janboddez / feed-reader Goto Github PK
View Code? Open in Web Editor NEWA very simple feed reader plugin for WordPress
Home Page: https://feedreader.site
License: GNU General Public License v3.0
A very simple feed reader plugin for WordPress
Home Page: https://feedreader.site
License: GNU General Public License v3.0
Not sure if there's feeds that have more than one (h-feeds often do, though), but we should eventually tackle this better.
Also, can we display them on the single-entry page, too? And, if so, avoid duplicates?
Simply add the callback also to wp_head
.
Example: https://colly.com/stream/any-space-whatever
The issue is the stripped down post body starts off with a string that == the title, and this makes the plugin think we're dealing with a note.
One solution I was thinking of is disable this for ~long posts and when the title isn't, e.g., obviously "cut short" (like, when it does not end in an ellipsis or something).
This is going to be a difficult one, probably. Maybe we should leave it to the Micropub clients. :-)
th
s are currently/by default center-aligned, td
s left-aligned (or right, I suppose, on rtl installs).
Probably a feed-reader.js thingy.
I.e., http_headers_useragent
.
This may something most easily accomplished by JS. Add display: block
to largish images, programmatically?
Doing so would "fix" captions that use, e.g., em
instead of figcaption
, and so on, by forcing them onto a line of their own.
Stole this idea from Monocle, way back: refresh unread counts once per minute or so (this part is optional, even, as navigating around would update the sidebar anyhow, and polling only occurs once every 15 minutes at best), using AJAX, and each time an item is marked (un)read.
Since we're already using JS for the (un)read buttons, shouldn't be to hard to trigger the function.
On single-entry pages, show the "featured" image(s) or video.
But only if they're not already part of the entry content.
This could be a plugin, once we have those filters in place. Or a native thingy. Or a separate option.
When a feed is first added, any items present are marked as read. This is how it should be.
But when importing OPML, feeds aren't polled until later, and "new" items won't show up as new.
We can either leave this as is or assign a random "last polled" date (when importing from OPML only) to work around this.
Wouldn't be too hard to deregister the stylesheet in question, but maybe we can offer a proper option.
Also, we should do this when posts are fetched, not on render.
We could use (some) microformats for styling, e.g., avatars, but then we can't strip them.
You know, to save on bandwidth. Although embedding 3rd-party resources ... well, you know the risks.
Right now, we fetch models early, and cache them. Then after we're done with them, we delete the cached model, so as not to run into issues with persistent object cache plugins. (We really only want to cache for the duration of the request.) Maybe static properties are the cleaner solution.
Replace current_user_can( 'edit_others_posts' )
with current_user_can( 'publish_posts' )
?
Add a Microsub endpoint, map categories to "channels," feeds to "sources," support at least marking entries as read.
Other items (e.g., following new feeds, etc.) to be considered.
Add a command or cron function to weekly or so to clean up old (read) entries.
It'd be nice if we could wrap tables in a selectable / scrollable div
so as to keep them readable on smaller screens.
Some site return a 403 when WordPress attempts to fetch their images. Should be okay with a "spoofed" user agent.
Video poster
attributes are not run through our "image proxy."
Way we currently detect categories is, if an outline does not have a URL, we consider it a category, and keep that in memory as the "current" category. (This also wrongly applies a category to uncategorized items that somehow follow a [closed] "category" outline.)
Think OPML v2 supports a comma-separated (?) category attribute on the actual feed items themselves, so we could try adding that in.
Since we support only one category per feed (which is different from WP's built-in taxonomies), we can either pick the first or the last one, for now.
We're suffering from this, too: https://publishpress.com/blog/wordpress-publishing/multisite-networks/#h-difference-3-the-administrator-has-less-power
I sometimes prefer to have an item's author name replaced with the name I gave the feed, for instance, although the icons help a lot recognizing the individual feed.
But I still might wanna "correct" longer or faulty names/feeds, etc.
So, make item name and author filterable, and then we'll see about everything else.
And push 'em down by, like, 1px?
Now, because it's likely rendered way ahead of the sidebar, we might wanna avoid needless queries en run the whole category/feed thing with all unread entries here, and cache it for the current request only.
Some authors name their notes "No Title," causing these notes to be interpreted as ... articles.
If we could somehow filter titles, perhaps provide the feed as an argument, we could allow site owners to "fix" this, for particular feeds.
Maybe there should be an option, whether a (reply, etc.) post should be draft or published.
Icons aren't deleted when a feed is.
It can take a little while for bookmarks and replies to go through ... We're probably better off immediately hiding the form, or graying out the submit button until successful submission is confirmed.
Seems to consider https://chriscoyier.net/2023/05/30/248/ a "note."
TL;DR don't use mb_detect_encoding()
. https://stackoverflow.com/questions/44571109/html-entity-decode-charset-ascii-not-supported
Like, we could allow them but strip atributes. Or look into HTMLPurifier to turn them into a more common element. (Although I'd rather not add another dependency.)
https://colly.com/stream/gordon-miller was added twice, although the content did change ever so slightly in between
Use the user profile?
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.