Return Infomation from Web Viewer?

[Salesman0Gil]

Is there a way to retunr information from the Web Viewer?

I just read this article: http://www.briandunning.com/wv/12

It gives you the Latitude and Longitude for address.

Is there a way to grab the info from Latitude and Longitude.

[Genx]

http://www.filemaker.com/help/FunctionsRef-38.html

[mr_vodka]

Do a search on these forums for "scraping". There should be a few threads on here regarding using the technique of web scraping via the function Genx has posted.

[Salesman0Gil]

Thanks. After much reading and tinkering. I finally got it.

Scraping! Wow! It really has opened up a bunch of new doors.

[Robert Collins]

I have read this document several times and checked in the manual.

I need a field to display the current url - I have named the web viewer object as "webviewer1"

but I can't get my calculation right - this is what I put:

GetLayoutObjectAttribute(webviewer1;Source)

can anyone tell me what I'm doing wrong -

Thanks

[mr_vodka]

GetLayoutObjectAttribute( "webviewer1"; "Source" ) in an unstored calc should do it.

[Robert Collins]

Thanks, that did work but the field doesn't update when I go to a different Url unless I click on the field that returns the Url - can I set this to update automatically?

thanks again

[aholtzapfel]

Is there any way to tell within a script if web page in the viewer is fully loaded?

[fabriceN]

Yes,

Patterncount ( GetLayoutObjectAttribute ( "webViewerName" ; "content" ) ; "

[aholtzapfel]

Thank you!

[aholtzapfel]

Here is an example of "scraping" a web page for information. This file takes an address, city, and state and returns address, city, st, zip, +4Zip, and County in the USPS Standard Address Format. It uses USPS.com loaded in a web viewer to do it.

Address_Check__zipped__Folder.zip

[lance123]

Hi To All,

I am using this technique to good effect and it's great!

One question I have is on how to detect a 'timeout'. This happens when the web engine tries to get info from the site and it takes more than one minute. This can happen simply because the site is running slow. I am dealing with it but it is set to a timed count of the loops which will add to more than that time.

If someone has a way, I would appreciate the tip.

Thanks,

Lance

[aholtzapfel]

in the lookup script, in the loop create a counter

set variable [$X , 0]

Loop

check for web page loading...

Exit loop if [loaded or $X > Number of secs for time out]

pause[1 sec]

end loop

If[ $X > Number of secs for time out]

Show Message[ It timed out!]

end if

or somthing like that.

Edited June 11, 2007 by Guest

[lance123]

Hi aholtzapfel,

Yes, that is pretty much what I am doing. It just seems that it is a bit clunky not being able to detect the state. I was hoping for a 'sure thing'.

Thanks,

Lance

[ConnieHull]

This sounds really like what I am looking for... One problem when I unzip it and open it I get an error "Unknown Script step" I've attached a screen shot of the script if anyone can tell me what is supposed to go here I would greatly appreciate it.

Is it possible that since I'm using 8 that this is unavailable to me?

Picture_1.pdf

[aholtzapfel]

Is it possible that since I'm using 8 that this is unavailable to me?

Yes, the example uses filemaker's webviewer (Availible in ver 8.5 and above)

Before ver8.5 came out I was doing this with a VBS script (would need to use applescript on a mac) and there are advantages to using an OS level script to scrape a page but it is alot more involved than using the web view.