Jump to content
UBot Underground

Net66

Fellow UBotter
  • Content Count

    559
  • Joined

  • Last visited

  • Days Won

    17

Posts posted by Net66

  1. Okay I think I have found a solution that I can live with for awhile.

     

    What I realized is that I needed to do the replacement on each URL string individually rather than attempting to do it via the write to file piece of my list generation.

     

    So when I load my list items that is the point that I do the search & replace.

     

    It's two extra steps but it does solve the problem.

     

    I still think at the point of scraping the integrity of the URL string should be left intact. Maybe that can be incorporated into a future version.

     

    Thanks everyone for the comments.

     

    Buddy

     

    I think that is as good a work-around as any. I like your choice of variable names in the example :-)

     

    Andy

  2. I gave up on uBot a few months ago as it was to hard to work out how to make simple bots. After reading a few php books and learning the basics I am back again with motivation to make some usefull bots.

     

    I need to know how to layout my bot properly.

    Instead of having every task in one big list, how do I seperate them as seperate scripts in the same bot.

    e.g. script 1. log in to gmail, script 2. Search for activation email, Script3. Click on activation link etc.

     

    Im sure this is an easy fix.

     

    There are some useful pointers in this thread too http://ubotstudio.com/forum/index.php?/topic/2944-learning-the-right-way-to-create-bots/

     

    Andy

  3. Well, if you really want an opinion: In my experience nag screens make me crack the application in question, sheerly out of annoyance :P But I don't think that's what you're looking for. Personally, I'm against nag screens in all shapes and forms. I think a note along with the download link about possible upgrades, or in the readme, is enough, and anything else is just needlessly annoying the user and detracting from the user's experience.

     

    If you wanted to avoid nag screens then depending on what your bot does you could just make a free 'lite' version that has a few limitations and people would pay to get the full leaded version. The lite version could also navigate to the order page on startup which would spell out the key advantages to paying you for the upgrade.

     

    Andy

  4. Hi there,

     

    I am trying to build a bot that will automatically click a thumbs up to various articles from my website.

     

    Basically my website is amoungst other competitors articles. I only want to give a positive vote to my article.

     

    <A href="LINKTOVOTEFORME">THUMBSUP</A> 
    <DIV class=article>
    <H3><SPAN><A href="http://www.REDIRECTLINK.com">NAMEOFMYARTICLE</A> </SPAN></H3>
    <P><STRONG>From:</STRONG> <DIV>mywebsite.com</DIV></p>
    

     

    The title of my article changes, the only way i can tell that it is my website is where it says <DIV>mywebsite.com</DIV>

     

    So basicly, I want to select the link to vote, but only where the containing area where <DIV>mywebsite.com</DIV>

     

     

    Any ideas?

     

     

    Cheers!

     

    Build two lists, one that contains the vote links and the other that contains the site name from the div tags. Now loop through the list of site names and whenever the list item contains your site name the same position in the other list will contain the vote link, add this to a new list whenever the condition is met.

     

    Loop through the new list navigating to each link.

     

    Andy

  5. Andy I really do appreciate your efforts! It still is not working. Makes me wonder if it is my setup.

     

    I am running the newest ubot version. I attached the error I am getting. I am glad they fixed the wrapping of the error message.

     

    Buddy

     

    I am not using the latest beta because the lack of insert node and a few other things were slowing me up too much so I rolled back to the current stable release - I don't know if this makes any difference.

     

    Andy

  6. I've had Ubot a week now (so I am still a newbie myself). As I come from a programming and web background it has been a little easier to get my head around what ubot does and how it does it. Having said that I did spend a few hours failing miserably at a fairly simple scrape task on my first day because I had misunderstood something.

     

    I thought I'd write up a few tips for learning ubot which might help newbies avoid some of the frustrations that learning something like this can bring.

     

    1) Watch ALL the tutorial videos you can. It becomes tempting to skip over videos because you already think you know how to do xyz but it is amazing what little things you pick up that the tutorial may not have set out to teach but get covered anyway.

     

    The training videos by Seth that you are linked to in the email when you purchase, the super videos at Aaron Nimocks site http://www.ubottutorials.com and whatever else you can find. I appreciate that many of these videos run to 45+ mins but its a wise time investment to watch even the videos that cover things you don't think you need to know.

     

    2) Always remember "there's more than one way to skin a cat". If something really isn't working as you thought see if you can come up with an alternative method and try that. I find it useful to open the site I am working with in an ordinary browser and see how it behaves, etc. Often right clicking and opening in a new window in a regular browser will give you some useful clues about where something is navigating to.

     

    3) Make your first bot attempts fairly simple. My first full bot scraped a friend list on a site that had very basic navigation (no ajax, etc) and then used the private message system to message each of them with a random pause between messages.

     

    4) Download and look at as many sample bits of bot code as you can. Looking at bots that have been written by experienced ubotters will give you ideas on the best ways to lay out a bot. Aaron Nimocks bots are nicely structured and kept looking tidy by splitting parts of the bot into subs.

     

    5) Keep notes! Whenever you come across a good part in a video or a nice example piece of code make a note of where you saw it, where in the video it was, etc. This makes referencing back to things much quicker.

     

    6) If you are learning ubot for the first time I'd stick with the official stable release rather than the beta so you don't get stuck because of a bug. Also you won't end up reporting things as bugs that are really user error from being so new to ubot.

     

    7) If you really can't solve something try taking a break. So many times I have spent hours looking for a problem in a program or script, finally taken a break, come back and spotted the problem right away. This doesn't just apply to ubot its good practise with any kind of scripting or programming.

     

    Above all... have fun with it and don't get discouraged if it takes you a while to figure this out - stick with it and you'll soon be turning out killer bots in record time.

     

    I hope this helps someone.

     

    Andy

    • Like 4
  7. Andy,

     

    Is this working for you? I followed your instructions but no go. The same error that I reported earlier still pops up.

     

    Buddy

    Yes it works for me. I was going to try to find time to see if I could scrape from the pop up window on the view all link but been a bit busy. Facebook is a nightmare to scrape from!

  8. I see a next page triangle.

     

    Sorry, should have been clearer!

     

    Login to Facebook. Account > Edit Friends.

    Left hand tab under "Lists" click "Friends"

     

    This loads the friends list on a main page rather than a pop up window.

     

    Run the bot sample.

     

    Andy

  9. Has anyone had any luck at scraping Facebook profiles from the friends section?

     

    I can scrape the first page but I cannot get it to advance to the next page, scrape and then advance until it is done.

     

    I end up having to page through it manually and executing my bot without initializing my list so I don't wipe it out.

     

    Any ideas?

     

    thanks!

     

    The attachment should help. You have to be on the first list of friends page. It scrapes the url of each friend and adds them to a friends list. Basically it is checking if the current url has changed or not after it has clicked the advance button. Because facebook keeps the button and its attributes present (it just greys it out) if it clicks it and the url does not change we can assume we were on the last page.

     

    You can then do a loop through each url in the friends list and scrape whatever data you require. For the sake of the demo this just runs through, builds the list and writes it out to a text file.

     

    Hope this helps

     

    Andy

    scrapefriends.ubot

  10. As for neobooks, UBot Studio was never intended to be an application development tool. The bot compilation system was intended to be a convenient way to use bots. I'm not saying that we won't expand it to be more friendly to bot resellers, but as of right now, that's not its intended purpose, as it is with neobooks.

     

    OK, I appreciate that. But if that is really the case why make such a big deal out of the compile feature on the sales page? Where it details the compile option as a wow feature and talks about its use for distribution and selling.

     

    If it isn't it's intended purpose then it maybe shouldn't be a key point?

     

    " #3 - Bot Compiler: Turn Your Bots into Independent Software You can Distribute or Sell."

     

    Having said that let me be clear - I love ubot and am already turning out some good stuff with it after only a few days of use.

     

    Best wishes,

     

    Andy

  11. Understand that when you sell a UBot compiled bot (which I assume is the reason you're upset about the branding), you're selling them about 90% UBot technology and 10% your own work in scripting. Having the powered by link isn't really too much to ask for.

     

    I can live with the link (not that I have a choice) but I think that's a bit of a weak arguement...

     

    Maybe Microsoft should force people to have "written using Microsoft Word" printed at the foot of every document? After all when you write a document it is 90% their technology that allows you to write it and 10% what you actually write? My Nokia phone doesn't announce to the person I text or phone that the call was made possible by Nokia, yet what I say or type is surely a tiny percentage compared to their technology that makes it possible.

     

    I own Neobooks excellent rapid application builder which I use to write and compile database apps. It sells for a similar price and they don't have any reference to the tool that built the product in the compiled version, yet a vast number of people who develop with it put 'Built with Neosoft' messages and buttons in their apps - and they don't even have an affiliate program!

     

    Given the choice, since I know I get an affiliate commission if they follow the link I'd probably leave it turned on anyway. It would just be nice to have the choice via a tick box or something.

     

    Just my 2 pennys worth.

     

    Regards,

     

    Andy

    • Like 1
×
×
  • Create New...