You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is your feature request related to a problem? Please describe
This project and the FeedMergeBride by @dvikan are amazing. But I’ve stumbled over a problem.
Sometimes two feeds provide the same article. See #2855.
The existing feature in FeedMergeBride is great, when the url provided by the feeds are identical. I'm having the problem that some of my feeds have a different url for the same article. The title however is identical. So filtering duplicates based on the title is possible.
I’ve provided an example below under "Additional context".
Describe the solution you'd like
it would be great if FeedMergeBride could filter based on title. The easiest solution would be to first strip out all duplicates based on url and then strip the remaining based on title.
It would be ideal if there were two checkboxes when setting up FeedMergeBride: One for "Filter by url" and another to "Filter by title". This would allow desired behavior to be configured by the user.
Describe alternatives you've considered
I've copied FeedMergeBridge.php, renamed the file and class, then added the resulting new bridge to my installation.
After that I went ahead and edited the new bridge, replacing all occurrences of 'uri' with 'title' in lines 96 to 109.
In the end it looked like that:
// Remove duplicates by using title as unique key
$items = [];
foreach ($this->items as $item) {
$index = $item['title'] ?? null;
if ($index) {
// Overwrite duplicates
$items[$index] = $item;
} else {
$items[] = $item;
}
}
See https://gist.github.com/DrErfinder/9a0ba483f897e21a0c4db0c6cae567f4 for full file.
So far this works flawless for my purposes.
However, I’m not good enough with PHP to know if that's the correct way or how to implement it without losing the existing deduplication.
Is your feature request related to a problem? Please describe
This project and the FeedMergeBride by @dvikan are amazing. But I’ve stumbled over a problem.
Sometimes two feeds provide the same article. See #2855.
The existing feature in FeedMergeBride is great, when the url provided by the feeds are identical. I'm having the problem that some of my feeds have a different url for the same article. The title however is identical. So filtering duplicates based on the title is possible.
I’ve provided an example below under "Additional context".
Describe the solution you'd like
it would be great if FeedMergeBride could filter based on title. The easiest solution would be to first strip out all duplicates based on url and then strip the remaining based on title.
It would be ideal if there were two checkboxes when setting up FeedMergeBride: One for "Filter by url" and another to "Filter by title". This would allow desired behavior to be configured by the user.
Describe alternatives you've considered
I've copied FeedMergeBridge.php, renamed the file and class, then added the resulting new bridge to my installation.
After that I went ahead and edited the new bridge, replacing all occurrences of 'uri' with 'title' in lines 96 to 109.
In the end it looked like that:
See https://gist.github.com/DrErfinder/9a0ba483f897e21a0c4db0c6cae567f4 for full file.
So far this works flawless for my purposes.
However, I’m not good enough with PHP to know if that's the correct way or how to implement it without losing the existing deduplication.
Additional context
Here’s an example:
German IT-News-Website Heise.de provides a whole host of RSS-Feeds (See https://www.heise.de/news-extern/news.html for full list)
Some of the articles in their Apple specific feed (https://www.heise.de/mac-and-i/feed.xml) are also published in their general news feed (https://www.heise.de/rss/heise-atom.xml)
The url of the article is
"https://www.heise.de/news/Apple-Studie-Logisches-Denken-von-KI-kaum-nachweissbar-und-sehr-fragil-9980855.html?wt_mc=rss.red.ho.ho.atom.beitrag.beitrag" in one feed and
"https://www.heise.de/news/Apple-Studie-Logisches-Denken-von-KI-kaum-nachweissbar-und-sehr-fragil-9980855.html" in the other.
The title is "KI und logisches Denken: Apple-Forscher zweifeln – und warnen" in both cases. I assume the url differs for some form of analytics/statistics
The text was updated successfully, but these errors were encountered: