Facebooker talk - Mar 2008 - I think I know why plugin installer is not getting http_multipart_post.rb

looking through the rails source, I think it skips the download of the  
file if it starts with http.

(links method of RecursiveHTTPFetcher which is used by plugin installer)

def links(base_url, contents)
     links = []
     contents.scan(/href\s*=\s*\"*[^\">]*/i) do |link|
       link = link.sub(/href="/i, "")
       next if link =~ /^http/i || link =~ /^\./
       links << File.join(base_url, link)
     end
     links
   end

Since the one file thats not getting downloaded is net/ 
http_multipart_post.rb

Can one of the committers rename the file to not start with http?

Once its done I can test to be sure, but im pretty sure this is the  
problem.

It will save lots of folks headaches, I know the gem is preferred, but  
if there is going to be a plugin, it should work, and its been broken  
for a long while now.

thanks
Joel

-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://rubyforge.org/pipermail/facebooker-talk/attachments/20080303/4b97966f/attachment.html

Hey Joel,

I am not seeing this problem.  I am on rails 2.02 and just did a fresh
plugin install and it came down fine.

Screen cast, longer than I wanted it to be, you can skip ahead about
2/3 of the way through:

http://screencast.com/t/sak9LJRUy



On Mon, Mar 3, 2008 at 5:14 PM, Joel Nylund <jnylund at yahoo.com>
wrote:
> looking through the rails source, I think it skips the download of the file
> if it starts with http.
>
> (links method of RecursiveHTTPFetcher which is used by plugin installer)
>
>
> def links(base_url, contents)
>     links = []
>     contents.scan(/href\s*=\s*\"*[^\">]*/i) do |link|
>       link = link.sub(/href="/i, "")
>       next if link =~ /^http/i || link =~ /^\./
>       links << File.join(base_url, link)
>     end
>     links
>   end
>
> Since the one file thats not getting downloaded is
> net/http_multipart_post.rb
>
> Can one of the committers rename the file to not start with http?
>
> Once its done I can test to be sure, but im pretty sure this is the
problem.
>
> It will save lots of folks headaches, I know the gem is preferred, but if
> there is going to be a plugin, it should work, and its been broken for a
> long while now.
>
> thanks
> Joel
>
>
> _______________________________________________
>  Facebooker-talk mailing list
>  Facebooker-talk at rubyforge.org
>  http://rubyforge.org/mailman/listinfo/facebooker-talk
>
>

Must be a rails 1.2.x problem

Joel

On Mar 3, 2008, at 7:28 PM, "David Clements" <digidigo at
gmail.com> wrote:

Hey Joel,

I am not seeing this problem.  I am on rails 2.02 and just did a fresh
plugin install and it came down fine.

Screen cast, longer than I wanted it to be, you can skip ahead about
2/3 of the way through:

http://screencast.com/t/sak9LJRUy



On Mon, Mar 3, 2008 at 5:14 PM, Joel Nylund <jnylund at yahoo.com> wrote:
looking through the rails source, I think it skips the download of the file
if it starts with http.

(links method of RecursiveHTTPFetcher which is used by plugin installer)


def links(base_url, contents)
   links = []
   contents.scan(/href\s*=\s*\"*[^\">]*/i) do |link|
     link = link.sub(/href="/i, "")
     next if link =~ /^http/i || link =~ /^\./
     links << File.join(base_url, link)
   end
   links
 end

Since the one file thats not getting downloaded is
net/http_multipart_post.rb

Can one of the committers rename the file to not start with http?

Once its done I can test to be sure, but im pretty sure this is the problem.

It will save lots of folks headaches, I know the gem is preferred, but if
there is going to be a plugin, it should work, and its been broken for a
long while now.

thanks
Joel


_______________________________________________
Facebooker-talk mailing list
Facebooker-talk at rubyforge.org
http://rubyforge.org/mailman/listinfo/facebooker-talk