View Full Version : Http_Post


Louis Nickerson
04-25-2008, 09:15 PM
I hope I'm heading inthe right direction here.

The Goal...

Take a company name from my database and search the Florida Division of Corportations for that company.

Upon finding that company, collect the names of the officers of the corporation from the site and populate the appropriate fields within the customer record.

If this info could be accessed directly through the URL, I believe I would have "most" of the battle won. Unfortunately, I can't see any way to query their servers through the URL.

The Approach...(I think)

The webpage located at this address http://www.sunbiz.org/corinam.html is the first stop along the way. This is a form where you can enter a company name and click a submit button which will return a new page with a list of likely matches.

I think if I can get this much accomplished, then I will be able to muddle my way through most of the rest of the process. (Submitting the rest of the required info to the next page, parsing the officer info from the resulting page and writing the data to the table).


If anyone has any experience with HTTP_Post() and would be willinig to take a look at this for me, it would be most appreciated.

Kindest Regards,


04-25-2008, 10:35 PM
I think you will find the following thread helpful


Looking at the web page

<INPUT id=entity_search maxLength=192 size=65 name=names_comp_name>

I think you will be able to use getelementbyid to fill the form.

If not try this


This should get you started.

Louis Nickerson
04-25-2008, 10:45 PM
Thanks John...I'll look into your suggestions.


04-26-2008, 01:16 AM
You might find this thread interesting as well.


Steve Wood
04-26-2008, 04:37 AM
http_post can send parameters to the target page and return the contents of the screen as a string, which you can then parse. The problem with this particular website is you have to either be very specific with the parameters to get the proper company (out of a sea of similar ones). If you had the exact parameters, the target page is about three pages deep after that first search page (which you would bypass). If you don't have exact parameters you are going to get back a bunch of irrelevent information. I can't see how you are going to automatically parse that information, find what you want, and fire another more targeted http_post request. What you get back in the http_post is called "body" and is the complete html of the page as if you were looking at it. Then you've got to parse the results; I use extract_string() often in this process.

The other issue, is that their copyright probably won't let you do this. I once created a similar process to extract info from the ABA and after a thousand hits or so, my http_post just returned a screen that said "knock it off or we will prosecute you for copyright violation". Their server had some means to track hits by IP address over a given period of time.

Louis Nickerson
04-26-2008, 06:55 AM

Thank you for responding.

Your explaination of "How" to accomplish this is right in line with what I expected to have to do. My problem is that I have limited HTML knowledge/experience and don't understand how to use the HTTP_Post() function to accomplish the goal.

As for the site's Copyright, I wouldn't knowingly violate a copyright and I suppose I should look into this, but my initial thoughts are that this is a Government owned website. All of the data is public information and anyone is welcome to use the site and assocoated data at will.

The state will even provide downloadable files or even CD's with all of this data at no charge.

If there is any way you could see your way clear to whip up an example of how to make this work for me, it would save me a ton of head scratching and frustration. I've read the help files, searched the forum and even tried to take the Amazon.com (http://www.amazon.com/gp/redirect.html?ie=UTF8&location=http%3A%2F%2Fwww.amazon.com%2F&tag=131369-20&linkCode=ur2&camp=1789&creative=390957) book search example from the help text and modify it to accommodate my needs, but I'm not having much luck.


Steve Wood
04-26-2008, 01:14 PM
Getting the content is pretty easy. The following will return the response from that leading page:

txt = http_post("http://www.sunbiz.org/corinam.html","").body

If you run that you will see that what you get from http_post is not the same as if you actually went to the page and ran a search.

But I ran a quick test and you will not be able to submit a company name for the search using the http_post method. Not unless you go about three pages deep and submit an exact query. And you won't know what to submit.

I think http_post is not going to get you anything useful and your options are to use a macro language to mimic what someone would actually physically type in while at their website or some way to simulate pressing their submit button.

Louis Nickerson
04-30-2008, 08:32 PM

As always, thank you for taking a look at this.

I will investigate other approaches unless someone else can suggest a solution.

Kindest Regards,