feedreader News Feed 
Saturday, June 28, 2008  |  From feedreader

For those of you who use my RSS FeedReader Web Part for SharePoint, it has been updated into an ‘official’ release for the project.  Version is now the latest release.  It incorporates all the changes from our team’s contributor, Ryan – thanks Ryan!!!  You can read the checkin on the Codeplex site or review this previous post for Ryan’s checkin notes.

As always, the source code is included on the site, licensed under the Ms-Pl.

There has been some emails/work items/debate over the installer.  The installation is two parts:

    • Installation of binaries
    • Deployment to SharePoint servers

It has been much debated with me in private emails about that the MSI installer should automatically deploy to the server.  Initially this is how feedreader did it.  And I was flooded with email complaints.  Why?  Well, turns out people don’t always install to the same locations, have SharePoint in the same locations, want to deploy only certain web parts to certain SharePoint site collections, etc., etc. – I could go on.  The bottom line is that every configuration was different enough that it didn’t make sense to me to put effort into the installer to either a) guess or b) ask configuration questions.

SharePoint already provides a tool to do this: stsadm.exe.  The readme (yes, there is a readme file) provides instructions for deploying the web part to your specific configurations.  I also received notes that the sample script should be more explicit.  Again, people haven’t always been installing the binaries to the same location so if I was explicit, I’d be wrong in some instances.  I can’t please everyone :-)

The first step (after running the MSI which installs the binaries to your machine) is to run the stsadm tool to add the web part solution:

stsadm -o addsolution -filename %YOUR_INSTALL_DIR%\SharePoint.WebParts.Rss.wsp

Note the %YOUR_INSTALL_DIR% is not a literal you should be typing.  This refers to where you installed the binaries.  If you didn’t change any settings it would be this:

stsadm -o addsolution -filename "C:\Program Files\Tim Heuer\RSS feedreader Web Part\SharePoint.WebParts.Rss.wsp"

Notice that if you have spaces in your path you must put them in quotes.

If you have problems/suggestions log them as a bug/work item on the Codeplex site please.  If you have an idea and would like to contribute, the source is available!

The web part package you may have noticed is not packaged in a WSP file for easier deployment to Windows SharePoint Services v3 and in fact, that is a new requirement (WSS v3) for this update.

This work is licensed under a Creative Commons Attribution By license.

Wednesday, June 04, 2008  |  From feedreader

So you want to read an RSS/Atom feed on the interwebs and saw the SyndicationFeed class you could use in Silverlight to give a nice RIA display of the syndicated data.  Great, no problem right, just wire up an WebClient, point it to the RSS feed on something like http://silverlight.net or something and boom, done.  Wait, what’s this 404 Not Found error?  In most cases this is going to be a result of a cross-domain issue.  If you haven’t started working with services yet, Silverlight requires a cross-domain policy file to be in place to access remote data not on the same site-of-origin of the Silverlight application.

If you want to learn more about this in further detail you can read this and view this.

Crap.  So now what do you do?  You don’t have a server that would enable you to write a proxy service and you don’t really have the time to do that.  Aha, enter some free services for you!


First, depending on what you are trying to do with the data, give Popfly a look.  Popfly contains several templates for importing syndicated information and displaying it in different visualizations.  For instance in about 4 clicks I can import an RSS feed, connect it to a visualizer and have this:

Check it out…you may not need to do anything else!

Feedburner and Yahoo! Pipes

Pipes is similar to Popfly but doesn’t really provide a breadth of possibilities of visualizations and ease of mashup of way different types of sources, but for this purpose I think it works well.  In Pipes, you can create an input feed and map it to an output, even merging various sources together.  The end result can be a new RSS feed for you.  And Yahoo Pipes already has a cross-domain policy file in place for Flash (which Silverlight supports).  You have to change your endpoint URI a little bit and it wasn’t clear until I searched, but for example, here is a RSS feed URL you could use for combining my blog and the Silverlight community blogs in one.

Feedburner is a syndication service that does a lot of statistics of your feed, helps you manage subscriber data and can save you some bandwidth as well.   It does RSS really well (and enclosure support, etc).  Best of all, it also supports cross-domain policies via the Flash format (again, which Silverlight supports). 

So if you find a feed that is on a site without cross-domain policy support, you can create a new Feedburner feed, Yahoo Pipe or Popfly mashup and be good to go!

A subtle workaround for getting data from sites that aren’t providing the policy files :-)

Hope this helps!

This work is licensed under a Creative Commons Attribution By license.

Thursday, January 17, 2008  |  From feedreader

feedreader, my sharepoint web part originally built for sharepoint 2003, has been consumed by quite a few people (which i think is cool -- it is a rare moment when a lowly d00d like me can fill a gap).  while sharepoint 2007 has better built-in support for syndicated feeds (actually 2003 did already with the xml web part), i still think feedreader has some advantages that can be leveraged.

a while back i put feedreader on codeplex, an open source sharing ground.  i used the microsoft public license so anyone can download, alter, and profit from the code.  i did this because i was getting feature requests weekly as well as some of the problematic issues that i never fixed (namely proxy server support needs to be better).  i thought to myself that there way better developers out there than me and can actually help out.  i wanted to prove my own theory wrong.

well, there wasn't a ton of contributions (more work items though), but over the past month i got another one of those requests.  and just the other day, the rss feed from my codeplex project source code check-ins delivered this to me:

1) Added the code that jdenicola suggested in the codeplex discussion forum to fix the object reference... error. The error is caused by the cache being empty. Which is weird since the PartCacheWrite line is being called, but when it reads it doesn’t return anything. Might be environmental, but I didn’t spend too much time trying to figure it out.

2) Changed the cache key from this.Parent.ClientID to this.ClientID (neither key affected the results of PartCacheRead)

3) Moved the assignment of the graphic to the web part code from the .dwp. This fixed the display of the icon at the top right in the title bar, but it still didn’t display the graphic in the web part library. Could be a SharePoint bug not reading the property. The properties PartImageSmall and PartImageLarge are obsolete. I used TitleIconImageUrl and CatalogIconImageUrl (but still couldn’t get the catalog icon image to show up)

4) Repackaged it in a .wsp for WSS 3.0. I included a new manifest file, a couple .ddf’s, and some stsadm commands to handle the install/upgrade from a build event.

5) Added .snk files so assemblies could be strong named and added to the GAC

6) Added setting for "Expand Headline Descriptions by Default" which, when checked, will expand the item descriptions when the page first loads instead of showing them as collapsed (assuming the headline descriptions are not hidden)

7) Added a div tag with a class called sg-item-description surrounding the item descriptions which the user can overload to control the look (background color, border, text size, etc.) of the description

sweet.  a contribution...and a good one at that.  several things were fixed and a few added.  the contributor, ryan mcintyre really stepped up because he saw some things he could fix and leverage for his own benefit as well.  i should also note that flickr4writer also got some contributions from josh holmes a few months back as well.  it is cool to see some progress on something i hadn't had the time to work on and to that i say 'thank you' to ryan for making feedreader a better experience!

Sunday, February 18, 2007  |  From feedreader

over the past few weeks i've been getting a few requests for if my web parts work on sharepoint 2007 (also referred to as moss).  well to be honest, time has been getting the better of me and i hadn't had a chance to test.

well tonight i fired up a windows sharepoint services 2007 instance and loaded the web parts AS IS to see what they did

after that i loaded them up.  here's the flash web part in action:


and here's the feedreader in action:


they rendered without issue for me.  there is something funkadelic in the upper left corner that i didn't bother to investigate (i'm sure a weird image reference).

as a reminder, the feedreader does not work well at all with proxy servers requiring authentication...and the recommended configuration for proxies anyway is to use the sharepoint web.config defaultProxy settings.

Sunday, August 13, 2006  |  From feedreader

it's official! for the past few weeks i've been able to use a new tool just released, called windows live writer.  at it's core, writer is a blogging client (i'm sure there will be other uses for it in the future, but for now, this is what i use it for).  after seeing the announcement of the tool, i was skeptic.  YABC is what i thought (yet another blogging client, for those not familiar with the YA* acronyms).


but i bleed windows blue, so i installed it.  i should point out at this point that i amwas a blogjet user and gladly paid for it as it iswas a great tool for me having all the features i needed.  i loved it.  really.

my initial reaction to writer was 'sweet' -- until i started digging into features.  the writing experience was great and familiar, and the style preview in real time is great.  but i quickly noticed some lacking features.  i pinged the team and pointed them out.  after going back and forth, i was happy to see they started to trickle in to the daily teams.  it was great to see the agility of this writer team and willingness to take feedback.

i then noticed a few glaring missing things: integration with tagging, integration with flickr, etc. -- the things i loved about blogjet.  have no fear, writer has a plugin model and is extensible.  sweet.  after reviewing the sdk i came up with a flickr plugin that allows you to look at your flickr images (or others) and insert a reference to it.  yeah, it's simple, but that's what i loved about blogjet's simplicity as well.  the great thing is that the writer extensibility layer is .net -- nice! 

i wrote a quick and dirty tagging plugin as well that allows you to template some items.  i suspect this will be replaced by a much better one, but for now it does the job for me.  here's a screen shot of the tagging plugin:


and i also wrote a flickr image plugin that looks like this:


i would encourage you to download writer now and start using it -- take it for a spin...i think you'll like it -- and the extensible model.  give the team feedback.  let me know if you have any questions.  write plugins.  I hope to have my plugins available on codeplex as soon as the project gets approved.  please subscribe to this blog to know when that happens.

what are you waiting for? 

tags: , , , , , , ,

Thursday, June 29, 2006  |  From feedreader

following releasing code for feedreader, the source for doclibutils is now available as well.

this is far less mature and has a few known scripting issues, but now those can tinker away!

codeplex: DocLibUtils

Wednesday, June 28, 2006  |  From feedreader

well, for those that have been emailing, i hope you stayed subscribed.  feedreader has finally found a home and the complete source code is available. 

as i’m sure most of you have reflected the code ;-) you’ll know it isn’t brain surgery, but that there are probably some things that could be cleaned up and we know there are areas that need to be fixed.

so visit the feedreader codeplex today and get the source released under the creative commons attribution-sharealike 2.5 license.

remember, the purpose of open source is to share :-D, so be sure to contribute back your fixes and log issues as well!

 feedreader News Feed 

Last edited Dec 7, 2006 at 11:16 PM by codeplexadmin, version 1


No comments yet.