Questions about web browser reverse-engineering

[HackerDude]

I'm working on some web-browser based JS stuff, and looking at the Kindle's USB partition, I can see a folder called .active_content_sandbox

In there, there is are cached Kindle store files, such as index.html, secondaryStore.html, e.t.c.

Most of them are obfuscated, but I can see some code that is related to a custom window.kindle property:

if(!kindle || typeof kindle === 'undefined') {
                    var kindle = window.kindle || top.kindle;
                }
                var host = kindle;
                
                if (host.dev.webkitLog) {
                    host.dev.webkitLog('on');
                }  
                
                var Locale = window.Locale || top.Locale;
                var DateTimeFormat = window.DateTimeFormat || top.DateTimeFormat;
                var NumberFormat = window.NumberFormat || top.NumberFormat;

and also

kindle.dev.loadResource&&(kindle.dev.loadResource(window.frameElement?window.frameElement.id:"externalPage","jquery"),kindle.dev.setSensitivity(!0,1500)),kindle&&(kindle.chrome.setTitleBar("",pageTitle)

However, when I try to observer window.kindle from the browser, it merely shows up as undefined, running the following code:

for (prop in window) {
  log(prop);
}

Shows that it does indeed exist, however, checking its type shows as "undefined" and it appears to be inaccessible


Does anyone know anything about how this works???
Thanks

[Kusuri]

i'm not really experienced that much in kindle hacking, but i know a bit of javascript (the code you posted here is js). and my guess is that window.kindle gets defined inside the application self (main kindle UI?), so if you try to access it outside of that context of the application where it is used, it isn't defined as a variable (window.kindle).. so it's undefined.

if i where you, i would try if i somehow could inject code into that main application where that code you posted is used to learn more about window.kindle and what exactly it contains. on a windows pc i would do something like "console.log(window.kindle);" but on a kindle you would need to save it somehow in a logfile i guess.

edit: maybe you could use the KindleTool to extract a kindle update to maybe find more of the code of the main application where the window.kindle gets defined that you can't normally access (in the filesystem of the kindle)?

[NiLuJe]

Historically, you'd have to look into a factory image and the diags partition to have a look at un (or less-) obfuscated code. But that was a hell of a lot of FW versions ago (and, like, 6 UI iterations ago? ).

TL;DR: No idea how this shit works since the React stuff.

[HackerDude]

My bad! It turns out, upon further investigation, that merely checking if window.kindle exists defined it (JS be like). It appears to be defined only when viewing the Kindle's "store" app, as the store app is basically just a glorified webview

I have also discovered that the Kindle's browser is the equivalent of Safari 5, so I can now locally test HTML and JS code

Unfortunately, as a result of this, it doesn't seem to be that useful to me

[luketheduke]

Originally Posted by Bluebotlabs

It turns out, upon further investigation, that merely checking if window.kindle exists defined it (JS be like).

Don't you love Javascript?

[HackerDude]

Originally Posted by luketheduke

Don't you love Javascript?

Unfortunately, I'm going to have to learn to for this current project...