Need some alpha testers for my hosted browser automation/scraping system

NathanRidley

New member
Jul 26, 2008
189
1
0
London, UK
In 2012 I periodically spoke about what I'm working on in this thread. Now, I'm planning on turning on an early version for a few people to try out, as some feedback and real world testing is needed. I can't charge for this if it crashes, chews up too much memory, behaves strangely or craps out some other way, and if it has bugs, I'm never going to find them all myself. I need a few others using the system.

The way it works is you get an account with an API key and you control your browser through HTTP calls. It's currently running on Chrome somewhere around version 18, but this will be updated in a couple of months when the next version of the underlying tech is released.

Currently the following functions are supported, but I can (and will be) adding more as needed:

  • Open Window
  • Load URL
  • Take Screenshot
  • Click Element (real low level click event)
  • Type Text (real up/down keypress events with realistic delays)
  • Wait For Text (to allow for unpredictable ajax page wait times)
  • Set Field Value (text, checkbox, radio, select)
  • Resize Window
  • Get Window Info
  • Extract Content (ultra powerful/flexible data/content extraction/transformation)

This is *alpha* software. Nobody has used it but me and there's plenty left for me, but I need some people using it. It has full proxy support and I'd prefer you use exclusive/private proxies, not open/public proxies you scraped from somewhere. I'll allow you to not use proxies if you're not doing any "dirty" scraping (CraigsList, Google, etc), but otherwise, proxies are required. If you are using proper exclusive proxies which don't reveal my server IP address, you can scrape whatever the hell you want and I won't give you any grief about it, as long as you're not costing me a bunch of money.

Finally, I'm not going to automatically accept everyone who asks. It would help if I've seen you around on the forum, or if you seem like you legitimately have use for this. It's a waste of my time to issue accounts that never get used, so only ask if you think this is something you'd actually use, or would spend enough time alpha testing to get me some feedback/usage data.

Finally, the scraping is awesome. I have a few finishing touches to put on it, but it essentially allows you to project anything on the page back into your own customised JSON structure, returned with the HTTP response.

Let me know! If I don't know you and you have no established credibility on WF, also let me know why I should pick you.

p.s. Mods; I'm not using this to get email addresses from WF users; I really just need testers. We'll keep feedback in this thread so it stays on WF. When I'm ready to charge, I'll pay for an ad or thread on WF.
 


Hmm I just realised there's a potentially confusing line in what I wrote. To clarify, this is a fully-hosted SaaS browser automation solution. There is nothing you need to do other than issue HTTP requests and get back results. No software to install, etc.

Nice. Available IP's on the backend? I'm interested.

Do you mean do *I* have any available proxies? I don't, but you can get them from PacketFlip.com or other vendors such as IPRental.com. You basically specify your proxy address and credentials (if any) when starting a new session and it gets used for all requests. Let me know if you'd like an account and I'll set you up and finish up some documentation quickly.
 
To those who've expressed interest, bear with me briefly while I finish up a bit of account self-creation code, then you can sign yourselves up at your leisure using an HTTP request. Also going to finish up some documentation. I hear APIs are hard to use if you don't know what requests to send!

Or proxy51.com!

Naturally! ;)
 
I'm definitely down for this... I need to scrape the shit out of some content for a test I'm doing.... Already got 20 private proxies paid for too!
 
Wow, writing documentation is time consuming :p
Oh well, here's what I've done so far, more coming:

Api Documentation « Systemizer API

I'll update this thread as I add more documentation. Right now we have the API overview, account creation, and some session-related calls. I'll be adding more in the morning when I get up, but I'm off to bed right now...
 
If you want a python API client, I can whip one up (no charge). It'll make me properly understand the system.
 
If you want a python API client, I can whip one up (no charge). It'll make me properly understand the system.

Awesome, thanks! When you do, stick it on GitHub and I'll link it up and credit you :)

What did you use for the design/theme/styling of your docs? Is that custom or something preexisting?

'Tis sexay

Thanks, I did it myself, though the layout is inspired by the Stripe API docs.
 
he hosts all the browser instances for you

Yeah basically no server overhead on your part, also my startup/teardown of individual sessions is a shitload quicker (and lighter) than Selenium. As long as I have idle workers, session start time is near instant once requested.
 
I have done a fair bit of scraping in my day and would be interested in giving you some freed back.

Feel free to hit me up on skype or pm me.

Yeah cool, well I am expecting to have it running properly over the weekend as I've almost finished all of the main documentation required to use the browser automation part of the product. It's taken the last week or so, but hopefully it will mean everything is clear and understandable so that there's less frustration when trying to the system out.
 
I don't have much rep on the boards since i'm very new here but i'm willing to alpha test if you need some extra people.

Thanks, actually I've defined the account limitations in a safe enough way that I'm not going to restrict who signs up, so have look here over the next couple of days and you'll be able to sign yourself up and give it a go.