Success?

[Dave Winer]

Has anyone gotten River3 to work?

Problems? Happiness?

Dave

[Andrew Shell]

I've started poking around but then I had to help move so I'll come back to it in a bit.

--
You received this message because you are subscribed to the Google Groups "frontier-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email to frontier-use...@googlegroups.com.
Visit this group at http://groups.google.com/group/frontier-user.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[john]

hi,

My initial attempt:

https://dl.dropboxusercontent.com/u/81715/river3/html/nytRivercopy.html

just uses your nytRiver opml

BUT 

the html file produced had:

var jsonFeed = 'http:/rivers/nytRiver.js?callback=';

Which I changed to:

var jsonFeed = 'https://dl.dropboxusercontent.com/u/81715/river3/rivers/nytRiver.js?callback=';

in a duplicated html file pointed to above.

Seems to work fine although this laptop will not be running the OPML app all the time.

I guess fargo is the best place to create a few more opml files?

cheers

john

[Mark Yeager]

I've seem to have it running on my EC2 instance with Dropbox newly installed, and I am seeing folders getting populated in Dropbox, but slowly. Since it is a 1st time synch, it might be a while before it finishes.

I added the NYTRiver.opml.xml to my Dropbox/river3/lists folder.Right now there are the folders Lists, Misc, Rivers, Templates I assume successfully generated by River3.

How do I force a "scan now" like you can in the OPML app for River2? I assume there will be an index.html in the Rivers folder that I will point my browser at? Thanks,

Mark

[Andrew Shell]

Dave answered that in another thread:

There's no command to do a manual refresh in River3.

You can do it by running the script river3suite.scanner. Just open it up and click the Run button.

I will consider adding it as a feature in a future update.

Cheers,
Andrew

[Andrew Shell]

It seems to be working.

https://dl.dropboxusercontent.com/s/0ny9q1bdw87nopo/andrewshell.html

Same as John I had to manually change the url for the .js file.

Next I'll play with saving to S3 and see if that fixes the issues with the river js file.

Cheers,
Andrew

[Dave Winer]

Looking at the code it pretty much assumes that you're saving the JSON file to S3 when building the river. 

I could probably make it work locally, but it wouldn't be worth much it seems if it didn't work publicly.

I am preparing an update that adds Scan Now and Open Folder commands to the sub-menu of the Tools menu.

Dave

[Eric Davis]

Just updated to the latest river3.root (0.44) but still no dice. When I run river3Suite.scanner it errors out with:

Couldn't create the folder "/Users/eric/Dropbox/River3/lists:", because "The folder "/Users/eric/Dropbox/River3/lists:" wasn't found."

Debugging my way through it, the 'file.surefilepath (listsfolder + "xxx")' line is where it bails -- if that helps.

I set the base folder to "/Users/eric/Dropbox/River3/" and I've tried every permutation of having the both River3 folder and the "lists" subfolder exist and not exist.

I'll keep playing with it. I imagine it's something simple I'm just overlooking.

[Dave Winer]

I must have missed the first report.

Can you narrate a little more, slowly and carefully. 

What steps you did that led to this result?

Dave

[Dave Winer]

The path probably should be:

Macintosh HD:Users:eric:Dropbox:RIver3:

Dave

[Eric Davis]

Woo-hoo! That was it.

Thanks.

[Andrew Shell]

It looks like it's working with S3.

My rivers are:
http://static.andrewshell.org/html/andrewshell.html
http://static.andrewshell.org/html/asdonews.html
http://static.andrewshell.org/html/web608.html
http://static.andrewshell.org/html/datecheckpro.html

The current River2 urls are:

http://river.andrewshell.org/ <- My fast river

http://news.andrewshell.org/ <- Slow river

http://www.web608.org/ <- Local Madison tech news

http://river.datecheckpro.com/ <- Grocery news for work

However most of these are slow rivers so it will take a few days for them to fill up.  At that time I'll swap out the source of my various rivers powered by River2 and then delete them from my server.

I'm not sure if I'm just being slow here, but how do you change the title and description?  I was thinking it would be the title and description from the OPML subscription list but that does not seem to be the case?

Also, sort of unrelated and very low priority but how on earth do I edit the title and description  of an OPML file from within the OPML editor?  I had to open the file with a text editor because I wasn't sure how to do it.  I tried double clicking the window and changing the items under /attr then saving but that didn't seem to save the changes.  Oh well.

Cheers,
Andrew

Dave

[Dave Winer]

So far I've been the only user, and I haven't needed to set titles and descriptions.

I basically tested and documented what I had. 

I'll try to fill feature request or bug fixes, as I can -- this is probably a little of both.

If you can help make that easier for me, I'll probably be able to do it sooner than later.

Dave

[Mark Yeager]

But there are no tabbed rivers. How does that work with River3? I can't even get one Dropbox page working, and something is getting deposited there.

Mark

[Theron Kelso]

Tried this morning, Error: Installed, updated the everything: opml.root, rivers3.root, etc.

Error:

"Can't decompile the table to an XML string because the table is empty."

Go To > takes me to rivers3Suite.buildFeedList and put the cursor on "file.writewholefile (f, xml.decompile (@listtable))"; immediately after the "(@listtable)".

Troubleshooting:

1) In OPML Server > River3 > Preferences > Scanning, I unchecked; Check this box to enable the aggregator. No error messages occured, but nothing happened.

2) rechecked box from 1, receive new message: "Can't find a sub-table named "item #1". Click on Go To > takes me to river3Suite.initFeed - new (tabletype, @adrdata^.prefs)

3) Reinstalled. Better results... no more error messages. It created four new folders: reiver3lists, river3misc, river3rivers, and river3templates at the *same* level as my river3 that I identified in Folder. Still nothing.

Configuration:

Folder: C:\Users\tkelso\Documents\Dropbox\Public\river3 (Note: in river3Data.root, root = C:\\Users\\tkelso\\Documents\\Dropbox\\Public\\river3)

Within: "C:\Users\tkelso\Documents\Dropbox\Public\river3", I have one subfolder: "lists", with one file: "nytRiver.opml"

Not using S3

tired....

Theron Kelso ~ theron...@gmail.com ~ (703) 237-3430

Dave

[Mark Yeager]

On Jun 24, 2013, at 10:01 AM, Theron Kelso <theron...@gmail.com> wrote:

> Tried this morning, Error: Installed, updated the everything: opml.root, rivers3.root, etc.

> 3) Reinstalled. Better results... no more error messages. It created four new folders: reiver3lists, river3misc, river3rivers, and river3templates at the *same* level as my river3 that I identified in Folder. Still nothing.

Make sure in the path to the dropbox folder in River3 prefs that you have at the terminus a "\" e.g.. "….\river3\" not "….\river3" and delete the other folders

Mark

[Dave Winer]

Are you still getting the error or did it clear up when you re-installed?

Dave

[Theron Kelso]

No more errors; but the folders never get created. The nytriver feeds and feed items showup in river3Data.root and running saveLog() creates the logs folder, with log showing import of 53 feeds. one step closer. 

Theron Kelso ~ theron...@gmail.com ~ (703) 237-3430

[Mark Yeager]

I switched over to my Mac because I couldn't get River3 to run on my EC2/S3 setup. River2 runs fine there.

On the Mac I get the same error as Theron described "Can't decompile the table to an XML string because the table is empty.", but I DO get folders appearing in my Dropbox rivers3 folder.

Mark

[Dave Winer]

I just released an update that will make sure it doesn't try to build the feed list if there are no feeds. 

[Dave Winer]

BTW, none of the output folders will be created until there's at least one OPML subscription list in the lists folder.

That's how it's supposed to function. Until you've given it at least one list, it has nothing to do.

Dave

[Mark Yeager]

Thanks-- no errors and folders all there and updating seems fine with nytRiver.opml in the lists folder. The nytRiver.html says "Can't display the river because the response from the JSON server is not valid"
With that, the hard part for me is what to do next… :-)

Any pointers or tutorials appreciated.

Mark

[Dave Winer]

Next thing to do: Set up the S3 connection.

Dave

[Mark Yeager]

I have an S3, just like for river2, and setup a river3 bucket. I even changed to new keys, but when I submitted, I used to get in OPML prefs for S3  " Amazon doesn't like security credentials", but I just tried and they did like them, so something has changed.

But nothing else has changed with river3.

[Theron Kelso]

In Short: Success! 

In Long: My Notes.

The start of a Washington Post River is here.

Thank you.

Theron Kelso ~ theron...@gmail.com ~ (703) 237-3430

[Dave Winer]

I'm glad it's working for you now! :-)

However -- what isn't clear from the notes is what you did to get it to work.

Dave

[Theron Kelso]

One of two things. First the path. my first attempt was without the trailing "\". It created some folder up a directory from where I expected. Deleting them, didn't automatically recreate them. OR "Check this box to enable the aggregator." I rushed this without the folder structure in place - because of course, I "knew" I wanted to enable the aggregator. It did leave me wondering if I just rushed the process - even this time. It seemed like there are points, where you just have to wait to let things happen. (Example: once the example opml file is in place, OPML Editor has to cycle through the schedule.)

Theron Kelso ~ theron...@gmail.com ~ (703) 237-3430

[Dave Winer]

Thanks for explaining. It helps me understand what happened on the other end. ;-)

Originally I was going to ship it without a UI at all.

The prefs interface was put together in about an hour. 

I didn't have a lot of time, but River3 was pretty well debugged and it seemed a shame to have people still running River2 with this so far along.

Dave

[Mark Yeager]

When I am running OPML on EC2 and I input my credentials, I get the above error. I have changed the keys several times, and waited a day to try again. This is new. When I run OPML on my Mac, the exact same credentials are accepted.
Mark

[Mark Yeager]

S3 credentials

On Jun 27, 2013, at 10:39 AM, Mark Yeager <yea...@me.com> wrote:

> When I am running OPML on EC2 and I input my credentials, I get the above error. I have changed the keys several times, and waited a day to try again. This is new. When I run OPML on my Mac, the exact same credentials are accepted.
> Mark
>

[Dave Winer]

Hard to know what the problem is.

Want to try doing it manually? If so...

1. On your desktop machine, bring the OPML Editor to the front, jump to the "user" table.

2. Put the cursor on s3.

3. Choose Export cursor part from the Misc menu.

4. Click OK on the confirmation prompt.

5. A window should open on the desktop and you should see a file called user.s3.fttb.

6. Somehow get that file onto your EC2 machine. (I generallly use email for this.)

7. On the server, drag the user.s3.fttb file onto the OPML icon. 

8. Click OK to all confirmation prompts.

Now you know that the server has exactly the same credentials as the desktop machine.

If it still doesn't work there's a real mystery.

Dave

[Mark Yeager]

Still does not work for the EC2 instance running OPML.

I did the reverse, moving the fttb file from EC2 to the Mac, dropping on icon in dock, and accepting the update. That works, giving the dialog in green "Amazon likes your security credentials."

[Dave Winer]

Have you tried creating another EC2 instance?

If I think of anything else to try I'll post another email.

Dave

[Mark Yeager]

Not yet. May try that later today.