I'm thinking about new JavaScript errors API which would work by overwriting
`window.errors` hook and provide a way to retrieve JS errors from page.
So, for example. I enable errors catching:
```ruby
Watir.catch_errors = true
```
Then, we add JavaScript (using `#execute_script`) like the one below in the
manner we run checkers (after going to url, refreshing page, clicking on
elements).
```js
window.jsErrors = [];
// make sure we run other callbacks
var callback;
if (window.onerror != null) {
callback = window.onerror;
} else {
callback = function () {};
}
// catch errors
window.onerror = function (msg, url, line) {
var error = {};
error.msg = msg;
error.url = url;
error.line = line;
window.jsErrors.push(error);
return callback();
}
```
After, we can retrieve array of hashes with error details by using
```ruby
Watir.errors
#=> [{ :message => "error", :url => "google.com", :line => 25 }]
```
There are several concerns however:
1. It would be better if we could only add script after page load (or
significant DOM change). I'm not sure if that's possible to do. @jarib Can we
get this info from Selenium-WebDriver?
2. I'm not completely sure about API methods naming. Use `errors` or
`js_errors` or `javascript_errors`.
3. Performance impact. I suppose we can make it faster by using
`Driver#execute_async_script`, but I haven't yet experimented.
Let me know what do you guys think.
---
Reply to this email directly or view it on GitHub:
https://github.com/watir/watir-webdriver/issues/191
_______________________________________________
Wtr-development mailing list
[email protected]
http://rubyforge.org/mailman/listinfo/wtr-development
