filklore

Same problem here.

I wondered if it could be anything to do with what Permalink settings we use. I use “month and name” (i.e. https://<site>/wordpress/2016/10/sample-post/). Do you two with the same problem use the same?

I notice the demo site uses numeric Permalinks (i.e. https://<site>/wordpress/archives/123). Could this make a difference? I tried changing mine (temporarily), but still had the problem.

According to wangguard.com tonight:

“After 6 years of work and only 6 donations,
I don’t see that people appreciate WangGuard, so I’m Shutting Down WGG”

Which means either the author has indeed given up and ended the service abruptly, or else the site has been hacked.

Time will tell, but I suspect the former.

Robin

Thank you for this – it led me to a solution to my own problem.

Having read your post, I investigated, and found this handy tool:

https://developers.facebook.com/tools/debug

Enter your wordpress post URL in it, and it will analyse it as if processing for a link (but without actually posting to Facebook). The great thing is that in doing so, it clears the cache, so you should run this after tweaking anything. Also it gives you some hints on what is going wrong.

Worth knowing for the future.

I think I have sorted out my problem, the following things are worth checking out for anyone with a similar problem. They may not solve your issue, but it’s worth a try.

1) I learned from another post here that Facebook caches images. So if you don’t get the right (or any) image the first time for a given post, no amount of reprocessing that post will change anything. You may make changes to WordPress that means the right image is now being submitted, but Facebook will ignore it, as it thinks it already has an image.

2) This can be overcome by using this debugging tool:

https://developers.facebook.com/tools/debug

Enter your wordpress post URL in it, and it will analyse it as if processing for a link (but without actually posting to Facebook). The great thing is that in doing so, it clears the cache, so you should run this after tweaking anything. Also it gives you some hints on what is going wrong.

3) Which lead me to the source of my problem. The above tool told me I had duplicate “og:url” headers in the post. As it uses the “og:image” header (which was also duplicated) to identify the image, this was the cause.

Why were they duplicated? Because I had two Facebook plugins: one to automatically forward my posts to my timeline as links; another to add “Link” and “Share” buttons for visitors. Seems they will not co-exist. Removing the “Link/Share” plugin solved the duplication.

4) And finally, my default image was a “normal” avatar/icon – 96 x 96. The debug tool told me that og:image sources need to be greater than 200 x 200 pixels (and experimentation showed me this does not mean 200 x 200 but greater than 200 x 200!). Otherwise, it ignores them.

So… I fixed my problem by only having one Facebook plugin, increasing the size of my default avatar/icon, and clearing the cache using the debug tool. I am now getting images in my link posts!

Glad it isn’t just me.

Tried this plugin today, and am happy with it, except for the lack of thumbnail. Have tried all 4 settings, only the Gravatar gave me anything – a low-res Facebook icon. After trying all the other options, I reluctantly got involved with Gravatar, and assigned one to my address. Now evern that setting gives me a blank thumbnail.

See https://www.facebook.com/chris.malme to see what I mean.

Subsequent to my last reply, there are other things that need to be tweaked in a similar manner – i.e. to point to the right place for avatars.

But the above does actually get the comments into WordPress, which I think is a good start. If I find time to do any more playing with it, I will post an update.

Meanwhile if you look through the code for similar incidences to those I mentioned, you should be able to plug further holes yourself. Just don’t do a global cut and paste!

I only came across this plugin today, and was disappointed about it not working if the blog is not in the root folder.

However, I seem to have come up with a fix, in that it is importing comments for me. It is an inelegent hard-coded fix, but very simple. It may be that it can be tidied up by someone else.

The problem is entirely in the lj_comments_import_js.php file, and in the last two functions.

The last function, lj_comments_adminpage_sync(), has a reference in it to

document.getElementById(“lj_sync_progress”).innerHTML = \'<img src=”/wp-content/plugins/lj-comments-import-reloaded/ajax-loader.gif” alt=”loading…” />\’;

Edit the img src to include your blog directory – for instance, my blog is in a directory labelled wordpress, so that would become:

document.getElementById(“lj_sync_progress”).innerHTML = \'<img src=”/wordpress/wp-content/plugins/lj-comments-import-reloaded/ajax-loader.gif” alt=”loading…” />\’;

Similarly the function before that, lj_comments_call_sync(need_to_handle_response) has two lines starting:

lj_comments_ajax_get(“https://&#8217;.$_SERVER[“HTTP_HOST”].’/wp-content….

again change that to include the name of the directory – in my case:

lj_comments_ajax_get(“https://&#8217;.$_SERVER[“HTTP_HOST”].’/wordpress/wp-content

This will get the plugin working.

I think you could probably improve my fix by using $_SERVER[“DOCUMENT_ROOT”] instead of $_SERVER[“HTTP_HOST”], and then it wouldn’t need the blog directory added. But I’ve got comments working, so I don’t think I want to play with it too much – I’ll leave that to others.