![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
I need a program that will automate web activity
- Thread starter matmga400
- Start date
MrGuvernment
Fully [H]
- Joined
- Aug 3, 2004
- Messages
- 21,808
how should it know something of interest you may like?
I meant to say that it should click on a particular button within the website.
Chris_Lonardo
[H]ard|OCP Storage Engineer & Editor
- Joined
- Feb 10, 2002
- Messages
- 1,726
Selenium should work for you- http://www.seleniumhq.org/
I already tried Selenium.
I think I used it for less than one day because I couldn't make it run at a certain time by itself and I think there were websites that it wouldn't log into. Then again, I was not an expert with this program and it's been a while since I tried it.
Are there some features I missed? Anything different I should have done with this program?
AutoIT for a generic solution, iMacros for a browser solution (http://imacros.net/overview)
Digital Viper-X-
[H]F Junkie
- Joined
- Dec 9, 2000
- Messages
- 15,115
autohotkey?
it's more simple, but if it's persistent you can record it and just replay it
eg
win+r (or a mouse click) / chrome.exe
alt+ent for maximize
tab tab tab until you land on your button
space
alt+f4 close
set on a timer
it's more simple, but if it's persistent you can record it and just replay it
eg
win+r (or a mouse click) / chrome.exe
alt+ent for maximize
tab tab tab until you land on your button
space
alt+f4 close
set on a timer
CaptNumbNutz
Fully [H]
- Joined
- Apr 11, 2007
- Messages
- 24,970
1. Create website full of ads
2. Go to forum and ask about programs to automate clicking activity
3. Start running program
4. Profit
5. Get investigated for fraud
2. Go to forum and ask about programs to automate clicking activity
3. Start running program
4. Profit
5. Get investigated for fraud
You write that part of the program yourself, or run the program as a scheduled task in your OS.
jojo69
[H]F Junkie
- Joined
- Sep 13, 2009
- Messages
- 11,267
1. Create website full of ads
2. Go to forum and ask about programs to automate clicking activity
3. Start running program
4. Profit
5. Get investigated for fraud
no, clearly OP is planning a crime and is trying to set up an alibi
I understand the part about Win 7 running Selenium as a scheduled task at a particular time. But how would I get Selenium to automatically run a saved file?
I assure you that I'm not trying to do anything illegal or unethical. It's just a few bad people that ruin it for the most of us. I'm just asking a technical question here and the topic of automated activity is fair game in the subject of programming.
I'll take it you are trying to do this for free as much as possible.
I use TestingAnywhere, a program created by AutomationAnywhere. Unfortunately it's not a free program.
Selenium is another option and you could also look into Oracle's OpenScript for web based functions. You'll have to code on the backend to schedule the tasks for the free options... however, TestingAnywhere lets you schedule to your hearts content.
I use TestingAnywhere, a program created by AutomationAnywhere. Unfortunately it's not a free program.
Selenium is another option and you could also look into Oracle's OpenScript for web based functions. You'll have to code on the backend to schedule the tasks for the free options... however, TestingAnywhere lets you schedule to your hearts content.
Tawnos
2[H]4U
- Joined
- Sep 9, 2001
- Messages
- 3,808
Easiest way (for me, at least) would be to use selenium webdriver to perform the actions I desire, then compile that into an executable that task scheduler would run. You can also use the older selenium RC and set up your command line params in the scheduled task.
StoleMyOwnCar
2[H]4U
- Joined
- Sep 30, 2013
- Messages
- 2,997
I haven't tried Selenium yet, thanks for whoever linked it. Here are some other options:
http://phantomjs.org/
However, I have found phantomjs to be a bit tricky when it comes to actual navigation and polling after navigation. I managed to hack it in with some Javascript tricks, but I think the better option is casperjs (though I haven't used it yet, it used phantomjs as its core):
http://casperjs.org/
The issue with phantomjs is that I had to basically create a settimeout that waited until the page was loaded, and even then it was a bit weird.
If you're familiar with Perl, there's WWW::Mechanize
http://search.cpan.org/~ether/WWW-Mechanize-1.74/lib/WWW/Mechanize.pm
(I believe this is included as core in Strawberry Perl, so you shouldn't have to use CPAN to grab it).
With WWW::Mechanize you can select a certain form and post to it. Then you can find which link you need and click on it. Beware that it doesn't support Javascript though. So for instance if it is a pseudo form that does some kind of override of the post method and does an Ajax post it may not work. I haven't personally tried it for that, but I simply do not believe it would work.
The advantage of phantomjs or casperjs (I think, I haven't tried the latter) is that either one lets you run actual javascript in the page. Since they also base their automation scripts on javascript, it's natural that they support any javascript happening in the page. You can practically do anything using either one. Opportunities are limitless. Just that if any requests happen via AJAX you should likely have a set timeout call in order to let the script have a bit of breathing room. It requires asynchronous polling and execution. I think almost every language has some kind of crawler library, so WWW Mechanize isn't really novel.
I mean technically since this is a form you're talking about, you could find out what the form posts to and then simply use curl to send a post request to that URL with your relevant info. I think curl lets you keep your login session active, so then afterwards you can use curl to grab the logged in page, parse out the link of interest, and then curl the link.
In all cases you might need to edit the user agent. The advantage of perl and phantomjs (and curl) is that they let you practically run it all from command line. Very easily made into a scheduled task.
http://phantomjs.org/
However, I have found phantomjs to be a bit tricky when it comes to actual navigation and polling after navigation. I managed to hack it in with some Javascript tricks, but I think the better option is casperjs (though I haven't used it yet, it used phantomjs as its core):
http://casperjs.org/
The issue with phantomjs is that I had to basically create a settimeout that waited until the page was loaded, and even then it was a bit weird.
If you're familiar with Perl, there's WWW::Mechanize
http://search.cpan.org/~ether/WWW-Mechanize-1.74/lib/WWW/Mechanize.pm
(I believe this is included as core in Strawberry Perl, so you shouldn't have to use CPAN to grab it).
With WWW::Mechanize you can select a certain form and post to it. Then you can find which link you need and click on it. Beware that it doesn't support Javascript though. So for instance if it is a pseudo form that does some kind of override of the post method and does an Ajax post it may not work. I haven't personally tried it for that, but I simply do not believe it would work.
The advantage of phantomjs or casperjs (I think, I haven't tried the latter) is that either one lets you run actual javascript in the page. Since they also base their automation scripts on javascript, it's natural that they support any javascript happening in the page. You can practically do anything using either one. Opportunities are limitless. Just that if any requests happen via AJAX you should likely have a set timeout call in order to let the script have a bit of breathing room. It requires asynchronous polling and execution. I think almost every language has some kind of crawler library, so WWW Mechanize isn't really novel.
I mean technically since this is a form you're talking about, you could find out what the form posts to and then simply use curl to send a post request to that URL with your relevant info. I think curl lets you keep your login session active, so then afterwards you can use curl to grab the logged in page, parse out the link of interest, and then curl the link.
In all cases you might need to edit the user agent. The advantage of perl and phantomjs (and curl) is that they let you practically run it all from command line. Very easily made into a scheduled task.