aartaka
commented
3 years ago There are several ways to interact with DOM elements as of now:
- Parenscript (basically, a JavaScript code that we compile from Lisp code and run in the context of current page) -- this is the simplest way, as you can write arbitrary JS code that will most probably do the thing you want.
- Element hints (things we use for keyboard navigation) -- a harder yet more pleasant way. To use these with DOM, you need to define methods acting on a CLOS class corresponding to the given DOM element (e.g.,
image-hintfor<img>,input-hintfor<input>). Then (and that's the part we haven't yet implemented, so you can help with that) you can add hints of the page (add-element-hints), find the one you need (look at the code ofquery-hintin element-hint.lisp for how we currently do that with minibuffer), and invoke the necessary method on it. But this is still clunky, so Parenscript is the best bet as of now. - You can use Plump to parse the page if the only thing you need is to obtain values (Dexador, Drakma, and other ways to interact with the webpage programmatically are in this category too).
cl-03
jmercouris
Which packages are used behind Nyxt that operates DOM elements，these actions like MOUSE-CLICK, that switch the window to the href-links。 I want to automate the Nyxt process which like selenium but not use selenium which refer to Java or Python。 I've been thinking about this for a long time。 would you mind teaching me？ Can I find it in the file 《*.asd》? Maybe the principle is that ,located the DOM element ,then find the links around it ,then invoke DEXADOR or DRAKMA to request the URL I guess.But how Nyxt knew the DOM element been actived and the cursor located on it？ It can be found that request a url until select DOM elements in the book 《common lisp cookbooks》,but without next steps that interact with DOM elements.