hrhuu

This plugin seems to be using REST API that is not anymore supported by Facebook – see similar discussion from here: https://github.com/jaredatch/EA-Share-Count/issues/57 ‘

In other words. It is not working until updated.

• This reply was modified 7 years, 9 months ago by hrhuu.

I’m not sure what’s going on…is that the way the urls look on your site when you are paging?

They are crawl errors in Google webmaster tools “Crawl errors”. Errors about not found pages (error code 404). Here is more comprehensive copy&paste from there:

2 industry-news-now/&pg=5?pg=8	404	20.10.2013
3 industry-news-now/&pg=5?pg=9  404 20.10.2013
4 industry-news-now/&pg=3?pg=11 404 15.10.2013
5 industry-news-now/&pg=5?pg=0  404 19.10.2013
6 industry-news-now/&pg=5?pg=6  404 19.10.2013
7 industry-news-now/&pg=5?pg=10 404 19.10.2013
8 industry-news-now/&pg=5?pg=11 404 19.10.2013
9 industry-news-now/&pg=4?pg=5 404 18.10.2013
10 industry-news-now/&pg=4?pg=7 404 18.10.2013
11 industry-news-now/&pg=4?pg=0 404 18.10.2013
12 industry-news-now/&pg=4?pg=10 404 18.10.2013

In other words for some reason the plugin provides for Google bot (and end users) links that are not working. And when GoogleBot crawls the site and follows those broken links – it returns error 404. Form of those links you can see there above.

However as told, unfortunately I cannot reproduce this error manually when browsing the site – I can find them only as errors in Webmaster tool. It is not good to provide continuously broken links to Googlebot crawler, it may affect on Google search results. Maybe you could conclude – when does the script produce those links that have “pg”-variable mentioned two times one after another?

Indeed, paging is still producing errors but less frequently.

This is copy&paste from latest Google webmaster tool Crawl errors in chronological order:
/&pg=3?pg=11
/&pg=4?pg=0
/&pg=4?pg=5
/&pg=4?pg=7
/&pg=4?pg=8
/&pg=4?pg=9
/&pg=4?pg=10
/&pg=4?pg=11
/&pg=5?pg=0
/&pg=5?pg=10
/&pg=5?pg=11
/&pg=5?pg=6
/&pg=5?pg=8
/&pg=5?pg=9

When now browsing and testing paging of RSS-multi-importer-result-page by myself – I cannot find out how Google Bot finds out URLs with “/&” instead of normal “/pg=”. It is not mentioned in sitemap_xml. Strange. But somehow it finds them…

I have faced very same issue as Edgeem – with Genesis 2.0 as well. So I guess it is just incompatible with genesis 2.0 currently.

Thank you very much! Paging problems are actually more crucial that they looked out – I had 250 crawling errors from Google bot because them according to webmasters tools. But it seems to work with 3 pages items what I have now!

Since I have modified template/default and wp-rss-multi-importer.php files for my needs (e.g. translated user interface texts and made some style modifications) – it would be helpful to know which files are modified in each version updates. Just if it was easy to let us know.

Secondly have to just ask, if there was easy way to update plugin instead of deactivate->delete->install new version?

Well, just asking for next update ??

I am not sure if this issue belongs to here – but page pagination seems to have a bug in this beta.

You can only proceed to the page 2. And after that it start to mix ?and add pg=1&pg2&pg… after already existing parameters.

My pagination settings:
Number of Entries per Feed: 80 (I liked it to be “nolimit”)
Number of Entries per Page of Output (Go here to see how to set this option): 20
Do you want pagination? (Note: this will override the Number of Entries per Page of Output) YES
Number of Posts per Page for Pagination: 20

Well, I am sure this bug can be found easily – if just having enough items for pagination (I have on test site – embed shortcode on a page and there are 200 items fetched into database).

OK – 2.70 sounds really great, waiting forward!

Great thank you very much – “author” works now OK!!

Probably unnecessary to remind but could be good to update rss_multi_importer_widget.php for 2.70 too, I think it still fetches feeds by every page load and if used as element displayed on every page it is especially heavily loaded.

Looks out generally very good! I would have couple of questions:

1. AUTHOR OF ITEM
It seems that Author of item is not working in this version (https://www.ads-software.com/support/topic/items-author-not-displayed-author-attribute-isnt-parsed-at-all?replies=12). Could the Author-fix be added here too?

1.1. Beta version OK to use?
Indeed does it make sense to start to use this Beta-version? Yes it does in that sense that scheduled feed-fetching is obligatory but how it will be updated and is it compatible with upcoming new version of official version? Is Beta its own version branch or same as public one and we will see these changes as updates in couple of days in the official public plugin?

2. META-information
Could list of feeds include information about 1) the timestamp when the specific feed is fetched last time, and 2) number of total items and number of items that were fetched to db this time. (Indeed, how does the script notice now if an item is fetched already in db or not?)

Alternatively/in addition, could each Item include information about fetching time. So that you could see when it is fetched – especially good for “archived” items that have disappeared already from original feed.

I think this kind of META-information would be user friendly for plugin-administrator. You would get some feedback of fetching process and knowledge that it works properly.

Could you please make couple of lines summary of the best practice – “How to CACHE feeds”?

Maybe something like?

1) fetch&cache RSS-feeds by CRON script instead of letting my web visitors to wait for any fetching. [OUTPUT: compilation-RSS.xml on my server]

2) And then just “parse&echo” this static complation-RSS.xml on my server for web visitors. [OUTPUT: Items of compilation-RSS parsed&displayed on WordPress pages according to user settings]

The general caching can be then done by dedicated plugins such as “W3 Total Cache”.

(currently getting regularly “server timeout”-error when fetching feeds from multiple sources; I could add server-timeout-value but it would not shorten the time that user needs to wait; and as principal – you should not let user/search engine wait anything but just printout pages he/she wants to see)

I will – just an addition.

Widget (/inc/rss_multi_importer_widget.php) seems to be separated script and it needed all same modifications as wp-rss-multi-importer.php (I modified it too and it thus works).

Indeed maybe it could make sense to refer to same core-code (rss_multi_importer_widget.php) in Widget since it is basically same code. Thus the plugin would be easier to keep updated.

Great, it works! Thank you very much for fixing it this fast!

Unfortunately this is not the case.

Indeed I had a string without @-character there at first – the result was very same. I changed the test string to email address for testing if email address is required there (email address has been at some point popular entry for RSS-item author attribute).

Now I updated author-attribute to “Aut-hor Name” (referesh https://feeds.feedburner.com/wp-rss-multi-importer-test-feed?format=xml). Please take a look if you could find out the bug.

Here is a test feed with item-author attribute:
https://feeds.feedburner.com/wp-rss-multi-importer-test-feed?format=xml

Correction: Version 2.66.5 of WP-RSS Multi-Importer plugin.

(I just noticed that there is now also 2.66.6 available. However according to change log – there is no changes related to this problem)