some_guy

just tested my yahoo bot and it doesn't work anymore, something about not entering the birthdate, the issue I had couple months back was that the country selector need to fire some javascript that wasn't documentated in the page, This fixed it var fireOnThis = document.getElementById('country'); if( document.createEventObject ) { fireOnThis.selectedIndex = 231; fireOnThis.fireEvent('onchange'); } 231 was the index of the country I was trying to select.

You have to do something funky with javascript, fire an event manually, let me go see if I can find my old bot.

If you're not prepared to do the above, quit now and save yourself some bother. The above attitude will do you good in so much IM related, not just ubot / programming.

Scaped these a couple of days ago and verified, dunno about know though 211.138.124.196:80 211.138.124.197:80 121.14.158.75:80 218.26.14.122:80 119.70.40.101:8080 222.92.117.241:80 193.136.138.15:3128 70.33.177.129:3128 217.176.200.250:80 94.100.90.22:80 202.28.38.32:8080 85.31.179.178:3128 91.121.201.130:80 77.94.32.171:80 206.224.254.11:80 192.133.83.235:80 61.145.112.86:80 130.245.191.60:3124 195.168.109.60:8080 216.218.211.156:80 147.102.82.32:3128 68.68.107.60:29505 222.165.130.82:80 193.191.148.227:3128 193.191.148.227:3127 143.205.172.12:3124 144.140.129.30:80 128.232.103.203:3124 7

I have one, it used to work, haven't tried it in a couple of months but don't see why it wouldn't. What you got to swap for it?

Have been trying to test it out in buzzfeed, but cannot get the flash thing to work at all. Just get a black box with a save and cancel button. From what I can tell, the code you have posted above is not relevant but tbh I don't really know. Will have another look next time reboot machine to see if that helps me run the flash.

My javascript skills are pretty rubbish, so can't help that much, but if you can get the ID of the control you want to fire, you can use something like var fireThis = document.getElementById('**ELEMENT HERE**'); if( document.createEventObject ) { fireThis.fireEvent('click'); } replace **ELEMENT HERE** for the element ID you want you can replace 'click' for any JS event, so onchange, mouseup, mousedown etc etc

The web browser control is IE based, it will not work with Firefox (if you were wanting to utilize FF plugins, you can't from what I understand). It might be technically possible for uBot to give the option of working with a mozilla based browser (geckofx), but don't think this is on the cards (atm).

Hi, sorry for not replying have been out of action last couple of weeks. I've had issue with copying nodes, especially subs and if changing the text in the name. As Alc hitting refresh should help, I also used to do something else which would mean it always worked, but can't remember what as am abit ubot rusty. If I remember I shall post it here

Very nice and thank you. How did you get round the gmx password javascript checker thing if you don't mind me asking?

yeah yeah you guys definately getting some love sent your way from me

how to make teh moniez please

Many thanks for this, will give it a good test when get some time.

Replied to the PM you sent, although in the other thread said I solved it, I was wrong, it did seemingly a couple of times, but then failed. So I gave up. Just went back to try and fix it and seems I deleted the almost finished bot in a fit of rage. The issue is to do with (I think) Javascript attached to the Javascript events onkeydown onblur Could look at trying to force these events in Javascript. Also is possible some events are required that are not noted in the source code. Could try and force onchange, onkeyup etc etc

Just a question, how do you guarantee? I can say safely you do not have a captcha rate of 100% and tbh saying so is unrealistic. Unless there have been some changes to your system in the last 2 weeks. 95%? Possibly, I did get a couple of wrong answers when decoding Gmail captchas but they are pretty tough (I only have a success rate of 50% when I do them manually ) perhaps if I sent another 90 through they rate would go back to up to 95%.

Solved my problem. Instead of selecting the submit button, then changing chosen attribute>click chosen, I use a Run Javascript node with var o = document.getElementById('signup-w-email'); o.submit(); where 'signup-w-email' is the ID of the form used.

sounds as if it is picking up on the google pages. If can understand the scripting, put a pause node after it scrapes the google pages, then test the list to see if it has picked up the links to the following pages. It could be that the code I used to scrape the google pages isn't working for you as google switches up the code. This is just a guess, as the scraper still works for me. Does this problem occur for every search term? If get some free time over the next few days will try and improve it. But feel free to dive into the code and fix it and post it up

Got a similar issue signing up to blinklist social site. The submit button if I manually click it works, if ubot clicks it doesn't work. Am intending to look at my problem tonight, if I can solve it will let you know as it may help.

I have a bot which has many different yet similar functions (social network signup) The signup sub is nearly the same each time with a few changes The catpcha routine is nearly the same. Should I be able to copy the sub from one script, and paste it to another. Then right click the sub name, select insert string and change the name of the sub to a unique variable. This then throws an error about a key. I click refresh and it carries on (or so I thought). I was pretty sure this was working (not 100%). Now when am testing the bot, the run sub is being skipped. I have to copy the sub name f

Perhaps something has gone muddled between the server and you. Are you using a proxy? Have you tried restarting browser / machine? Have you tried to sign up in a browser without ubot?

If that doesn't work, next look at the source code and see if there is any javascript that is fired, if not, try and force a JS event onclick / onchange for that element.

To solve your issue with the b tags use a strip_html javascript call, there is a lib uploaded to the forums for this. However you cannot use the cite tags for scraping because they may not have the full URL <cite>http://www.somereallylongdomainandthensomehere.com/folder......../webpage.html I uploaded a bot to the buy / borrow / steal board with a Google Serp Scraper, have a look in that to see what I scraped the URLS with, cannot remember off the top of my head

Magic Or on a more serious note, I know PHP can store information in a cookie or a session variable. The session variable sits on the server so wouldn't be cleaned by ccleaner. Purely speculating here as I don't know, but when you go to site, there must be something kind of data swap for sessions to work, I am guessing something is passed over that authenticates you that doesn't change (normally) when you go the site. This can be used to identify you (unless you log out, close browser etc).

I'll have a look at the captchabot API and see what can be done. When was tested this bot last friday night, it was firing back the answers in seconds (friday night uk time so mid afternoon US time). When was revising the bot saturday midday UK time, it was taking 1-2 mins as posted above. I assume that deathbycaptcha uses US based web sites to break the captchas (enter this captcha to see boobies kinda thing).

Posted a bot in the string lib thread that shows you how to do this