We Keep Coding

minko get object from dom

I create an object in my javascript function, and I'd want to retrieve it in c++ from dom class for change some values, but I can access only by id, tag or class that are part of css syntax. Is there the possibility to get my object and set values or send to him those value?

First of all I think you should always get DOM elements by id or class name because IMHO it's the most versatile way to get things from this kind of tree.
Anyway, just like any other DOM Minko provides the childNodes and parentNode properties if you want/have to browse the tree :
AbstractDOM::childNodes()
AbstractDOM::parentNode()
When you've found the right DOM element, you can then use the other DOM methods to get its content, set its value, etc... Everything you need should be in the AbstractDOM base class definition.

can I use document.prototype.getElementById to get an element in another element, not in the whole document

I recently found on this javascript tip :
element_number = Array.prototype.indexOf.call(element_1, element_2);
It allows developers to use the indexOf method on an object wich is not an Array.
I would like to know if it is possible to use a similar syntax to call the getElementById method but not on the whole document (document.getElementById), just on an element like this :
my_div_2 = document.prototype.getElementById.call(div_1, "id_of_my_div_2");
The idea is that my document contains tabs and elements having the same id can be present several times in the document.
If it is not possible, did somebody write a function doing that :
Search in an element another element by id.

No, but you can accomplish what you want using querySelector instead.
my_div_2 = div_1.querySelector("#id_of_my_div_2");
Not supported in IE6/7.
Docs: MDN element.querySelector()
That said, duplicate IDs are invalid.
This technique is useful if the same script is used on different pages where the ID may or may not be in a particular container.

No. The getElementById method must be applied on objects of type Document, otherwise it would throw a WRONG_THIS_VALUE exception. You could try it with
myDiv1 = document.getElementById("div1");
myDiv2 = document.getElementById.call(myDiv1, "div2"); // no ".prototype"!

This makes no sense as only one element with a given id is possible in a document.
So the only right syntax would be
my_div_2 = document.getElementById("id_of_my_div_2");
Don't reuse an id in a document. You probably can achieve the desired result using classes instead of id.

Control references in jQuery

function eegetdropdownvalue_str(ctl){return ctl.selectedIndex>=0&&ctl[ctl.selectedIndex]?ctl[ctl.selectedIndex].value:''}
The above function is called with
co.p1A10=eegetdropdownvalue_str(document.formc.p1A10);
I want to switch the call over to jQuery to drop the document.form reference however doing this
co.p1A10=eegetdropdownvalue_str($('p1A10'));
Does not reference the control correctly - How should I do this?

There's two things wrong with your code.
First, $('p1A10') references nothing.
jQuery selectors work almost identically (if not completely identically) to the way css works.
So, just ask yourself how you would reference the object(s) in question in CSS and you're half way there.
I'm assuming that p1A10 is the name or id of an object. Since we're using CSS/jQuery syntax, this should be an id, although you can select by other attributes such as $("select[name='p1A10']") .
To reference an object by ID we use the # character (again, just like in CSS). So we can select your node via $('#p1A10').
The second problem is that your function is expecting a DOM object not a jQuery object. To keep your code intact, we need to say $('#p1A10')[0] where 0 is the first element within the collection of jQuery elements.
I've provided two examples to explain this a little better. One uses your existing infrastructure and one replaces it.
http://jsfiddle.net/TD6Uu/5/
Hope it helps.

Given a form with id formc and a select with name p1A10 you could e.g. use:
o.p1A10 = eegetdropdownvalue_str($('#formc select[name="p1A10"]').get(0));
If this doesn't do it, please provide use with the exact HTML structure

ExtJs 4, is there a way to find control? (not by "id")

I have button with id = 'someid', i can find it by Ext.getCmp('someid')...
Is there a way to set some kind of "extraId" to that button, and find that button by "extraId'?

The question is where you want to search to get your item.
Another way could be the itemId.
An itemId can be used as an alternative way to get a reference to a component when no object reference is available. Instead of using an id with Ext.getCmp, use itemId with Ext.container.Container.getComponent which will retrieve itemId's or id's. Since itemId's are an index to the container's internal MixedCollection, the itemId is scoped locally to the container -- avoiding potential conflicts with Ext.ComponentManager which requires a unique id.
Source and further documentation: http://docs.sencha.com/ext-js/4-0/#/api/Ext.AbstractComponent-cfg-itemId
The ComponentQuery (mentioned by wombleton) is also a good way but maybe not as performant as the itemId.

You can search for it by doing an Ext.ComponentQuery.
Example that will work in the console on the linked page:
Ext.ComponentQuery.query('button[text=OK]')
You can replace text=OK with the attribute you are searching for.

I don't have access to the ExtJS 4 API docs, but I know that in 3.X and previous verions have some other ways of getting references to your compoents without the use of the DOM ID.
If your components are part of a form or form panel, you should have access to a method called findField(). It is inherited from Basic Form. This method can take an id, dataIndex, name, or hiddenName to get a reference to your component.
There is also a helper method called findByType(xtype) that you can use to try and find your buttons. Another option would be to extend the button you use to place a unique property that you can use as a reference.

An alternative to id is itemId and it is limited to the scope of the container in which it is defined. So, in this case, there is not need to define globally unique itemId as done in case of id. We can use component query to get the component object having the itemId. Please follow this link where i have posted about how to use itemId and what is its benefit.
http://jksnu.blogspot.in/2012/11/use-itemid-instead-of-id.html

Unique identifier for HTML elements

Besides the ID, if you say you want a unique identifier for an HTML element (let’s say a div).
I browsed the DOM for something (like a number or string) that was unique for each element; but the DOM was big and I failed to find that on the Internet.
Is there a property (in the DOM obviously) that is unique only to that element? (Other than the ID and also you don't specify it, but it comes when the DOM is constructed.)

Depending on the objective, here are two suggestions.
Unless you actually need to express the id as some kind of string, you can save the normal DOM reference.
If you do need to express it as a string for some reason, then you'll need to assign a unique id yourself.
var getId = (function () {
var incrementingId = 0;
return function(element) {
if (!element.id) {
element.id = "id_" + incrementingId++;
// Possibly add a check if this ID really is unique
}
return element.id;
};
}());

The only other identifier I can think of is the XPath of the element in the document.
For instance, the title link inside the heading of this very page has an XPath of
/html/body/div[3]/div[2]/div/h1/a
But like Pekka already said, it depends on what you want to do. And I don’t think you can get the XPath easily from the DOM in JavaScript, despite XPath being available nowadays in JavaScript engines.

Internet Explorer has a property, "uniqueID", for every element. The problem is that the other browsers don't support it.

You can use a library or roll your own to create a unique identifier. jQuery has .data():
Store arbitrary data associated with the matched elements or return
the value at the named data store for the first element in the set of
matched elements.

I just encountered the same situation, and while I was looking into some DOM elements in the Chrome developer tools inspector, I noticed that they all seem to have a property like jQuery11230892710877873282 assigned with a unique number.
Obviously the number after 'jQuery' is different every time you load the page. My guess is that jQuery is generating this internally every time it tries to access or manipulate any DOM element.
I played a little bit with it, and it looks like elements that are never accessed/manipulated by jQuery may not have this property, but the moment you do something like $(elem), the property will be there. So, since we're using both jQuery and Lodash, I devised the following function to return a unique ID regardless of whether the element actually has a DOM id attribute.
_.reduce(
$(elem),
function(result, value, key) {
if(_.startsWith(key, 'jQuery'))
return value;
},
0)

There is the name attribute that can be addressed by document.getElementByName.
I don't think other unique identifiers exist - even though you could simulate one by setting a property (like title) to a unique value, and then query for that. But that is kludgy.