stringer-rss / stringer Goto Github PK

Not sure how - but I ended up with a few orphaned stories. Here's the error:

https://gist.github.com/darron/5564903

Turns out there's a couple stories still in the stories table:

http://d.pr/i/nncC

But that feed is gone:

http://d.pr/i/yY7V

I deleted a number of feeds when I migrated the DB: https://github.com/swanson/stringer/pull/66

But given the :delete_all - they shouldn't be there.

As soon as I removed them - the interface loaded again - before it just died with a 500.

Any other way you can think of that they'd still be around?

None of the flash messages (sessions_controller, feeds_controller, etc) are using i18n

013-05-03T17:58:49.200892+00:00 app[web.1]: D, [2013-05-03T17:58:49.200677 #12] DEBUG -- :   Feed Load (2.3ms)  SELECT "feeds".* FROM "feeds" WHERE "feeds"."id" = 89 LIMIT 1
2013-05-03T17:58:49.218083+00:00 app[web.1]:    app/views/partials/_story.erb:65528:in `__tilt_19335440'
2013-05-03T17:58:49.218083+00:00 app[web.1]: NoMethodError - undefined method `[]' for nil:NilClass:
2013-05-03T17:58:49.212424+00:00 app[web.1]: D, [2013-05-03T17:58:49.212193 #12] DEBUG -- :   Feed Load (6.6ms)  SELECT "feeds".* FROM "feeds" WHERE "feeds"."id" = 41 LIMIT 1
2013-05-03T17:58:49.218083+00:00 app[web.1]:    /app/app/models/story.rb:7:in `headline'

at least that's what I think the problem is.

Using groups/folders from greader is breaking the importer

Should be able to add single feeds and remove feeds.

Should be able to view a list of all feeds you are subbed too

Any chances of adding these to the documentation please?

Regards,
Deepak Kundu

I think it would be nice to have a way to categorize feeds into groups.

I currently use Feedly, and would like to move away from it, but I rely a bit on groups to focus my reading since I have anywhere from 700-800 newsfeed entries a day, and its a bit rough having them all just in 1 list.

First: This is a beautiful piece of code.
Second: Is there documentation somewhere listing all of the supported shortcuts?

Thank you!

I don't have a link off hand, but something similar to how heroku allows you to quickly filter your apps but with javascript that fuzzy searches the titles of all the stories.

Use medium as a starting point for post styles

feeds without proper timestamps are getting extras loaded

Something is not correct with the pulling only new entries

Here's what I did:

1. Pull the repo (this morning).
2. Migrate the db, foreman start
3. Launch the web app
4. Skip Google Reader import
5. Add a feed (http://mdswanson.com, of course)
6. Run rake fetch_feeds
7. Observe seemingly good things happening in the console
8. Refresh the UI
9. Observe RSS Zero™

In step 9, I expected to see at least the latest article from http://mdswanson.com.

If I have (4) unread, and I mark one as read, it should update the page title to show (3)

https://github.com/rstacruz/sinatra-assetpack

Chrome has super aggressive caching - so I want cache busting at least. Might as well minify and compress while we are at it.

Have a look at the screenshot below, you can see that the delete button is there but it's just not in the right place.

Browser: Google Chrome 28.0.1500.6 (Official Build 199111) dev
OS: Mac OSX 10.8.3

Heroku automatically writes these, but for VPS we need to set them up.

Also update instructions for VPS to make sure it is run in prod env - not dev.

Have a look at the screenshot below, you can see that the feed status icon is not in the right position.
Browser: Mozilla Firefox 20.0.1
OS: Archlinux X64

• create username/pw
• prompt to import from google reader (skippable)
• ajax-y spinner while we parse imported feeds
• help overlay on the feeds page

Will responsive be painful? Serve seperate phone UI? Need to figure out the UI for a phone.

Just like in the BIG_FREE_READER we need a "Keep unread" checkbox/button (hotkey "m") to keep actual item unread after leaving them.

Subscribing to feeds is still quite complex for 'normal' people: copying a URL is hard.
Would you be interested in adding registration for SubToMe? This way, subscribing becomes as easy as a click of a button!

We'd be more than happy to help!

Handle 404s, parsing errors.

Display the status of a feed (last fetched, success/failure) so user can diagnose or report

j/k to move up down
? to show all shortcuts

what else?

git pull / heroku push / rake db:migrate etc etc

After trying to make a fresh install in order to test my Translation Commit, i run into this when i first start the app:

ActiveRecord::RecordNotFound at /setup/password
Couldn't find User with id=1

Anyone did a fresh install recently?

The colored dots on the feeds list have some meaning - but it is not clear without looking in the code.

Green: feed parsed successfully
Yellow: there was an error parsing, but we have parsed this feed successfully before (site is temporarily down, some entry is causing problems)
Red: we have never parsed this feed successfully (404, malformed feed)

A small tooltip or something is probably adequate.

Hey Matt! Just playing around with Stringer. Looks great, btw I can't login. Can you please explain how to create an user to login? Thank you!

http://www.reddit.com/r/ruby/comments/1dmgcg/a_selfhosted_antisocial_rss_reader_written_in_ruby/c9s7s8i

I tried it out and it doesn't work with the hacker news rss feed. Does anybody else have this problem?

Hello everyone,

First of all, congratulations for this! I already have a working copy that I'm using daily.

What do you think about adding i18n support? I can work on that.

Thanks!

I just recently installed stringer and am enjoying using it. One thing I miss from Google Reader is the ability to see stories you've already read. Would that feature be of interest to include in the app?

I've got a crude version working locally, but it will need some more work before I'm comfortable pull requesting. I'm not a very experienced developer and this is the first time I've even attempted to contribute a feature to open source, so bear with me :)

I deleted a feed after rake fetch_feeds, so some stories got orphaned (not belonging to specific feed) and news page failed to render while getting feed title of new story at app/models/story.rb#L19.
I locally added nil-check code, but I think there may be similar issues elsewhere.
Maybe it is better to remove all stories belonging to the feed.

Anti-Social: a pervasive pattern of disregard for, and violation of, the rights of others that begins in childhood or early adolescence and continues into adulthood.

Asocial: the lack of a strong motivation to engage in social interaction or the preference for solitary activities.

I seriously doubt your rss reader is built to intentionally cause harm. The word you are looking for is asocial.

I just installed Stringer locally and played around with it a bit. It's nice. Not my preferred UI (I prefer three-column to river-of-news) but that's not really relevant.

I just thought I'd make a quick feature request: if you have plans to add a client API in the future, you might be interested in basing it on my Standardized Open Feed API proposal. (https://github.com/redwallhp/SOFA) It's new, and pretty much unused by anyone thus far, but I'd like to see it gain traction.

Fix the logic with first_run_completed? as per #12

Subscribe to a public hub (eg. superfeedr.com) and have them watch your feeds updates. Then you could implement a simple (optional setting for the user) script that would be called every time the hub get updates and inserts them into the DB.
So the fetch_feeds script will only check the DB for updates and not every feed you are watching.

The user could provide his superfeedr.com credentials through the app. Then subscribing/unsubscribing from the hub it's easy through their public API.

(i have implement it allready but in PHP in a private repository. I dont know much of Ruby. But i would be more that glad to help you. :) )

The README mentions they are there, but doesn't say what they are.

I think it would be great to replace URLs like /add_feed, /delete_feed or /mark_as_read with REST resources (/stories, /feeds) that get accessed through semantic HTTP verbs (ex. POST on /feeds to add a new feed, DELETE on /feeds/:id to delete one).

@swanson would you consider a pull request in this sense?

The frontend code is written in plain old jQuery and stores state in the DOM. This can get out of hand quickly, how about using a framework like Backbone.js or AngularJS?

Use a webhook to auto refresh periodically - update favicon with unread count?

Investigate further: https://gist.github.com/plasticine/5175588

Add instructions for deploying to heroku and setting use the scheduler jobs/dyno poker

I reckon this would be awesome with keyword group filtering. I tried to do this before with https://github.com/hamstar/feedmonster but ran out of time.

Here's how it would work

1. Create a keyword group and give it a name (e.g. dodgey bastards)
2. Add keywords to the group (e.g. facebook, security, privacy, breach)
3. Select to match on all or any keywords
4. Select to match on just the title or body as well
5. Each story is matched against the keyword group
6. You can click on your named keyword group to see all the stories that match it

Possible improvements

• regex matching
• filter syntax matching (e.g. dogs, cats, !bears, (hawks|eagles) )

Domain language?

I was trying to think of what to call this in the language of the domain... it's called stringer... so maybe a keyword group could be called a rope (a collection of strings)? Or maybe just tag groups or something...

Potential issues

• retrospective matching could be resource intensive