When using the above script in a macro it often returns a large amount of text. Is there any way to keep the first 10 or so lines of text in a variable and discard the rest?
Currently I'm writing a temp file to my home directory and using a sed command. It works but I'm just wondering if there's a better way?
tell application "Safari"
tell front document
return paragraphs 1 thru 10 of (get source)
end tell
end tell
That'll normally just be metadata and maybe some scripting stuff though -- and, of course, will depend on how the raw HTML is laid out (if someone's written it without line breaks you're in for a surprise!). Are you after something in particular?
Well I have a macro for subscribing to various RSS feeds. Normally I use the Switch/Case action based on the URL to proceed but needed a way to determine if the site is a Discourse forum or not. This works a treat by just grabbing the metadata at the beginning of the page without saving the entire source. Thank you @Nige_S!
Footnote – for more focused extractions from HTML, the XQuery 1.0 syntax is simple and powerful.
(you can use it from AppleScript and JavaScript for Automation as well as from browser JavaScript)