Super Dave

I've been taking a hit over here from facebook. This is really a two part question. First, has anyone else had a slap across bot accounts? Like, massive sweeps of accounts being gone overnight? Second, I'm having trouble identifying different variables that help FB determine my accounts are bots. I've tried localizing registrations to the IPs, tons of IPs, regular status updates, irregular updates, different times between registration and confirmation, profile details, experimented with # of outbound friend requests, experimented with # of organic or inbound requests, pictures, linking the

http://imgur.com/36N9c.jpg Here's the thready with an explanation

Lots of FB experience here. Account Creation/Confirmation, Status Updates, Wall Posts, Direct Messages, Profile updates, etc What are you having the most trouble with?

Hey thanks Frank! The output that you read is hand-coded after I finish each sub. As you may recall I've been lobbying for a print/export feature specifically so I can track changes in larger bots. Right now I code the ubot portion, then the sudo code version, then let my SVN keep track of the changes. If you don't mind, take a look into the other thread and comment on my post at the end. It'd be great to get some feedback from others who use advanced architecture in their bots!

Great video Frank and very similar to how I set up my bots. here's what I do.. First I have my controller bot, it's the one that handles all the big chores like navigating here or there, moving data from one routine to the next etc. Those routines however are nested into smaller bots. I'm also very specific with my naming conventions as well so that I don't lose track of sub routines or variables that belong to them. Basic structure is like this... "Module-Page-Action-variable" Here's how it fits together I hope the code is easy enough to follow... Mainbot.ubot include lib.facebook.ub

In the meantime I've found that putting the code for commenting inside of a new sub, with a "run javascript "/** With Comments **/" in there as well to explain why the code is commented. lol I can't tell you how many times I've made a small change only to continue development and then come back later and not know wtf I wrote that for.

I've created a timer sub that takes days,hours,min, and or seconds as input then after calculating the seconds needed it'll loop for that number of seconds. Inside my loop however I set variable to display through a UI stat mon so I get a visual display of how many seconds I have to wait. And sharing is caring, so see attached. Inputs are times as well as a message. Use "Resting %time%s..." to display "Resting 20s...", "Resting 19s..." in the #debug variable That said, put this at the end of a loop and you'll have a visual countdown until the next iteration of your loop PS If you rewr

Just two lists, and a loop. input: #input-file clearList %input-file addToList(%input-file, listFromFile(#input-file)) loop(listTotal(%input-file) { set #this-loop-line = nextListItem(%input-file) clearList %this-loop-line-parts addToList(%this-loop-line-parts, listFromText(#this-loop-line, ":")) set #field1 = listItem(%this-loop-line-parts, 0) set #field2 = listItem(%this-loop-line-parts, 1) set #field3 = listItem(%this-loop-line-parts, 2) // Do Stuff }

As Pftg4 mentioned a while loop would be the best and most flexible. Here's how I handle the multiple attempts as well as setting a max limit set #num-tries = 0 set #max-tries = 5 set #test-pass = 0 WHILE both #test-pass = 0 #num-tries < #max-tries inc #num-tries // do captcha stuff waitFinish if searchPage("Something Specific to the 'Success' page") then inc #test-pass END WHILE if searchPage("Something Specific to the 'Success' page) then // More code here now that captions are solved else // Something else here if attempts went over the max allowed.

Have a look here for an additional line to run that deletes the folder used to store Flash Cookies http://ubotstudio.com/forum/index.php?/topic/4239-ultimate-cookie-solution/page__p__16860__fromsearch__1#entry16860

Hey guys I've been pretty deep into building Facebook libraries to act as the core to a marketing platform I'm developing. I'm putting the finishing touches on an extensive list subs that plug right into your script to handle login/logout, friend requests (sending and accepting), targeted friend finder (Location or Interested Targeted), status updates, posting to pages, and friend walls. I'm putting the feelers out there now because I'd like to know what some of your biggest difficulties are in these areas so that when I release them I've got everyone covered. So with that said, what are

Needed a solution to removing duplicate lines in massive text files (+15mb) before processing in uBot. uBot's own ignore-list-duplicate function worked but with the file sizes I was dealing with it wouldn't scale. I've used 'sort' to do this through a terminal session but now I needed a way to use sort with ubot. After a little research I found a developer's site with this description: For a little perspective, I killed a bot after an hour into processing a 600k line file... with the "sort.exe -u" utility processed the file in less than 15 seconds. Enjoy: http://sourceforge.net/project

set #nav_url "http://whereiwannago.com" set #search_str "Something specific to the target page" set #nav_timeout "as long as you want to wait" set #max_tries "how many times to try before skipping?" set #is_loaded = 0 set #nav_tries = 0 START while both: #is_loaded = 0 #nav_tries < #max_tries nav #url wait no wait for #search_str timeout #nav_timeout if searchPage(#search_str) set #is_loaded = 1 else inc #nav_tries [ Put a delay in here to wait a certain time before trying again] END while START if not #is_loaded = 0 [do some cool stuff now that the page is loaded] else [page di

var raw = '{1}'; var matches = raw.match(/\(?(\d{3})\)?[- ]?(\d{3})[- ]?(\d{4})/g); if ( matches != null) { alert(matches[0]); } I'm trying to match a phone number in many formats. I've worked very hard to figure that regex out and tested it with different sandboxes http://gskinner.com/RegExr/ http://www.pagecolumn.com/tool/regtest.htm Ubot throws an error that doesn't give much explanation, just the text from my code (shown above). Anyone able to shed some light on this?

I agree. As the community grows it'll be more and more important to keep help and reference sections separated. Sections along the lines of script library, help requests, and work requests come to mind.

cmd.exe /c - CMD.EXE is window's dos prompt. The /c switch will append the following command in the same process rmdir /s /q - This is Windows "Delete Directory" command. /s switch is used to delete all sub directories and files while the /q sets the process to quiet mode with no output back to the screen. "%APPDATA%\Macromedia\Flash Player\#SharedObjects\" - %APPDATA% is the system folder for applications, on my machine this points to "C:\Documents and Settings\Customer\Application Data" you can see where yours points by typing %APPDATA% as a run command. The rest of the path points spe

Ahhh... super cookies, flash cookies, local stored object cookies! Cookies cookies and more cookies. Why can't we just delete the cookie jar? http://imgur.com/36N9c.jpg I confirmed delete of cookies through Macromedia's Support/Settings site. Testing with that site comes with a catch tho, it's data is stored in a folder outside of the objects folder. This means once you clear cookies with the function above all the sites will still be listed but the "used" calculations will now show "-". To manually delete the settings references run [cmd.exe /c rmdir /s /q "%APPDATA%\Macromedia\Flash Pl

BotNinja, Check this out http://ubotug.com/index.php?/topic/4162-sudo-code-export-method I made a suggestion for an export function in plain text so I can track changes with SVN and another guy commented that an import function would work for his needs. Sounds like something like this would also work for your needs? If we get support behind this feature perhaps the timetable on implementation can be moved up.

This has been discussed and packed here.. http://ubotug.com/index.php?/topic/2205-library-strings/

select field set #thisBid eval "prompt('Enter the bid amount');" send keys chosen field #thisBid

Just wasn't using the right commands. Here's what worked.. set #cycle True while #cycle = True { set #cycle eval confirm('Continue?'); } Now there seems to be a bug with javascripts breaking after a nav command =\

Setting a count wont work for this instance. I want it to loop forever, and each look request feedback from the user in the form of a Yes/No decision. Is that possible? How do I set a variable from the TRUE/FALSE decision of a javascript confirm() ?

First I tried to choose by attribute with wildcards but that was no good. I could not choose id=*captcha_dialog*style*none I eventually just used select by ID. It works for not, but doesn't work the way it was requested. We'll see if it hold up!

use javascript to enforce that the element has style="visibility:none"

That's a good question. So far I haven't noticed anything but here's where I'm confused... I run the bot in the creation studio and it works fine. Compile it and run in on that same PC, still good. As soon as I take that compiled copy to another PC and run it - the search function doesn't trip and end us breaking the process at this particular stage. Does the search function have dependencies that might be different between the workstations?