webautomationlab

See if you can select the tag above the input tag. Then you can use a wildcard on the OUTERHTML which *should* be unique from the other input fields.

What happened here, and it happens often, is that on each page load, if you are trying to trigger JS or select elements, UBot may outpace the loading of the page, which results in commands getting skipped. I *always* add a 30 second wait finish, followed by a 2 second delay. If the script runs well with them, then I may experiment with taking them out, if speeding up the script is necessary (say, I loop 100 or 200 times). In those cases, an extra 3 or 4 seconds per loop can add up. But for short routines, adding another 4 or 6 seconds to a script, or 3 or 4 seconds for each page load, isn

Without an URL, this is hard to troubleshoot, since you haven't shown us the complete code, and your description of what is going on is sort of disjointed. If you don't want to show the URL publicly, PM it to me and I will take a look tomorrow.

I don't believe so.

Chrome/Safari do not support many javascript-based rich editors.

For anyone wondering, the solution to this was STEP ONE Page Scrape the button <input id="comment_submit_840378" class="button" value="COMMENT" Red left side Blue right side Store that in a variable. Set Var #UniqueID => Page Scrape (as above) STEP TWO Run the javascript in the button (onmouseup) Run Javascript comment_controller_{1}.postComment('/comment/create?layout=off'); Where {1} is #UniqueID

Look in the user options on those sites, you may be able to switch from a rich text editor to a plain text editor.

Javascript. Not Java. Can you scrape this LEFT SIDE <input id="comment_submit_ RIGHT SIDE " class="button" value="COMMENT" onmouseover="this.className='button button-r'" onmouseout="this.className='button'" onmouseup="comment_controller_ To get the numerical value? Then store it in a variable, and then use that to target the button? If you don't want to publicly ost the site, if you PM me an URL, I'd be happy to take a quick look for you.

If Eval Current List Item = END then STOP or something like that.

If you want a ubot which can access that page, fill in the form, submit it, then scrape the results in text posted at the bottom, yes. If you want something that does not work off that page, you will have to provide clearer details.

I don't need it now, since I encountered these issues last year and have come up with workarounds like disabling stuff in IE and other software, but when I need it, I am definitely going to buy it. It's a great idea to pre-process. UBot has added the ability to download which I have not fully tested yet, but may provide similar functionality to what W/A does. @OP, I'm curious how Speedbot works with rendering xml (or non-HTML) to the screen. If you watch his video, he is already talking about doing that down the line.

That is totally awesome. It solves a major issue with browser rendering a lot of urls in succession in IE. You should definitely offer to disable external CSS, which would solve the background images issue, and for the purposes of scraping, unstyled HTML is still fine. I am very impressed.

Do a search on here for Yahoo. I remember people having issues with this or something similar at Yahoo a couple months ago. If you're still stuck after that, let me know and I will try to give you a hand.

Get a copy of Expresso http://www.ultrapico.com/Expresso.htm

Unfortunately no, UBot only works in IE.

No problem. Pro tip. I wish I knew it a year ago, because I really struggled with tinymce and iMacros.

That's tinymce. It is a javascript based editor. You can probably search the forum for help with tinymce. However, in my experience it is easier (though not always possible) to edit your account preferences on a particular site, and choose not to use ENHANCED or RICH TEXT editors, and choose a PLAIN TEXT editor instead which would allow you to post straight html. Since I am not very familiar with tumblr, I don't know if that option is available. If you can't find it, let me know and I will look.

Should be doable. The best way to do it is either two scripts where Script 1 scrapes all of the expired listings then Script 2 goes to each expired listing scraped and grabs what you want. Or you could put it all in one big script. Depends how long it will take to run, how automated you need it to be etc. Either way, you can fit it in one bot. Scripts are subprograms within a UBot.

Is the list of words constant for each batch of sites, or does the list change, site to site? Either way, it is doable. The issue I have with doing anything with browsing large lists of URLs is that UBot crushes itself loading big pages in succession. If we could download the pages, and parse the page code without rendering it (as WinAutomation does) it would make things a lot more efficient. In other words, when we want to work at the HTML level (scraping, parsing), we don't need to render the HTML in the browser. Subsequent operations might require rendering, to interact with elements

Action Commands => Run Javascript

lol

Please provide more detail of the screen output of the password and/or the blog system. Right now your query is so vague, no one can offer any sort of authoritative answer.

+rep

http://ubotstudio.com/forum/index.php?/topic/2672-auto-start/page__hl__auto__fromsearch__1

Greencat is awesome and helpful! Thank you greencat!