PlkrData
Channel Import/Export

What is PlkrData?

PlkrData is a small utility for importing and exporting channel data. "Channels" are the way Plucker refers to web-site presets; they define the URL, the name of the resulting channel, the depth, and so on.

Why do we need it?

Channel data includes the starting URL, which may not be intuitive and may be quite long, as well as depth options, stayonhost/stayondomain options, defining icons, and more. This information would have to be deduced, or at least cut-and-pasted into multiple fields on a form, to load it from a web site into Plucker. PlkrData allows content providers or aggregation websites to create easily-downloaded .plkrdata files that may then be executed to create the channel in Plucker.

Still To Come

The primary To Do left is IPC with Plucker Desktop, such that Desktop refreshes when a channel is loaded via PlkrData. This awaits some development time from the Desktop team (possibly including me<g>).

PlkrData also does not yet support programmatically determining the current JPluck fileset or exporting from JPluck .jxl files to .plkrdata files.

Hey! It's missing some fields!

Possibly. I went through the Plucker manuals to try to discern which fields to include and which to exclude. There are only two required fields, home_url and doc_name, and 14 optional fields. Other fields are ignored (or prohibited.)

Adding additional fields is absolutely trivial. They go into the ieEntries aggregate, but please simply email them to me at . (Can you tell I get too much spam?)

What about translations?

That too is trivial. The source file plkrdata_english.h contains all the translatable strings. (Strings that shouldn't be translated aren't in it.) So grab that, create a new plkrdata_language.h file, and I'll add it to the mix.


Usage: PlkrData [channelname] [filename] [/help || /load || /save || /jxl=filename || /register]
Your OS may use a different switch-character, such as "-".

Channel NameSpecifies the channel to save (export) from Plucker.ini, or to load from filename. "*" specifies all. Wildcards are not supported; * is a special case.
Warning: "* /save /jxl=filename can result in a very large number of warnings filling the screen if several imported channels use spider-limiting features not supported by JPluck.
File NameFile to load or save to. If not specified, assumes "channelname.plkrdata".
/helpPuts up the command-line help
/loadLoads the data from the specified .plkrdata file into Plucker.ini
This is what the execution action is.
/saveSaves (exports) channel data for specified channel to a .plkrdata file, named after filename if specified or after the channel otherwise.
If you do not fully qualify a destination file, it goes into $PLUCKERHOME.
/registerAdds PlkrData to the (Windows) registry so that files may be point-and-clicked on. Note that some browsers may need additional configuration to recognize this.
The resulting key is HKEY_CLASSES_ROOT\plkr_data\shell\open\command with the value Install directory\PlkrData.exe "%1" /load. You can check this in RegEdit. For JPluck, you would want to predetermine a JPluck file, putting that in the value like c:\Palm\Plucker\PlkrData.exe "%1" /load /jxl="c:\JPluck\MyStuff.jxl"
/unregisterRemoves PlkrData from the (Windows) registry. Essentially uninstalls without deleting files.
/jxl=filenameOutputs channel to a JPluck-compatible JXL file.
Note this only currently works with /load; exporting from a JXL is not yet supported. Also the JXL format does not support prioritizing of inclusions/exclusions, so there is only a single level of them available.


Examples

To export the channel
channelname
from Plucker to a .plkrdata
plkrdata channelname /save
To load the channel
channelname
from PlkrData to Plucker.ini
plkrdata channelname /load
OR plkrdata channelname.plkrdata /load
To load the channel
channelname
from PlkrData to JPlucker file toto
plkrdata channelname /load /jxl=c:\toto.jxl
OR plkrdata channelname /load /jxl=c:\toto.jxl
To load all channels from PlkrData to JPlucker file c:\JPluck\Channels.jxlplkrdata * /load /jxl=c:\JPluck\Channels.jxl
To save all channels from Plucker.ini to file c:\output.plkrdata
Note that the resulting .plkrdata will not autoload from a web browser.
plkrdata * c:\output.plkrdata /save
You can export from Plucker.ini and import into your JPluck jxl file in one step by using the Plucker.ini explicitly as your source... i.e. as your .plkrdata file. Like this...
Note that you should fully qualify Plucker.ini, although it should find it just fine
Import/Export in oneplkrdata Forbes d:\palm\plucker\plucker.ini /load /jxl=c:\toto.jxl

History

21-Dec-02, Build V2 Beta 5 - adds

19-Dec-02, Build V2 Beta 4 - adds

17-Dec-02, Build V2 Beta 3 - adds

13-Dec-02, Build V2 Beta 2 - adds

12-Dec-02, Build V2 Beta 1 - adds

21-Nov-02, Version 1.0 - adds

20-Nov-02, Beta 3 - adds

Files

Windows binary: PlkrData.exe

The source code is complete and current, but requires TinyXML and wxWindows to compile. TinyXML is copyright (c) 2000-2002 Lee Thomason (www.grinninglizard.com) - Thanks Lee! - and included in the ZIP file; wxWindows must be downloaded/installed separately.

Source code: plkrdata_src.zip


Sample .plkrdata files

The following files can be loaded into your Plucker.INI or *.jxl file directly from your webbrowser by PlkrData. First download PlkrData and let it register the filetype. If you're using JPluck, you will have to customize the registration for now by adding the /jxl=your main jxl filename inside the registration area. After completing this, some browsers will let you point-and-click to load the .plkrdata files. Other browsers may require an additional step telling them to let the OS handle this filetype.

Site NameDescriptionComments
SlashdotNews for Nerds, Stuff that MattersUses exclusions, which are not yet supported by JPluck
BBC NewsNews out of England
New York TimesNews out of America
NorthWest Cable NewsNews out of Seattle
USA TodayAlmost looks like news
Business WeekBusiness news

The Windows User Guide, which is still a work-in-progress, is here. I got a good start on it, and then got delayed by travel and by the belief that items such as the above-utility would do more for hapless users than a user's guide would.

This next page has since presumably been folded into CVS tree, something I don't have access to do. It is ImageParser.htm, allowing the prevention of alternate images.

And from before that, this Spider page implements "--stayondomain". This one has made it into CVS by now.