Open garritfra opened 3 years ago
Fiddling around with this, I found that it's not as easy as I expected. In order to query non-standard attributes (like category
), the raw entry would need to be stored. Another way would be to add an "extra tags" field, that itself is a key-value map.
Is there a simpler way to query ever field in the original entry?
For what it's worth, having anticipated the usefulness of this for The Guardian's rather large news feed in my own (forthcoming) independent implementation of Miniflux's API, I simply had rules try to match against the title, and in the absence of a match, each category name until a match is found. So, a keep rule of ^dev-blog$
would suffice.
The downside to this is that since Miniflux doesn't actually expose RSS/Atom categories, it might be confusing for users. I'd be keen to discuss the best, least confusing way to expose such a feature.
Since the parser for the
Please implement this and update the Documentation.
This was implemented here: #2526 Just documentation needs to be updated
This was implemented here: #2526 Just documentation needs to be updated
Please correct me if I am mistaken, but that PR only addresses global filter rules. The request here is for fields used in filter rules applied to individual feeds. The fields used to filter within specific tags (e.g., EntryTitle
, EntryContent
) cannot be used in a feed's keep or block rules, only in the global filter rules.
@shanewstone , Yes, that's correct. Individual feed filtering needs to be refactored to use the same logic (EntryTitle, EntryContent). For my personal use, I used this patch https://github.com/miniflux/v2/compare/main...jacob-faber:miniflux-v2:filter-content to match the content of individual feeds.
As a user, I want to be able to add a filter rule to a feed based on a specific xml tag. Currently, filtering is only possible on the title.
Concrete example
Considering a feed that has these entries:
I want to be able to only subscribe to entries marked with the category "dev-blogs".
Possible solution
Add the possibility to specify tags in Keep/Block rules. If the rule is a tag (regex/raw character parsing?), query the entire entry. If not, continue as previously implemented, by just querying the title.
Resources
This feature could be implemented here: https://github.com/miniflux/v2/blob/de7a61309878e335fe99c7afbbe6a40cc097b0ae/reader/processor/processor.go#L86
Context
I stumbled upon this issue when I wanted to subscribe to the dev-blogs of a game. The authors only have a single rss feed for all their content.