Log In
Home
(0 votes) 9:53 pm
May 1, 2023
Offline
According to the log, the plugin generates an AI image from Stable Difussion BEFORE it checks the article against the filter words.
The result is that if the article doesn't match my mandatory words or phrases and is not posted, I still use image generation credits.
Can this be changed so that the test of the filters is moved to earlier in the work flow so that if the decision is made not to create a post, then a request to Stable Diffusion to create an image is not generated.
Thank you.
10:15 pm
July 2, 2009
Offline
I see your point and yes, that's a reasonable suggestion. I'll check if it can be implemented without disturbing the current logic of the code.
8:48 pm
May 1, 2023
Offline
Thanks, I appreciate this!
10:20 am
July 2, 2009
Offline
The requested change has been implemented in version 10.083. Please note that this change affects only the images added via the 
Login to see this link are still added before the post content filleting operations, because the post text content is not finalized at this stage.
3:49 pm
May 1, 2023
Offline
Thank you! This solved the problem. As a bonus, it processes much faster as well without the uneeded calls...
9:48 pm
May 1, 2023
Offline
I have now added some spanish-language feeds, and if the plugin finds one that meets my critereia (i.e.: filters) then I want it translated into English using the Deep-L API.
This works great, but suffers from the same problem as th image generation did!
CyberSEO is fully translating every article before determining if it matches my filter criteria. I just got one article, but the eight previous articles that failed my filters were all fully translated. I have three such feeds, and they are fairly busy news feeds. I expect that I will likely have to translate quite a few full-text articles daily before I get one or two posts. This will very quickly use up my DEEP-L API credits.
Can you please look into making a similar change for translation? I've disabled translation in the meantime.
Thanks.
9:51 pm
July 2, 2009
Offline
If your criteria checks the text, it can be applied to the text that has been translated and is ready to be checked. Obviously, it can't be applied earlier. This would break the whole logic.
10:14 pm
May 1, 2023
Offline
Not at all, my filter terms are in Spanish. This is exactly so I don't have to translate every article to test the terms. That doesn't really make sense to me because of the translation overhead it involves.
10:20 pm
July 2, 2009
Offline
I see your point, but what about all the other users who want to make sure that the spun and translated content does not contain the words they don't want to see in their posts? They don't know how it will be translated and what will be generated as a result. The filtering feature will be useless for them in this case?
10:29 pm
May 1, 2023
Offline
I don't believe so... For example, if i wanted to add some feeds in German (which I do not speak), I would write down my filter terms (negative and/or positive) and then use DEEP-L to translate them into German. Then I'd add them as filter terms in the plugin. I don't think I need to be able to speak or understand the language of the feed.
This makes far more sense to me than fully translating possibly dozens of articles a day that don't meet my criteria.
10:32 pm
May 1, 2023
Offline
As a compromise (although I know this a much bigger coding job!) you could provide an option under the Translate section of the Feed options to translate the feed pre or post the filter term check.
1:10 pm
July 2, 2009
Offline
The requested option was implemented in version 10.087. Note the "Apply filters before" drop-down box on the "Content filtering" tab.
8:57 pm
May 1, 2023
Offline
Thank you for your prompt response to this!
Unfortunately, I think this change broke something. Now every post is rejected because "The post date is older than a specified period .... The post will not be added." This happens with every post in every feed, and rejects new posts from today when I'm allowing posts that are three days old.
Next, I have a question. Once this is fixed, I am curious about the first option in the drop-down: "full article extraction." Can you tell me what data is available to test against the filters before the full-text extraction takes place. I assume it is the title and the extract, but I would appreciate it if you could confirm (or correct) this assumption.
Thanks!
9:23 pm
July 2, 2009
Offline
Only the actual feed content is available before the full-text extraction. Some RSS feeds contain full-text articles, others contain only abbreviated versions.
As for the date filtering issue, please update your plugin manually on the CyberSEO Pro "General Settings" page. It should work properly now.
All RSSMost Users Ever Online: 541
Currently Online:
16 Guest(s)
Currently Browsing this Page:
1 Guest(s)
Top Posters:
ninja321: 84
s.baryshev.aoasp: 68
Freedom: 61
Pandermos: 54
MediFormatica: 49
B8europe: 48
Member Stats:
Guest Posters: 337
Members: 2852
Moderators: 0
Admins: 1
Forum Stats:
Groups: 1
Forums: 5
Topics: 1640
Posts: 8352
Newest Members:
torontomark48, info.ckmedianetwork, contact.mybeautystar, samuelbodde, john.prush, creightonnick0Administrators: CyberSEO: 3947