[SOLVED] HTML read tag

[Pe3s]

Hello forum users, is it possible to read the content of tags from an html file and display them in ListView?
Regards

[waltfair]

Since html is a text file , you could of course parse the text and extract the tags, but
I assume you're asking for something more elegant, but I can't help you there.
Good luck,
Walt Fair PHD, PE

[dsiders]

Hello forum users, is it possible to read the content of tags from an html file and display them in ListView?

Regards

FPC has several options for parsing HTML:

* fasthtmlparser (simplistic but fast for tag or text values)
* THTMLReader (Fires events when document, element, or attribute components are found)
* THTMLDocument (A DOM document where you do all the heavy lifting)

Demo code or example usage in fcl or the Lazarus IDE available for each.

Adding the results from one of these to a list view depends on the parser/analyzer selected. But that should be the trivial part of the exercise.

[dsiders]

Hello forum users, is it possible to read the content of tags from an html file and display them in ListView?

Regards

FPC has several options for parsing HTML:

* fasthtmlparser (simplistic but fast for tag or text values)
* THTMLReader (Fires events when document, element, or attribute components are found)
* THTMLDocument (A DOM document where you do all the heavy lifting)

Demo code or example usage in fcl or the Lazarus IDE available for each.

Adding the results from one of these to a list view depends on the parser/analyzer selected. But that should be the trivial part of the exercise.

I was curious how each of the options performed, so I put together the attached demo. Ironically, the option with the word "Fast" in its name... wasn't really all that fast. Go figure.

[Pe3s]

Thank you @waltfair, dsiders

[madref]

You can also take a look at this thread:
https://forum.lazarus.freepascal.org/index.php/topic,61206.0.html

[wp]

I was curious how each of the options performed, so I put together the attached demo. Ironically, the option with the word "Fast" in its name... wasn't really all that fast. Go figure.

Thank you, dsiders, for this demo. This is the first time that I see the sax reader in action explicitely.

Nevertheless, I doubt whether this speed test is of any value to see the speed differences between the three parsers. This is because you add the parsed results to a listview, and this introduces a lot of overhead. Moreover, the tests are not written inconsistently, in one test clearing the listview is included in the measurement code, as well as extraction of the html from the memo.Text (but this is negligible in comparison with the first issue).

I rewrote the demo so that the results are first written to a stringlist. Now the time to parse your (small) demo file is displayed as 0 ms for all three cases. Found an html file in the docs/html/lazutils folder which is greater than 1 MB (uni936c.html), and for this the times are measured as follows:

• fasthtmlparser: 31 ms
• SAX parser: 187 ms
• DOM parser: 656 ms.

(The time to display the output in the listview, however, is "endless", about 1 minute so so. Should do this in virtual mode, but did not want to spend so much time on it).

[BeniBela]

Or the parser from my internet tools.

Like a dom, but  130 ms:

Code: Pascal  [Select][+][-]

1. uses simplehtmltreeparser;
2. procedure TForm1.Button1Click(Sender: TObject);
3.   procedure ProcessNode(ANode: TTreeNode);
4.   var
5.     Value: String;
6.   begin
7.     Value := Trim(String(ANode.getStringValue()));
8.     if Value = '' then Exit;
9.  
10.     {$IFDEF STRINGLIST}
11.     FOutputList.Add(ANode.value+ '=' + Value);
12.     {$ELSE}
13.     with ListViewDoc.Items.Add do
14.     begin
15.       Caption := ANode.ClassName;
16.       SubItems.Add(String(ANode.NodeName));
17.       SubItems.Add(Value);
18.     end;
19.     {$ENDIF}
20.   end;
21.  
22.   procedure ProcessElement(AElement: TTreeNode);
23.   var
24.     a: TTreeAttribute;
25.     n: TTreeNode;
26.   begin
27.     {$IFDEF STRINGLIST}
28.     FOutputList.Add(AElement.ClassName + '=' + AElement.value);
29.     {$ELSE}
30.     with ListViewDoc.Items.Add do
31.     begin
32.       Caption := AElement.ClassName;
33.       SubItems.Add(String(AElement.TagName));
34.     end;
35.     {$ENDIF}
36.  
37.     for a in AElement.getEnumeratorAttributes do
38.       ProcessNode(a)
39.     ;
40.  
41.     for n in AElement.getEnumeratorChildren do
42.       if n.typ = tetOpen then ProcessElement(n)
43.       else ProcessNode(n);
44.   end;
45.  
46. var
47.   tp: TTreeParser;
48.   AStart, AEnd: QWord;
49.   t: String;
50.   tnode: TTreeDocument;
51. begin
52.   t := memo1.Lines.Text;
53.   {$IFDEF STRINGLIST}
54.   FOutputList.BeginUpdate;
55.   {$ELSE}
56.   ListViewReader.BeginUpdate;
57.   {$ENDIF}
58.   try
59.     {$IFDEF STRINGLIST}
60.     FOutputList.Clear;
61.     {$ELSE}
62.     ListViewReader.Clear;
63.     {$ENDIF}
64.     AStart := GetTickCount64;
65.  
66.  
67.     tp := TTreeParser.Create;
68.     tp.parsingModel := pmHTML;
69.     tnode := tp.parseTree(t, '', 'html');
70.     ProcessElement(tnode);
71.     //caption := query('count($_1//*)', [xqvalue(tnode)]).tostring;
72.     //t := tnode.outerHTML();
73.     tp.free;
74.  
75.  
76.     AEnd := GetTickCount64;
77.     StatusBar1.SimpleText := Format('TTreeParser (%d msec)', [AEnd - AStart]);
78.   finally
79.     {$IFDEF STRINGLISt}
80.     FOutputList.EndUpdate;
81.     DisplayOutput(ListViewDoc);
82.     {$ELSE}
83.     ListViewDoc.EndUpdate;
84.     {$ENDIF}
85.   end;
86.  
87. end;
88.