]     QQ:75492895     *)
(* **************************************************** *)
unit DxHtmlParser;

interface
uses Windows,MSHTML,ActiveX,DxHtmlElement,Forms;

type
  TDxHtmlParser = class
   private
    FHtmlDoc: IHTMLDocument2;
    FHTML: string ;
    FWebTables: TDxTableCollection;
    FWebElements: TDxWebElementCollection;
    FWebComb: TDxWebCombobox;
     procedure SetHTML( const Value: string );
     function GetWebCombobox(AName: string ): TDxWebCombobox;
   public
     constructor Create;
     destructor Destroy; override ;
     property HTML: string read FHTML write SetHTML;
     property WebTables: TDxTableCollection read FWebTables;
     property WebElements: TDxWebElementCollection read FWebElements;
     property WebCombobox[Name: string ]: TDxWebCombobox read GetWebCombobox;
   end ;
implementation

{ TDxHtmlParser }

constructor TDxHtmlParser.Create;
begin
  CoInitialize( nil );
   // 創建IHTMLDocument2接口
  CoCreateInstance(CLASS_HTMLDocument, nil , CLSCTX_INPROC_SERVER, IID_IHTMLDocument2, FHtmlDoc);
  Assert(FHtmlDoc <> nil , ' 構建HTMLDocument接口失敗 ' );
  FHtmlDoc.Set_designMode( ' On ' ); // 設置爲設計模式，不執行腳本
   while not (FHtmlDoc.readyState = ' complete ' ) do
   begin
    sleep( 1 );
    Application.ProcessMessages;
   end ;
  FWebTables : = TDxTableCollection.Create(FHtmlDoc);
  FWebElements : = TDxWebElementCollection.Create( nil );
  FWebComb : = TDxWebCombobox.Create( nil );
end ;

destructor TDxHtmlParser.Destroy;
begin
  FWebTables.Free;
  FWebElements.Free;
  FWebComb.Free;
  CoUninitialize;
   inherited ;
end ;

function TDxHtmlParser.GetWebCombobox(AName: string ): TDxWebCombobox;
begin
    if FWebElements.Collection <> nil then
    begin
     FWebComb.CombInterface : = FWebElements.ElementByName[AName] as IHTMLSelectElement;
     Result : = FWebComb;
    end
    else Result : = nil ;
end ;

procedure TDxHtmlParser.SetHTML( const Value: string );
begin
   if FHTML <> Value then
   begin
    FHTML : = Value;
    FHtmlDoc.body.innerHTML : = FHTML;
    FWebElements.Collection : = FHtmlDoc.all;
   end ;
end ;

end .

全部代碼下載

不得閒

發佈了81 篇原創文章 · 獲贊 148 · 訪問量 79萬+

他的留言板關注

Delphi 通過MSHTML實現一個HTML解析類

【簡寫Mybatis-02】註冊機的實現以及SqlSession處理

手繪二維碼

.NET藉助虛擬網卡實現一個簡單異地組網工具

判斷是否是64位操作系統

BIOS 和DOS中斷大全

CPU信息的獲得，比較全面的

關於Sqlite工具 DxSqliteManagement

使用MSHTML解析HTML代碼

Mac下配置sublime實現LaTeX

https://yachay.unat.edu.pe/blog/index.php?comment_area=format_blog&comment_component=blog&comment_co

linux以太網驅動總結