A BB Code to use og: metas

[darnoldy]

Folks-

I want to use the og: data to display links to external pages. I thought I could make a BB code that uses javaScript to go get the info. It turns out I can't (at least without some help). It has been many ears since I tried to write javascript—and I'm really rusty.

First, I tried:
function getMetas (theURL){
var theTarget = new document()
theTarget.location.href = theURL
alert(theTarget.location.href)
if(theTarget.document.getElementsByName("meta").length>0){
alert("meta tags "&& theTarget.document.getElementsByName("meta").length)
}else{
alert("no meta tags")
}
}
which gave me an error that document is not a constructor.

When I tried:
function getMetas (theURL){
var theTarget = window.open()
theTarget.location.href = theURL
alert(theTarget.location.href)
if(theTarget.document.getElementsByName("meta").length>0){
alert("meta tags "&& theTarget.document.getElementsByName("meta").length)
}else{
alert("no meta tags")
}
}
The script does not give an error message, but the first alert displays the message "about:blank"—I assume because the page hasn't had time to load.

Am I missing something small or am I on a fool's errand? Any helpful hints you can share?

[kh99]

Well, I'm not a javascript expert so I don't know exactly why that isn't working. You're probably right that it's because the document hasn't loaded yet. But I'm not sure that having a user's browser load the page in a separate window every time one of those links is in a post is a good idea. Anyway, I did 5 minutes of research on the subject so I'm obviously not well informed, but it seems like there is a function that will create a new document from a url, but it's only in newer browsers. But without using that, I think you could parse it as XML or even just as a string, instead of loading it into a DOM object.

Doing a search for "open graph protocol parsers" I notice that there are parsers for php, so you might be able to figure out a server-side solution. Maybe when the post is being saved check it for one of those bbcodes and if it's found, parse the document for the og data and do a replacement. But I'm not sure what would happen if the user wanted to edit the post later.

[darnoldy]

But I'm not sure that having a user's browser load the page in a separate window every time one of those links is in a post is a good idea.Yeh, that would be a bad idea! My thought was that the script would run at the time the message was created, and then the proper data would become part of the message (like how the quote tag works). Having the target page open in a new window while the script is running (then close the window) is not ideal?it was just how I could get it to work with my limited skill.

but it seems like there is a function that will create a new document from a url, but it's only in newer browsers. But without using that, I think you could parse it as XML or even just as a string, instead of loading it into a DOM object.hmm..have to do more research.

I notice that there are parsers for php, so you might be able to figure out a server-side solution.I was trying to avoid php?mostly because I am even less familiar with that than javascript.

Thaks for your help.