Extract URL from Mail message

Question

Hello Applescript-Meisters,
I have the following problem:
Every week we receive an email which contains a secure URL (https://) with always different username and password to download a ".tar" file.
Since I'm new to Applescript I tried to extract the paragraph of the email message with the line which contains the URL, but doesn't seem to work:
tell application "Mail"
set theMessages to message 1 of mailbox "EXAMPLE" of account "AccountTest" whose subject begins with "Data request"
set theContent to content of theMessages
set someData to paragraphs of theContent
repeat with someData in theContent
if someData begins with "https://" then
set theURL to paragraph of someData
else
set theURL to paragraph 18 of theContent -- this works but this line can change weekly
end if
end repeat
end tell
Any help or suggestions would be greatly appreciated.
Cheers,
gilles

Answer 1

The most obvious problem is how you're trying to iterate through the paragraphs.
You start off with:
set theContent to content of theMessages
which is fine - theContent now contains the text of the email.
You then extract the paragraphs of that text:
set someData to paragraphs of theContent
But you then reset 'someData' to be the iterator in a loop:
repeat with someData in theContent
Additionally, you're iterating through theContent, which is a single block of text.
I think what you mean to do is iterate through the paragraphs of theContent, like:
repeat with someData in (get paragraphs of theContent)
Once you have this, you know someData is the paragraph you're currently looking at so you can:
set theURL to someData
Finally, you do not want an 'else ... set theURL' statement because that will reset theURL every time you find a line that doesn't begin with "https://" (e.g. if paragraph 10 starts with https:// you set theURL to that paragraph, but as soon as you move onto line 11 you reset theURL because the paragraph doesn't begin with https://
To address this you should just iterate through the paragraphs and only set theURL to paragraph 18 if you didn't find any matching lines.
Putting it all together:
set theURL to "" -- default value for theURL
theContent to content of theMessage
repeat with someData in (get paragraphs of theContent)
 if someData begins with "http://" then
  set theURL to someData
 end if
end repeat
if theURL = "" then
 try
  set theURL to paragraph 18 of theContent
 end try
end if
Note that I've included the 'paragraph 18' line in a try block. That's so that the script doesn't fail if there are less than 18 paragraphs in the message (you can't get paragraph 18 of a 17-paragraph document).