Pinterest urlencode

[niccol]

Yes, I understand the process of rendering. That is not the issue. But thanks for the clarification.

I understand about html entities. That is not the point either.



The point is about how pinterest parses the 'url' get parameter in their links. Not the URL the 'url' get parameter.

That is the main point.

But secondly, I am interested in how the search engines actually use the canonical link. If it gets rendered to the screen as in SERPs then obviously there is no issue. Because the encoding is dealt with by the browser. But is this really the only way the canonical link gets used? What the canonical link says is that all pages in that group can be reached through that particular URL. If I make a http request to that page ( by whatever means ) I do not get the html returned that I am expecting.

So, if a search engine makes an http request to that URL then it doesn't get the response from the server that I want it to.





@Melanie
check http://shop~~bloved.com/index.php?ma...roducts_id=171 and see if that pins ok for you . Just remove the ~~, It works fine for me and has been tested by the client. Let me know.

[mprough]

Indeed, but I want to use a custom button =)... but none the less you lead me in the right direction... we actually need it encoded in the source to post.

So I use NUMINIX's Pinterest script to drive the image grabbing and url grabbing....

so this

<?php echo urlencode($piniturl); ?>

becomes this

<?php echo urlencode(str_replace('&amp;','&',$piniturl)); ?>

and now it works

https://outdoorflagstore.com/index.p...roducts_id=685

In my reading I think there is a way using rawurlencode as well... but working is working and it's Miller time here

[niccol]

I think that may have been what I said in post#8

[mprough]

I think that may have been what I said in post#8

yep, but it does not work with $canonicalLink

[niccol]

here's my secondary worry.

Imagine I am a search engine. There are lots of pages I know are equivalent. They all have the same canonical tag. So after a while I decide to see what is in them. The obvious thing is to send a http request to the canonical url, right? Because that is the common URL for all these pages. That is the one I am being told I should index and not the others. I am not a browser, I am a search engine, so I just send the request. I don't render anything at any point.

But if I do that then I get a response that is the 'product not found' page. So as a good search engine I index that page. Which has got none of the relevant information on it.

Please, someone tell me that this logic is wrong ....

(Try it with curl if you want to see what I mean.)

[niccol]

SO, did you try the link that I sent to the shopBloved site? That works fine for me. I'd be interested to know if it doesn't work for you.

[mprough]

SO, did you try the link that I sent to the shopBloved site? That works fine for me. I'd be interested to know if it doesn't work for you.

Yes, "indeed" in post 15.... But it does not work for me with $canonicalLink. I tried several sites 1.5.0 & 1.5.1 without any luck

[niccol]

I think we will put that down to weirdness in that case. The code I posted is exactly what I am using.

How odd :-) As you say. If it works it works. Miller time.

[mprough]

I think we will put that down to weirdness in that case. The code I posted is exactly what I am using.

How odd :-) As you say. If it works it works. Miller time.

Yep Miller Time =)

[lhungil]

Yup, that appears to be working! Perhaps form validation was failing because I was using a dummy domain / image when testing ;)