htmlvalidator-help Mailing List for HTML Validator for Mozilla and Firefox (Page 13)
Brought to you by:
mgueury
You can subscribe to this list here.
2006 |
Jan
|
Feb
|
Mar
|
Apr
(1) |
May
(65) |
Jun
(52) |
Jul
(33) |
Aug
(33) |
Sep
(2) |
Oct
(10) |
Nov
(8) |
Dec
(3) |
---|---|---|---|---|---|---|---|---|---|---|---|---|
2007 |
Jan
(48) |
Feb
(22) |
Mar
(25) |
Apr
(5) |
May
(4) |
Jun
(2) |
Jul
(4) |
Aug
(2) |
Sep
|
Oct
|
Nov
(1) |
Dec
(1) |
2008 |
Jan
(3) |
Feb
|
Mar
(4) |
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
From: [LoN]Kamikaze <LoN...@gm...> - 2006-05-20 11:56:56
|
Marc Gueury wrote: > It is still very good. But I would prefer that you make the same in the= > HTML validator source code > and send me the change. I cannot do this until I get Tidy running on my system. I hope I won't see the segfault of the last version that I couldn't track down. |
From: [LoN]Kamikaze <LoN...@gm...> - 2006-05-20 11:52:16
|
I have addressed the problem that line numbers interfere with copy and paste, even when their display is turned off, by writing a function that moves the code back into the old format. I think this is the most flexible solution so far and the performance impact is acceptable. There are 2 problems left that I cannot solve. But they're acceptable, as far as I see it: - Marked code gets unmarked, when display is switched between numbered and non-numbered display. - With line wrapping on, the code can break at different places, because of there being different amounts of space for the code, with and without line numbers. This can force the user to scroll back to the code of interest after switching. |
From: Marc G. <mg...@sk...> - 2006-05-20 10:34:47
|
Hi all, Seen it seems some of you want to make the compilation on Linux before me :-) I was not expecting this. So, I have rerelease the source code of the beta. http://users.skynet.be/mgueury/mozilla/tidy_08x_source.zip Here are the difference with last time: - this is the version 0.83 with the install code to work too in Mozilla Suite 1.7 - the OpenSP code is inside the source and at the right place - corrected the error given in the list - makefile with OpenSp in place of OpenSP - declaration of the i variable in nsOpenSP.cpp In the C part of the extension, you will find a build_all.bat file that run the build of all the C part in Windows. The same needs to be done on Linux. Then the XPI repackaged, like explained here <http://users.skynet.be/mgueury/mozilla/compile.html> I think this time all is there and it should be possible to make it work on Linux and others OS. The only things I am not sure: - it could be that the XPI will need to be installed by a user who has right on the Firefox directory - the line numbers in the source could be not in sync with the line... Marc |
From: [LoN]Kamikaze <LoN...@gm...> - 2006-05-19 18:32:04
|
Here is my temporary 2nd demo. I'll need the full Firefox output to validate it, but the algorithm really didn't need any changes. I just hat to exchange 'code' for 'line1'. This demo has buttons to show and hide line numbers, as well as buttons to switch between the pre and normal display. |
From: Spectroscope <spe...@gm...> - 2006-05-19 18:03:18
|
| > > 1) Line numbering doesn't work properly | |> Are you using the wrapping in the source code ? | |Oh, indeed, if I disable line wrapping it works. Are these settings |mutually exclusive? Ok, forget this part of my mail, I finally found the other thread :-) |
From: [LoN]Kamikaze <LoN...@gm...> - 2006-05-19 17:57:29
|
Can you send me your example code in an attachment? What you gave me is not complete. |
From: <spe...@gm...> - 2006-05-19 17:52:28
|
> > 1) Line numbering doesn't work properly > Are you using the wrapping in the source code ? Oh, indeed, if I disable line wrapping it works. Are these settings mutually exclusive? > > 2) The disabling of the checkmark "Enable Validation in the View > > Source" has no effect. Validation is always turned on and can only > > switched off with CTRL-U. Ah, now I see. I wasn't aware of the fact that you make a difference between the different ways of opening the source. My fault. > Do you think it is not intuitive ? Yes and no :-) Intuitive means you could guess how it works. Then I have to say:no, not intuitive. But on the other hand I only was used to change between keyboard shortcut, right-click-menu and double click on the icon with no special plan because I expected them to do the same. Now that I understand how the new mechanism works I like it. On a second thought why not additionally introducing a keyboard shortcut for "view source with validation", let's say CTRL-SHIFT-U? That way we could hit CTRL-U for a plain view and CTRL-SHIFT-U for a validated view. For us keyboard lovers that would be a simple way to launch the desired view. Yes, I think I'd like that. And one small other question: When validation in the source is turned off and I open the source, I can see a flickering at the part of the screen where the validation would be displayed (if it would have been turned on). It looks like the progress bar expanded to the height of the validation area that disappears when the source has fully loaded, so it displays longer when the page source to display is larger. Not a big thing, I only like to know if this is regular behaviour or if this means that HTML Tidy validates but only doesn't show the results. If so, doesn't this slow the display down? |
From: [LoN]Kamikaze <LoN...@gm...> - 2006-05-19 17:52:04
|
Marc Gueury wrote: > I like your idea a lot. Just that it is not as simple as this. > The source of the page is written in HTML in ""several"" pre tags. > Inside the pre-tags, there are a lot of span tags with colors. >=20 I experimented with <pre> tags, it's pretty easy to cope with that. > You can get an idea by doing this: > - take a page in firefox > - look the source > - select all > - copy the source in the clipboard > Open Mozilla composer or NVU > - create a blank page > - paste the HTML in the HTML preview > - look the HTML source. >=20 > It gives something like this: >=20 I'll give your example a try. >=20 > The problem with your approach is that > - it does not take into account the pre and the span. > And I do not see how to adapt this algorithm for it. Mostly when > there are several pre > tags. > - also copy paste of the source copy the line number. What is not expec= ted. >=20 Yeah, but copy and paste is rather rare in this case, isn't it? You normally work on the original source of your page (especially if it is dynamically generated). If you want to report problems on a page it's even useful to be able to copy the line numbers. It's also easy to switch the numbers off on the fly, with a little CSS. > Whatever, the idea is bright. >=20 > Marc >=20 Thanks. |
From: Marc G. <mg...@sk...> - 2006-05-19 17:23:36
|
I like your idea a lot. Just that it is not as simple as this. The source of the page is written in HTML in ""several"" pre tags. Inside the pre-tags, there are a lot of span tags with colors. You can get an idea by doing this: - take a page in firefox - look the source - select all - copy the source in the clipboard Open Mozilla composer or NVU - create a blank page - paste the HTML in the HTML preview - look the HTML source. It gives something like this: <html> <head> <meta content=3D"text/html; charset=3DISO-8859-1" http-equiv=3D"content-type"> <title></title> </head> <body> <pre id=3D"line1"><<span class=3D"start-tag">html</span>><<span class=3D"start-tag">head</span>><<span=20 class=3D"start-tag">meta</span><span class=3D"attribute-name"> http-equiv</span>=3D<span=20 class=3D"attribute-value">"content-type" </span><span class=3D"attribute-name">content</span>=3D<span=20 class=3D"attribute-value">"text/html; charset=3DUTF-8"</span>><<spa= n class=3D"start-tag">title</span>>Google</<span=20 class=3D"end-tag">title</span>><<span class=3D"start-tag">style</span>><!--<br>body,td,a,p,.h{font-family= :arial,sans-serif;}<br>.h{font-size:=20 20px;}<br>.q{color:#0000cc;}<br>--><br></<span class=3D"end-tag">style</span>><br><<span=20 class=3D"start-tag">script</span>><br><!--<br>function=20 sf(){document.f.q.focus();}<br>function rwt(el,oi,cad,ct,cd,sg){var e =3D= =20 window.encodeURIComponent ? encodeURIComponent : escape;var=20 oi_param=3D"";var cad_param=3D"";if (oi) oi_param=3D"&oi=3D"+e(oi);if= (cad)=20 cad_param=3D"&cad=3D"+e(cad);el.href=3D"/url?sa=3Dt"+oi_param+cad_par= am+"&ct=3D"+e(ct)+"&cd=3D"+e(cd)+"&url=3D"+e(el.href).replace= (/\+/g,"%2B")+"&ei=3D3v1tRMGUGbS8wgGo5YWbBg"+sg;el.onmousedown=3D"";r= eturn=20 true;}<br>// --><br></<span class=3D"end-tag">script</span>><br></<span=20 class=3D"end-tag">head</span>><<span The problem with your approach is that - it does not take into account the pre and the span. And I do not see how to adapt this algorithm for it. Mostly when=20 there are several pre tags. - also copy paste of the source copy the line number. What is not expecte= d. Whatever, the idea is bright. Marc [LoN]Kamikaze wrote: > Marc Gueury wrote: > =20 >> Hi Change, >> >> It is designed like this. The 2 features : >> - wrap long line >> - line number are incompatible. If one is set the other is disabled. >> >> There is a good reason behind it. The source code is a HTML page. >> The line numbers are placed in a css "frame" and the source code in hi= s >> own css "frame" >> >> There is no relation between the 2. Just that the number of lines is t= he >> same. >> If the wrapping is set to true. I have no way to match the line number= s >> and lines in the source code. >> >> I tried other algorithms: >> For ex, putting the lines number at the beginning of each line. But it >> did not work well. >> It was a lot too slow and was ugly with wrapping lines. >> >> So, in short, not possible to fix. And it is intended to work like thi= s >> until that a better algorithm is found. >> Hope it helps, >> >> Marc >> >> =20 > > Here is my solution. I find the performance acceptable. The demo > provides buttons to switch between nowrap and normal mode on the fly. > =20 > > -----------------------------------------------------------------------= - > > <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"> > <html> > <head> > <meta http-equiv=3D"content-type" content=3D"text/html; "> > <title>SELFHTML: Stylesheets / CSS-Eigenschaften / Ausrichtung und=20 > Absatzkontrolle</title> > <link rel=3D"stylesheet" type=3D"text/css" href=3D"../../src/selfhtml.c= ss"> > <meta name=3D"description" content=3D"Welche CSS-Eigenschaften es gibt,= um=20 > Text auszurichten und Texteigenschaften wie Zeilenh=C3=B6he und Einr=C3= =BCckung=20 > zu kontrollieren."> > <meta name=3D"keywords" content=3D"SELFHTML, CSS, Stylesheets, Cascadin= g=20 > Stylesheets, HTML, CSS-Eigenschaften, Texteinr=C3=BCckung, Ausrichtung,= =20 > Horinzontalausrichtung, Vertikalausrichtung, Zeilenh=C3=B6he, Textumbru= ch,=20 > Zeilenumbruch, text-indent, line-height, vertical-align, text-align,=20 > white-space"> > <meta name=3D"author" content=3D"Redaktion SELFHTML, selfhtml81@selfhtm= l.org"> > <meta name=3D"DC.Publisher" content=3D"Stefan M=C3=BCnz"> > <meta name=3D"DC.Date" content=3D"2005-03-20T00:57:10+01:00"> > <meta name=3D"DC.Identifier"=20 > content=3D"http://de.selfhtml.org/css/eigenschaften/ausrichtung.htm"> > <meta name=3D"DC.Language" content=3D"de"> > <meta name=3D"DC.Rights" content=3D"../../editorial/copyright.htm"> > <meta name=3D"DC.Date.created" content=3D"2001-10-27T08:00+01:00"> > <meta name=3D"SELF.Pagetype" content=3D"page"> > <link rel=3D"shortcut icon" href=3D"../../src/favicon.ico"> > <link rel=3D"author" title=3D"Impressum" href=3D"../../editorial/impres= sum.htm"> > <link rel=3D"contents" title=3D"Inhaltsverzeichnis"=20 > href=3D"../../navigation/inhalt.htm"> > <link rel=3D"index" title=3D"Stichwortverzeichnis"=20 > href=3D"../../navigation/stichwort.htm"> > <link rel=3D"search" title=3D"Suche" href=3D"../../navigation/suche/ind= ex.htm"> > <link rel=3D"help" title=3D"Hilfe" href=3D"../../editorial/index.htm"> > <link rel=3D"copyright" title=3D"Urheberrecht"=20 > href=3D"../../editorial/copyright.htm"> > <link rel=3D"top" title=3D"SELFHTML" href=3D"../../index.htm"> > <link rel=3D"up" title=3D"CSS-Eigenschaften" href=3D"index.htm"> > <link rel=3D"next" title=3D"Au=C3=9Fenrand und Abstand" href=3D"randabs= tand.htm"> > <link rel=3D"prev" title=3D"Schriftformatierung mit Schriftartendatei"=20 > href=3D"schrift_datei.htm"> > <link rel=3D"first" title=3D"Schriftformatierung" href=3D"schrift.htm"> > <link rel=3D"last" title=3D"Filter (nur Microsoft)" href=3D"filter.htm"= > > </head> > <body> > > <table cellpadding=3D"4" cellspacing=3D"0" border=3D"0" width=3D"100%"> > <tr> > <td colspan=3D"2" class=3D"nav"><a class=3D"an" name=3D"top"><img=20 > src=3D"../../src/refkap.gif" width=3D"16" height=3D"13" alt=3D"Teil von= "></a>=20 > <a href=3D"../../index.htm">SELFHTML</a>/<a=20 > href=3D"../../navigation/index.htm" target=3D"_parent"=20 > class=3D"navh">Navigationshilfen</a> <img src=3D"../../src/refkap.gif"=20 > width=3D"16" height=3D"13" alt=3D"Teil von"> <a=20 > href=3D"../index.htm">Stylesheets (CSS)</a> <img=20 > src=3D"../../src/refkap.gif" width=3D"16" height=3D"13" alt=3D"Teil von= "> <a=20 > href=3D"index.htm">CSS-Eigenschaften</a></td> > </tr> > <tr> > <td class=3D"doc" width=3D"110"><a href=3D"../../index.htm"><img=20 > src=3D"../../src/logo.gif" width=3D"106" height=3D"109" border=3D"0"=20 > alt=3D"SELFHTML"></a></td> > <td class=3D"docbot" width=3D"100%"><h1 class=3D"ph1">Ausrichtung und=20 > Absatzkontrolle</h1></td> > </tr> > <tr> > <td class=3D"doctop"> > <img src=3D"../../src/dokx.gif" width=3D"30" height=3D"20" vspace=3D"6"= =20 > alt=3D"Informationsseite"> > </td> > <td valign=3D"top" nowrap=3D"nowrap"> > <p> > <img src=3D"../../src/down.gif" width=3D"14" height=3D"10" alt=3D"nach=20 > unten"> <a href=3D"#allgemeines">Allgemeines zu Ausrichtung und=20 > Absatzkontrolle</a><br> > <img src=3D"../../src/down.gif" width=3D"14" height=3D"10" alt=3D"nach=20 > unten"> <a href=3D"#text_indent">text-indent</a> (Texteinr=C3=BCckung)<= br> > <img src=3D"../../src/down.gif" width=3D"14" height=3D"10" alt=3D"nach=20 > unten"> <a href=3D"#line_height">line-height</a> (Zeilenh=C3=B6he)<br> > <img src=3D"../../src/down.gif" width=3D"14" height=3D"10" alt=3D"nach=20 > unten"> <a href=3D"#vertical_align">vertical-align</a> (Vertikale=20 > Ausrichtung)<br> > <img src=3D"../../src/down.gif" width=3D"14" height=3D"10" alt=3D"nach=20 > unten"> <a href=3D"#text_align">text-align</a> (Horizontale Ausrichtung= )<br> > <img src=3D"../../src/down.gif" width=3D"14" height=3D"10" alt=3D"nach=20 > unten"> <a href=3D"#white_space">white-space</a> (Textumbruch) > </p> > </td> > </tr><tr><td colspan=3D"2" class=3D"doc"> <a href=3D"#bottom"><img=20 > src=3D"../../src/down.gif" width=3D"14" height=3D"10" border=3D"0" alt=3D= "nach=20 > unten"></a> </td></tr> > </table> > > > > <h2><a class=3D"an" name=3D"allgemeines">Allgemeines zu Ausrichtung und= =20 > Absatzkontrolle</a></h2> > > <p>F=C3=BCr Textbl=C3=B6cke oder Textabs=C3=A4tze gibt es absatztypisch= e=20 > Eigenschaften. Abs=C3=A4tze k=C3=B6nnen beispielsweise links- oder rech= tsb=C3=BCndig=20 > ausgerichtet sein. Bei Tabellenzellen oder anderen Elemente, die eine=20 > berechnete oder definierte H=C3=B6he haben k=C3=B6nnen und mehr Platz l= assen als=20 > der Text darin einnimmt, wird h=C3=A4ufig vertikale Ausrichtung oben,=20 > mittig oder untenb=C3=BCndig erforderlich.</p> > > <p>Zu den typischen Eigenschaften der Absatzformatierung geh=C3=B6ren=20 > ferner die Merkmale Einzug und Zeilenh=C3=B6he. Und schlie=C3=9Flich au= ch noch=20 > die Angabe, wie der Flie=C3=9Ftext innerhalb des Absatzes umgebrochen=20 > werden soll: ob gar nicht, automatisch, oder so, wie er im Editor=20 > eingegeben wurde.</p> > > > > <table cellpadding=3D"4" cellspacing=3D"0" border=3D"0" width=3D"100%"> > <tr> > <td class=3D"doc"> <a href=3D"#top"><img src=3D"../../src/up.gif" width= =3D"14"=20 > height=3D"10" border=3D"0" alt=3D"nach oben"></a><a href=3D"#bottom"><i= mg=20 > src=3D"../../src/down.gif" width=3D"14" height=3D"10" border=3D"0" alt=3D= "nach=20 > unten"></a> </td> > </tr> > </table> > > > > <h2><img src=3D"../../src/css10.gif" width=3D"30" height=3D"30" alt=3D"= CSS=20 > 1.0"><img src=3D"../../src/netsc4.gif" width=3D"30" height=3D"30"=20 > alt=3D"Netscape 4.0"><img src=3D"../../src/msie3.gif" width=3D"30"=20 > height=3D"30" alt=3D"MS IE 3.0"><img src=3D"../../src/op512.gif" width=3D= "30"=20 > height=3D"30" alt=3D"Opera 5.12"><img src=3D"../../src/ffox1.gif" width= =3D"30"=20 > height=3D"30" alt=3D"Mozilla Firefox 1"><img src=3D"../../src/konq33.gi= f"=20 > width=3D"30" height=3D"30" alt=3D"Konqueror 3.3"><img=20 > src=3D"../../src/saf12.gif" width=3D"30" height=3D"30" alt=3D"Safari 1.= 2"> <a=20 > class=3D"an" name=3D"text_indent">text-indent (Texteinr=C3=BCckung)</a>= </h2> > > <p>Sie k=C3=B6nnen f=C3=BCr einen mehrzeiligen Flie=C3=9Ftext bestimmen= , dass die=20 > erste Zeile einger=C3=BCckt wird. Dieses Gestaltungselement ist in B=C3= =BCchern=20 > und Zeitschriften sehr verbreitet.</p> > > <h3 class=3D"xmp">Beispiel:</h3> > > <p><img src=3D"../../src/dokf.gif" width=3D"15" height=3D"10"=20 > alt=3D"Beispiel-Seite"> <a=20 > href=3D"anzeige/text_indent.htm">Anzeigebeispiel: So sieht's aus</a></p= > > > <pre> > <html><head><title>text-indent</title> > </head><body> > > <p style=3D"text-indent:1em;">Dieser Textabsatz hat eine > Texteinrückung von 1em. Texteinrückungen dieser Art sind an > die Ästhetik von Büchern oder Zeitungen angelehnt, wo > häufig mit dieser Formatierung gearbeitet wird. Damit das auch > wirklich sichtbar wird, enthält dieser Textabsatz > schließlich noch diesen überflüssigen Satz, der > sich aber hoffentlich wenigstens flüssig liest.</p> > > <!--usw.--> > </body></html> > </pre> > > <h3 class=3D"xpl">Erl=C3=A4uterung:</h3> > > <p>Mit <code>text-indent:</code> k=C3=B6nnen Sie eine Einr=C3=BCckung f= =C3=BCr die=20 > erste Zeile bestimmen. Erlaubt ist eine <img src=3D"../../src/dok.gif"=20 > width=3D"15" height=3D"10" alt=3D"Seite"> <a=20 > href=3D"../formate/wertzuweisung.htm#numerische">numerische Angabe</a>.= </p> > > <h3 class=3D"inf">Beachten Sie:</h3> > > <p>Mit einem negativen Wert f=C3=BCr <code>text-indent</code> bewirken = Sie=20 > eine Textausr=C3=BCckung in der ersten Flie=C3=9Ftextzeile.</p> > > > > <table cellpadding=3D"4" cellspacing=3D"0" border=3D"0" width=3D"100%"> > <tr> > <td class=3D"doc"> <a href=3D"#top"><img src=3D"../../src/up.gif" width= =3D"14"=20 > height=3D"10" border=3D"0" alt=3D"nach oben"></a><a href=3D"#bottom"><i= mg=20 > src=3D"../../src/down.gif" width=3D"14" height=3D"10" border=3D"0" alt=3D= "nach=20 > unten"></a> </td> > </tr> > </table> > > > > <h2><img src=3D"../../src/css10.gif" width=3D"30" height=3D"30" alt=3D"= CSS=20 > 1.0"><img src=3D"../../src/netsc4.gif" width=3D"30" height=3D"30"=20 > alt=3D"Netscape 4.0"><img src=3D"../../src/msie3.gif" width=3D"30"=20 > height=3D"30" alt=3D"MS IE 3.0"><img src=3D"../../src/op512.gif" width=3D= "30"=20 > height=3D"30" alt=3D"Opera 5.12"><img src=3D"../../src/ffox1.gif" width= =3D"30"=20 > height=3D"30" alt=3D"Mozilla Firefox 1"><img src=3D"../../src/konq33.gi= f"=20 > width=3D"30" height=3D"30" alt=3D"Konqueror 3.3"><img=20 > src=3D"../../src/saf12.gif" width=3D"30" height=3D"30" alt=3D"Safari 1.= 2"> <a=20 > class=3D"an" name=3D"line_height">line-height (Zeilenh=C3=B6he)</a></h2= > > > <p>Sie k=C3=B6nnen f=C3=BCr Elemente mit viel Flie=C3=9Ftext die Zeilen= h=C3=B6he=20 > bestimmen. Diese Angabe ist vor allem in Verbindung mit einer Angabe=20 > zur <img src=3D"../../src/dok.gif" width=3D"15" height=3D"10"=20 > alt=3D"Seite"> <a href=3D"schrift.htm#font_size">Schriftgr=C3=B6=C3=9Fe= </a>=20 > interessant.</p> > > <h3 class=3D"xmp">Beispiel:</h3> > > <p><img src=3D"../../src/dokf.gif" width=3D"15" height=3D"10"=20 > alt=3D"Beispiel-Seite"> <a=20 > href=3D"anzeige/line_height.htm">Anzeigebeispiel: So sieht's aus</a></p= > > > <pre> > <html><head><title>line-height</title> > </head><body> > > <p style=3D"line-height:1.4em; font-size:1em; margin-left:20%;=20 > margin-right:20%"> > Die Zeilenhöhe ist ein altbekanntes DTP-Feature zur Formatierung v= on > Absätzen. Um damit richtig umzugehen, sind jedoch Kenntnisse in de= r > typografischen Wirkung von Text hilfreich. ... </p> > > </body></html> > </pre> > > <h3 class=3D"xpl">Erl=C3=A4uterung:</h3> > <p>Mit <code>line-height:</code> k=C3=B6nnen Sie die Zeilenh=C3=B6he be= stimmen.=20 > Erlaubt ist eine <img src=3D"../../src/dok.gif" width=3D"15" height=3D"= 10"=20 > alt=3D"Seite"> <a=20 > href=3D"../formate/wertzuweisung.htm#numerische">numerische Angabe</a>.= =20 > Auch Prozentangaben sind erlaubt. Prozentwerte beziehen sich dabei auf=20 > die Schriftgr=C3=B6=C3=9Fe des Elements, f=C3=BCr das die Zeilenh=C3=B6= he bestimmt wird.</p> > > <h3 class=3D"inf">Beachten Sie:</h3> > > <p>Bei erzwungener Zeilenh=C3=B6he ist es m=C3=B6glich, dass ein=20 > interpretierendes Programm dieser Angabe den Vorrang einr=C3=A4umt und=20 > Elemente abschneidet, die h=C3=B6her sind, beispielsweise eine im Flie=C3= =9Ftext=20 > referenzierte Grafik.</p> > > > > <table cellpadding=3D"4" cellspacing=3D"0" border=3D"0" width=3D"100%"> > <tr> > <td class=3D"doc"> <a href=3D"#top"><img src=3D"../../src/up.gif" width= =3D"14"=20 > height=3D"10" border=3D"0" alt=3D"nach oben"></a><a href=3D"#bottom"><i= mg=20 > src=3D"../../src/down.gif" width=3D"14" height=3D"10" border=3D"0" alt=3D= "nach=20 > unten"></a> </td> > </tr> > </table> > > > > <h2><img src=3D"../../src/css10.gif" width=3D"30" height=3D"30" alt=3D"= CSS=20 > 1.0"><img src=3D"../../src/netsc4.gif" width=3D"30" height=3D"30"=20 > alt=3D"Netscape 4.0"><img src=3D"../../src/msie3.gif" width=3D"30"=20 > height=3D"30" alt=3D"MS IE 3.0"><img src=3D"../../src/op512.gif" width=3D= "30"=20 > height=3D"30" alt=3D"Opera 5.12"><img src=3D"../../src/ffox1.gif" width= =3D"30"=20 > height=3D"30" alt=3D"Mozilla Firefox 1"><img src=3D"../../src/konq33.gi= f"=20 > width=3D"30" height=3D"30" alt=3D"Konqueror 3.3"><img=20 > src=3D"../../src/saf12.gif" width=3D"30" height=3D"30" alt=3D"Safari 1.= 2"> <a=20 > class=3D"an" name=3D"vertical_align">vertical-align (vertikale=20 > Ausrichtung)</a></h2> > > <p>Sie k=C3=B6nnen nebeneinanderstehende Elemente mit unterschiedlicher= =20 > H=C3=B6he, zum Beispiel Tabellenzellen in einer Tabellenzeile oder=20 > Textpassagen mit unterschiedlicher Schriftgr=C3=B6=C3=9Fe innerhalb ein= er Zeile,=20 > im Verh=C3=A4ltnis zueinander ausrichten.</p> > > <h3 class=3D"xmp">Beispiel:</h3> > > <p><img src=3D"../../src/dokf.gif" width=3D"15" height=3D"10"=20 > alt=3D"Beispiel-Seite"> <a=20 > href=3D"anzeige/vertical_align.htm">Anzeigebeispiel: So sieht's aus</a>= </p> > > <pre> > <html><head><title>vertical-align</title> > <style type=3D"text/css"> > .oben { vertical-align:top; background-color:#CCCCCC; } > .mittig { vertical-align:middle; background-color:#DDDDDD; } > .unten { vertical-align:bottom; background-color:#EEEEEE; } > </style> > </head><body> > > <table border=3D"1"><tr> > <td height=3D"200" class=3D"oben"><b>Text oben</b></td> > <td height=3D"200" class=3D"mittig"><b>Text mittig</b></td> > <td height=3D"200" class=3D"unten"><b>Text unten</b></td> > </tr></table> > > </body></html> > </pre> > > <h3 class=3D"xpl">Erl=C3=A4uterung:</h3> > > <p>Mit <code>vertical-align:</code> k=C3=B6nnen Sie die vertikale=20 > Ausrichtung bestimmen. Folgende Angaben sind m=C3=B6glich:</p> > > <p> > <code>top</code> =3D obenb=C3=BCndig ausrichten.<br> > <code>middle</code> =3D mittig ausrichten.<br> > <code>bottom</code> =3D untenb=C3=BCndig ausrichten.<br> > <code>baseline</code> =3D an der Basislinie ausrichten (oder=20 > untenb=C3=BCndig, wenn es keine Basislinie gibt).<br> > <code>sub</code> =3D tieferstellen (ohne die Schriftgr=C3=B6=C3=9Fe zu=20 > reduzieren).<br> > <code>super</code> =3D h=C3=B6herstellen (ohne die Schriftgr=C3=B6=C3=9F= e zu=20 > reduzieren).<br> > <code>text-top</code> =3D am oberen Schriftrand ausrichten.<br> > <code>text-bottom</code> =3D am unteren Schriftrand ausrichten.<br> > </p> > > <p>Bei Flie=C3=9Ftext ist auch eine prozentuale Angabe m=C3=B6glich. Di= e=20 > Ausrichtung orientiert sich dann an der elementeigenen Zeilenh=C3=B6he.= </p> > > <p>Das folgende Bild veranschaulicht das Modell der vertikalen=20 > Ausrichtung:</p> > > <p><img src=3D"ausrichtung.gif" width=3D"300" height=3D"150" alt=3D"ver= tikale=20 > Ausrichtung"></p> > > <p>Die gelben K=C3=A4sten stellen Kindelemente des gro=C3=9Fen Kastens = dar, die=20 > freistehenden schwarzen Texte stellen Zeicheninhalt des gro=C3=9Fen Kas= tens=20 > dar.</p> > > > > <table cellpadding=3D"4" cellspacing=3D"0" border=3D"0" width=3D"100%"> > <tr> > <td class=3D"doc"> <a href=3D"#top"><img src=3D"../../src/up.gif" width= =3D"14"=20 > height=3D"10" border=3D"0" alt=3D"nach oben"></a><a href=3D"#bottom"><i= mg=20 > src=3D"../../src/down.gif" width=3D"14" height=3D"10" border=3D"0" alt=3D= "nach=20 > unten"></a> </td> > </tr> > </table> > > > > <h2><img src=3D"../../src/css10.gif" width=3D"30" height=3D"30" alt=3D"= CSS=20 > 1.0"><img src=3D"../../src/netsc4.gif" width=3D"30" height=3D"30"=20 > alt=3D"Netscape 4.0"><img src=3D"../../src/msie3.gif" width=3D"30"=20 > height=3D"30" alt=3D"MS IE 3.0"><img src=3D"../../src/op512.gif" width=3D= "30"=20 > height=3D"30" alt=3D"Opera 5.12"><img src=3D"../../src/ffox1.gif" width= =3D"30"=20 > height=3D"30" alt=3D"Mozilla Firefox 1"><img src=3D"../../src/konq33.gi= f"=20 > width=3D"30" height=3D"30" alt=3D"Konqueror 3.3"><img=20 > src=3D"../../src/saf12.gif" width=3D"30" height=3D"30" alt=3D"Safari 1.= 2"> <a=20 > class=3D"an" name=3D"text_align">text-align (horizontale Ausrichtung)</= a></h2> > > <p>Sie k=C3=B6nnen Textabs=C3=A4tze und in anderen Block-Elementen enth= altenen=20 > Flie=C3=9Ftext oder Inline-Elemente ausrichten.</p> > > <h3 class=3D"xmp">Beispiel:</h3> > > <p><img src=3D"../../src/dokf.gif" width=3D"15" height=3D"10"=20 > alt=3D"Beispiel-Seite"> <a=20 > href=3D"anzeige/text_align.htm">Anzeigebeispiel: So sieht's aus</a></p> > > <pre> > <html><head><title>text-align</title> > </head><body> > > <p style=3D"text-align:left; margin-left:50px; margin-right:50px">Der=20 > Absatz, > den Sie gerade lesen, wurde mit <b>text-align</b> linksb=C3=BCndig=20 > ausgerichtet. > Ferner wurde ein Rand von 50px eingestellt, just for fun oder auch, > damit Sie sehen, wie die einzelnen Stylesheet-Angaben zusammenspielen.<= br> > Noch eine Textzeile<br> > Noch eine Textzeile</p> > > <!--usw.--> > </body></html> > </pre> > > <h3 class=3D"xpl">Erl=C3=A4uterung:</h3> > > <p>Mit <code>text-align:</code> k=C3=B6nnen Sie Textinhalte von=20 > Block-Elementen ausrichten. Folgende Angaben sind m=C3=B6glich:</p> > > <p> > <code>left</code> =3D linksb=C3=BCndig ausrichten (Voreinstellung).<br> > <code>center</code> =3D zentriert ausrichten.<br> > <code>right</code> =3D rechtsb=C3=BCndig ausrichten.<br> > <code>justify</code> =3D als Blocksatz ausrichten.</p> > > <h3 class=3D"inf">Beachten Sie:</h3> > > <p><code>text-align</code> bezieht sich nicht nur auf Textinhalte,=20 > sondern auf alle <b>inline</b> dargestellten Elemente (wie z.B.=20 > <code><img></code>). Der MS Internet-Explorer wendet diese Eigenschaft=20 > jedoch f=C3=A4lschlich auch auf Block-Elemente an.<br> > Block-Elemente selbst werden =C3=BCber <img src=3D"../../src/dok.gif"=20 > width=3D"15" height=3D"10" alt=3D"Seite"> <code><a=20 > href=3D"randabstand.htm">margin</a></code> ausgerichtet, wobei gleiche=20 > Werte f=C3=BCr <code>margin-left</code> und <code>margin-right</code> e= ine=20 > Zentrierung bewirken k=C3=B6nnen.</p> > > <p>Eine ehemals eingef=C3=BChrte CSS-Eigenschaft namens=20 > <code>alignment</code> ist mittlerweile aus dem CSS-Standard=20 > verschwunden. Verwenden Sie nur noch <code>text-align</code>.</p> > > > > <table cellpadding=3D"4" cellspacing=3D"0" border=3D"0" width=3D"100%"> > <tr> > <td class=3D"doc"> <a href=3D"#top"><img src=3D"../../src/up.gif" width= =3D"14"=20 > height=3D"10" border=3D"0" alt=3D"nach oben"></a><a href=3D"#bottom"><i= mg=20 > src=3D"../../src/down.gif" width=3D"14" height=3D"10" border=3D"0" alt=3D= "nach=20 > unten"></a> </td> > </tr> > </table> > > > > <h2><img src=3D"../../src/css20.gif" width=3D"30" height=3D"30" alt=3D"= CSS=20 > 2.0"><img src=3D"../../src/netsc6.gif" width=3D"30" height=3D"30"=20 > alt=3D"Netscape 6.0"><img src=3D"../../src/op512.gif" width=3D"30"=20 > height=3D"30" alt=3D"Opera 5.12"><img src=3D"../../src/ffox1.gif" width= =3D"30"=20 > height=3D"30" alt=3D"Mozilla Firefox 1"><img src=3D"../../src/konq33.gi= f"=20 > width=3D"30" height=3D"30" alt=3D"Konqueror 3.3"><img=20 > src=3D"../../src/saf12.gif" width=3D"30" height=3D"30" alt=3D"Safari 1.= 2"> <a=20 > class=3D"an" name=3D"white_space">white-space (Textumbruch)</a></h2> > > <p>Sie k=C3=B6nnen festlegen, wie der Zeilenumbruch innerhalb von Eleme= nten=20 > mit Text erfolgen soll.</p> > > <h3 class=3D"xmp">Beispiel:</h3> > > <p><img src=3D"../../src/dokf.gif" width=3D"15" height=3D"10"=20 > alt=3D"Beispiel-Seite"> <a=20 > href=3D"anzeige/white_space.htm">Anzeigebeispiel: So sieht's aus</a></p= > > > <pre> > <html><head><title>white-space</title> > <style type=3D"text/css"> > #editor { white-space:pre; } > #langezeile { white-space:nowrap; } > </style> > </head><body> > > <p id=3D"editor"> > Der Absatz, den Sie gerade lesen, > wurde als predefiniert definiert. > Zeilenumbr=C3=BCche sollten dabei so interpretiert > werden, wie sie im Editor eingegeben wurden.</p> > > <p id=3D"langezeile">Der Absatz, den Sie gerade lesen, darf nicht > umgebrochen werden. Es sollte kein automatischer Zeilenumbruch=20 > erfolgen. Bei > diesem längeren Text müssen Sie vermutlich quer scrollen, um=20 > alles zu lesen.</p> > > </body></html> > </pre> > > <h3 class=3D"xpl">Erl=C3=A4uterung:</h3> > > <p>Mit <code>white-space:</code> k=C3=B6nnen Sie das Verhalten f=C3=BCr= den=20 > Zeilenumbruch festlegen. Folgende Angaben sind m=C3=B6glich:</p> > > <p> > <code>normal</code> =3D automatischer Zeilenumbruch (wie bei allen=20 > HTML-Tags au=C3=9Fer <pre> und <nowrap>).<br> > <code>pre</code> =3D Zeilenumbruch wie im Editor eingegeben (so wie bei= =20 > <code><pre>...<pre></code>).<br> > <code>nowrap</code> =3D Kein automatischer Zeilenumbruch, Umbruch=20 > m=C3=B6glich durch entsprechende HTML-Tags.<br> > </p> > > <h3 class=3D"inf">Beachten Sie:</h3> > > <p>Diese Angabe geh=C3=B6rt zur CSS-Version 2.0. Der Internet Explorer=20 > interpretiert nur den Wert <code>nowrap</code>, nicht den Wert=20 > <code>pre</code>. Netscape 4.x interpretiert den Wert=20 > <code>pre</code>, nicht aber <code>nowrap</code>. Erst Netscape 6.x=20 > beherrscht alle Angaben.</p> > > > > <table cellpadding=3D"4" cellspacing=3D"0" border=3D"0" width=3D"100%"> > <tr><td colspan=3D"2" class=3D"doc"> > <a href=3D"#top"><img src=3D"../../src/up.gif" width=3D"14" height=3D"= 10"=20 > border=3D"0" alt=3D"nach oben"></a> > </td></tr> > <tr><td class=3D"doc"><a href=3D"randabstand.htm"><img=20 > src=3D"../../src/next.gif" width=3D"10" height=3D"10" border=3D"0" hspa= ce=3D"10"=20 > alt=3D"weiter"></a></td> > <td width=3D"100%"><img src=3D"../../src/dok.gif" width=3D"15" height=3D= "10"=20 > alt=3D"Seite"> <a href=3D"randabstand.htm">Au=C3=9Fenrand und Abstand</= a> > </td></tr> > <tr> > <td class=3D"doc"><a href=3D"schrift_datei.htm"><img=20 > src=3D"../../src/prev.gif" width=3D"10" height=3D"10" border=3D"0" hspa= ce=3D"10"=20 > alt=3D"zur=C3=BCck"></a></td> > <td><img src=3D"../../src/dok.gif" width=3D"15" height=3D"10"=20 > alt=3D"Seite"> <a href=3D"schrift_datei.htm">Schriftformatierung mit=20 > Schriftartendatei</a> > </td> > </tr> > <tr><td colspan=3D"2" class=3D"doc"> </td> > </tr> > <tr> > <td colspan=3D"2" class=3D"nav"><a class=3D"an" name=3D"bottom"><img=20 > src=3D"../../src/refkap.gif" width=3D"16" height=3D"13" alt=3D"Teil von= "></a>=20 > <a href=3D"../../index.htm">SELFHTML</a>/<a=20 > href=3D"../../navigation/index.htm" target=3D"_parent"=20 > class=3D"navh">Navigationshilfen</a> <img src=3D"../../src/refkap.gif"=20 > width=3D"16" height=3D"13" alt=3D"Teil von"> <a=20 > href=3D"../index.htm">Stylesheets (CSS)</a> <img=20 > src=3D"../../src/refkap.gif" width=3D"16" height=3D"13" alt=3D"Teil von= "> <a=20 > href=3D"index.htm">CSS-Eigenschaften</a></td> > </tr> > </table> > > <p>=C2=A9 2005 <img src=3D"../../src/dok.gif" width=3D"15" height=3D"10= "=20 > alt=3D"Seite"> <a href=3D"../../editorial/impressum.htm">Impressum</a><= /p> > > </body> > </html> |
From: Marc G. <mg...@sk...> - 2006-05-19 16:14:10
|
spe...@gm... wrote: > Hi, > > some remarks regarding the 0.83 beta > > 1) Line numbering doesn't work properly > When you open view-source there's no line numbering. You can enable it > by menu but this setting is forgotten with the next source-window you > open. > Interestingly, the checkmark is remembered (there is a checkmarkin the > menu) but no line numbers appear. > Hi, Are you using the wrapping in the source code ? > 2) The disabling of the checkmark "Enable Validation in the View > Source" has no effect. Validation is always turned on and can only > switched off with CTRL-U. > The way it works is this: - if you disable the validation in the source code. Pressing CTRL+U or doing View Page Source does not show the validation Double clicking the icon always show the validation. I changed this behavior due to a lot of request. The way people were seeing this is that you click on the icon when the icon is not green to see the error. So the validation should be on. If you just look the source, the setting is used. Do you think it is not intuitive ? Thanks by advance, Marc > 3) The enabling of the validation of the view-source should depend on > the icon state (validation in the browser turned on / turned off): > -- When the Icon shows enabled validation a doubleclick on the icon > (or a regular "View source") should open the view-source with > validation enabled. > -- When the Icon shows disabled validation the view-source window > should open with validation disabled. > > Thanks for you work. > > > ------------------------------------------------------- > Using Tomcat but need to do more? Need to support web services, security? > Get stuff done quickly with pre-integrated technology to make your job > easier > Download IBM WebSphere Application Server v.1.0.1 based on Apache > Geronimo > http://sel.as-us.falkag.net/sel?cmd_______________________________________________ > > Htmlvalidator-help mailing list > Htm...@li... > https://lists.sourceforge.net/lists/listinfo/htmlvalidator-help > > > |
From: <spe...@gm...> - 2006-05-19 16:02:46
|
Hi, some remarks regarding the 0.83 beta 1) Line numbering doesn't work properly When you open view-source there's no line numbering. You can enable it by menu but this setting is forgotten with the next source-window you open. Interestingly, the checkmark is remembered (there is a checkmarkin the menu) but no line numbers appear. 2) The disabling of the checkmark "Enable Validation in the View Source" has no effect. Validation is always turned on and can only switched off with CTRL-U. 3) The enabling of the validation of the view-source should depend on the icon state (validation in the browser turned on / turned off): -- When the Icon shows enabled validation a doubleclick on the icon (or a regular "View source") should open the view-source with validation enabled. -- When the Icon shows disabled validation the view-source window should open with validation disabled. Thanks for you work. |
From: [LoN]Kamikaze <LoN...@gm...> - 2006-05-19 12:58:16
|
Marc Gueury wrote: > Hi Change, > > It is designed like this. The 2 features : > - wrap long line > - line number are incompatible. If one is set the other is disabled. > > There is a good reason behind it. The source code is a HTML page. > The line numbers are placed in a css "frame" and the source code in his > own css "frame" > > There is no relation between the 2. Just that the number of lines is the > same. > If the wrapping is set to true. I have no way to match the line numbers > and lines in the source code. > > I tried other algorithms: > For ex, putting the lines number at the beginning of each line. But it > did not work well. > It was a lot too slow and was ugly with wrapping lines. > > So, in short, not possible to fix. And it is intended to work like this > until that a better algorithm is found. > Hope it helps, > > Marc > Here is my solution. I find the performance acceptable. The demo provides buttons to switch between nowrap and normal mode on the fly. |
From: Marc G. <mg...@sk...> - 2006-05-19 07:27:56
|
Hi Change, It is designed like this. The 2 features : - wrap long line - line number are incompatible. If one is set the other is disabled. There is a good reason behind it. The source code is a HTML page. The line numbers are placed in a css "frame" and the source code in his own css "frame" There is no relation between the 2. Just that the number of lines is the same. If the wrapping is set to true. I have no way to match the line numbers and lines in the source code. I tried other algorithms: For ex, putting the lines number at the beginning of each line. But it did not work well. It was a lot too slow and was ugly with wrapping lines. So, in short, not possible to fix. And it is intended to work like this until that a better algorithm is found. Hope it helps, Marc Hi, Are you aware that no line numbers are shown while using "wrap long lines"? Since the code is more readable and scan-able to me when I use "wrap long lines", I'm always using this. I couldn't find the line numbers in 0.8x until I disabled this feature. Would it be possible to have line numbers while it is enabled in a future version of this beta? Thanks, Change. |
From: Marc G. <mg...@sk...> - 2006-05-19 07:18:11
|
Hi all, I have not tried to compile that version on Linux yet. But here are the missing piece in short. (I will try to do it too when I find the time) The OpenSP directory 1.5.2 should be downloaded from the CVS of OpenJade and place in mozilla/extensions/tidy/opensp. No other version since the 1.5.1 beeing the last prod version does not compile on Windows at all. I think that it is not even enough and that some change need to be in it done to compile it too .... I hope it is better for Linux than Windows. Then compile OpenSP... The result is osp152.dll (or .so) Then compile the extension nstidy.dll (.so) (see the http://users.skynet.be/mgueury/mozilla/compile.html ) Then repackage the extension with the 2 compiled library. It should be it. Marc Ps: thanks for the 2 small mistakes already. I will try to reupload soon a source version with the 2 fout corrected and the opensp code at the right place. Just that the source will be huge then. [LoN]Kamikaze wrote: > The 'spvalid' target fails, because '../opensp/include/config.h' is missing. > > |
From: [LoN]Kamikaze <LoN...@gm...> - 2006-05-18 11:15:43
|
The 'spvalid' target fails, because '../opensp/include/config.h' is missi= ng. |
From: [LoN]Kamikaze <LoN...@gm...> - 2006-05-18 11:06:19
|
In line 220 in the for loop 'i' is missing 'int'. |
From: [LoN]Kamikaze <LoN...@gm...> - 2006-05-18 10:52:23
|
On Unix systems compile fails because a file is named 'nsOpenSP.cpp' instead of 'nsOpenSp.cpp'. |
From: Change <ch...@zo...> - 2006-05-17 17:36:07
|
Hi, Are you aware that no line numbers are shown while using "wrap long lines"? Since the code is more readable and scan-able to me when I use "wrap long lines", I'm always using this. I couldn't find the line numbers in 0.8x until I disabled this feature. Would it be possible to have line numbers while it is enabled in a future version of this beta? Thanks, Change. |
From: Fabio R. <ml...@ro...> - 2006-05-17 13:22:21
|
Hi, I just succeded in compiling OpenSP for Linux. I am able to generate x86 and amd64 binaries. Just tell me what you want :-) Ciao, Fabio |
From: Fabio R. <ml...@ro...> - 2006-05-17 12:45:50
|
Hi, I just encountered an annoying bug in HTML Validator 0.8 for Windows. Open the attached file in firefox: no errors / warning given. Now select "Advanced / Validate after JavaScript..." and see the warning. Mainly, it removes the "/>" inside the <input> line. It does the same with "<img>", "<br>" and so on... Any hint? Thanks, Fabio |
From: Fabio R. <fa...@ro...> - 2006-05-17 12:06:24
|
Hi, I have seen that the new HTML Validator 0.8.x beta is out for Windows. I'd like to contribute in compiling the OpenSP package for Linux, since it seems the biggest problem for having HTML Validator 0.8 on linux. I have already downloaded the lates OpenSP source code. How would you like it to be compiled? The "simple" shared library is ok? Ciao, Fabio Rotondo |
From: Marc G. <mg...@sk...> - 2006-05-16 05:27:13
|
Test |
From: Marc G. <mg...@sk...> - 2006-05-16 05:26:01
|
Hi Bryce, I think you are a good candidate for the 0.8x beta version I released today if you are a Windows user. This version has a lot of new features but also validation of HTML pages after Javascript or Ajax execution. It works not in the same way than View Rendered Source. But with the innerHTML of the page what allow to find more errors. You can download it here: http://users.skynet.be/mgueury/mozilla/preview_080.html There are test files and explanations here: http://users.skynet.be/mgueury/mozilla/user_guide.html#AJAX Enjoy, Marc Bryce Nesbitt wrote: > Marc; > > HTML Tidy for FireFox is very very useful -- I can't imagine debugging complex websites > (which are built from many HTML fragments) any other way. > > Now, what about Javascript for the future? With many websites there are really *two* > totally separate websites that need to be debugged. First, the markup as presented to > the browser, e.g.: > > <body> > <p>This is a test</p> > <script> > document.write('<td>Hello World!</td>'); > </script> > </body> > </html> > > > Then, the markup after javascript functions have run, e.g.: > > <body> > <p>This is a test</p> > <td>Hello World!</td> > </body> > </html> > > > Is it technically feasible for HTML Tidy to validate both versions of a web page? Often the > really hairy stuff is built by javascript. > > I know of one extension that does something close: > http://jennifermadden.com/scripts/ViewRenderedSource.html > > |
From: Marc G. <mg...@sk...> - 2006-05-14 20:06:06
|
Hi Bryce, I think you are a good candidate for the 0.8x beta version I released today if you are a Windows user. This version has a lot of new features but also validation of HTML pages after Javascript or Ajax execution. It works not in the same way than View Rendered Source. But with the innerHTML of the page what allow to find more errors. You can download it here: http://users.skynet.be/mgueury/mozilla/preview_080.html There are test files and explanations here: http://users.skynet.be/mgueury/mozilla/user_guide.html#AJAX Enjoy, Marc Bryce Nesbitt wrote: > Marc; > > HTML Tidy for FireFox is very very useful -- I can't imagine debugging complex websites > (which are built from many HTML fragments) any other way. > > Now, what about Javascript for the future? With many websites there are really *two* > totally separate websites that need to be debugged. First, the markup as presented to > the browser, e.g.: > > <body> > <p>This is a test</p> > <script> > document.write('<td>Hello World!</td>'); > </script> > </body> > </html> > > > Then, the markup after javascript functions have run, e.g.: > > <body> > <p>This is a test</p> > <td>Hello World!</td> > </body> > </html> > > > Is it technically feasible for HTML Tidy to validate both versions of a web page? Often the > really hairy stuff is built by javascript. > > I know of one extension that does something close: > http://jennifermadden.com/scripts/ViewRenderedSource.html > > |
From: Marc G. <mg...@sk...> - 2006-05-13 10:01:33
|
Hi Bryce, I think you are a good candidate for the 0.8x beta version I released today if you are a Windows user. This version has a lot of new features but also validation of HTML pages after Javascript or Ajax execution. It works not in the same way than View Rendered Source. But with the innerHTML of the page what allow to find more errors. You can download it here: http://users.skynet.be/mgueury/mozilla/preview_080.html There are test files and explanations here: http://users.skynet.be/mgueury/mozilla/user_guide.html#AJAX Enjoy, Marc Bryce Nesbitt wrote: > Marc; > > HTML Tidy for FireFox is very very useful -- I can't imagine debugging complex websites > (which are built from many HTML fragments) any other way. > > Now, what about Javascript for the future? With many websites there are really *two* > totally separate websites that need to be debugged. First, the markup as presented to > the browser, e.g.: > > <body> > <p>This is a test</p> > <script> > document.write('<td>Hello World!</td>'); > </script> > </body> > </html> > > > Then, the markup after javascript functions have run, e.g.: > > <body> > <p>This is a test</p> > <td>Hello World!</td> > </body> > </html> > > > Is it technically feasible for HTML Tidy to validate both versions of a web page? Often the > really hairy stuff is built by javascript. > > I know of one extension that does something close: > http://jennifermadden.com/scripts/ViewRenderedSource.html > > |