Alex said:
All regular benefits of web application:
(1) easy to deploy with my application
(2) multiple browsers compatibility
Plus, if source code is available, I can invoke it from my js
application when it is exactly in the state I wanted to inspect plus I
can pass the reference to the exact object I wanted to inspect.
[...]
Here is a 'DOMwalk' script I wrote some time ago as a quick 'n dirty
tree dump. If you give domReport() an id it will walk down from there.
If no id is given, it starts from the html element. I've tested it in
most modern browsers and it seems to be fine.
Everything is listed in a table with a node reference, with a bit of
effort a better presentation is possible. Nodes could be listed as <li>
elements and each parent node could have an associated <ul> element with
an onclick that collapses the branch. Node attributes could be in a
<span> beneath each node and be shown/hidden when the node is clicked on.
I had intentions of doing at least some of the above, but it seemed
pointless given that a perfectly serviceable DOM inspector comes free
with Geko browsers (there are probably some for other browsers too but I
haven't bothered to look).
Let me know if you want the above functionality added.