Gabriel Torres

Google usually doesn't honor robots.txt exclusions like those, but you can disallow those directories in..... hmm... either your Analytics or Webmaster Tools interface (I forget which one has that feature).

When articles are edited in Pages, they show up as a recent activity in the activity streams, however the date displayed is their original published date, and not the edited date. This looks weird, because we end up have items with really old dates in the middle of a list containing very recent items.
A detailed description of this issue is discussed here: 
 

Thanks @Jim M, I was unaware of that.

Hi @Marc Stridgen, I see, here it is configured as 
Update the last action: When the item is edited only
However, as I said, it looks weird that the item is listed with its original published date, because it shows an old article in the middle of the articles. I believe this could be improved to show "updated" and the updated date.
I will open a topic in the suggestions forum as per your suggestion.
Cheers.

@Marc Stridgen Since we run a very large community, let me know if you need any report or access to our Google Search Console. I'd gladly give you full access.

It wouldnt do that if the store module itself is still active

The question is a bit off topic, but related. Is it worth adding the extra lines
Disallow: /store/ Disallow: /clients/ Disallow: /settings/ Disallow: /messenger/ to robots.txt?
For the messenger page, google somehow indexed the page.

 

Thank you for bringing this issue to our attention! I can confirm this should be further reviewed and I have logged an internal bug report for our development team to investigate and address as necessary, in a future maintenance release.
 

Thank you for your report. I have added this to our internal bug tracker

Thank you for bringing this issue to our attention! I can confirm this should be further reviewed and I have logged an internal bug report for our development team to investigate and address as necessary, in a future maintenance release.
 

Thank you for bringing this issue to our attention! I can confirm this should be further reviewed and I have logged an internal bug report for our development team to investigate and address as necessary, in a future maintenance release.
 

@Daniel F Can you please provide the exact language key? All language strings containing "topic" are translated here, hence me opening this topic.
The language key 'topic' defaults to "Topic" (with a capitalized "T") and it is translated in our install, so it isn't this key.
mysql> SELECT * FROM core_sys_lang_words WHERE word_key='topic' AND lang_id='2'; +---------+---------+----------+----------+--------------+-------------+----------------------+---------------------+---------+-------------+-------------+------------+----------------+ | word_id | lang_id | word_app | word_key | word_default | word_custom | word_default_version | word_custom_version | word_js | word_export | word_plugin | word_theme | word_is_custom | +---------+---------+----------+----------+--------------+-------------+----------------------+---------------------+---------+-------------+-------------+------------+----------------+ | 18630 | 2 | forums | topic | Topic | Tópico | 101016 | 101024 | 0 | 1 | NULL | NULL | | +---------+---------+----------+----------+--------------+-------------+----------------------+---------------------+---------+-------------+-------------+------------+----------------+ 1 row in set (0.00 sec) Cheers.

@Marc Stridgen Solved this by editing core > global > global > includeMeta and replacing device-width and device-height with width and height, respectively.
Before:
<link rel="apple-touch-startup-image" media="screen and (device-width: {expression="$image['width'] / $image['density']"}px) and (device-height: {expression="$image['height'] / $image['density']"}px) and (-webkit-device-pixel-ratio: {$image['density']}) and (orientation: {$image['orientation']})" href="{file="$image['url']" extension="core_Icons"}"> After:
<link rel="apple-touch-startup-image" media="screen and (width: {expression="$image['width'] / $image['density']"}px) and (height: {expression="$image['height'] / $image['density']"}px) and (-webkit-device-pixel-ratio: {$image['density']}) and (orientation: {$image['orientation']})" href="{file="$image['url']" extension="core_Icons"}"> This should be fixed in a coming update! 🙂
Gabe.

@Marc Stridgen same result, it generates lots of these invalid apple-touch-startup-image links.

@SeNioR- Not the same issue. The rel="nofollow" is present, but the tags also have rel="noopener" and per HTML standard links cannot have two rel=, they must be combined into one.

Hi,
I wonder if the message "Post #0 unset as best answer" is the correct expected value in the moderator logs (in the ACP and in the topic itself). It seems weird to me. The language key is modlog__best_answer_unset which defaults to Post #%s unset as best answer. I believe that %s should be replaced with the post id (pid).

If we check the logs for setting the best answer, we can see that the post id (pid) is being set correctly:

Upon inspecting the source code, I found the bug at 
/system/Content/Controller.php
Where it is:
\IPS\Session::i()->modLog( 'modlog__best_answer_unset', array( $item->$solvedField => FALSE ), $item ); It should be:
\IPS\Session::i()->modLog( 'modlog__best_answer_unset', array( $comment->$idField => FALSE ), $item ); Cheers,
Gabe.

This is regarding records in databases in Pages.
My Activity Streams > Create new stream
Content items only Content types: (selected all databases, which contain articles) Read status: everything Ownership: everything Following: everything Time period: last 7 days Sorting: newest activity first To my surprise, very old articles showed up. The reason being: I recently moved these articles from one category to another. IMO, these articles shouldn't be listed. Simply moving an article from one category to another shouldn't trigger it as being a "new activity".
If this is the intended behavior, then the text should be about the item being moved, not about it being posted.
Edit: this was caused because the records were edited. However, I don't think a simple edit shouldn't trigger this.

Thanks

As you may noticed from my recent topics, I am trying to improve our website's SEO.
One thing that caught my eye in our website's Google Search Console was the Coverage > Blocked due to access forbidden (403) report, with 38.7k entries.
I noticed that most URLs have repeated patterns such as ?advanced_search_submitted=, ?do=reportComment, and ?do=markRead.
My question is very simple: for SEO strategy, should we add those patterns to our robots.txt, or this wouldn't matter at all?
My logic here: if we place these in robots.txt, Google won't try to crawl these pages, saving crawling time/bandwidth.
Cheers!

Few instances were fixed, but this one was indeed missed.
Thanks for the report @Gabriel Torres, I have fixed it for an upcoming release:)
 

Hi,
Validating our website with the W3 validator tool, I found a glitch in the core > front > sharelinks templates, where the tag "rel=" is duplicated (there is a rel="nofollow" and a rel='noopener'; they should be combined as rel="nofollow noopener"). For example, the twitter template is:
<a href="http://twitter.com/share?url={$url}" rel="nofollow" class="cShareLink cShareLink_twitter" target="_blank" data-role="shareLink" title='{lang="twitter_text"}' data-ipsTooltip rel='noopener'> <i class="fa fa-twitter"></i> </a> But it should be:
<a href="http://twitter.com/share?url={$url}" rel="nofollow noopener" class="cShareLink cShareLink_twitter" target="_blank" data-role="shareLink" title='{lang="twitter_text"}' data-ipsTooltip> <i class="fa fa-twitter"></i> </a> The same happens in the linkedin, pinterest, and reddit templates.
Cheers.

Hi,
Running the W3 HTML validator on our website, it complains about this line:
<input type="checkbox" name="remember_me" id="remember_me_checkbox" value="1" checked aria-checked="true"> This also happens when we audit www.invisioncommunity.com
Please advise.
Cheers.

@AlexWebsites Yes, and the latest version doesn't deal with the patterns I am talking about, hence me opening this topic.. 😉
 

@AlexWebsites Yes, and the latest version doesn't deal with the patterns I am talking about, hence me opening this topic.. 😉
 

@Stuart Silvester brainfart here. I was aware of that, I even exchanged a couple of messages with you regarding robots.txt. We have in fact this directive in our robots.txt. I edited the title of this topic to reflect our current (and correct) discussion.
# Block profile pages as these have little unique value, consume a lot of crawl time and contain hundreds of 301 links Disallow: /profile/ I was intrigued on why Google was still indexing my own profile and found this:
https://developers.google.com/search/docs/advanced/robots/intro
That is exactly what is happening here.
Therefore, based on what you wrote, besides the nofollow, I believe that probably it should be added the noindex directive to the user profile pages.
However, in order for the noindex to work, the pages must not be blocked in robots.txt:
https://developers.google.com/search/docs/advanced/crawling/block-indexing
Could you please take a look into this?
Thanks.

Hi,
I wish we had a way to specify the order of the databases displayed in the Search and in the +Post features of the community. Currently, there is no way to specify the order we want them to be displayed, so they are listed according to the primary_id_field. I wanted to change the order, so the most used/popular databases are listed at the top and the least used/popular are listed at the bottom of these lists. Examples:

 
In these two places, I wanted to change the order of the databases.
Proposed solution
A new 'order' field in the databases.
In the ACP > Pages > Databases screen, the databases would have the drag & drop reordering feature that is present in several other places in the ACP (e.g. ACP > Pages > Blocks).
Wherever in the front-end that databases are to be displayed, they would follow the same order set in the ACP.
I hope you can implement this.
Cheers,
Gabe