Limit of 4K
in [JavaScript]
|
From: sheldonlg on 29 Oct 2009 19:13 First, I know just a little Javascript -- enough to be dangerous :-). I have come across a problem that Firefox discards all but the first chunk of 4096 on a return from an AJAX call. I searched Google and found that it is a known problem and found a suggested solution. I tried it but it doesn't work. I am asking here on what I have to do to my code to make it work transparent to the particular browser. Here is the pertinent code segment. The getNodeText function was the suggestion I found on Google. The commented line in handleResponse is what was there before (and also only got 4096). function getNodeText(xmlNode) { if(!xmlNode) return ''; if(typeof(xmlNode.textContent) != "undefined") return xmlNode.textContent; return xmlNode.firstChild.nodeValue; } function handleResponse(pObjRequest) { if (pObjRequest.readyState==4) { if (pObjRequest.status==200) { //resp = pObjRequest.responseText; resp = getNodeText(pObjRequest); condition_code = parseMultiXML(resp); pObjRequest=null; // dispose of the now un-needed object. } } } Any help would be appreciated.
From: Stefan Weiss on 29 Oct 2009 19:19 On 30/10/09 00:13, sheldonlg wrote: > I have come across a problem that Firefox discards all but the first > chunk of 4096 on a return from an AJAX call. I searched Google and > found that it is a known problem and found a suggested solution. I > tried it but it doesn't work. I am asking here on what I have to do to > my code to make it work transparent to the particular browser. Here is > the pertinent code segment. The getNodeText function was the suggestion > I found on Google. The commented line in handleResponse is what was > there before (and also only got 4096). Call normalizeDocument() on the response XML. cheers, stefan
From: Stefan Weiss on 29 Oct 2009 19:50 On 30/10/09 00:19, Stefan Weiss wrote: > On 30/10/09 00:13, sheldonlg wrote: >> I have come across a problem that Firefox discards all but the first >> chunk of 4096 on a return from an AJAX call. I searched Google and >> found that it is a known problem and found a suggested solution. I >> tried it but it doesn't work. I am asking here on what I have to do to >> my code to make it work transparent to the particular browser. Here is >> the pertinent code segment. The getNodeText function was the suggestion >> I found on Google. The commented line in handleResponse is what was >> there before (and also only got 4096). > > Call normalizeDocument() on the response XML. Sorry, that was probably too terse to be useful. The normalizeDocument() method should be called on the responseXML's document element - if the method is available. Here's some example code; 'xhr' is the XMLHttpRequest object, and error handling has been omitted for clarity: var docEle = xhr.responseXML.documentElement; if (typeof docEle.normalizeDocument == "function") { docEle.normalizeDocument(); } This will join adjacent text nodes into single text nodes, avoiding the 4K limit in Mozilla's XML parser. It will also perform other normalizations; the full description is available here: <http://www.w3.org/TR/2004/REC-DOM-Level-3-Core-20040407/core.html#Document3-normalizeDocument> HTH, stefan
From: Martin Honnen on 31 Oct 2009 08:40 Stefan Weiss wrote: > Sorry, that was probably too terse to be useful. The normalizeDocument() > method should be called on the responseXML's document element - if the > method is available. Here's some example code; 'xhr' is the > XMLHttpRequest object, and error handling has been omitted for clarity: > > var docEle = xhr.responseXML.documentElement; > if (typeof docEle.normalizeDocument == "function") { > docEle.normalizeDocument(); > } normalizeDocument is a method of the Document interface, not of the Element interface so you will never be able to call it on the documentElement. And normalizeDocument is DOM Level 3 while the method 'normalize' is DOM Level 1 and does all you need so doing xhr.responseXML.documentElement.normalize() should suffice and have better support than the normalizeDocument method. -- Martin Honnen http://msmvps.com/blogs/martin_honnen/
From: Stefan Weiss on 31 Oct 2009 15:59
On 31/10/09 13:40, Martin Honnen wrote: > Stefan Weiss wrote: > >> Sorry, that was probably too terse to be useful. The normalizeDocument() >> method should be called on the responseXML's document element - if the >> method is available. Here's some example code; 'xhr' is the >> XMLHttpRequest object, and error handling has been omitted for clarity: >> >> var docEle = xhr.responseXML.documentElement; >> if (typeof docEle.normalizeDocument == "function") { >> docEle.normalizeDocument(); >> } > > normalizeDocument is a method of the Document interface, not of the > Element interface so you will never be able to call it on the > documentElement. > > And normalizeDocument is DOM Level 3 while the method 'normalize' is DOM > Level 1 and does all you need so doing > xhr.responseXML.documentElement.normalize() > should suffice and have better support than the normalizeDocument method. Thanks, I got that mixed up when I tried to simplify existing code. I wasn't aware that normalize() was better supported; very interesting. cheers, stefan |