damog / feedbag Goto Github PK

Would you be open for sponsorhip for this open source project? We only ask for a logo + link to Superfeedr which is a feed fetching/parsing platform as a service...

Please, let me know!
Thanks,

I'm trying to install Feed 0.9.4 using

gem install feedbag

but I'm having troubles. I always receive this error:

ERROR:  While executing gem ... (Gem::UnsatisfiableDependencyError)
    Unable to resolve dependency: 'feedbag (= 0.9.4)' requires 'nokogiri (~> 0)'

I'm using Fedora 20, (don't know if it's matter) Rails 3.2.3 and Ruby 1.9.2. When I run gem list, I got:

*** LOCAL GEMS ***
actionmailer (4.1.4, 3.2.3)
actionpack (4.1.4, 3.2.3)
actionview (4.1.4)
activemodel (4.1.4, 3.2.3)
activerecord (4.1.4, 3.2.3)
activeresource (3.2.3)
activesupport (4.1.4, 3.2.3)
addressable (2.3.6)
arel (5.0.1.20140414130214, 3.0.3)
bigdecimal (1.1.0)
builder (3.2.2, 3.0.4)
bundler (1.6.2)
bundler-unload (1.0.2)
erubis (2.7.0)
excon (0.33.0)
executable-hooks (1.3.2)
gem-wrappers (1.2.5)
heroku-api (0.3.18)
hike (1.2.3)
hpricot (0.8.6)
i18n (0.6.11)
io-console (0.3)
journey (1.0.4)
json (1.8.1, 1.5.5)
launchy (2.4.2)
mail (2.5.4, 2.4.4)
mime-types (1.25.1)
mini_portile (0.6.0)
minitest (5.4.0, 2.5.1)
multi_json (1.10.1)
nokogiri (1.6.3.1)
polyglot (0.3.5)
rack (1.5.2, 1.4.5)
rack-cache (1.2)
rack-ssl (1.3.4)
rack-test (0.6.2)
rails (4.1.4, 3.2.3)
railties (4.1.4, 3.2.3)
rake (0.9.2.2)
rdoc (3.9.5)
rest-client (1.6.7)
rubygems-bundler (1.4.4)
rubyzip (0.9.9)
rvm (1.11.3.9)
sprockets (2.12.1, 2.1.3)
sprockets-rails (2.1.3)
thor (0.19.1, 0.14.6)
thread_safe (0.3.4)
tilt (1.4.1)
treetop (1.4.15)
tzinfo (1.2.1, 0.3.40)

Do you know what can I do to properly install the latest version of Feedbag?

P.S: Please, forget my other issues, I'm trying to install Feedbag in a different "gemset" now (the "default" gemset).

I did a gem install feedbag, and it seems to have also installed hpricot as a dependency. The docs say it's only dependent on nokogiri.

Did something go wrong here?

$ gem install feedbag
Fetching: hpricot-0.8.6.gem (100%)
Building native extensions.  This could take a while...
Successfully installed hpricot-0.8.6
Fetching: feedbag-0.9.2.gem (100%)
Successfully installed feedbag-0.9.2
Parsing documentation for feedbag-0.9.2
Installing ri documentation for feedbag-0.9.2
Parsing documentation for hpricot-0.8.6
Installing ri documentation for hpricot-0.8.6
Done installing documentation for feedbag, hpricot after 3 seconds
2 gems installed

See: #20

A way to override the request headers smartly is necessary.

Since feedbag sends network requests, it'd be nice to have a control over timeouts to avoid long requests.

As feedbag uses URI.parse and Open::URI we've started getting tons of errors like the following:

URI::InvalidURIError: URI must be ascii only "https://www.rehabsociety.org.hk/zh-hant/\u4E3B\u9801/feed/"

Workaround here is to escape the URL before passing it to feedbag, but it'd be really useful if feedbag will be able to work with non-ASCII URIs (aka IRI)

Thanks in advance

Feedbag.feed?('https://aeon.co/culture') returns true while this link is actually html page.
BTW, https://validator.w3.org/feed/ says that this link looks like a web page rather than feed.

We recently faced an issue with long detection of the feeds (already 6 hours on the moment of writing).

The url of the feed is "http://superkuh.com/feeds6.html"

After inspecting the details, we discovered that the long detection happens during inspecting url present on that page: https://link.springer.com/search.rss?facet-content-type=Article&facet-journal-id=41116&channel-name=Living+Reviews+in+Solar+Physics

It seems that existing regexp for looks_like_feed? may be slow in some cases.

I'm not entirely sure how to fix this regex, but it'd be great to eliminate such problem. It'd also be great to inspect other places for similar problems.

It will be possible to leverage Regexp timeouts of Ruby 3.2 to prevent this from happening as soon as they release, but would be great to find a solution before that.

Hello,

First of all, thanks for your gem, it's really awesome!
Secondly, I'm having some issues with your gem that I hope you can solve soon (if not, will be fine too :P )

I'm using a gem to detect if a given name is an URL:
url = Twitter::Extractor::extract_urls(self.name)

After that, I do something like that:

if !url.empty?
     feed_url = Feedbag.find(url)
end

This works almost all time, but when I try put my own Blogger url ("http://fpaladini.blogspot.com.br/") in self.name, I received this error:

URI::InvalidURIError (bad URI(is not URI?): ["http://fpaladini.blogspot.com.br/"]):

But it isn't a bad URL, you can type that in your browser and you'll see this opens. What can I do to solve this issue?

Currently when there is an error it outputs to $stderr, but it would be nice if we could control the place the output goes to, such as by pointing it at the Rails logger.

Hello guys!

First of all, thanks for this amazing gem, is really helping me. I'm here because I'm having troubles with some very used sites, like Gizmodo, Kotaku and others famous websites.

I want know, why Feedbag can't find the Feed RSS for this popular websites? Can you do something about that, I mean, it's a problem with Feedbag gem or this just can't find the Feed of these websites? If is the last one, what could be done to get the feed rss from these websites?

Thanks and keep doing the wonderful work!
Fernando Paladini.

Hi.
I tried get existing feed on the page

Feedbag.find 'https://habrahabr.ru/hub/ruby/'

but got an error

Error occurred with `https://habrahabr.ru/hub/ruby/': 403 Forbidden

On the web page source I see the feed tag

  <link title="Хабрахабр / Все публикации подряд / Ruby" type="application/rss+xml" rel="alternate" href="https://habrahabr.ru/rss/hub/ruby/all/"/>

After migrating our app to Ruby 2.7 we see the following warning:
/usr/local/bundle/ruby/2.7.0/gems/feedbag-0.10/lib/feedbag.rb:100: warning: calling URI.open via Kernel#open is deprecated, call URI.open directly or use URI#open

Could you change the code to hide the warnings?