Get the address behind a link in email

Question

Hey all
I have a problem I can't seem to work out.
One of the email systems that I have to use automatically adds the word "blocked" to any html link in an email. So for example, if you get an email from Expedia with a link to your itinerary that says "Click here for your trip details", it will change the underlying link address from "http://expedia.com/" to "blockedhttp://expedia.com".
To get around this, you have to manually copy the link, and paste it into the browser, then remove the "blocked". Kind of a hassle.
I want to script this process as a Service. I've written Services before, and have experience in AS, but I can't figure out how to do it. Parsing the link and sending to my browser isn't a problem. The problem is that I haven't found a way to get Mail to pass me the value of the LINK instead of the TEXT.
I've even tried using Automator and telling it (as a Service) "Service receives selected URLs in Any Application" - but of course it doesn't recognize "blockedhttp" as a valid URL so it gets ignored.
Mail has the right-click "copy link" function - if I could figure out how to script that, I could grab the link of the clipboard and parse it. Alas, no dice there either.
Anyone have any clever thoughts about how I can extract the link from that href in Mail, or execute the "Copy Link" function?
Message was edited by: stephen.bradley

Answer 1

Well, for anyone in the future who might have a situation like this to solve, here's the solution. At least, the only solution I could come up with, I'm sure a more brilliant programmer would do something different.
There seemed to be no way of grabbing the embedded link from the email text and getting it into a custom Service with a simple right-click. The suggestion to parse the file as raw source was interesting, but it would have meant parsing all of the sometimes several dozen emails I get everyday, looking for the few with links.
So, I went a different way. I wrote a script to handle the "on open location" event, and had it parse out the 'blocked' portion of the link, then fed that cleaned link back to the Finder to open in my default browser.
Save the Applescript as an application so it produced a bundle, then went into the bundle's info.plist file and set it up as the default handler for protocol 'blockedhttp' and 'blockedhttps'. It was added to LaunchServices (which OS X took care of for me). Now clicking a link formatted that way causes my AS app to be called as the handler, it does its simple little dance and passes it on.
Truth is the solution was simple, once I looked at it differently.
Hope that helps someone in the future!
on run
display dialog ¬
"This script applet will respond to webpage links beginning with:"¬
& return & return & ¬
"blockedhttp(s)://" buttons {"OK"} ¬
default button 1 with title "AKO Blockage Handler" with icon 1
tell application "Finder" to update (path to me)
end run
on open location this_url
set theLength to length of this_url
set newURL to (characters 8 through theLength) of this_url as string
tell application "Finder"
open location newURL
end tell
end open location
To make the info.plist file handle my new "protocol" I just added a couple of keys:
<key>CFBundleURLTypes</key>
<array>
<dict>
<key>CFBundleURLName</key>
<string>BlockedURL Helper</string>
<key>CFBundleURLSchemes</key>
<array>
<string>blockedhttp</string>
<string>blockedhttps</string>
</array>
</dict>
</array>
The CFBunderURLName is just a friendly name for the protocol, and the URLSchemes defines the protocol identifier.
Message was edited by: stephen.bradley
Message was edited by: stephen.bradley