Net66

Use 'Wait For' which waits for something you know will be on the page it redirects to. For some reason 'wait for' does not appear in the list of flow commands or in the script reference but it is there if you search for it. Andy

You are using scrape selected attribute but not choosing anything by attribute first. Attached is same bot that selects the input field first and it works. Andy simple_scrape_fixed.ubot

Very nice bot John - good work. Andy

OK this topic comes up a lot and I have a fairly decent solution that I am using for myself. My solution combines hardware locking with 'phoning home' and stops the easy recreation of a page and a hosts file fiddle. It also stops the potential for sharing a key. At the moment it is not in a user-friendly format, I had no intention of releasing it so I didn't make install routines or anything pretty. But maybe it is worth my time to tidy it and package it with a help file or video and sell it. At the moment it integrates with Clickbank for payment processing and once they pay and downloa

You probably want to open a support ticket for this.

I don't know enough about proxies to give you any suggestions at all. I can tell you that I currently pay $29.95 a month for 10 private (and uber-fast) proxies and I can choose locations (from a limited range) and change them once a month. I am using these people for my proxies. Andy

Can you give a link to the page you are trying to work with? Andy

I don't know about genius lol. Anyhoo.. here's a neat trick I figured out... shell cmd.exe /c dir {1} /b >{2}\filelist.txt {1} is variable with file path you want to get directory listing for {2} document folder If you do that and check in the documents folder and open filelist.txt you will see it contains a list of files in the folder you asked for. You can read that file into a list and do what you like. You could of course do cmd.exe /c dir c:\specificpath\myfile.txt /b >c:\documents\files.txt Which looks for a specific file. Read the text file into a variabl

Read the file into a list then go through each line of this list and add it to a new list using "list from text" with "," as the delimeter. That way you it doesn't matter that there is no comma on the end of a line. Andy

You can store it in csv or plain text which ever is more suited to the particular task. It is also possible with the shell command to grab a directory listing (use dir with the /b switch to get a bare format listing) of a chosen folder, you can then read that list of file names into a ubot list and process it that way (I do this in a bot I wrote for blogging). Spin works for blocks of text. Andy

Check your email John :-) Andy

Sure - andy [at] netodex.com I got a manic afternoon though but will try to look at it later today. Andy

Yes - no problem at all. Ubot comes with a moneyback guarantee... take the plunge and try it for yourself, I'd be 99.9% sure you'll keep it! Andy

You are comparing apples to oranges here. A script that checks a username and password is valid is easier to create and simple to get around. I could buy a script that uses this and simply pass it around to my friends with a note of my user details. A hardware script locks the licence to one physical computer and while it is still possible to hack/cheat it makes it much harder and stops all the novice/numptie/opportunist copiers. As you put it - "simple". Sadly simple to use/implement also means simple to rip off/break/bypass/crack. A programmer friend of mine protected his app with a

I am not a twitter user so I've not looked at how they return results but I can give you an idea from a script I did for something else. In the site I was scraping there was a 'next' button for as long as there was a next page of results. On the last page of results this next button was not there. So I did a while loop constructed something like this.. Set #nonext = 0 Navigate to first page of results While #nonext = 0 scrape stuff if searchpage does not find "next button" then set #nonext=1 else navigate to next page (end of while loop) Andy

Go to the tutorials page http://ubotstudio.com/tutorials.aspx and under the section titled "Skills Series" there is a video called "creating user interface". Check that out first. Also check out the excellent tutorials on Aaron Nimocks site http://www.ubottutorials.com/category/tutorials - while the tut may not specifically be about ui you will see him use ui stuff extensively in his bots. Andy

Cool - the useragent thing will make a big difference. I was chatting to a good friend at the weekend who does the security to stop people scraping data from a big marketing site. He says it is a losing battle and things like capatchas are just making the web more un-user friendly because bots soon get round them (as well we all know) and they just get in the way of normal users. His companies approach is to simply look for repeated hits in a short space of time from an IP or range of IPs, but they have to take into account user agent because they have had universities legitimately reques

Neat trick - thanks for sharing. Andy

Ubot can handle straight text and also csv files. So I would save from excel into comma delimited format. You can then load the entire file into a "list". Once in a list you can go through sequencially OR if you want a specific entry like the 15th you can access that (although list positions start at zero so for the 15th entry you'd access list position 14). So in a nutshell - yes you can do what you are asking and it's really pretty simple for ubot. Andy

If you are selecting those inputs by attribute (the most common method) and they move them but don't change the name/ids of those then yes you won't have to edit your bot at all. The nice thing about ubot is that if a web page does radically change it really is probably only a few minutes work to get your bot up to date. Andy

Good tip! Andy

You could have just edited out your login details and uploaded a script without them in. But in the absense of that I gave this a quick go and it worked. Select by attribute, click chosen works just fine for the login button. See attached:digglogin.ubot Andy

For the same money from the guys I use you only get 10 but they are private to you and they are FAST. By rotating them while scraping with ubot I get passed all the 'you have made too many requests in a short space of time' blocks. I have a bot that scrapes from ezine articles. If you try doing that with no proxy it soon catches on (even if you put delays and stuff in) but with just 10 proxies and switching randomly I don't get caught at all. So while I see the appeal of 600+ proxies for cheap I don't think you need that many and my experience of shared proxies is that you can hit a site a

Ubot Studio will get the job done fast. To do what ubot does in macro languages would take a lot more scripting and know-how. Ubot is designed to automate web tasks and it does it very well indeed. There is a slight learning curve but if you watch the various videos and give it a go you soon pick it up. And the users in this forum are pretty helpful if you get stuck too. Andy

I use private proxies and they work all the time without subwindows. But mine don't require username and password credentials (I get them from yourprivateproxy). Where are you getting your proxies from? Andy