- From: Laura Holmes <holmes@google.com>
- Date: Mon, 20 Aug 2007 14:09:01 -0400
- To: "Sean Owen" <srowen@google.com>
- Cc: "Abel Rionda" <abel.rionda@fundacionctic.org>, public-mobileok-checker@w3.org
- Message-ID: <135a9f560708201109i31379021o662f9c9505e4d196@mail.gmail.com>
Indeed this seems to be the fix! Thanks! The code is already switched, and I will be making a commit soon. I really appreciate all your help. Hoping this change doesn't cause more issues to arise, Laura On 8/20/07, Sean Owen <srowen@google.com> wrote: > > Yes let us have a call tomorrow to discuss this, other open issues, > give status, make sure we are all set for our next F2F meeting, and > more. > > (So is this solved if Saxon is only used where we have to, and that is > to re-parse the original document to obtain line numbers? Laura and I > were discussing a related issue today that is fixed by doing so.) > > Sean > > On 8/20/07, Abel Rionda <abel.rionda@fundacionctic.org> wrote: > > > > > > > > > > Hi Laura and everyone, > > > > I have been studying this strange problem and I have realized the > > following: > > > > *I think that the problem rises with pages of small size. For example > if > > you try > > to check these two mobile pages (they are modifications of > google/xhtml): > > > > -http://idi.fundacionctic.org/bk/google_bad.xhtml > > -http://idi.fundacionctic.org/bk/google_ok.xhtml > > > > you will see that doc content is built correctly in the second case > and > > the only difference > > between them is that the second one has two extra lines. > > > > *It seems that the problem is the DocumentBuilderFactory > Implementation. > > Currently, saxon > > implementation is used > > (net.sf.saxon.dom.DocumentBuilderFactoryImpl) but if xerces > > is placed > > instead everything seems all right. > > So you can replace this line in parseDOM method in HTTPXHTMLResource > > > > final DocumentBuilderFactory dbf = new > > DocumentBuilderFactoryImpl(); > > > > with the default included in Java 5 (Xerces - > > com.sun.org.apache.xerces.internal.jaxp.DocumentBuilderFactoryImpl) > > > > final DocumentBuilderFactory dbf > > =DocumentBuilderFactory.newInstance(); > > > > Please try this and tell me if it solves the problem. > > > > Regards, > > > > Abel. > > > > PS: Is a call planned for tomorrow? Miguel and Nacho are on holidays > but I > > will try to attend it. > > > > > > > > > > -----Mensaje original----- > > De: public-mobileok-checker-request@w3.org en nombre de > > Laura Holmes > > Enviado el: lun 13/08/2007 23:31 > > Para: public-mobileok-checker > > Asunto: strange elements in DocContent of moki > > > > > > Hey all, > > I'm encountering some phantom elements in the moki document. Basically, > > when the original source is dumped back into the moki, I get some > content > > that looks like: > > > > <html xmlns="" xmlns:xsi=" > > http://www.w3.org/2001/XMLSchema-instance" > > version="-//WAPFORUM//DTD XHTML Mobile 1.0//EN"> > > <head profile=""> > > <title>Test</title> > > </head> > > <body style="background:url( 'Hand.gif' );" > > xml:space="preserve"><p style="font-size: 30pt; position: absolute; > > left:450px" xml:space="preserve">Test</p> > > > > The code continues, but this snippet exemplifies some of the issues: > > 1) no namespace! > > 2) a mysterious empty 'property' attribute in the <head> tag > > 3) many other tags contain the 'xml:space="preserve" attribute. > > > > For purposes of comparison, here's the text it should be outputting: > > <html xmlns="http://www.w3.org/1999/xhtml"> > > <head> > > <title>Test</title> > > </head> > > <body style="background:url( 'Hand.gif' );"> > > <p style="font-size: 30pt; position: absolute; left:450px">Test</p> > > > > Does anyone know why this is going on, or if there's a quick fix? > > > > Thanks, > > Laura > > > > > > > > > > >
Received on Monday, 20 August 2007 18:09:34 UTC