Hi, I wanted to write a C++ code that visits a particular URL and then enters a pre-defined value in it and presses "Enter" so that it can navigate to the next page and then get the data present in that page by parsing that page. This is similar to something that can be done by selenium, but that's mostly testing and I want to specifically use C++ so that I can gain some fluency in C++ too. Could you suggest some libraries /Framework that can help us get this task done?
Try Casablanca (aka C++ Rest).
Originally Posted by Adak
Oh God, is that a Microsoft product?
*vomits*
OP, would libcurl do what you want?
Edit : Yay, I finally figured out how to do links!!!! Turns out, it's not [a href="..."][/a] but [url="...
Last edited by MutantJohn; 06-28-2015 at 12:08 PM.
Oh God, is that a Linux C API?
*vomits*
OP, would you ignore this idiot?
Hey, they have some Windows stuff on their download page.
>_>
<_<
Don't care, linux lover. My current library (guess what it is - it's actually Casablanca!) suits my purposes just fine.
You say Linux lover like it's a bad thing XD
Out of curiosity, how would one go about this without using an external library?
If you want to request the HTML content of a webpage, you obviously have to use sockets, right? How do you make a page request?
You connect to the website using your OS APIs and follow the HTTP protocol.
No, it's a cross-platform library/API, you dingbat.
Just out of curiousity, even with Casablanca or libcurl - how would automate the whole thing?
this is not just about creating a simple http request, it's actually parse the HTML, set some details in some DOM element , send the request again and parse the result again.
I just don't get how can you recomend an HTTP library for a Web automation task, which is not the same thing.
Unfortunatly, something like Selenium does not exist (from what I know) for C++, and honestly? I don't see why doing it when you can do it in Selenium with simple JavaScript.
You go to your page, boot up your local http debugger which is probably built in to your browser, then look at the http requests sent from the page as you push different buttons and fill in different values. Then you just mimic those requests from your http library. Not so hard.
For parsing, you can use htmlcxx. Fetch the page, then parse it using htmlcxx to get your data. It doesn't need to be complex parsing. Just get to whatever element the data is contained within and extract the relevant data.
I would say, just switch to Java and use jsoup lol.
or Node.js , which is a bit more suitable.
I doubt anything javascript-based would be better for anything, but regardless, the OP asked specifically for C++.
this is very language-beagot saying.
and I don't agree, node.js has excelent html parsing libraries where you can basically excess the DOM as it meant to be excessed (by JS).
