Feature Requests

[lwindolf]

Put all your feature requests and ideas here!

Please make one comment per feature and like existing feature comments you want to vote for!

[horstjens]

Synchronization

this may have been asked a million times before, but is there a step-by-step tutorial of how to host the .lifera database on a web server / a google drive so that i can view my feeds synced from my desktop computer as well as from my android mobile devices ?

[sjehuda]

DecSync

Horst, I reply to your comment because it is similar to what I have suggested at #1256, to add DecSync.

This is a project from @39aldo39 which I want to start a discussion about.

https://github.com/39aldo39/DecSync
https://github.com/39aldo39/libdecsync

See also #956

[RafaelLinux]

Column Category

Please, it's important to add (or let user enable/disable) the "Category" field on each post, next to "Date" column.

Thank you

[lwindolf]

The problem here is that there is no single category per post. Some feeds might have none or just one category, while other have 20.

Having this in a shared column makes the feature unusable :-(

[sjehuda]

Perhaps the first category or category that appears the most should appear?

[johnno66]

Caching by time period

(in Preferences ->Feeds -> Feed Cache Handling)
Please add option to keep items based on age, e.g. only keep items that are <1 week old
This would (in my usage scenario at least) significantly decrease database size.

[sjehuda]

This would (in my usage scenario at least) significantly decrease database size.

This comment is concerning the database size.

I suppose that Liferea compares entry ID and if not available, then it would find a match of Link or Title, in order to determine whether that entry has been downloaded or not.

In order not to download again an entry, that was read, as a new entry, Liferea has to store a value to identify that entry.

So to make what you want, and I think it's a good idea, Liferea can delete summaries.

---

UI Consideration

If Liferea would hide entries:

• No need for UI consideration.

If Liferea would not hide entries:

• Because the rest of the information, like Title, Link and Date, would not be deleted, it might be good to keep those entries visible in the UI.

• This would mean that upon clicking on an item that is more than a week old, Liferea would show no title for it.

• This means that a UI change would have to be made.

• Upon selecting an entry that is older than 7 days, Liferea would display this text "This item is older than 7 days and therefore has been deleted. Click the download button (that button does not exist) to display (download) the item summary".

• • The download process would be inefficient, and even slow, especially when a feed has thousand of entries, because then Liferea would have to (1) Download the feed (2) Lookup for the entry (3) Catch and display the entry summary.
• • Perhaps archiving old summaries in a compressed archive would be a solution to save storage space.

[payrim]

Duplicate subscription

remove duplicate subscription links? i imported from 3 different .opml files and some subscription in between are duplicate but reside in different folders

[sjehuda]

payrim, when importing an .opml file, my news bot doesn't import a feed that already appears in its database, but this is because my bot has no support for categories.

Should there be an exception to that if a subscription link appears in a different category? (i.e. a subscription link that integrates both news articles and podcast)

[lwindolf]

@payrim I've created issue #1333 to adress this. We actually have the feature already when creating a single subscription, but the check is not triggered for OPML.

[sjehuda]

Select Feeds From Online Database

This is my attempt to restore a service similar in nature to Syndic8.

Provide a functionality to search for feeds from a database, be it from a remote (online) or local (offline) database.

Here is a database I have composed. The database is public domain.

I suggest to make that database accessible via IPFS.
I also suggest to make the practice of a remote feeds database a common practice with any other Feed Reader available.

I think that this is a crucial feature because it would make RSS Readers independent and not subjected to the whims of web browser vendors.

Thud has this functionality, yet it appears to be using a centralized source to do so. I have opened a similar ticket at aerotoad/Thud#55.

---

Warning! (Government)

Because this database is intended to be used with my project, and because it has less than 10K entries, I don't make any selection, which means that some sources might not be appropriate (i.e. nsfw).

Personally, I might have a legal issue with that database at the moment, because I also add sources of neighbouring countries. Some of these sources appear to be censored by the ISP providers here, and this is another reason of why I have an interest of making that database available via a decentralized mean (e.g. IPFS).

Suggestion to allocate feeds via IPFS

There would be several databases allocated by language code and country code.

Unrestricted

• de would include all feeds for use by German speaking audience that is not subjected to restrictions poised by any other German speaking countries (Austria, Germany, Liechtenstein, Luxembourg, Switzerland).

Restricted

• de-at would include any feed in German, excluding those that are forbidden by Austrian government.
• de-ch would include any feed in German, excluding those that are forbidden by Swiss government.
• de-de would include any feed in German, excluding those that are forbidden by German government.
• etc.

Unrestricted databases (those that do not have country code), like de, will either be unavailable or would be available by enabling an option.

[lwindolf]

BTW I've outlined my current design goals here https://github.com/lwindolf/newsagain/blob/main/research.md

[sjehuda]

I did not have an intention to create a PWA, though I definitely intend to try.

Incidently, I have just received this link from a friend on XMPP https://feedle.world/

I will also create an HTML service based on Python and pywebview, and I will further inform you.

We have several databases of 3K feeds or so.

It will be interesting!

[sjehuda]

BTW I've outlined my current design goals here https://github.com/lwindolf/newsagain/blob/main/research.md

Is the statement at lwindolf/newsagain about replacing Liferea serious?

Once Liferea would allow to modify its feeds tree with a plugin, prepare to unexpected features that would make Liferea an interesting publishing software.

[lwindolf]

@sjehuda Yes, the statement is serious. I'm sorry to disappoint. I believe you that with support for feed list modification a lot of functionality can be build, but I do not want to spend my time on it. I hope you understand this. I wrote about the decision https://lzone.de/liferea/blog/20-years-of-Liferea-and-the-future

[sjehuda]

This is a good post!

The concerns of GTK+ were the elements that have deterred me from writing a desktop software in GTK+, albeit that I have desired for it it greately for a long period of time.

The note of libxml2 is good and understandably conveys the message of maintenance concerns.

This is for example different with another important library for Liferea: libxml2. The API was practically 95% stable for 20 years. Zero maintenance effort!

The interface of Liferea, is a "no nosesnse" interface. It always was beautiful and it will always remain beautiful. It should be the standard.

If the plugin system would have more permissions to manipulate data, I will create several plugins to integrate Liferea with XMPP.

[sjehuda]

@lwindolf I ask for your permission to add 3rd-level (i.e. ###) titles to posts.

Synchronization #974 (comment)

Column Category #974 (comment)

Caching by time period #974 (comment)

Duplicate subscriptions #974 (comment)

[lwindolf]

@sjehuda I'm sorry I do not understand what you want to do? Do you want to format the linked comments?

[sjehuda]

I want to add titles like I did here #974 (comment).

It appears that you gave my profile the privilege to do so.

[lwindolf]

Feel free to do so :-)

[sjehuda]

I added a title to this post #974 (comment) for realization

[sjehuda]

Thank you