public virtual MemoryStream

Thanks for fixing the "Open In New Window" bug. It was a pet peeve of mine since FireBird is my default browser. Now it's consistent.

Regards

Senkwe

SharpReader 0.9.2.1 has been released

Equal justice...What!
I haven't addressed this issue yet, but...after reading this site. I have to give my 2 cent$ I have no opinion on his guilt or innocence, given that there are no facts yet out about the case. What I do...

New upgrades
Well.. it seems as if some people out there have been very busy lately.. I wish I could say the same thing myself ;( Anyhow,...

Luke -

Seems like a shift-click on a link (open in new window) now takes forever. It used to be instant. Is this because you're checking to see which browser to launch? Could you do this once, at time zero, and remember the answer?

Ole

Luke,

I'm not sure when this happened (probably with this version) but Open in New Window from the browser pane context menu doesn't open a new window any more - the page opens in the last used window instead.

I'm using Internet Explorer 6 as my default browser and I'm sure this used to work. Any chance of a fix, it's a really quite irritating bug.

thanks for the excellent software.

Algunas novedades de RSS
Al RSS se le está prestando demasiada atención. Lo que originalmente servía para tener un acceso diferente a las novedaddes...

Every time I start up SharpReader, none of the feeds are updated. Every icon has the lightning bolt on it. If I go to Task Manager, the app is taking up about 170 Meg of memory, 19 threads and 510 handles (FWIW ;) ).

This could be an issue on my box, but I've never seen this behavior until I installed the new version. Luke, any ideas?

Regards,

Jason

Time to upgrade - again (SharpReader)

New SharpReader Released
On Tuesday Luke Hutteman released a new version of SharpReader. SharpReader is a .Net based news aggregator. At this point,

I still get the same "non refreshing" feeds problem I reported before. Bit like Jason, but not as severe.

And now you've pointed it out, I see I'm hitting the cached copy of feed bug in the version I'm stuck with. I've missed a number of comment threads cos of that in the past but put it down to our proxy/bad timing. Now I can see it is that bug :(

Opening in the default browser is great, thanks. However, I also would prefer being able to have open in a new window work. In my install, it also opens in the most recent browser window, and if provided the option, I would choose to have a new browser window opened each time. (Call-out to Simon's post, above).

Is not having the spacebar scroll the webpage displayed for items without a description just the tradeoff we have to make to not have crashes (I'll take that tradeoff!), or a, um, unintended side-effect of the fix?

Geek Weakly 08.01
A 250 mile race across the dessert by robot cars. A whole set of artistic robots. And now, a robot grill at McDonald's. Those of you who scoffed at that 50-year vision of a robotic nation from last week's report may want to think again! At least we can...

Are you by any chance doing the "is it new, old, or changed" test before you do some security/fixup changes that make it into the saved item? I'm seeing /item/9619/ in revjim.net/rss20/ as a changed item every read, and other than some inline style in a form that you might be securing, I can't see why.

(I suppose showing a (optional) diff view on changed items would require a big change to how you save items, to allow for saving two versions?)

Phil: that particular blog-entry contains a hidden input-parameter like

<input type=\"hidden\" name=\"bBlog\" value=\"11ded6cc539fbbf83543a7e0e599cefe\" />

it uses a different regenerated random value on every refresh, which is why SharpReader marks the item as changed.

After Mark applied the Winer Watch to his own blog, I did consider putting diff-like functionality in SharpReader - though only for previously read items that have since changed. Unfortunately I'm not aware of anything like difflib or diff.py for .NET so I'd have to dust off my old college text books for the algorithm first.

Since in this particular entry, the change is in a hidden input-parameter though, I'm afraid it won't have shown even with a diff implementation.

Oh and the spacebar scroll-down not working for items without a description is a bug that has nothing to do with the crash-fix. I'll see if I can get that fixed for the next release.

Simon: in Internet Explorer, go to Tools|Internet Options|Advanced and make sure the option "Reuse windows for launching shortcuts" is not checked. That should fix your problem.

Owen: Since you like the default browser functionality, I take it you're not an IE user. In case you're using Mozilla/Firebird, Phil has some great tips on setting that up for use with SharpReader.

Ole: Starting a new browser is now done similarly to typing a URL in the Start-menu's "Run..." dialog. Does that go slow for you as well?

Interesting. I was on the verge of calling you crazy, since a random hidden input was the first thing I looked for, when I thought to view source in the embedded browser rather than Firebird. For reasons beyond the ken of mortal man, at least this copy, if not this version, of Firebird doesn't display a type="hidden" input in view source, or include it when it saves the page. Very odd, and if it's not just something with this install, very buggy.

I tried that page in my Firebird this morning (had only checked through view-source from SharpReader last night) and a view-source did show the hidden field for me. Then I tried it again to verify that the value had changed, and the hidden field was gone! So I tried Internet Explorer instead and again, no hidden field.

Then I checked the site's cookies, and saw a long hex-string in there just like the hidden input field showed the first time. After removing this cookie, the next page-refresh contains the hidden input again (and set the cookie again).

So this is obviously some kind of session tracking, and the reason it always shows in SharpReader is because SharpReader ignores cookies.

Some of My Favorite Things
SharpReader 0.9.2.1 has been released, and has fixed a bug that wouldn’t use my default browser (Firebird) under certian circumstances. I’ve now switched my desktop to use it as my default as well. My favorite instant messenger Trillian Pro...

Suggestion: Mouse buttons for back and forward navigation are not working in the browser view. I am addicted mouse page navigation.

Lovely software, Luke. Minor refinement request: a hot key to open or close the feed property box.

Luke,

There's a really annoying bug I've found...
Sometimes, when you delete an item, it doesn't actually get deleted (!)
Altough it disappears momentarily from the list, the count in the feeds tree is still the same, and when you go to another feed and back, the item is there again...

I'm having the same memory problem Jason Bock mentioned above on July 31.

It started with release 0.9.2 and was not rectified in 0.9.2.1.

I use it on two machines, one Windows XP Home Edition, the other Windows 2000 Professional Edition. On both machines it regularly eats more than 100 megs (!) of memory, and frequently I just have to kill the process, because it keeps creeping up and up, a few dozen kb at a time. I've watched it creep up to 156 megs on one machine. I usually kill it before then, though.

This problem has made SharpReader unusable.

Also, the Auto-Purge appears not to work at all.

--g

I just found this program, through links on http://rss.lockergnome.com/ (I spend way too much time hanging around Lockergnome), and through recommendations of a couple friends. Already having .NET Framework 1.1(I'm learning C#), for me, it was a very small download for what I consider to be a great program.

Let me say that I am new to Blogging, but old - very old - to computers (suffice it to say I worked on computers during the Johnson Administration. I have become a fan of reading blogs and use RSS as my resource for a number of what I used to receive by email. Sharpreader was a major part of the drive for me to embrace RSS. I would like to suggest that the option switch for opening a link in an external browser be added to thye context menu as well? Just a thought.

Thank God! I thought I was the only on the planet who spent too much tiume hanging out in the Lockergnome feeds!

SharpReader cannot be ngen'ed. And it's bad since the startup time is really...well really annoying.

In fact ngen keeps complaining about a dependency with nunit.framework.

I am using sharpreader for external feeds (using proxy) and for internal ones (developed inhouse).
Would it be possible that sharpreader reads also the proxy exception from IE or at least add a "skip proxy" to the properties of each feed?
This way I would not have to switch the proxy on and off all the time depending on which feed I want to look at.

Only thing I really notice on a regular basis is that after #Reader updates, often some entries that were previously read will show up again as being "new" but still in italic font indicating that's it's "read." And it seems to choose random feeds to do this to. I thought perhaps a trackback entry somewhere might be causing it, but that does not seem to be the case.

I can't load #reader. It just doesn't show up. I am wondering if this is because I am using the .NET binary from VS.NET instead of Windows Update. Are these the same?

Hey, Luke, you're SharpReader.net URL is in an infinite loop and keeps going back to your Weblog. I can't get it (SharpReader) if I can't find it.

I have a problem installing SharpReader. At first I ran the .exe file and it complained it couldn't find netfx.msi, so I figured, I'd better reinstall .Net framework, so I first removed both .Net 1 and 1.1 then went to WindowsUpdate to reinstall. After that I get an error saying: Application has generated an exception that could not be handled."

I'm on Win XP Pro with all service packs and updates installed.

Sounds like SharpReader is really cool, I'd like to try it out...

i have the loop to, would be a good thing to place a download here button on the side
CAN NOT DOWNLOAD SHARPREADER

-Hannes

Enhancement request: I would like to store the configuration and cache files in another configurable directory. "<user name>\Application Data\SharpReader" seems to be hardcoded. Or am I wrong?

Sorry about sharpreader.net's unavailability - I was trying to setup a redirect yesterday from hutteman.com to hutteman.com/weblog but this inadvertently also made sharpreader.net redirect there (they're both hosted on the same account).

The redirect has been removed and sharpreader.net is available again now.

Lovely program!

Bugfix:
- When an item is locked, the context menu ought to read "unlock item" to be consistent with mark read/unread semantics.

Some suggestions:
- Hotkey for Feed Properties.
- A way to "save" the internal state and open it on another SharpReader instance (think work & home PCs, and not wanting to go through all the feeds twice marking/checking/reading the articles).
- Better (some?) docs / outline of the functionality. For instance, what is "lock Item"
- Source code donwload? :D

All in all, a great program, and helpfull to boot. Mant thanks

Thank God! I thought I was the only on the planet who spent too much tiume hanging out in the Lockergnome feeds!

Hi, Luke. I love SharpReader. It saves the labor to go round Web sites dramatically. Thank you.

I have one request.
I run closed (staff only) blogs and Wikis, which demand usename and password to enter (BASIC auth.).
If SharpReader supports a format "http://username:password@domainname/", I can access RSS of these pages with SharpReaer...
Could you consider it?

Hey Luke,
Can you read sometimes your mail :p

(sharpreader@hutteman.com)

This is the only place I found to post this (is there a whish list?):

I woulk like to view my yahoogroups messages using RSS, but their version isn't parsed by SharpReader. I get a parsing error. Could you make SharpReader a little more "tolerant" :)

You can view any group messages by going to http://groups.yahoo.com/group//messages?rss

Nice work!

Which group are you trying to access? It seems like groups that have their archives set as "for members only" will return a login form to SharpReader instead of the actual feed. Since SharpReader does not store any cookies and does not allow for user-interaction when downloading the feed, it is unable to do anything with this response.

For instance this group has public archives and is parsed just fine in my SharpReader, while this one has its archives set as "for members only" and returns with a login screen which subsequently causes SharpReader to display a parsing error.

If the feed you are trying to access does have public archives, please let me know what Yahoo group it is for so that I can try to figure out what the problem may be for that particular feed.

Would love to have an option added to SR that, if turned on, would select the first unread story in a feed when that feed is selected in the left column.

It would be nice to have a feature where SR would automatically move feeds not updated in a long time say two weeks to a separate category.

This is my first experience with RSS -- SharpReader is great! I'm loving it already.

I do have one problem with the program (have the latest version): I can't seem to drag more than one feed into a category. Does anyone else have this problem?

Figured out the folder thing -- it was user error. {sheepish grin} Once a folder has one item in it, you can no longer add items by dragging them on to the folder. You have to drag them on top of an item inside the folder.

Kyle: After you drop the first feed on the folder you have to drop all the other feeds onto an item in the folder, not on the folder itself. Otherwise there wouldn't be a way to move a feed to be right before an existing folder 'cause you wouldn't know where to drop it.

Luke,
I was working on a project that has (roughly) similar in-memory dataset requirements to Sharpreader. The solution I'm planning to implement might be usable by Sharpreader to cut down on the memory requirements.

What if Sharpreader used multiple serialized datasets and only loaded the ones necessary for display. For instance:
All Feed names and subjects could be in a primary dataset
The Feed data (the body of the post) could be in separate datasets, one for each feed
- or -
The Feed name, subject and body could be stored together but separate the datasets based on the user category. That way we could categorize things to cut down on the memory (if we could turn off the global feed aggregation view). (I rarely want to look at the newest stuff in ALL of my subscriptions at the same time.) As long as we broke the categories into managable sizes we could avoid having Sharpreader use up 100MB of our RAM. All uncategorized feeds could go into a separate dataset as well. Even with global feed aggregation you could probably do this - I've found that deserializing a dataset is pretty quick so it would only be a minor delay for the user.

In either scenario I'm sure the overall feed info (RSS URL and options) would need to be contained in a separate dataset.

What do you think? Decent workaround until something like SharpHSQL is production ready?

Date Tag Issue:

There are many site's who are using tag for date field.
plasticbag.org, wired, sun are those.

SharpReader will not support that tag ?

Conan: I'm not sure what kind of tag you are talking about, but the three feeds you mention only contain a title, link and description per item. They do not contain a date in the feed. In the absense of a date field, SharpReader defaults it to the date/time it is parsing the feed.

Any chance of intergrating IE security & credentials so you can access pages stored on company intranets which are protected with Intergrated Authentication?

Thanks!

Jack: yes - http authentication is planned for the next release.

A couple of changes, since you mentioned a new release is coming:

1. Better use of the Tab key. I'd like to have only four potential 'landing places' when pressing Tab, in a circular order: the URL box, the current element in the subscriptions list, the current article in the list, and the article window. Focus should shift accordingly.

2. Better memory of what has focus. Any time SR loses focuses (such as when I ALT+TAB away to another app) and then comes back, focus is set to the current article in the list (in the top right window, that is) even if before losing app focus, focus was set to the article window. Even using CTRL-F to search within the current article causes this change.

Thanks again!

Does Robert Scoble ever have nothing to say?

Does Robert Scoble ever have nothing to say?

I'd like to add my vote for http authentication as soon as possible. We just discovered RSS recently and now can't get enough for ourselves and clients we support. SharpReader is the tool we are recommending to everyone, but we have a number of private intranet sites where basic authentication is required. So... please. Any scheduled release date???

Again Date Tag:

Sorry. I omitted exact URL's.
They are as follows:
http://slashdot.org/slashdot.rss
http://www.wired.com/news_drop/netcenter/netcenter.rdf
http://www.plasticbag.org/index.xml

Anyway, I think they use RSS 0.91 or 1.0, while SharpReader use RSS 2.0
But, It seems that only Date Tag makes problem. If SharpReader support Tag, everything will be OK.

Another critical problem:
SharpReader stops refreshing after using a few weeks. It does not refresh URLs and cannot get feeds. I should remove the URL's and re-open them.

Conan: SharpReader supports RSS 0.9x, 1.0 and 2.0 and accepts both <dc:date>'s and <pubdate>'s in a variety of date-formats. Of the three feeds you mention, only slashdot actually contains a date (wired and plasticbag only have item, link and description) and it shows up fine in SharpReader for me.

I'm not familiar with the problem you mention where SharpReader stops refreshing. Please email me your sharpreader.log file (from C:\Documents and Settings\[USERNAME]\Application Data\SharpReader) and I'll see if I can figure out what's going on there. This may also help me figure out why you can't see the dates for the slashdot feed.