Re: More table questions

[Robert O'Connor]

> What is the relation
> between bit depth and the table? I'm missing something here obviously.

The large pannable tables are drawn using the large 
pannable image code at runtime in the viewer.

The code for the sliceable images will likely get code 
reuse for large pannable tables.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Danish updated for plucker-desktop

[Robert O'Connor]

On 10 Mar 2003 at 10:32, Claus Hindsgaul wrote:

> I have attached an updated translation of Plucker-desktop (and wxwin)

Thanks!

> Please include in cvs.

Will do.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Please announce new versions in Plucker-announce

[Robert O'Connor]

> It would be helpful if new versions of Plucker software were announced in
> Plucker-announce official mailing list.

Hi Ken,

Sure thing.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Table questions

[Robert O'Connor]

Hi Laurens,

> 2) The table cell data function (0x97) has a "columns" and "rows" parameter.
> I can't figure out what their purpose is. Are these needed for nested
> tables?

Perhaps they are colspan and rowspan for the cells.


Rowspan and colspan are both well supported in Chris's table code.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Large pannable tables >64kb (Addition to proposed solution to 64k image limitation)

[Robert O'Connor]

I noticed when testing tables the other day, that if a big pannable table is spidered 
at a high 
enough bit depth, there will be no longer a thumbnail link put into the pannable 
table, instead 
just a [TABLE XX] placeholder is inserted, where XX appears to be the record number.   

Can the image slicing code work for large pannable tables?

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: proposed solution to 64k image limitation

[Robert O'Connor]

I have a fuzzy memory of Bill mentioning on the list that the parser had the capacity 
to crop 
out a specified section of an image (in the capacity to crop to some 150x150 part of a 
big 
image and show it on the palm, instead of shinking the whole image to 150x150). I 
think that is 
what it was about. Here is the message I was thinking of:

http://www.mail-archive.com/[EMAIL PROTECTED]/msg01864.html

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: HiRes Viewer: Possible image display bug

[Robert O'Connor]

On 7 Mar 2003 at 14:40, Ken Stuart wrote:

> Hello,
> 
> I use Plucker Desktop and the HiRes Viewer from hires.plkrg.org to view web
> sites on my Clie NR70/u (OS 4.1).   I generally have the latest code (and this
> morning, I installed both the latest Desktop and the latest Hires Viewer.
> 
> One of the sites has a weather radar image, which is larger than the PDA
> screen.   I have Plucker Desktop set to "Always link the thumbnail to a second,
> full-size copy".
> 
> The image is way less than the 60kb limit, and somewhat larger than a Palm
> screen, but not grossly so, in other words, an ideal test case.
> 
> I hope that either I have overlooked a switch/preference somewhere, or else it
> is a simple bug to fix...

Hi Ken,

Thanks for your letter.

As Adam mentioned, there is currently a 60K uncompressed limit on images. I tested it 
out on 
the CVS parser. Here are the settings I used, which are includeded below (can 
copy/paste into 
your .pluckerrc/plucker.ini). I tried in on 2bpp and seems to work as expected (don't 
have a 
color device so can't test). If doesn't work for you, perhaps was a parser bug that 
has been 
since fixed in CVS. Note in the listing below, that you can control which algorithm is 
used to 
shrink oversized images: either try_reduce_dimension, or try_reduce_bpp. 

As an aside, you can note why the interface declares it as "always link to a full 
sized 
version", since the alt_maxwidth and alt_maxdepth are each set to a million pixels.

Hope that helps,

Best wishes,
Robert

[Testalwayslinkifbigger]
after_command=
alt_maxheight=100
alt_maxwidth=100
alt_text=1
anchor_color=#FF
backup_bit=0
before_command=
big_icon=
bpp=2
category=
charset=
compression=zlib
copyprevention_bit=0
copy_to_dir=
depth_first=0
directory_on_card=
doc_file=channels/Testalwayslinkifbigger/Testalwayslinkifbigger
doc_name=Test always link if bigger
exclusion_lists=
handheld_target_storage_mode=0
home_maxdepth=2
home_stayondomain=1
home_stayonhost=0
home_url=http://printer.wunderground.com/radar/station.asp?ID=MAX19
home_url_pattern=
image_compression_limit=50
indent_paragraphs=0
launchable_bit=0
maxheight=250
maxwidth=150
no_urlinfo=1
owner_id_build=
referrer=
small_icon=
status_line_length=60
try_reduce_bpp=1
try_reduce_dimension=0
update_base=2003-03-08T17:59:00
update_enabled=0
update_frequency=1
update_period=daily
user=Rob O'Connor
user_agent=
verbosity=1
tables=1
close_on_exit=1
close_on_error=1

___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Why is the indent so large for elements in CVS parser?

[Robert O'Connor]

I built the CVS parser today, and I must say, very slick, especially the 
tables--brilliant.

I also had a spin of the hires viewer, very sharp work. I like the reorganization of 
the prefs 
into a PalmOS standard mechanism, and the pannable image arrows especially.

My only question is about the  lists Why do we indent so much (about 2-3 
characters) with 
these on the handheld? The  lists seem to only indent a more reasonable amount (a 
bit over 
1 character or so).

Best wishes,
Robert



___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Commerical Applications

[Robert O'Connor]

> Our releases are pretty silent to the user, they either see no dialogs or
> dialogs where they can just keep clicking next and accept all the defaults.
> We will pre load a channel into Plucker Desktop for them.  We also plan to
> use links on a web site to allow adding more channels if needed (via
> plkrdata).  We also have the requirement to make everything as simple as
> possible - a one click install with no options and then sync the content
> when hotsync is pushed.  We've been successful so far with our previous
> release.

Sounds good.

> Also, Robert has the Windows installer Innosetup
> script available too.

It also is in the Plucker CVS.

> I agree with Robert in that the other parsers haven't been tested nearly as
> much as the Python one.  In my case, I just plan to test it on our specific
> pages for now.

When you feel it is ready, feel free to drop a note, and we can set in the CVS.

> Also, since the Plucker Desktop menus/dialogs/etc are based on XML resource
> files, it might even be possible to disable all the options you don't want
> people messing with just by modifying the XML resource files.  I'm not
> certain of this however.

Most of the text and image resources are in XML resources for better maintainability. 
Quite a 
bit of the strings though, and configuration options were included in the code, these 
will be 
brought outside the executable in the next CVS update, for easier maintainability.

Best wishes,
Robert
 
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Commerical Applications

[Robert O'Connor]

>   If you change the icons to suit your business logos, or the forms to
> suit your business needs, distributing those changes wouldn't serve to help
> the community much, and generally don't fall under the redistribution clause
> of the GPL. If someone used "Your" distributed icon in their product, it
> would be a copyright (and likely trademark) violation against your company.
> 
>   The important thing is that if you change the _source_ to the
> project, that those changes be available to the community, so they can
> benefit from them. For example, if you added support for image maps within
> Plucker, or parsing forms, or talking to a back-end database of content..
> _those_ changes are required to be set back into the community, per the
> license.

I agree 100%. It is fundamental that the extra functionality be returned back to the 
project. 
With regard to trademarks, IIRC some projects (RedHat, Mozilla for some reason are the 
ones 
coming to mind) trademark their icon/logo and require a redistributed version to bear 
a 
different name and logo. IIRC that was the reason why a regular mozilla distribution 
doesn't 
have the red star/Shep Fairey T-Rex on the splash screen.
 
Companies or governments I think can have a part of the project, as their need for 
solutions 
can help further some different functionality, as long as the terms of the GPL are 
held.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

re: Commerical Applications

[Robert O'Connor]

> Installing
> Python, Python Parser, and Plucker Desktop may be cumbersome for the tech
> staff and IMHO prone to more things going wrong.

> Jpluck only requires the latest version of Java 

Of course, that is your opinion, but I am going to have to spend time to put in 

a few comments.


In both cases, there is a script and a VM to read it, and either would require 
the suitable VM (Java or Python) to be installed, and of a high enough version. 
In an uncoupled solution as the Plucker package, you can use whatever parser 
needed, python, a java one, a perl one, or a C++ one with no VM or scripts at 
all needed to be installed. And has been discussed on plucker-team, the Linux 
will in the future have a single package installer for convenience, so with 
regards to items to install, that is not an item really. Furthermore, since 
python is Free and unencumbered, the relevant parts can be cut up and packaged 
all-in- one, as is done on MSW, so that there is just a single set up click 
wizard without worrying about tracking down virtual machines.  


The Plucker python parser (and some of the others), have a multi-year history 
behind them of removing bugs and optimizations for distribution, vastly 
decreasing the number of errors in a deployment.  


The caveat I guess would be compilations on obtuse platforms, but the original 
poster mentioned that the end users were having too much compexity in setting 
up an AvantGo channel, it is unlikely to be their task to compile things or use 
commandlines at all. Just want the simple installer, which we provide. If he is 
distributing to rare low percentage platforms outside Posix/OSX/Microsoft that 
would be better fit for a pure interpreted language solution like Java or 
other.  


With regard to choice of languages in general, the Plucker project because of 
its goals, is unencumbered by the non-Free languages. While Delphi/Kylix, QT, 
Java, or .NET are able to make applications that run across some different 
OS's, they are all non-Free and coupled with their own set of license 
restrictions on use or distribution (and QT won't work with the Free compilers 
on Windows). Each of these have a single company at the helm holding the reins 
that can step and decide to induce a fee at any time for further development, 
or change their license agreement for futre versions, as they control what 
happens at the top level. Especially relevant since that is what AvantGo also 
did: take in people for free and then decide later to chage the rules of what 
they control (in their case a few-thousand fee for >8 users). Ditto for UniSys 
and GIF format. It is not that the holding company is inherently evil, it is 
that their shareholders want money, and the company then bows to that pressure 
when things get rough. And Sun's current situation isn't that sunny: I don't 
know how long they are going to maintain their current path without going to 
the more dire ways of making money from the things that they control.  


Best wishes,
Robert

.---~~---.   

   /  \
  /\MedicalMnemonics.com
 |__|
.---+`  `+---.  A free non-profit online searchable
|   | () |   |  database of medical mnemonics to 
help
`---+.__.+---'  remember the important details.
(|  OO  |)
 ^\/^   http://www.medicalmnemonics.com
   \\// [EMAIL PROTECTED]
\/   

 `-_  _-'
~~

___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Commerical Applications

[Robert O'Connor]

On 28 Feb 2003 at 13:51, Kevin Bolduan wrote:

> Greetings,
> 
> I'm new to the list and apologize for bringing the big, bad specter of commercial 
> applications to the great open source project you've all created. Simply put, I'm 
> looking for a viable alternative to AvantGo's "Custom Channels" for a client who 
> wants to make their product information and 
website available on Palm OS devices. Currently their site is essentially 
brochureware, with about 300 products (the HTML-for-Palm version of the site I made 
has about 1,000 pages.)
> 
> The Plucker Desktop application is great, and would work very well for our purposes. 
> The initial roll-out of the content would be only to dealers (around 1,500), but 
> eventually we'd like a way to have consumers be able to access the information as 
> well. The problem is that the dealers are, 
well, not all that technologically sophisticated... Put it this way - the sign-up and 
account process for AvantGo was deemed too cumbersome and complex... :-\
> 
> My vision is a customized version of the Plucker Desktop. For example, the 
> "Showcase" would be removed, 25-30 channels (each representing different sections of 
> the website) would be pre-populated (like the Clinical Exam) with predetermined 
> update schedules, and references to "Plucker" and the 
rubber chicken and pacifier icons would be replaced with my client's name and logo. Of 
course anywhere it was possible, we'd have "Powered By Plucker" or whatever was 
necessary. All the tabs for configuring a channel (Spidering, Links, Formatting, etc.) 
would be removed and/or uneditable and pre-
configured with the proper information for our particular application.
> 
> Another useful option would be to "disable" a channel. I know I can delete them, but 
> it would be nice to allow users to temporarily turn off a specific channel for the 
> next sync.
> 
> I'm a marketer and website developer from the Dark Side (ASP, mostly) - Python and 
> the development of applications like this is well outside my realm of expertise. 
> Rather, I'm looking for a partner to modify the Plucker Desktop to fit a very 
> commercial application. Given the open source nature 
and GPL licensing, I don't even know if this is possible - any information would be 
greatly appreciated.
> 
> Thank you for your time!
> 
> -Kevin Bolduan
> [EMAIL PROTECTED]
> ___
> plucker-dev mailing list
> [EMAIL PROTECTED]
> http://lists.rubberchicken.org/mailman/listinfo/plucker-dev
> 

.---~~---.   

   /  \
  /\MedicalMnemonics.com
 |__|
.---+`  `+---.  A free non-profit online searchable
|   | () |   |  database of medical mnemonics to 
help
`---+.__.+---'  remember the important details.
(|  OO  |)
 ^\/^   http://www.medicalmnemonics.com
   \\// [EMAIL PROTECTED]
\/   

 `-_  _-'
~~

___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Cygwin filename case insensitivity in gcc (Was: patch to viewer/configure.in)

[Robert O'Connor]

> Indeed, the Plucker Team has ignored the possibility to build plucker
> on Windows.  Who cares?  Despite that oversite, it can be done now,
> thanks to Cygwin.  (That's why Cygwin exists!)  Moreover, it's
> entirely possible that if header-file-naming was the only impediment
> to a Cygwin build, then they might have made the necessary changes.
> So I don't think the Plucker Team should be criticized for this issue.
> 
> My 20p.

I have built the Plucker viewer and parser in Cygwin before. Although, that is
-With an older Cygwin (beta20).
-Haven't built the new hires branches, with the extra SDKs.

The cygwin just needed a few extra tools to download that weren't part of the 
distribution back 
in their beta20 days.

There were some question marks on the lists over the past year about how to build with 
Cygwin. 
Thanks very much for your advice on the changes needed to do it with the new hires 
branches and 
a new cygwin.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Tables

[Robert O'Connor]

On 9 Feb 2003 at 17:23, Chris Hawks wrote:

> I just checked in changes to TextParser.py and PluckerDocs.py to support
> things like DIV and CENTER in table cells. Also better support for nested
> tables.

Cool! The tables are certainly going to be a highlights of the next Plucker release.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Errors Compiling Plucker Desktop 1.3 (wxMac 2.4)

[Robert O'Connor]

On 9 Feb 2003 at 1:04, Alexandre Enkerli wrote:

> I'm trying to build Plucker Desktop 1.3 (same with CVS) using wxMac 2.4 
> on Mac OS X 10.2.3, 

> hhictrl.h:155: syntax error before `*'
> plucker_controller.h:322: syntax error before `*'

Hi Alexandre,

Thanks for your letter. Nice catch! There should be a Mac include in 
palm_installer_gtk.h. Line 
39 should have a mac condition appended at the end (so the class gets included in the 
compile) 
so it reads:
#if defined(__WXGTK__) || defined(__WXX11__) || defined(__WXMOTIF__) || 
defined(__WXMAC__)

This is now updated in CVS. 

If there is anything else I can help you with, fell free to drop a note off-list 
anytime. My 
address is rob[at]medicalmnemonics.com

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Screen Scraping lawsuits?

[Robert O'Connor]

On 8 Feb 2003 at 14:21, David A. Desrosiers wrote:

>   Very interesting threads going on related to "screen scraping"
> (basically unauthorized spidering). Worth a read, follow the threads:
 
My understanding wehen I looked at the rules over this issue, was that:

(1) If web content is posted the internet http:// on port 80, there is implicit 
permission to 
access it and read it locally. AFAIK deep-linking cases have never gotten off the 
ground very 
much, since it is the server's responsibility to redirect if they are coming from 
somewhere 
else.

(2) You can use whatever agent you want to download it (not just a well-known 
webbrowser), even 
agents that strip out ads, replace offensive words with asterisks, whatever you want. 
If 
someone wants to see their movie wearing sunglasses so the content they is filtered to 
a darker 
color, that is their fair use right.

(3) You just cannot redistribute the edited/scraped work without permission. There was 
a home 
movie company that is cutting out the violent scenes from movies to sell to families 
with small 
kids. I doubt they will survive. More likely, the original content producers will 
start their 
own DVD option, so they can sell more copies to viewers who want to watch as a family 
with 
small kids.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Plucker Desktop Feature Request

[Robert O'Connor]

> > > My proposal is this:
> > > --Have ability to leave proxy password blank and ask for it every time
> > > it is needed (or every session),
> > 
> > This can be done quite easily. Perhaps a 
> > [ ] Prompt for username/password
> > checkbox for the proxy pages, as the UI?
> 
> That would be perfect.
> 
> > > and/or
> > > --Have ability to use a PAC file

Let's go with the prompt, as I think you are right, a PAC file will be more work/code 
to 
maintain.

Please add this to bugs.plkr.org when you have a moment, so it can be tracked and in a 
timely 
manner.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Plucker Desktop Feature Request

[Robert O'Connor]

On 3 Feb 2003 at 17:37, Brent Haag wrote:

> My proposal is this:
> --Have ability to leave proxy password blank and ask for it every time
> it is needed (or every session),

This can be done quite easily. Perhaps a 
[ ] Prompt for username/password
checkbox for the proxy pages, as the UI?

> and/or
> --Have ability to use a PAC file

I don't know what a PAC file is. Is the format described somewhere?

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: plucker-desktop feature suggestion

[Robert O'Connor]

Hi Eugene,

You'll get good service today ;-)

Has been fixed, and now exits silently (as per documentation says). Will be part of 
the next 
package.

Best wishes,
Robert

On 2 Feb 2003 at 7:37, Eugene Y. Vasserman wrote:

> in the windows version of plucker_desktop, if one calls it with -d
> (update all due channels) and no channels are due, it pops up a very
> annoying dialog box informing the user that no channels are due. there
> should be a command-line flag to eliminate this dialog box (in case
> plucker_desktop was called from a scheduler of some sort). i don't know
> if a similar problem exists on non-windows OSes...
> thanks,
> eugene
> 
> -- 
> Eugene "Gman" Vasserman
> [EMAIL PROTECTED]
> http://uranium.cataclasm.org/~eugene/
> ___
> plucker-dev mailing list
> [EMAIL PROTECTED]
> http://lists.rubberchicken.org/mailman/listinfo/plucker-dev
> 

.---~~---.   

   /  \
  /\MedicalMnemonics.com
 |__|
.---+`  `+---.  A free non-profit online searchable
|   | () |   |  database of medical mnemonics to 
help
`---+.__.+---'  remember the important details.
(|  OO  |)
 ^\/^   http://www.medicalmnemonics.com
   \\// [EMAIL PROTECTED]
\/   

 `-_  _-'
~~

___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Plucker Desktop Windows conduit

[Robert O'Connor]

> Okay I think I've found another wierd little bug. I wouldn't be surprised if 
> it was just my system, but I have rebooted a few times to check the obvious. I 
> think the plucker-desktop.exe process doesn't end when run from the command 
> line with the progress dialog option selected and the never close on completion 
> is not selected.

> If I select never close, and hit okay to close the dialog, the process seems > to 
>end fine. If I select console output it seems to work fine, even with the > always 
>close option selected.

Yes this is a bug, reported sometimes on the list. It is 99% likely to be an error on 
my part 
somewhere in the flow. Bug report entered, and I will try to uncover the cause this 
weekend.  

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Plucker Desktop Windows conduit

[Robert O'Connor]

> So, putting these all together, I made a conduit in about an hour that runs the 
> plucker-desktop.exe --update-due process during sync to have all the waiting 
> channels plucked and installed. Plucker_desktop even pops up a nice progress 
> dialog. Seems to work great on the small channels I tried.

Nicely done Bill! That's great work! Will certainly have to roll that into the 
downloads (your 
desktop viewer also, and the other nice additions) for the next release.
 
> I do have a couple of items though:
> If no channels are due, a message box pops up telling the user that 
> there were no channels due. The docs say it should just silently end. I would 
> like this behaviour instead of waiting for the ok button to get pressed. (I do 
> have the auto close always option selected and it works fine when an update is 
> done).

Thanks for catching this. I also prefer in commandline mode that it would just end 
silently 
(but keep the message dialog for if in GUI mode, and they pressed a button, get a 
message to 
tell them what happened). I will add it to the buglist and correct it straight away.
 
> Will plucker_desktop.exe always be found at 
> plucker_install_dir/plucker_desktop/plucker-desktop.exe? If so, I believe I 
> can find it based on the registry key for the Plucker path.

Yes. 
 
You may wish to put in a if condition when the conduit is called, to look at the 
autoupdate 
mode in the plucker.ini to see if it is a '3'. 
A [PLUCKER_DAEMON] section was set aside with a key of autoupdate_mode, with 4 numbers:
0=Never
1=Immediately
2=Deferred
3=OnSync
The 3 is updating the due chanels on OnSync (what you have kindly done).
These correspond to the radio boxes on Preferences > Autoupdate.
The 4th radio box is there but currently hidden, as the capacity isn't there until 
your work. 
If do a check for the '3' value, then that way regular folk can turn the conduit 
behaviour on 
and off.
Sound OK?

> Are there any objections to creating a registry key The Plucker 
> Team/PluckerCond/Command to hold the executable command to run? This way an 
> advanced user could edit the registry to run a program other than plucker-
> desktop during the sync.

Sounds very reasonable. Nice to have different parts of the chain be able to work with 
one 
another.
Perhaps you may wish to put it in the Plucker.ini to make it more portable though 
across Linux 
and Macintosh also. (And also easier editing)
Perhaps as a:   
[PLUCKER_DAEMON]
sync_commandline=
 
> Does it matter if the hotsync is being run for one user and the channel 
> is scheduled to be installed for another? I think the update-due command 
> option will update all channels for all users. I don't think this will be a 
> big problem so I'm prone to ignoring the issue for now. 

Yes, probably not a big problem, as most people are just one person using. 
However, probably worthwhile to consider a workable strategy for restricting the 
updates a 
single user.

On the commandline for the desktop it is easy enough to add an option modifier for 
restricting 
to a certain user, for example something like user-filter (whatever name for the 
modifier you 
think would be descriptive):
plucker-desktop --update-due --username-filter "Bill Nalen"
Then as part of the filtering of only due channels, it would also look to see if that 
user was 
one of the ones listed as a destination. (I think that it is fair though that it 
should go to 
all that channels destinations if one user updates, otherwise people wouldn't have put 
the 
destinations together in the same channel).
That is one approach, though it is a bit more difficult then to use a fixed entry in 
the 
registry for a commandline to execute, since the user-filter value will change 
depending on who 
is Syncing.

Another option would be to for the conduit to write a value of a user to the ini for 
what the 
value of username-filter would be. For example:
[PLUCKER_DAEMON]
autoupdate_mode=3
username_filter=Bill Nalen

Then have the fixed commandline in the registry as just:
plucker-desktop --update-due --userame-filter 
and if no value for username-filter is given, it will get the value from 
PLUCKER_DAEMON > username_filter and filter to that user name.

Please advise on what you think with regards to strategy.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Single instance [was: Re: feature request, kinda...]

[Robert O'Connor]

Hi Eugene,

Thanks for your letter. 
You can add it on the bugs list at bugs.plkr.org as a feature request. That will give 
the best 
forum for tracking, etc. The implenation is pretty straightfoward--there is a 
singleinstancechecker class ready to go.

Best wishes,
Robert

On 28 Jan 2003 at 9:06, Eugene Y. Vasserman wrote:

> I think we need a "allow only one instance" for plucker desktop, due to
> the problem of overwriting .ini file when one version of desktop updates
> or adds a channel, and another one does not. Either that, or .ini file
> needs to be re-parsed on quit, and merged with current into in memory...
> This seems pretty important, as information on newly
> added/deleted/updated channels can be lost very easily. Any thoughts?
> Thanks,
> eugene
> 
> -- 
> Eugene "Gman" Vasserman
> [EMAIL PROTECTED]
> http://uranium.cataclasm.org/~eugene/
> ___
> plucker-dev mailing list
> [EMAIL PROTECTED]
> http://lists.rubberchicken.org/mailman/listinfo/plucker-dev
> 

.---~~---.   

   /  \
  /\MedicalMnemonics.com
 |__|
.---+`  `+---.  A free non-profit online searchable
|   | () |   |  database of medical mnemonics to 
help
`---+.__.+---'  remember the important details.
(|  OO  |)
 ^\/^   http://www.medicalmnemonics.com
   \\// [EMAIL PROTECTED]
\/   

 `-_  _-'
~~

___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Showcase modifications

[Robert O'Connor]

On 20 Jan 2003 at 20:23, Eugene Y. Vasserman wrote:

> I have a quick question: is there currently a way to modify the global
> showcase for plucker? there are seversal sites that I've found that are
> very mobile-friendly and can be plucked easily. Some require
> customization. I'd like to share what I've found. Is there any way to
> send in showcase URLs? If so, it would be really cool if showcase
> updates could be pulled down to plucker desktop from a server (updated
> ever so often) so new users can get benefits of pre-written showcases 
> without re-installing plucker. If not, that would be a great improvement
> (read: feature suggestion).
> Thanks,
> eugene

Agree 100%. This was in before actually, and subsequently taken out. It would probably 
be best 
to use an export of a live database, most likely Davids openurls.plkr.org

The feature, as you describe, will make its way in, in time

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: problem with 'mobile' espn.com

[Robert O'Connor]

On 23 Jan 2003 at 6:58, David Han wrote:

> I don't understand exactly how the priority part of the exclusion lists
> work though. For 'exclude everything with low priority (.*)' I put that
> in as priority 0. For the 'high priority' settings, I put them in as
> priority 1. Is that what you meant?

Yes, you are correct.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

There seems to be a no-cost PocketPC development environment (for MSW)

[Robert O'Connor]

They seem to have an IDE, and a modded version of VisualC++ for free download, for 
making 
PocketPC applications.

http://msdn.microsoft.com/downloads/sample.asp?url=/msdn-
files/027/001/963/msdncompositedoc.xml&frame=true

It is pretty massive, weigning it an a rather compact 300+Megs.

I remember back when they tried to charge for the development tools, and they were 
ignored even 
more than they are now. (This is the reason why I have not much interest in developing 
for the 
platform--I remember when Outlook used to be free too, until they had enough users 
that they 
could start turning the screws to their users).

Could be useful though if someone wants to do a native PocketPC port of Plucker, 
perhaps using 
Bill J.'s explode library as the base.

Best wishes,
Robert

___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Does explode handle > and <

[Robert O'Connor]

On 21 Nov 2002 at 21:06, Bill Janssen wrote:

> Probably a bug.

Thanks Bill, I'll add a bug report, so I don't forget about it later.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Does explode handle > and <

[Robert O'Connor]

Hi Bill J. and Bill N.,

I notice that when explode expands documents into HTML, the '>' in Bill N.'s viewer 
doesn't 
make substitute > for a '>', so tags get thrown off by the stray '<' and '>'.

Is there any workarounds etc? I don't know if it is part of the underlying explode, or 
other. 
Was support for > and < put into explode?

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

(Fwd) Re: Kweping full eferences to links in plucker documen

[Robert O'Connor]

On 21 Nov 2002 at 22:16, [EMAIL PROTECTED] wrote:

> Hello
> 
> I installed Plicker not so long ago and I must say that it is a great
> alternative to several other solutions.
> 
> I have added my netscape bookmarks page onto my list.  The only problem
> is that as I want to know what they point to, the full url reference
> itself is not included in the document.  Maybe I missed out on an
> option, but I do not think so.  Would it be possible to add this?

The URL info for retrieved and unretrieved links can be turned on and off.
It is described in the docs under --no_urlinfo.

If you are configuring your .ini file using the Plucker Desktop, there is a 
checkbox on the channel dialog > Output options page:
[ ] Include the URL of each page in the output file.

Pehaps the text of that checkbox should be updated though, as it also switches 
on/off the URL information of unretrieved links.

Best wishes,
Robert
--- End of forwarded message ---.---~~---.   

   /  \
  /\MedicalMnemonics.com
 |__|
.---+`  `+---.  A free non-profit online searchable
|   | () |   |  database of medical mnemonics to 
help
`---+.__.+---'  remember the important details.
(|  OO  |)
 ^\/^   http://www.medicalmnemonics.com
   \\// [EMAIL PROTECTED]
\/   

 `-_  _-'
~~
.---~~---.   

   /  \
  /\MedicalMnemonics.com
 |__|
.---+`  `+---.  A free non-profit online searchable
|   | () |   |  database of medical mnemonics to 
help
`---+.__.+---'  remember the important details.
(|  OO  |)
 ^\/^   http://www.medicalmnemonics.com
   \\// [EMAIL PROTECTED]
\/   

 `-_  _-'
~~

___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: newline in env

[Robert O'Connor]

On 20 Nov 2002 at 14:56, Bill Janssen wrote:

> > I also think it is incorrect rendering.
> > Is there an alternative way to have a 1st column of preformatted font, and the 
>rest of the line 
> > in normal font in the current parser (or is CSS required?)
> 
> Whoa!  "column" is a loaded phrase.
> 
> I'd think
> 
> one this is the rest of this line
> 
> should work, but haven't tried it.
> 
> Bill

That is exactly what I was looking for. Thanks!

Best wishes,
Robert 

___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: newline in env

[Robert O'Connor]

> > 
> > a
> > b
> > c
> > 
> > is displayed as
> > 
> > a b c
> > 

> Seems wrong to me.  Please file a bug report.

I also think it is incorrect rendering.
Is there an alternative way to have a 1st column of preformatted font, and the rest of 
the line 
in normal font in the current parser (or is CSS required?)

ie, is it possible, without using pre tags, to have a line in the viewer that renders 
as:
this-word-in-preformatted-font this-word-in-normal-font

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Exclusion lists (Was Re: PlkrData Alpha Up)

[Robert O'Connor]

> >Probably a exclusion INI entry would have easier maintainabilty vis-a-vis 
> >a blob of binary
> >data.
> >Yes, there would be a few lines added to the Python parser in the 
> >ExclusionList.py file to add
> >the entries to the exclusions. The Desktop would read/write to the INI 
> >instead of a
> >exclusionlist for that channel.
> >
> >As a general strategy, one could define:
> >
> >parse_exclusion_entry(self,line)
> > m = re.match (r"([-+]?\d+):([-+]):(.*)", line)
> > if not m:
> > error ("ExclusionList: Cannot parse line: %s\n" % orig_line)
> > else:
> > prio = int (m.group (1))
> > action = m.group (2) == '+'
> > regexp = m.group (3)
> > new_item = (action, regexp)
> > if self._items.has_key (prio):
> > self._items[prio].append (new_item)
> > else:
> > self._items[prio] = [new_item]
> >
> >by taking the above out of the load_file, calling it 
> >parse_exclusion_entry(), and then write a
> >new entry that loops the keynames from the config. So it would be
> >
> >def load_file(self,filename)
> > ...(existing loop of number of lines in file)
> > line = (the existing strip of the /n)
> > parse_exclusion_entry(line)
> >
> >def load_exclusion_entries_from_ini(self)
> > 
> > line = read ini to get line
> > parse_exclusion_entry(line)
> >
> >Best wishes,
> >Robert
> 
> Okay, is this a proposal or merely a discussion?  I ask because if the 
> former, what I'd like to do is have PlkrData move the exclusion from 
> whichever source (INI or file) it is on the source, to the INI on the 
> destination.  But that will require users upgrade to the newest Plucker 
> Desktop (i.e. current + 1), which would include not just INI exclusion 
> editing support in the dialogs but also a new parser prior to the normal 
> (slower) release schedule.

All sounds reasonable.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Exclusion lists (Was Re: PlkrData Alpha Up)

[Robert O'Connor]

> Therefore, it was put forward, to include exclusions 
> >as numbered keys,
> >using the existing method of an exclusion list entry. For example:
> >exclusion_item1=0:-:.*\.mp3$
> >exclusion_item2=0:-:http://.*www.x10.com/.*
> >The parser would then loop through a section, looking to see if there was 
> >the next incremented
> >number of key, and if so, add that exclusion to the active exclusions.
> 
> This raises an issue of interfaces.  Certainly we could move exclusions to 
> Plucker.INI, but for much of my parser testing (like when I was adding 
> --stayondomain) I ran Plucker from the command line without a corresponding 
> INI channel.  Doing that, I can load exclusion lists from the 
> command-line.  So I certainly wouldn't want to completely deprecate 
> exclusion files.

Absolutely. Would always continue the existing support for dedicated exclusion lists, 
for both 
compatibility and because it is helpful.

> On the other hand, I also agree that all channel-specific configuration 
> data should go in a single place, and the INI file is a good contender for 
> that.
> 
> But either way, that's not in the Parser nor would Desktop see them or save 
> exclusions there itself, so there's not much point in doing it in PlkrData 
> right now.  The exclusion list has GOT to wind up in a file on the other 
> side.  So packaging it as an encoded blob is probably faster than 
> enumerating through a series of keys.

Probably a exclusion INI entry would have easier maintainabilty vis-a-vis a blob of 
binary 
data. 
Yes, there would be a few lines added to the Python parser in the ExclusionList.py 
file to add 
the entries to the exclusions. The Desktop would read/write to the INI instead of a 
exclusionlist for that channel.

As a general strategy, one could define:   

parse_exclusion_entry(self,line)
m = re.match (r"([-+]?\d+):([-+]):(.*)", line)
if not m:
error ("ExclusionList: Cannot parse line: %s\n" % orig_line)
else:
prio = int (m.group (1))
action = m.group (2) == '+'
regexp = m.group (3)
new_item = (action, regexp)
if self._items.has_key (prio):
self._items[prio].append (new_item)
else:
self._items[prio] = [new_item]

by taking the above out of the load_file, calling it parse_exclusion_entry(), and then 
write a 
new entry that loops the keynames from the config. So it would be

def load_file(self,filename)
...(existing loop of number of lines in file)
line = (the existing strip of the /n)
parse_exclusion_entry(line)

def load_exclusion_entries_from_ini(self)

line = read ini to get line
parse_exclusion_entry(line)

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: PlkrData Alpha Up

[Robert O'Connor]

On 19 Nov 2002 at 11:29, Fringe Ryder wrote:

> Since I haven't gotten any more feedback, I finished up icon functionality 
> my way and have tossed her up on the web.
> 
>   http://kittycentral.net/Plucker/PlkrData.html
> 
> Both sources and a Windows binary are there.
>
> All the basics work - registration under Windows (not Linux yet), saving, 
> and loading channel data and icons.
>
> Still to-do are things that we didn't think of in the discussions - 
> packaging up exclusion lists and unregistering - plus the Linux 
> registration.  Cross-process communications with Desktop may have to wait, 
> since I can't find any sign that Desktop would be expecting a message sent yet.
> 
> Feedback would be very appreciated.

Looking through, I think it is very nicely done. Great work! 

For exclusion lists, this has come up before with regards to how to package these with 
descriptions of sites, since that is arguably the most important part, as it is the 
part that 
most people don't know how to set up on their own.

Exclusion lists are about one of the few areas of the config that would benefit from 
an XML 
format. However, XML configs isn't doable now. Also, it would be beneficial to allow 
the 
exclusions to be commandline accessbile for future tools that wanted to call the 
parser with 
specific exclusions. Therefore, it was put forward, to include exclusions as numbered 
keys, 
using the existing method of an exclusion list entry. For example:
exclusion_item1=0:-:.*\.mp3$
exclusion_item2=0:-:http://.*www.x10.com/.*
The parser would then loop through a section, looking to see if there was the next 
incremented 
number of key, and if so, add that exclusion to the active exclusions.

Best wishes,
Robert


___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Non-HTML (was Re: Back/foward/home function code)

[Robert O'Connor]

On 20 Nov 2002 at 0:55, MJ Ray wrote:

> Robert O'Connor <[EMAIL PROTECTED]> wrote:
> > -Others: maybe the "marquee" tag from MSIE 3.x days, and all the other non-
> > standard crap that has come and gone over the years.
> 
> Yes!  We want  in Plucker now!

I enjoyed the Python comment next to the 'blink' on the list of currently 
unimplemented tags in 
the Parser (TextParser.py, around line 855). Props to whoever wrote that in.
 
> Seriously, has anyone who asks for support for standards-breaking actually
> thought the effect of it through to a conclusion?  I think Plucker should be
> praised for taking a reasonably pure-standards approach on this.  It's the
> only way that you can hope to keep things sane *and* moving forwards.

I agree 100%. 
I am amazed that any web designer which has struggled with the pain of having to deal 
with the 
woes of proprietary browser tags in years gone by, and is now working on Plucker with 
a bona-
fide chance to shape their own future of their distribution medium tools, would opt 
for 
pollution of them by standards-breaking non-HTML.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: VFS Best Practice

[Robert O'Connor]

On 25 Oct 2002 at 10:35, Wayne A. Arthurton wrote:

> What are people doing to install your Plucker DBs to a memory card w/o 
> having to manually change the location on each one?
> 
> I'm considering making all my DBs start with pkr e.g. pkrSlashDot and 
> then have in script run a move all DBs that start with pkr from the 
> Install dir to the Memory Stik install dir in the Palm Desktop.

There is support deduced for SD-Cards. What is the 
mechanism for Memory Stick, and can possibly support 
that also. 

-Do the files go to a specific Memory Stik dir on the 
harddrive to be transferred to device?
-Is it enough just to put them in the dir, and they get 
installed, or is there other items to be done also, 
like flicking a memory switch?

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Back/foward/home function code

[Robert O'Connor]

On 19 Nov 2002 at 11:12, David A. Desrosiers wrote:

>   Ugh, the worst of the offenders. There are no forms on the Onion's
> website, yet they still need to rely on this pods://avantgo/back syntax?
> Useless. On that page, _ALL_ articles reference index.html, and to all
> articles, index.html is their parent. The << "The Onion" link that brings
> you back could just as easily been a link to index.html, just like the
> _millions_ of other fully functional websites around the world do it. I
> guess this is the AvantGo'ish hack for history.back() in Javascript, and on
> this site, is a completely moot example, since it doesn't actually explain
> the use of it any further.

I agree 100%.
The only possible reason that I can see for it, is that it is one less link to queue, 
because 
it isn't a true hyperlink, it is an embedded command. 
I don't know how it is in AvantGo, since I haven't used it in about a year, but in 
Plucker, 
many of the sites require breadth first, and breadth first maintains a queue of all 
links until 
the end until it sees if it has already been retrieved. 
So in the onion case:
-With regular links to index.html: 10 extra links are queued and get time is spent 
checking, 
and seeing that it already been downloaded.
-With "back" commands: no extra links queued, no memory required to store this link 
until the 
end.

It is bit dubious though as far as whether functionality goes though. 

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Back/foward/home function code

[Robert O'Connor]

On 18 Nov 2002 at 21:43, Laurens M. Fridael wrote:

> complying with W3C standards is hardly a priority or
> even a consideration. 

I would tend to differ on that point.

Over the next few years, Plucker is going to become the dominant force as it continues 
to 
mature and the commercial alternatives wither. AvantGo, in particular is on the 
downgrade, with 
a delisted stock, layoffs, CEO resignations, and sagging sales.

As Plucker continues to rise, I would offer that Plucker chooses the best practices, 
in the 
form of documented W3C and ISO standards, instead of picking up the bad practices of 
the 
withering commercial off-line browsers.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Back/foward/home function code

[Robert O'Connor]

> A useful addition to the Viewer would be a function code (in a text record)
> that instructs the Viewer to go back or  forward in the document's history,
> or to the document's home page.
> 
> This way a parser can support AvantGo pods:// URLs. For instance the link  href="pods://avantgo/back"> instructs Avantgo to go one page back in its
> history. It would be nice if the Viewer supported this as well. There are
> some AvantGo sites that use these links. (The Onion for example.)

I agree that it would have some use, as far as a smaller queue for the breadfirst 
parser to 
have (since many pages will have a link to home, and perhaps a back in some places).

However, I would much prefer to see it as a  That 
is the 
W3C standard way for a back hyperlink. The pods:// protocol is not any standard, it is 
something proprietary and closed. It may be better for Plucker to encourage the 
standards 
instead of the proprietary things, especially from companies that may fold up shop 
soon.

I don't know if there is a javascript for home though. It isn't really a home in most 
browser 
senses though, since the home can be different in Plucker, since there can be a list 
of links 
to many channels as the home, and the "home" link at the bottom of one individual 
website  is 
usually meant as a link back to the main page of that website's channel, not back to 
the 
homepage of the list of links.

Best wishes,
Robert

___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: accesskey attribute support

[Robert O'Connor]

On 17 Nov 2002 at 21:21, Michael Nordström wrote:

> - how should the accesskey be entered on the Palm without interfering
>   with the gesture actions?

> Any suggestions on how to enter the accesskeys?

This is reasonable: there would be some keys that would be off-limits if just 
scribbled the
letter in directly. For example, can't use "I" since that is the down gesture.

One possible solution would be to have a prefixing stroke to go into accesskey mode. 
Currently,
for menu keystrokes, if there is a up-right stroke, followed by an "I" then that does a
"Details..." successfully (english viewer uses "I" for details menuitem), instead of 
being a
downward gesture. For accesskey, it would work similar: choose a different stroke for 
starting
accesskey mode, then the next stroke is the accesskey. The implementation could 
include having
a reverse-text "Accesskey:" prompt in the screen corner, similar to what the 
"Command:" does
now for menucommand mode.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Inline launchable database icon RFC (Was RE: Plucker Channel Addition Automation)

[Robert O'Connor]

On 17 Nov 2002 at 9:56, Fringe Ryder wrote:
 
> >Another option would be to directly use TBmp's inline, such as
> >big_icon_inline=\000\040\000\040\000\004\000\..
> >small_icon_inline=\000\017\000\011\000\...
> >and then just add an extra "if" condition after the "get_icon" function in 
> >PQAAppInfo.py of the
> >Python parser to check to see if there is a value for big_icon_inline. If 
> >there is, use the
> >big_icon_inline value, else use the default icon.
> 
> That sounds fine too.  Historically there -used- to be a limit on the 
> length of an INI string.  If wxWindows is not handicapped by that, this 
> will work.  Otherwise I was thinking of just tossing the pseudo-binary 
> images in at the end, with a length word in front of them.  But inline 
> encoding is fine too - easier to read, equally hard to edit by hand.

This would I think be the nicest solution. 
 
> > > Sheesh, Robert, there's a forest
> > > hidden in those trees!
> >
> >Probably not the altogether most helpful comment to aid discussion.
> 
> Perhaps not; I apologize for the tone. 
>

No problemo. We all raise tones sometimes. Glad we can stay focused instead around 
here for the 
99% of the time.

> of a tool to simplify adding sites to Plucker you would expect Joe-User to 
> build their .plkrdata tool by hand-coding it and then running a bitmap 
> mapping tool to append to the file a encoded icon.  The whole idea of this 
> is to make it easy for corporates and other non-techies to put up 
> easy-to-use Plucker Channel files on their sites.

Sorry, perhaps I was making things unclear, which I sometimes do with my low number of 
words. 
Just to sort that out, I was wagering that not many of the Joe-Users would be willing 
to hand-
code a .plkrdata file. (The previous outing of the Windows distro used manual 
configuration of 
ini files for the different channels, and it was not what many of the users were used 
to doing, 
unless they remembered their Soundblaster/DOS game days). The .plkrdata should also 
allowed to 
be hand-coded though, because INI editing is familiar to some people. So the Joe-User, 
no ini-
writing-needed approach would be, to generate a .plkrdata file is:

-Start Plucker Desktop.
-Set up a channel with the parameters that are desired for the site. If a custom icon 
file is 
wanted, Joe fills in the filename of the bitmap in the boxes for large and/or small 
icon.
-From the menu, click Import/Export > Export selected channel(s).
-A prompt is asked for a filename to save it as, with a suitable default filename 
given, such 
as a generated .plkrdata. As part of the export, the software writes a 
key of the 
icon in a text TBmp format.

Now there is a .plkrdata file, ready to be put online for one-click downloads by 
webusers via 
the 1-click tool. That .plkrdata file can also be sent via email to someone, or shared 
with 
others who want to load them on. Either a double-click on a .plkrdata file on the 
desktop adds 
the channel to plucker.ini, or the web download as you described.

An export from openurls.plkr.org/my.plkr.org would work the same, except instead of 
GUI widgets 
to fill in the parameters, there is form elements to fill.

The non Joe-User, hand-code way, is just to open up a textfile and write the keys by 
hand. The 
big_icon_inline is certainly hard to code by hand, but that key (like the others) is 
optional 
anyways.

Because it is in .ini format, shall we allow support for multiple channels inside, or 
only a 
single one in a .plkrdata file? If multiple, then perrhaps before the addition of the 
sections 
to the .ini, if there are multiple sections, have a prompt to check off those in the 
wxCheckListBox that are desired to be installed. (And have an option to turn off this 
prompting 
if so desired). This would make it easier for people to distribute a list of their 
known 
subject, such as all the astronomy PDA sites if they were an astronomy expert and 
PDA-savvy?

The other possiblilty for an icon is just to use the big_icon keys, and start 
supporting in the 
parser http:// URLs to the icons, and the parser downloads and converts using the 
image 
parsers? The downside is that many sites aren't too careful about not breaking their 
URLs to 
things: many can't even keep their main URL consistent for more than a few months.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Inline launchable database icon RFC (Was RE: Plucker Channel Addition Automation)

[Robert O'Connor]

On 16 Nov 2002 at 20:44, Fringe Ryder wrote:

Hi Tony, thanks for your letter.

So anyways, in summary then, 2 keys in the ini in a section to specify either a b64 
encoding or 
a PNM of images. 

Probably new keys instead of overloading the existing ones used for icon filenames, 
something 
like big_icon_inline= / small_icon_inline= or similar. 

Another option would be to directly use TBmp's inline, such as 
big_icon_inline=\000\040\000\040\000\004\000\..
small_icon_inline=\000\017\000\011\000\...
and then just add an extra "if" condition after the "get_icon" function in 
PQAAppInfo.py of the 
Python parser to check to see if there is a value for big_icon_inline. If there is, 
use the 
big_icon_inline value, else use the default icon.

Any others want to weigh in their thoughts during the design stage? 

On a somewhat related note, I had considered updating the large default launchable 
icon to 
instead of a Plucker icon with a tiny page icon in the corner, to have a corner bent 
page icon, 
with a plucker icon on the left side partially on the page, as that is what most OS's 
use to 
denote a document of a mimetype. Visually, it would be laid out like the ASCII layout 
below. 
Anyone object or have strong desire for having something else?

  +---
  |  ++
 ++   |
 ||   |
 ++   |
  |   |
  +---+ 

> Sheesh, Robert, there's a forest 
> hidden in those trees!

Probably not the altogether most helpful comment to aid discussion.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

(Fwd) RE: Plucker Channel Addition Automation

[Robert O'Connor]

> At 04:20 AM 11/16/2002  +0000, Robert O'Connor wrote:
> 
> > > 3. Creator (if needed) - given a channel name, convert that into the
> > > appropriate format (e.g. .plkrdata file).  This could be as simple as
> > > grabbing the relevant fields from the relevant channel section of the INI
> > > file or it could require packaging of that up to a different format.
> >
> >Probably not needed. Best just to use a single .ini format (less headaches).
> 
> You could have convinced me, if not for you adding the bitmap icon several 
> paragraphs down.  How do you figure Joe-User is going to plug a bitmap into 
> an INI file? 

I don't see much difficult about it. It is either a b64 encoding of the image 
or just a PNM-format to have an image of 1 and 0 characters of the pixels. In 
the part that you snipped, I said that the .ini could be generate
d from the Desktop export. There is already a widget to specify a custom large 
or small icon filename. The Desktop would just zip up the file into b64, or put 
it inline, PNM-style. I don't see how a casual computer use wo
uld have any problems with it.

> >(2) User clicks on link, and browser knows to do something with that 
> >plkrdata file, to launch
> >the one-click utility include with an argument of the . .plkrdata file. 
> >(Is the part that is
> >likely the most source of headache, dealing with the different possible 
> >browsers to have them
> >set up right).
>
> Don't most pass the file to the OS if they don't know how to handle it 
> themselves?  Popping the type in the registry for Windows and the 
> /proc/sys/fs/binfmt_misc/register for Linux should handle this for most 
> installations, no?

I don't know how all browsers do it. Sorry.
 
> >(3) The one-click utililty reads the lines of the .plkrdata file and 
> >transfers them to the
> >plucker.ini. They should probably be sanity checked for at least being 
> >proper config format, so
> >can't bring down the entire parser by a bad file. If using a wxWindows 
> >C++, this could be
> >accomplished by a wxFileConig Read() and Write() to read into the 
> >.plkrdata file, but then set
> >active config to plucker.ini and Write() it there. wxConfig would cough on 
> >the way in, so it
> >wouldn't possibly write illegally formatted stuff to plucker.ini.
> 
> I'm a bit leery of dragging in something as big and convoluted as wxWindows 
> solely for ini file processing.  The source file (.plkrdata) will have a 
> handful of required fields that can be loaded using a simple 
> GetProfileDataString() or equivalent; if they're missing, spit out an error 
> rather than corrupting Plucker.INI.  But I agree that sanity checks are 
> essential.

Actually big and convoluted is what will happen when writing something to work 
cross-platform, and have it work properly on the platforms. wx has already 
wrapped the functions that are need. I think the whole app could be written in 
about 40 lines of code, because there are already cross-platform classes for 
everything needed.
 
> >(3.5) Wesley had suggested to include the large and small icons, which I 
> >think is a good idea.
> >These would be b64 encoded into the .plkrdata textfile. The timing of when 
> >to unzip the b64 and
> >dump the icons into files ready for the parser is up in the air (do it on 
> >addition of a new
> >channel from showcase/.plkrdata, or as part of process before starting a 
> >channel, or put into
> >the parser to handle an icon as a text rendition (similar to how the 
> >default icon is in the
> >sourcecode in Python)?
> 
> This is a new one to me.  I hadn't even noticed the icon feature until just 
> now, looking through the docs to figure out what you were talking 
> about.  The help says that icons must be monochrome; is that still 
> true?  Anyhow, popping in the icons requires a bit more data than a 
> standard text INI can handle... need an encoder.  Not a problem, but does 
> reinforce the need for a creation function.

The help is correct. The icons must be monochrome. It is a Palm requirement for 

the .pdb format.

The icons can go into a standard .ini. b64 or PNM or other. I don't think the 1-

click tool should have to worry about this or any other setting really though, 
it just be flexible enough to copy links. The only required sanity check is 
that it is a legal .ini file. The rest doesn't matter. If someone wants to hand-

generate a goofy .ini file with bpp=7, or bpp=m, then it should only just stop 
that channel from working. However with an illegal .ini section such as an open 

ended [ in the title, adding that to the plucker.ini will break the entire 
plucker.ini from re

RE: Plucker Channel Addition Automation

[Robert O'Connor]

> Rob, I came across this comment on /. regarding conversations between you 
> (apparently) and a borderline-Plucker-user...

Yes I remember this gentleman that sent me this. 
 
> >The other item is a one-click "sign up for this channel" functionality -- 
> >basically just a metadata file on our Web server containing URL, depth and 
> >follow/nofollow directives, probably in XML. This requires some more work 
> >on the desktop app and installer.

> I think it's a fantastic idea, if we extrapolate to the next obvious 
> level.  We could, once we put this structure in place, allow the same data 
> to be downloaded from the Plucker website that David has redesigned so 
> effectively for any channels listed.  (i.e. users could simply 
> point-and-load rather than having to configure the channel using the data 
> on the website about each channel.)  This could also eventually extend or 
> replace the "Showcase" arena, which I'm ashamed to say I've never used.

This is the plan.

The current Desktop Showcase format (XML-kinda) will be deprecated, and a regular 
plucker.ini 
format used instead. If someone wants to put in the time, openurls.plkr.org can export 
to 
plucker.ini file format. David said he would have a look, but I am sure he would 
welcome a 
hand. Perhaps an XML file would be better eventually for a format instead of .ini, but 
that 
isn't near term, as it requires a python upgrade. And XML doesn't add much context 
really, 
since all the properties are "flat" because they need to be shared with commandline 
parameters.

The Plucker Desktop code gets simpler to maintain because the same plucker.ini parsing 
is used 
both for the regular channels list, and the showcase. Moreover, the listctrl derived 
class gets 
recycled, since both will have the same columns.

For a file extension, Bill J. uses ".plkr" extension for the Plucker databases, and it 
is 
registered as a MIME type now. It is reasonable to pick a extension for a description 
of the 
channel contnet, like ".plkrdata" as you suggest, or similar

Using Plucker regular.ini files, the content provider can then just describe his 
channel with 
widgets for color depth etc, either online at plkr.org with web forms or with the 
Desktop 
software. Either avenue would export/generate a .ini file of the channel without 
worrying about 
what each key is called. Can then drop that .ini file on the server. Alternatively, a 
joe-user 
can make a list of channels to share or install elsewhere, by selecting the channels 
in his 
list and chooseing "Import/Export > Export selected channels" from menu (this is also 
useful 
for content that wants to be retrieved both at home and work, without resetting up all 
the 
channels again on the 2nd machine).

The current TODO mechanisms would allow a new channel to be add in such an ini file 
via a 
"Import/Export > Import channel" menu command.
The drag and drop of a file into the window would also look at extension, and if it 
was the 
say, .plkrdata file extension, instead of using that filename as a starting point, it 
would 
import it instead.

These are cross platform ways, and cross browser.

> but some form of interface (not necessarily graphical) is needed to simplify > this 
>to the 
> level of windows users or to automate it for the convenience of everyone else.
 
I agree.

> I will write a utility to parse the data and store it in the Plucker INI 
> file.  This would have three functions...
> 
> 1. Installer  - would update the Windows registry or the Linux 
> /proc/sys/fs/binfmt_misc/register to recognize the unique extension 
> (.plkrdata ? something else) so that it can be launched from any 
> browser.   Also asks for or confirms defaults for the non-site-specific data.

Yep.

> 2. Parser - Read one of these files, converting to ini format and storing 
> the data in plucker.ini

This would actually also not cause problems when Plucker Desktop is running for adding 
channels. Though the Desktop reads the .ini file, then maintains an internal state of 
the keys, 
and if modify then .ini file, then close Desktop, it will write the state keys. 
However, it 
will leave sections alone that it didn't have a key for in the internal state.
 
> 3. Creator (if needed) - given a channel name, convert that into the 
> appropriate format (e.g. .plkrdata file).  This could be as simple as 
> grabbing the relevant fields from the relevant channel section of the INI 
> file or it could require packaging of that up to a different format.

Probably not needed. Best just to use a single .ini format (less headaches).

The rundown for operation I guess then would be:
(1) Content provider has link to ".plkrdata" file on their website's HTML page.

(2) User clicks on link, and browser knows to do something with that plkrdata file, to 
launch 
the one-click utility include with an argument of the . .plkrdata file. (Is the part 
that is 
likely the most source of headache, dealing with the different possible browsers to 
hav

Re: ~/.pluckerrc vs. ~/.plucker/desktop.ini

[Robert O'Connor]

We had a few discussions on the plucker-dev lists regarding .ini files and where to 
put things. 
This would have been in fall 2001, back at the design stage and planning stage. I 
spent a good 
time asking about ini format, as there is a zero sum of available time, and I did not 
want that 
time redesigning config files in a different way after release. 

It was decided to use a single .ini file, arranged by section, so that there was only 
one .ini 
file for tools to track down and manipulate. Channels were decided to be put in the 
.ini file 
also, to enable "-s" overrides of sections. I was ambivalent about either approach 
(inclusive 
vs. separate files), but specifically didn't want to start one way, then go back and 
completely 
rewrite/redebug again with the other approach.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Updates to explode - resend without attachments

[Robert O'Connor]

> >I should see this.
> >
> 
> >
> >I should see this too.
> 
> I thought that text withint the  tags should be ignored by the
> browser and not shown at all.  If I don't make the tag change, the wxHtml
> window shows the text (which includes lots of elements explode doesn't
> handle).

I agree, the stuff between  should never be displayed by the wxHtmlWindowr. 
Tags inside 
the  should also not have any effect. I am just having trouble isolating a 
case of it 
not working correctly.
What is an example of a line of HTML that makes text appear when it should not?

Example one HTML;

I should see this.


I should see this too.

Example one output, correctly shown in the wxHtmlWindow:

I should see this. I should see this too.

Example two HTML:


Normal size


Example two output, correctly shown in the wxHtmlWindow:

Normal size




Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Updates to explode - resend without attachments

[Robert O'Connor]

On 9 Nov 2002 at 20:31, Bill Janssen wrote:

> > I also made a change to the  would not be correctly ignored by the 
wxHtml parser?

The Desktop has a HTML editor. When I type this into the editor window, and then flip 
the tab 
to the preview window, it seems to render correctly:


I should see this.



I should see this too.


Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Windows Plucker-build (was Re: Downloads available on the web site - descriptions)

[Robert O'Connor]

> I'll send along the conduit stuff in a few days.  I just need to do the
> progress dialog.  Do you have a Windows machine to test it on?

Sounds great. I do have a machine to test it on.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: wanted features for viewer

[Robert O'Connor]

> 3. Show underline to the current line to help browsing.
>  Move the line with Jogdial or automatically. When the line reaches 
> to the bottom of the page, it scrolls.

> 5. Insert comments between lines.

> 8. Bookmark list in Library list.

Hi Matto,

All interesting ideas. 

I am a bit unclear as to what the above three would involve, perhaps you could explain 
them 
further?

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: PyPlucker 1.1 - 1.2 differences?

[Robert O'Connor]

> Well, it didn't work!
> What exactly the 1.2 PyPlucker needs to run which is different from the 1.1 
> windows packages? Is it something that "sits" in the latest (or desktop) 
> Plucker.ini?
> I know, I'll switch to desktop, but I would like to understand this better.

Hi Branko,

What is the problem/error message that happens?

Best wishes,
Robert
 
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Windows Plucker-build (was Re: Downloads available on the web site - descriptions)

[Robert O'Connor]

> >Yes. It is an installer for all the Windows components of the windows
> Package. The script is in
> 
> Great.  I guess if you want to add the conduit to the mix, I can send along
> my Inno script once I finish off the conduit (hopefully in a week).

Absolutely!

> Does this mean the Desktop is actually using none of Dirk's stuff? I.e. if
> I have the Desktop installed, can I delete everything (the exe's and dll's
> I mentioned earlier) except the desktop exe (and resources) and the
> pyplucker stuff?  

Yes that is correct. There isn't a dependency between Plucker Desktop and the 
commandline tools 
kindly made by Dirk. (Partly too since the source for some of Dirk's tools for some 
reason 
aren't avaialable).

> If this is the case, then perhaps there is no need for
> Dirk's stuff anymore (other than to support those using the command line
> versions)?

Perhaps, though there hasn't been many people upset by continuing to include them, to 
provide 
another usage option, and also for servers, etc that have scriptc calling the tools. 
It would 
save about 300K or so from the install.
 
> >The Microsoft API only allows proper killing
> >of a process by sending a close event to a window, and a windowless
> application obviously has
> >no windows.
> 
> Hmm, I thought there was something in CreateThread that would allow the
> thread owner to kill it.  I guess you would have to spawn 2 processes then
> instead of 1.

Could well be. You are welcome to look through os_msw.cpp to see how things are done 
at the 
moment and the older revisions in CVS that had other functions for other 
approaches--if there 
is better ways to do things, they are most certainly appreciated. I ended up spending 
about 7 
full days researching and working on different approaches that wouldn't fail for 
various 
reasons (for example:  isn't in one of the Windows kernels) and 
probably won't 
spend much more time on working around that poor API. The current end result is that 
95/98 uses 
the ::TerminateProcess() API with its built-in shortcomings and NT/2000/XP uses a hack 
to go 
inside a process and kill it from within (it is pretty well commented).

Best wishes,
Robert 
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Windows Plucker-build (was Re: Downloads available on the web site - descriptions)

[Robert O'Connor]

> Who currently builds the Windows stuff?  

I have built the Windows package for the last little while, though it has been Dirk 
who has 
kindly built them before then.

> Robert are you saying that your
> Inno Setup for Plucker Desktop does everything you outlined?

Yes. It is an installer for all the Windows components of the windows Package. The 
script is in 
CVS under the plucker_desktop/installer/msw but it is in fact an installer for 
everything, 
since easier to find and install things together.
The .spec file in plucker_desktop/installer/linux is just for Plucker Desktop, since 
Linux has 
proper package management.

Both Dirk and I agreed to put all the components into one installer for the Windows 
platform, 
since that is the par for that OS.

> Are we saying that the Windows Plucker Desktop installer is using Dirk's
> last build?  

It is using the tools to provide a separate commandline stub, as well as the toolchain 
for 
whatever batchfile/automatic execution are already in use.

I see the following programs from Dirk:
> DBTool.exe - probably not needed since Plucker Desktop takes care of
> channels now

Yes, cut.

> EditStub.exe - no idea what this does

Yes, cut. I think it just opened up Notepad.exe with a file.

> Plucker.dll - no idea what this does
> Plucker.exe - not sure if this is used

Don't recall. Not needed though.

> plucker-build.exe - this is what Plucker Desktop calls

Dirk's commandline shell for plucker-build. It is in the start menu listed as 
"Commandline 
Plucker (plucker-build.exe)". It started without arguments it flies up a box to enter 
some 
arguments, otherwise it executes python spider with the arguemtns. 

Left in for people who are using it in the batch files.

The Desktop bypasses it and just calls the python script directly, since the Windows 
API is 
[still] deficient in regards to killing a process. The Microsoft API only allows 
proper killing 
of a process by sending a close event to a window, and a windowless application 
obviously has 
no windows. Terminating a windowless application (plucker-build.exe) that spawned 
another 
windowless application (python.exe) couldn't be done with such a poor API. Using 
python.exe 
directly made it possible to terminate python.exe. In Plucker Desktop, there is a hack 
added 
for NT/Me/XP versions that allows a "killing from within" that will overcome the 
deficient API.

> RunSync.exe - used to run the parser from the short cut - probably not
> needed anymore

There is a RunSync.exe and a /Conduit/Conduit.exe. These are used by the older system 
to put it 
in the install directory for a Palm. Also kept in case people have it set up in their 
batch 
files or custom setups.

There is also a /Bmp2TBmp/Bmp2TBmp.exe that is used for the older windowspil parser, 
in case 
people want to use that parser (though pil2 gives better results without requiring it).

Best wishes,
Robert

___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Downloads available on the web site - descriptions

[Robert O'Connor]

On 8 Nov 2002 at 10:05, David A. Desrosiers wrote:

> 
> > ..if we have the sources lying around.
> 
>   That's the crux exactly. Unless someone has the source, we're stuck.
> As far as I know, nobody has actually received the source Dirk is using for
> those tools. Anyone else happen to have the source from Dirk?

Dirk has a lot of sources in CVS. The things missing are the sources for 
plucker-build.exe stub 
program and the Inno Setup installer script. A new Inno Setup installer script has 
everything I 
think would have been in the original.

> > Could you add something to the Plucker Desktop section in the left column
> > that says it includes the "distiller/parser".  Something like, if you want
> > to prepare your own content (as opposed to the sample ones you have), the
> > Plucker Desktop contains everything you need?  And maybe something to the
> > viewer section that says that if you only want to read already prepared
> > material, the viewer is all you need?
> 
>   Sure. This sounds doable.

Sounds good, perhaps just call it a full Windows package or something. Perhaps list 
the 
components inside the package? Somewhat similar to Bill's OSX package. The components 
to 
install can be unchecked/checked in the installer, in ways that don't result in broken 
installs. 
There will be a component in the package for doing each task. At the moment these are:
- The Plucker distiller, in Python (together with an appropriate Python setup, to have 
it 
running right away).
- The Palm Plucker viewers, both normal res and high-res versions, in all the 
translations. 
- Plucker Desktop (GUI), and translations.
- plucker-build.exe: A commandline only interface for the Python distiller.
- Help files

> Having a good, responsive mailing list full of helpful users, good
> facilities to getting the source and reporting bugs has improved the
> "responsibility" of the project considerably, and the users appreciate it.

I agree that the mailing lists are quite well managed and filled with helpful people.
 

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

RE: Unexpected char in declaration "<" error

[Robert O'Connor]

On 7 Nov 2002 at 14:06, Blake Winton wrote:

> > The problem is msword embeds loads of pseudo directives in
> > the form of (something like) 
> > and Plucker is choking on these.

I finally also happened to track down the origin of that garbage the other day.

There is also a bug note about this.

Best wishes,
Robert


___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Superscripts (etc)

[Robert O'Connor]

Michael had a prototype viewer including super/subscripting that he mentioned on the 
list, with 
a link to it IIRC. It would have probably been about 12 months ago.

He may still have it kicking around, or perhaps there is a diff or link to the viewer 
in the 
archives.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Alternative-to-AvantGo slashdot article

[Robert O'Connor]

Seems yet another Alternative-to-AvantGo question has been posted (as supposed to 
posting the 
answer a few weeks ago, but alas).

Anyways, can see the feedback on Plucker 1.2 from a few users here:
http://ask.slashdot.org/article.pl?sid=02/11/05/0934244&mode=thread&tid=95

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: wxWindows viewer available

[Robert O'Connor]

On 5 Nov 2002 at 7:15, Fringe Ryder wrote:

> can we PLEASE keep attachments off the mailing 
> list?  

David, I, and others have always had very small screenshots as part of a discussion. 
This 
attachment was specifically made to be small. I am one of the ones paying by the 
minute to 
download things, so certainly want things to be done as quick as possible, so that 
they can be 
downloaded in a burst, and then read offline, so I have always appreciated having one 
small, 
cropped, well-compressed file included directly.

Feel free to set up your mail client not to retrieve attachments from plucker-dev, 
though, if 
it so suits you.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: wxWindows viewer available

[Robert O'Connor]

On 5 Nov 2002 at 8:27, Bill Nalen wrote:

> I've put the source out at
> http://bill.nalens.com/programs/wxpluckerviewersrc.zip
> 
> There's not much there but I thought Robert might be interested in it for
> Plucker Desktop.  I've included my VC projects in case you use that.

Absolutely. Will make a fine addition. Perhaps a right-click menu entry of "view this 
channel 
output" in the main list (and a menubar entry of the same). If it has never been built 
yet, ask 
if want to build it now?

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: wxWindows viewer available

[Robert O'Connor]

On 4 Nov 2002 at 23:21, Bill Nalen wrote:

> Let me know what you think

Whoa, that is some outstanding work! Definately will be a nice feature to have, very 
nicely 
done!

I have attached a small 9kb screenshot so others can have a look right away also. The 
screeshot 
is of a Plucker .pdb of the CBC mobile news website (www.cbc.ca/mobile).

Best wishes,
Robert

The following section of this message contains a file attachment
prepared for transmission using the Internet MIME message format.
If you are using Pegasus Mail, or any another MIME-compliant system,
you should be able to save it or view it from within your mailer.
If you cannot, please ask your system administrator for assistance.

    File information ---
 File:  wx_desktop_plucker_viewer.gif
 Date:  5 Nov 2002, 13:27
 Size:  9454 bytes.
 Type:  GIF-image

<>

Does explode explode Palm images to another image type (Was Re: Patch for unpluck on Windows using MS VC 6.0)

[Robert O'Connor]

On 4 Nov 2002 at 14:32, Bill Janssen wrote:

> > -Image renders are built in, all have to do to support PNG, BMP, JPG, etc is to 
> 
> I doubt that Palm image renders are built in.  I think you'll have to
> go through pretty much the same dance I did in the GTK+ viewer.

Hi Bill,

Does your explode library explode the Palm images into either a JPG/GIF/PNG/BMP, that 
are then 
included as a link in the  tag? 

I thought that is what explode did for exploding images, though my memory is getting 
hazier as 
I get older ;-)

Best wishes,
Robert

___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: hires progress

[Robert O'Connor]

On 1 Nov 2002 at 20:15, Adam McDaniel wrote:

> I've made a couple screenshots of hires+ working in plucker :)
> 
> http://hires.plkr.org/screenshots/nr70_library.png
> http://hires.plkr.org/screenshots/nr70_document.png

Wow that is pretty cool! 
That second screenshot is quite an awesome example of color+hires working in action. 
Definately 
would make a good screenshot on the plkr.org screenshots page, as soon as you think 
the public 
can get ahold of the .prc.

Best wishes,
Robert


___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Patch for unpluck on Windows using MS VC 6.0

[Robert O'Connor]

> >You still can.  There's nothing stopping the desktop from popping up
> >the document in the viewer after it's plucked the home URL.
> 
> >The text widget in GTK+ seemed much more capable.
> 
> The GTK+ version needs a 5 meg download for Windows and installs a bunch of
> dll's, so it's kind of a heavy weight to just view Plucker documents.  Plus
> it wouldn't have the same look and feel 

It would be an interesting project to wrap up Bill's excellent GTK+ viewer in a 

general framework so that it could be a native application on not just 
GTK, but OSX, and MSW, as well as a little X11 executable for embedded Linux 
devices.

There is an wxHtmlWindow widget that will render simple HTML (for example the 
"preview" tab on the HTML editor, and the online HTML help).
-Forward and back behaviour is built in already.
-A similar feature set is available: font color, strike, bold, underline, H1-
H6, alignment, horizontal rules, etc.
-Image renders are built in, all have to do to support PNG, BMP, JPG, etc is to 

add a single line of the code to add in the ImageHandler for that image type.

Seems like all that would be needed would be to have a wxFrame with a 
wxHtmlWindow. Opening uses the explode to explode to a temp dir and do a 
SetPage() on the the 
wxHtmlWindow to show the document. Of course, a more detailed approach would 
probably be better, especially for very large documents--I haven't taken a 
close look at the GTK+ viewer sourcecode yet, but will do so today.

Best wishes,
Robert
.---~~---.   

   /  \
  /\MedicalMnemonics.com
 |__|
.---+`  `+---.  A free non-profit online searchable
|   | () |   |  database of medical mnemonics to 
help
`---+.__.+---'  remember the important details.
(|  OO  |)
 ^\/^   http://www.medicalmnemonics.com
   \\// [EMAIL PROTECTED]
\/   

 `-_  _-'
~~

___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Plucker Bookmark feature enhancements

[Robert O'Connor]

> > Any chance of a option to order the bookmarks, or if not just a
> > "Alphabetize" command, to put the bookmarks in alphabetical order?
>
>   How about two orders, "Natural", where it orders them by location in
> the document(s) (i.e. a bookmark 'Zed' at 10% would appear before a bookmark
> 'Aardvark' at 30%, when ordered "Natural"), and an Alphabetic one.

Sounds like a very good idea.

>   The problems with alphabetic begin when you deal with umlauts and
> such in the content itself. An English 'a' may not come first when compared
> to a Gaelic 'ae' combination or 'ä'. I'm not sure we have a table of
> precedence for that yet.

A good point. I am afraid I don't know much about that, though there is probably some 
rountines
kicking around somewhere for this, as it would be something that a lot of international
programs would require.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Plucker Bookmark feature enhancements

[Robert O'Connor]

On 31 Oct 2002 at 11:54, David A. Desrosiers wrote:

>   As I was testing Matto's new sub-search capabilities in the viewer,
> I noticed a few things in the Bookmarks form that we could enhance to make
> it much more useful. 

>   The floor is open for comments.

Any chance of a option to order the bookmarks, or if not just a "Alphabetize" command, 
to put 
the bookmarks in alphabetical order?

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Feedback and sugestion

[Robert O'Connor]

> Hello guys!
> In general I find plucker great! A bit slow on a 46MHz Visor, but still great!

Drop the color-depth in the viewer to 1 and things will render faster. 46MHz is more 
than ample 
speed.

> One quite important function that's missing is selection and copy to the
> clipboard... this is very helpful if one uses a dictionary or likes to take notes
> for example.

A patch is available. It may be part of a future release.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Due date behaviour (Was Re: JPluck Conduit based on Plucker Desktop)

[Robert O'Connor]

> Robert wrote:
> >So to decide if a channel is due, just have to read an update_base, and
> add on the frequency *
> >period, then compare that calculated time to current time and see if it
> thus due.
> 
> >After the channel is updated, the update_base gets advanced by
> update_periods until its next
> >calculated due date is in the future. (The update_base doesn't just
> advance one update_period
> >alone, to handle the case where the computer has been abandoned for
> awile).
> >So the conduit would just write the new update_base key and it wouldn't be
> due.
 
> I think the above statments conflict.  I think what you mean is that to
> decide if an update is due I just compare the update_base to the current
> time.  Then after I do the update, I add on frequency * period to the
> current time and store it as update_base.

The behaviour currently isn't quite to add on a (frequency*period) onto the current 
time, to 
handle the case where every update might not take place on its exact time. 

I'll have another go of a exposition and see if I can be less obtuse ;-).
By "due date" I mean the calculated result of a 
[update_base + (period*frequency)].

The current behaviour is set so that the update_base has some meaning. For example, I 
want to 
update The Onion every Wednesday, since the new edition comes out on Wednesday 
morning. So an 
update_base is some Wednesday of whatever month it is, period is a week, and frequency 
is 1.

However, I may sometimes not have my computer on, during Wednesday. Say I miss a 
Wednesday, and 
not update it until that Friday instead. If I just added a (frequency * period) onto 
the 
current time during Friday's update, and stored that as the new update_base, then 
updates would 
start happening on Fridays instead of my desired Wednesdays. So I add a (period * 
frequency) on 
to the update_base, and store that result as the new update_base. 

However, now what if I miss 3 successive Wednesdays before I turn my computer on, and 
happens 
to be a Friday that my computer is turned on? I don't want it to just advance the 
update_period 
by a single (period * frequency), since that would result in it still being due (and 
since I 
just plucked it, I don't need another copy of that same document--I want to wait until 
next 
week's edition comes out). Therefore, I want to keep advancing the update_base by a 
(period * 
frequency), until it results in a due date that is in the future.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Table rendering problem

[Robert O'Connor]

On 25 Oct 2002 at 0:30, David McNab wrote:

>   * http://www.rebirthing.co.nz/temp/pluckersample.html

It's not a problem with the table rendering, that is just how Plucker renders tables. 
It will 
be pretty hard to make sense of that one though.

> Is there any way to do a table so that it shows correctly in the viewer?

Option 1: Make a .png image (may want to bold up the headers and left column to make 
it easier 
to read.

Option 2: Use the  tag and align up the numbers in table. Example
1  2  3  4  5  6

___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: *Sniff*

[Robert O'Connor]

On 21 Oct 2002 at 15:09, Claus Hindsgaul wrote:

> Hi Robert,
> 
> *Sniff*
> 
> Where did my Danish translation go instead of into the 1.2 release?
> 
> I see only a French .mo-file (at least in the Linux binary package) and
> it doesn't seem to be in the CVS.
> 
> :-(  *Sniff*

Hi Claus,

I am pretty sure it is in CVS, unless there is something acting up. It was put in 
about 6 days 
ago when you first sent it. Here is the history:
http://cvs.plkr.org/index.cgi/plucker_desktop/langs/da/plucker-desktop.po

You will also appear as the top part in the 'About' box:
http://cvs.plkr.org/index.cgi/plucker_desktop/resource/html_about/plucker_desktop_translators.ht
m

I certainly give the highest respect for the hard work of people kind enough to 
translate items.

A new Windows package was waiting on Adam's hires version, which would then package 
the final 
1.2 hires, the final 1.2 normal-res viewer, and your Danish translation, instead of 
reuploading 
each package separately. I live in Ireland, and have to pay by the minute for accesss, 
so tried 
to minimize my minutes spent uploading 8 meg files, so that the lights won't go out 
from high 
bills, which will cut productivity. Next year I will cease to be unemployed though, so 
this 
will be less of a concern in the future ;-) 

I don't have a Linux computer at the moment (all I have is a very old laptop). I asked 
someone 
if they would like to rebuild the .spec file though a few days ago.

Best wishes,
Robert

___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Plucker Logo (Was Re: New Web Site)

[Robert O'Connor]

> But I have to admit I cringe somewhat when I see its 'pacifier' icon on
> my Palm screen. I'm not sure what the connection is between a superb
> off-line hypertext browser and a tool for calming crying babies.

Why, you 'Pluck' the pacifier out of their mouths, and that's what causes the crying.

Most logos don't have much to do with their product, that makes them stronger 
differentiator 
logos in the field. An 'apple' icon doesn't have much to do with computers, but less 
computer 
competitors can make a computer with an apple-like thing on it. (There is some 
background 
possible stories about that icon though, such as it is a bitten apple to represent 
knowledge, 
or perhaps even the knowledge of good and evil).

The pacifier does raise a few questionmarks in corporate deployments that otherwise 
might 
choose Plucker, though, if not for a pacifier. [You may have noticed that the alt text 
on the 
website's logo is "What's the deal with the pacifier.] Some like the pacifier though. 
Some even 
like the chicken (which I guess has a stronger connection to 'plucking' since you can 
plucker 
feathers, but even that took me a few months to tie together).

I wouldn't mind a logo submission contest, and either choose a simplified stylized 
pacifier or 
another logo, and use it consistently across everything. Some prototypes submitted, 
then 
community votes. Similar to the Debian logo decision some years back. 

I liked Shep Fairey's (obeygiant.com) stylized T-Rex logo that he did for Mozilla 
(little 
socialist obviously when it is in the star form, but that is pretty much what he 
does). The 
GIMP one isn't bad either, whatever that brown animal is.

On the ToDo for the Desktop is to have consistent logos with the viewer, website, 
docs, etc. 
This is not a high priority, though. So that I won't have to do it a third time, I 
will wait 
until there is a consensus on a Plucker final logo, typeface, and whether the written 
name is 
'Plucker' or 'plucker'.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: documentation on the Web site?

[Robert O'Connor]

On 19 Oct 2002 at 0:41, Michael Nordström wrote:

> On Fri, Oct 18, 2002, Bill Janssen wrote:
> > Right now it takes one to the HTML version of the user manual, which
> > is buggy and out-of-date.
>
> The Viewer chapter is up-to-date ;-)

The documentation for Plucker Desktop is up to date (this is mostly due to my aversion 
to
repeatedly answering a similar question, so things are in docs instead):
http://desktop.plkr.org/docs/user/plucker-desktop_contents.html

This was generated from the PluckerDesktop.tex file. It was written with conditionals 
to allow
it to be a chapter in a larger TeX manual as well as the help used inside the 
application. I
don't mind using some other base format like SGML, as long as I can generate the 
needed output
easily on any platform.

If there is a list o' links somewhere for docs, may want to toss in the Source docs 
too, as
perhaps they will woo some coders, if they can see that the code is fairly well 
organized and
straightforward:
http://desktop.plkr.org/docs/developer/index.html

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: New Web Site

[Robert O'Connor]

On 18 Oct 2002 at 18:10, Chris Hawks wrote:

> Looks GREAT!

Absolutely! 

Looks and works beautifully. Outstanding work David!

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Accept the license or don't *use* Plucker? (was: owner_id_build vs. copyprevention_bit)

[Robert O'Connor]

I should also mention that, that this accept/decline GPL is just for the Windows 
installer. In 
the Linux RPM, there is just a copy of the GPL in the docs directory, and IIRC an 
entry was put 
into the .spec file to say the licence is GPL, so that will show up in the package 
manager. And 
the GPL is in the online help.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Accept the license or don't *use* Plucker? (was: owner_id_build vs. copyprevention_bit)

[Robert O'Connor]

On 18 Oct 2002 at 15:24, Michael Nordström wrote:

> > User has to select the 'I accept' radio button on the license screen,
> > if they want to continue with rest of install. If don't like the GPL,
> > then they can choose not to install.
>
> One problem with that "license agreement" is that you don't have to
> accept GPL to *use* the program. The license only applies when you
> want to distribute the program (or a modified version.)

This is valid. Though in reality, as you say the most relevant part says that if you 
modify and
distribute this software, you have to release the source. If they are just planning to 
use,
then that the clause doesn't apply, the answer is then "Yes", they accept the 
agreement, and
the program continues. It might be a bit of wading through a lot of clauses to find 
that out
though ;-) I am a bit leery of making "short form summary" of the GPL though, since it 
is long
so that it has adequate protection when people try to break it.

> I still think it is a good idea to display the license to the user,
> but I have my doubts about the accept/decline options ;-)

The installer framework has two slots available: One is a license agreement slot with
accept/decline radiobuttons, the other is just a non-interactive regular message. GPL 
can be
moved over from one to the other if needed, though it would be nice to keep the 
message slot
open for a changelog type of description of what is new.

There is also a copy of the license as part of the online help.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Hiya :-)

[Robert O'Connor]

> 1 - Get wierd results trying to download v1.2beta (instantly downloads
> 1.25Mb then goes on to download a further 2.25Mb into an unusable
> .tar.gz - but that's IE for you - yet to try other browsers) 

That browser is braindead. Either that or it a stunt to try and keep Windows users 
away from 
Free software, which is more likely to be in tar.gz that the Window equivalent of .zip.

It expands the .gz to a .tar but keeps the filename of .tar.gz. Rename the file to 
just a .tar 
extension and it is fine.

If someone is minding a FAQ, this would be a good addition.

> 2 - Had you thought of providing 2bit/4bit greyscale imaging? :-)

It is available. Set the parser to greyscale and the viewer to grayscale.


Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Feedback, and need help with compiling

[Robert O'Connor]

On 16 Oct 2002 at 11:13, David McNab wrote:

Hi David,

Thanks for your letter. I can help you with two of these:

> I've tried all the speed settings and pixels-per-scroll settings. When I
> set 1px/scroll, and turn up the speed, the CPU thrashes, and the scroll
> is too slow for reading. At higher px/scroll settings, the display is
> too jumpy. If you can't optimise the autoscroll, perhaps another
> approach might be to go for the line moving down the screen, with the
> current page above the line and the previous page below the line.
> 
> I guess SuperDoc wins the autoscroll war - such a shame it's shareware
> and that some of its code can't be plundered

There is an open source smooth-autoscroll-for-low-clockspeeds in CSpotRun, that you 
may wish to 
use.

The reason why this isn't in Plucker, is because of the code bloat required, since it 
requires 
offscreen caching and sliding up of the caches. And CSpotRun is a much simpler render, 
Plucker 
does a lot more. So in addition to the code bloat, by the time the additions are made 
for 
Plucker, there is no so many calculations required that it isn't smooth anymore 
anyways.

The current method has almost zero extra code, it just acts similar to if a timed 
person 
punched the scroll button. 1pixel is smooth, and since OS5 is now out, there will be 
smoothness 
without extra bloat to code or maintain. (We tested what OS5 would be like by using 
overclockers to bump up Palm Vx's, back when this was written, and was smooth).
 
> 3) Compiling. If there are any of you using Debian GNU/Linux, can you
> please advise what I need to do to compile the stuff in plucker_desktop?

I don't know how well our make system is for it yet, since it hasn't been tested much 
in the 
field yet. There is a standalone makefile though (Makefile-wx-standalone), that you 
can use.
There are detailed step-by-step instructions here:
http://desktop.plkr.org/docs/developer/
specifically, the file:
http://desktop.plkr.org/docs/developer/docs_building_linux_gtk.html

If there are some parts that are unclear or there should be debian specific additions, 
just 
drop a note. If I can help in any other way, I will do my best.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Plucker Desktop URL cant be processed

[Robert O'Connor]

On 11 Oct 2002 at 17:30, Daniel Braunschweig wrote:

> Hi there,
> 
> I am not able able to proceed most files from the URL
> http://www.kino-berlin.de/*
> i.e.
> http://www.kino-berlin.de/021010/programcinema3685762.html
> 
> the error occurs also when i proceed the file itself offline.

Hi Daniel,

The most likely is:
- You need a version of the software from recently, as there was a switch from depth 
first to 
breadth first. Depth-first would not retrieve all pages.

2nd most likely is:
- Make sure you don't have a filter turned on such as stayonhost, or stayondomain.

Best wishes,
Robert 
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Hi there - Plucler Desktop

[Robert O'Connor]

On 15 Oct 2002 at 0:05, Matt Gillard wrote:

> Just a couple of questions. I am running the latest Beta of Plucker
> Desktop.. It seems to work pretty well. A couple of things I have
> noticed is that:
> 
> 1) No way to change the name of your channel once it is created without
> re-creating the channel (the name displayed in the desktop tool does not
> match the name of the pdb it creates). It seems to use the name you
> entered when it asks for a descriptive name for the channel - but that
> name is not editable afterwards.

Hi Matt,

Yes that is correct. A unique section name is generated based on the desctiptive name 
you 
enter, stripped of illegals. The section name is the section of the .ini where the 
settings are 
stored. The file names matches this one. That section name stays the same regardless 
of what 
the channel name is (or it if has one). You can read the archives of the whys that 
this is 
required, I probably won't explain it again. If someone asks a 2nd time, I'll put 
together a 
short FAQ though.

> 2) I cannot get it to successfully pluck http://www.smh.com.au/handheld
> (in the examples list with plucker desktop) or
> http://www.theage.com.au/handheld this is using the standard config with
> link depth set to 2.

Sometimes those newspapers screw up their handheld site, so it might have just been on 
the 
blink.
You could try cranking it up to maxdepth=3 though. I forget if a server jump from 
http://www.smh.com.au/handheld 
to 
http://www.smh.com.au/handheld/
counts as a depth.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: owner_id_build vs. copyprevention_bit

[Robert O'Connor]

On 10 Oct 2002 at 18:29, David A. Desrosiers wrote:

>   That would be an incorrect assumption. In fact, in my meeting with
> the CEO of a company found to be violating the GPL with Plucker source code
> (with our FSF-appointed attorney present), I came up with a method, which
> was agreed-upon, whereby the GPL could be presented as a "clickwrap" type of
> license, and if you disagreed, the application would not launch. I'll go
> into detail off-list, if you are interested, RMS.

For reference, this is currently what happens in the Windows installer package of 
Plucker.

The first page of the installer wizard is a splash screen, pressing "Next>" goes to a 
license 
screen, with the GPL on it.

User has to select the 'I accept' radio button on the license screen, if they want to 
continue 
with rest of install. If don't like the GPL, then they can choose not to install.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Sony devices writing to Palm/Install as well as Memory Stick (Was RE: destination dir (Plucker Desktop) )

[Robert O'Connor]

On 8 Oct 2002 at 12:30, Kundu, Partha wrote:

> I connect the Sony to my PC as a removable medium (the memory stick appears
> as a drive on the PC). I direct the Plucker desktop to write the channels
> into this directory (G:\blah). 

Good to know, thanks. Makes the fix a few weeks back to allow writing to a root drive 
more useful.
 
> So, the question then is, how does it show up (in addition) in the Install
> directory (of Palm)?

The only way Plucker Desktop would put the file into the
/Palm//Install 
directory is if you have put a handheld destination on the Channel dialog > 
Destinations page. 

The relevant part of the source is in palm_installer.cpp and palm_installer_msw.cpp 
files in CVS.

If you don't have a handheld destination specified as well as a directory, it must be 
either:
(a) You are running an "after_command" somewhere along the line in your plucker.ini 
that copies it. I don't know how that would happen though, as you would have 
remembered purposely doing it.
(b) Something Sony is doing.
(c) Other unknown aspect.

Perhaps someone else with a Sony device could shed some light?

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: destination dir (Plucker Desktop)

[Robert O'Connor]

On 8 Oct 2002 at 11:51, Kundu, Partha wrote:

> 
> 
> This is probably a question (comment/bug) for Robert O'Connor.
> 
> I notice that I have specified the memory stick as the destination for all
> 'plucked' files. However, in addition to copying to the MS, it also leaves
> the file in the Install area - with the result that the large plucked files
> are jammed into the memory (not only taking lots of Hotsync time, but also
> depleting precious MB from the 16MB device I have). Any way, the Install
> area is not written to, if the dest area specified is anything but?
> 
> Thanks for any help.

Hi Partha,

Sorry I don't understand what you are doing (I don't have a Sony device). 

The only directory that the .pdb always appears in by itself is the 
/channels//.pdb
The rest are the ones that you set up yourself.

If you can provide some more info about the the step-by-step of what your destinations 
are and 
how they work on a Sony/Memory stick, I may be able to provide some help.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Plucker 1.1 Release Article

[Robert O'Connor]

On 6 Oct 2002 at 14:49, Sudip Pokhrel wrote:

>   unexpected char in declaration: '<' Parsing failed.
>    all 0 pages | retrieved and parsed 
> 
>Just wondering where the problem lies - with Plucker or in the page
>itself? Is there any work around?

Likely illegal HTML.

There is a bug report on this (I don't know if this was yours), that shows a similar 
case of 
illegal HTML.

Work around is to have legal HTML.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Plucker 1.1 Release Article

[Robert O'Connor]

On 6 Oct 2002 at 14:49, Sudip Pokhrel wrote:

>   unexpected char in declaration: '<' Parsing failed.
>    all 0 pages | retrieved and parsed 
> 
>Just wondering where the problem lies - with Plucker or in the page
>itself? Is there any work around?

Likely illegal HTML.

There is a bug report on this (I don't know if this was yours), that shows a similar 
case of 
illegal HTML.

Work around is to have legal HTML.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Plucker viewer: czech

[Robert O'Connor]

And this is a review of the viewer, in czech. Quite extensive, with many screen 
captures of the 
viewer in action.

http://svetpda.cz/svetpda/svetpda.nsf/v/269082378977405FC1256C3600602684

Best wishes,
Robert

___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Plucker Desktop: czech

[Robert O'Connor]

Thanks go out to these guys for passing the word on to the Czech users of Plucker (one 
of the 
language for which the viewer is available is Czech).

http://svetpda.cz/svetpda/svetpda.nsf/v/CCF1E066A768B75FC1256C4300666384

If the new plkr.org website has a section to give a link back for thanks to places who 
are 
actively helping to pass the word about to people about Plucker, perhaps these guys 
could be 
part of that. 

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Need Input/Decision! (was Re: JPluck beta 1 release)

[Robert O'Connor]

> > We shouldn't agree on the language, we should agree on the syntax.
 
> With four or more different parser/spiders, they will have feature 
> divergence at some point.  Just as a minor example, we already have two 
> distinct Python spiders because Robert is distributing my changes adding 
> --stayondomain but Bill hasn't inserted it into his area of cvs, and it 
> probably hasn't been propagated into the Java parser and certainly not the 
> older Perl parser.  But --stayondomain support was -already- in the Plucker 
> Desktop and documented in the help long before I ever heard of Plucker.
>
> That means we -can't- agree on the "syntax"...  we can't even get changes 
> propagated into a single platform.  (e.g. Python in this case.)

I think it's great to have distillers/parsers available in different languages to 
dovetail with 
whatever else it needs to work with, and viewers for different platforms.

--stayondomain is actually a pretty good example of agreeing on the syntax in advance. 
There 
was a feature, and on the discussion lists, a name for the key/option was decided 
upon, what 
sort of value it would take, and what it is used for. Thus, when a desktop tool is 
written, the 
syntax is known in advance so can keep going, and when it is added to the parser, it 
is ready 
to work already with it and other tools, since it uses the agreed-upon syntax. 

I would certainly agree that different language distillers might have some added 
features, but 
using commmon keys are used for the common features makes things more manageable. 


Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: JPluck Conduit based on Plucker Desktop

[Robert O'Connor]

On 3 Oct 2002 at 10:59, Bill Nalen wrote:

> I've successfully hooked up my Java conduit (on Windows) to the JPluck
> parser code.  This is what it does:
> 
>   user hits hotsync button
>   read plucker.ini and grab all the section headings from plucker desktop
>   run JPluck parser with the info (url, doc name, etc.) from the ini section
>   sync the resulting records to the palm

Hi Bill,

This sounds great.

> To do:
>   find some way to select which channel should run automatically using
> the conduit (Robert?)

There is an autoupdate_mode= key in the PLUCKER_DAEMON section that was decided to ba 
a good 
generic place to put the mode of what to do to update. There is a current_mode=3 for 
update due 
channels when I hit the hotsync button. Perhaps there should be one more also, 
autoupdate_mode=4 for update all channels when I hit the hotsync button. 

[PLUCKER_DAEMON]
autoupdate_mode=3

You may wish to check to see if autoupdate_mode =0, =1, or =2 and not do updates on 
syncs if 
the key is set to one of those, so that your users don't do a double update, (an 
earlier pre-
sync update of the channel, and then also an update when press hotsync button).

For deciding on whether a channel is due, there are 4 keys, similar to the RSS 
standard.

update_enabled=1

update_frequency=1

update_period=daily

update_base=2002-10-02T22:57:00

update_enabled means it is scheduled to become due at some point. 0 would mark it as 
never due.

The update_base, frequency, and period are the components to describe a due behaviour. 
The base 
date in this case is oct 02 at 22:57, and it becomes due every 1 day from that time. 
The 
update_base uses the recommended W3C spec for time (it has a capital T in the middle, 
between 
end of date and start of time).

So to decide if a channel is due, just have to read an update_base, and add on the 
frequency * 
period, then compare that calculated time to current time and see if it thus due. 

After the channel is updated, the update_base gets advanced by update_periods until 
its next 
calculated due date is in the future. (The update_base doesn't just advance one 
update_period 
alone, to handle the case where the computer has been abandoned for awile).
So the conduit would just write the new update_base key and it wouldn't be due.

Best wishes,
Robert


___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: List moderation

[Robert O'Connor]

> On Wed, Oct 02, 2002, Mark Lillywhite wrote:
> > Mail to plucker-dev that seems relevant (development) that's from
> > unsubscribed users, I tend to let through.

On a somewhat related note, why does the viewer send mail to plucker-team, and not 
plucker-
list? 

I thought plucker-team was for team discussions, and most things coming from the 
viewer are 
questions on how to use something.

I know the menu item says 'Contact the developers', so plucker-team is the best 
address for 
that description, but perhaps the menu item could be (or have another menu entry of) 
something 
like 'Contact users mailing list'

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Errors plucking with Plucker_1_2 (20020924)

[Robert O'Connor]

On 30 Sep 2002 at 16:20, Bill Janssen wrote:

> > self._auto_scale = config.get_bool('auto_scale_images', 0) or
> > config.get_bool('try_reduce_dimension', 0) or config.get_bool('
> > try_reduce_bpp', 1)
> 
> This change isn't in the current code because the older code didn't
> work that way, and I'm not sure it makes sense to automagically alter
> images for people.

Will people know to fix the key in their .pluckerrc/.ini file? Right now it gives a 
warning to 
reduce the file but doesn't say how. Perhaps a more informative crash message to tell 
them to 
set a key of how to shrink? In the desktop, a key is written for every channel, but if 
using 
commandline only to create/distill, then it will always crash if hit a big image and 
forget to 
put in one of these keys set to 1. 

Or should the pil2 derived class set one of these to 1 if neither is specified, and 
leave the 
base class as is, with both set to zero?

I ran aground with this crash, but eventually figured it out. Wesley now hit it too a 
few days 
later and I could say what fixes it to shave some hours off him also tracking it down, 
but 
others will probably hit it too in the future. You have done so much excellent stuff 
on the 
distiller, I just hate to see people get a crash on it part way through. There is a 
bug open on 
it though, feel free to close it if you like. 

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Errors plucking with Plucker_1_2 (20020924)

[Robert O'Connor]

> self._auto_scale = config.get_bool('auto_scale_images', 0) or
> config.get_bool('try_reduce_dimension', 0) or config.get_bool('
> try_reduce_bpp', 1)
> 
> I'll let you know how it goes.

Yes that is what you need, that I had in mind

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Does Plucker work with MIME type RFC 1437?

[Robert O'Connor]

Does Plucker already support this, or is a patch needed for the current parser? 
Here is description of the MIME type:
http://www.faqs.org/rfcs/rfc1437.html

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Building Plucker-Desktop on OSX

[Robert O'Connor]

On 29 Sep 2002 at 0:51, Alexandre Enkerli wrote:

> [Robert O'C]
> > What is the problem with building the XRC library?
> % pwd
> /Volumes/Media/NuDown/wxWindows-2.3.3/contrib/src/xrc
> % make
> make: *** No rule to make target `xmltok.o', needed by 
> `../../../lib/libwxxrc.so.0.1.0'.  Stop.
> %
> 
> Maybe I did something wrong. I'm not sure which dir I should be in when 
> I issue the configure and make commands.
> 
> Alex

Hi Alex,

The file that isn't being built is part of expat. This file is located as a subdir of 
/contrib/src/xrc. 

Expat (fastest XML processor) is very portable, unlikely to have a problem building 
it. The 
error is from make not knowing where to find the file, which is most likely because 
configure 
wasn't run properly somehow for the library.

There is a brute force workaround if that is holding you up. You can just try editing 
the 
makefile, so that it knows where to find the expat files. Or can try moving them into 
the 
/contrib/src/xrc dir and just adding them to the list of source files after all the 
xh*.cpp 
files.

However, this would be a good question for [EMAIL PROTECTED] Just put the 
subject 
something like 
[wxMac] How to build XRC library on OSX?
and Stefan or one of the other Mac gurus can give you the lowdown. 

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Help request: for someone with perl knowledge and has latex2html on their system

[Robert O'Connor]

I made a mention on the latex2html about adding support in latex2html for a  switch to 
allow 
making of a contents file to build a standalone cross-platform help viewer with client 
side 
search of the pages, etc. This is the help viewer that is used in the Plucker Desktop 
help. The 
latex2html people thought it was a good idea and liked the cross platform help viewer, 
and said 
they would make it a part of the latex2html distro if support would be written. One 
guy signed 
up for the module, and wrote it. Attached is the module, make_html_helpfiles.pm

If someone has latex2html installed and can we test it for Plucker's documentation, 
that would 
be great. All that would be needed would be to build the Plucker docs with this 
module, and zip 
up the output (including the Plucker html files, the .hhp file, and the .hhc file).

These are the authors notes, that he sent over to me:


I'm not sure of the conventions with LaTeX2HTML and plug-in modules, so 
for now I've assumed the new module is in a new directory,
$LATEX2HTMLDIR/HTML_help.

To use the module, create the above directory, copy the module there, 
and add the following code snip to your LaTeX2HTML init file:

sub document_post_post_process {
require "$LATEX2HTMLDIR/HTML_help/make_html_helpfiles.pm";

&make_html_helpfiles;

}


Then re-process your files.

Caveats:
I've made assumptions about contents of the hhp file (compatibility, 
text search etc.)
I haven't added any documentation/pod.
I haven't worked out how to get the title to place in the table of 
contents. I haven't looked at the keyword stuff - haven't got info at the moment as to 
where 
this data is stored during the post processing.
--

Best wishes,
Robert



sub make_html_helpfiles {
print "\nProcessing HTML_help files\n";
my $sHhc_file = "$FILE.hhc";
open HHP, "> $sHhc_file" or die "Cannot open $sHhc_file";

print "\nProcessing $sHhc_file\n";

print HHP<







 


EOD

my $nPrev_indent;
foreach my $sKey (sort keys %section_info ) {
my ($nIndent, $sNode, $sText) = split '%:%', $section_info{$sKey};
# don't output the level 0 item
if ($nIndent) {
# IF indent has increased
# print 
if ($nIndent > $nPrev_indent) {
print HHP<
EOD
}
# ELSE IF indent has increased
# print 
elsif ($nIndent < $nPrev_indent) {
print HHP<
EOD
}

$nPrev_indent = $nIndent;

print HHP< 
  
EOD
}

print ' ' x $nIndent . "$sText\n" if $VERBOSITY;
}

print HHP<
EOD

close HHP;

print "\nFinished $sHhc_file\n";

# process the .hhp file
# I haven't found the variable which holds the title to go at the top of the 
contents tree, 
# so this is hard-coded here to 'Title' for now.
my $sHhp_title = 'Title';

my $sHhp_file = "$FILE.hhp";
open HHP, "> $sHhp_file" or die "Cannot open $sHhp_file";

print "\nProcessing $sHhp_file\n";
print HHP<

Re: Plucker-Desktop on MacOS X

[Robert O'Connor]

> I've been trying to get everything together so I can build 
> plucker-desktop on OSX (10.2 "Jaguar") and I'm still stuck with 
> wxWindows not being able to build some parts of xrc. Does anyone have a 
> binary available?
> Thanks!

Hi Alexandre,

Sounds great! What is the problem with building the XRC library?
(It may be something I can help you with. If not, the wxWindows OSX developers and 
users at wx-
[EMAIL PROTECTED] might have the answer for you.)

As a bonus, a full-on French translation is available, thanks to NH. You could test 
out the 
charsets, etc on MacOSX and see how things work out on the platform.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Anchor Colors

[Robert O'Connor]

On 27 Sep 2002 at 20:44, Shehu Dikko wrote:

> I still see this in a number of sites including the BBC 
> . 
> 
> I can't control how people write their sites so I'd like to be able to 
> overwrite.

> anchor_color=#FF
> anchor_color_override=1
> 
> or any other combination of the proposed switches?

I agree this is annoying. I don't know why BBC, CBC and The Onion all insist on this 
for their 
otherwise very well designed mobile sites.

It would probably be better maintenance in the long run though, instead of both doing 
stylesheets and "override" keys, to instead fold all these into a single style sheet, 
and use 
the strategy of modern desktop browsers use to flick on a switch of "use my stylesheet 
to 
override" so can override all the annoyances/bugs of people's choices for tags, 
colors, etc. 

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

[Desktop] Solaris package available, Visual C++ project/instructions available, MSW installer available

[Robert O'Connor]

For Plucker Desktop:

A Solaris package is available for download. This is due to virtuosity by Mr. Wesley 
Mason, who 
built it with GTK an Solaris 8. It can be downloaded from htp://desktop.plkr.org 
(under the 
linux section at the moment). A big thanks to Wesley for kindly providing this for 
other users 
of Plucker and Solaris.

A VisualC++ project makefile is in CVS, along with some instructions for that 
compiler. Some 
code corrections have also been made so that it compiles now out of the box with VC++. 
The VC++ 
build instructions are at: http://desktop.plkr.org/docs/developer/ 
This is due to the hard work of Mr. Tony McNamara, a big thanks for this.

A MSW installer is available that fixed the image_parser and other keys (from a few 
days back). 

The CVS also has the new tips from Greg M. (a big thanks for these, they are great), 
and a fix 
for scheduling from Tony McNamara (due to some nice detective work on his part). These 
will be 
part of the next installer.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Recipe for a MSW/MAC conduit for updating-when-I-press-the-sync button (Was Re: Development in Windows)

[Robert O'Connor]

On 24 Sep 2002 at 21:21, Bill Nalen wrote:

> One question, although I don't need the answer for a while :-), is there a
> date/time stamp that needs to be updated for the last sync?

Hi Bill There is an update_base key for the channel in the plucker.ini/.pluckerrc. It 
is in the 
W3C recommended format for datetime.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Uploaded MSVC Instructions

[Robert O'Connor]

On 24 Sep 2002 at 13:34, Fringe Ryder wrote:

> At 09:00 PM 9/24/2002  +0100, Robert O'Connor wrote:
> >On 24 Sep 2002 at 11:59, Fringe Ryder wrote:
> >
> > > The document for setting up an MSVC compile of PluckerDesktop is at:
> > >http://www.kittycentral.net/Plucker/Plucker_msvc.html
> >
> >This looks great Troy. Nice work!

> It's TONY. 

Ack! My pardons.

Added to the TODO.txt is to get book an assessment for my reading comprehension and 
retention.

;-)

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Uploaded MSVC Instructions

[Robert O'Connor]

On 24 Sep 2002 at 11:59, Fringe Ryder wrote:

> The document for setting up an MSVC compile of PluckerDesktop is at:
>http://www.kittycentral.net/Plucker/Plucker_msvc.html

This looks great Troy. Nice work! 
Your instructions and makefile will get put into the project straight away.

Best wishes and thanks again,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Recipe for a MSW/MAC conduit for updating-when-I-press-the-sync button (Was Re: Development in Windows)

[Robert O'Connor]

> IMHO, I think what is missing is a conduit (probably using
> the Plucker desktop channels & options) to make it a one step process (once
> the channels are set up).  I think we are close to this too.

Hi Bill, 

You are 100% on the mark.

A conduit now for update-when-hit-button is pretty short in terms of code (at least in 
terms of 
a visual version with the progress dialog, as it will track the termination of the 
processes). 
I have not picked up the Palm conduit kit or have any experience in it, but perhaps 
someone 
with good Conduit skill can have a look.

For MSW, this is what would be needed. AFAIK for conduits a Mac would be similar, so 
it might 
be workwhile to make as a base class with implementations for Mac and MSW to separate 
the 
implementation specifics like a registry.

Start with whatever a minimal conduit project usually is.

When a sync is started:

(1) Get the location of the system's plucker installation directory and store it. It 
is in the 
registry. Doing a search for "The Plucker Team" in the registry will show the folder 
of Plucker 
keys, and one of them is the plucker directory. Usually C:\program files\plucker or 
something 
similar.
(2) Use that stored plucker directory, adding on "\plucker.ini" to get the filename of 
plucker.ini.
(3) Read plucker.ini config file, and look up the value of the autoupdate mode. for 
Plucker. It 
is stored in
[PLUCKER_DAEMON]
autoupdate_mode=3
(4) If autoupdate_code is 3 (update due whenever I hit sync button) then
(5) Use the plucker directory location to get the filename of plucker-desktop, which 
will be an 
addition of "\plucker_desktop\plucker-desktop.exe"
(6) Do a synchronous shell execute command of plucker-desktop.exe with the commandline 
option 
of --update-due
for example the entire string would be something like:
"C:\program files\plucker\plucker_desktop\plucker-desktop --update-due"
By a sychronous shell execute, I mean that the the program flow of the conduit logic 
is halted 
until the shell command terminates. 

For this to work, the conduit would need to be able to run before software gets 
installed to 
the palm (I don't know if that is possible in the Palm conduit). 

There is a Desktop UI widget ready to go, it is the 4th radio button in Preferences 
dialog > 
Autoupdate tab: this reads/writes the autoupdate_mode key.

Best wishes,
Robert

___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Desktop Java option

[Robert O'Connor]

On 24 Sep 2002 at 9:05, Bill Nalen wrote:

> Robert, can we have an option to use the Python parser or Bill's Java
> parser?  As I mentioned earlier, the Java parser seems to allow me to pluck
> content from my intranet using authentication and the Python one doesn't.

Sure thing. All that would be needed is the commandline that is required.

> I think it would also allow end users to not have to install Python if they
> have Java already installed.  I may also be able to compile the Java code
> to a native Win32 library to take away the need to install an interpreter
> at all.  I haven't gotten the source yet, so I'm unable to offer a patch to
> do this.

External libraries don't work very well in wxWindows as they need to match the 
compiler it was 
made on. If there is a possibility that could wrap the java library into an exe, or an 
exe that 
just loads and gives a public interface to the DLL that work great. 
 
> Also, would it be useful to view the pdb after it has been completed?  Like
> a view now option?  This could use the code from Bill's GTK viewer perhaps?
> Just a thought.

It would he handy to allow a desktop viewer, for reading docs, and also for 
development.

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Re: Plucker Desktop HTML editor's icons

[Robert O'Connor]

On 24 Sep 2002 at 19:20, Nicolas Huillard wrote:

> I'm looking for somewhere to find pretty icons for the editor (blue on gray 
> ones are not very good; not that would make better ones myself) :
> * is there somewhere a library of free icons to pick from ?
> * did someone already picked on this ?
> * is there a special requirement for the format of these icons (PNG, 32bpp, 
> 21x22) ?

I agree they aren't very pretty. I was thinking of moving to some white on blue/purple 
like in 
the online help, which I think are pretty enough, and still readable on low bit 
depths. 

It is beneficial, though, if they can render legibly on systems with colors set to 256 
colors 
(like mine). The dither is horrible and unreadable even on 256, so solid colors are 
used.

IIRC, higher bitdepths were needed for transparency on the PNGs. There are curretly 5 
image 
handlers that are included in the Desktop, which you can use for icons:
-BMP
-GIF
-JPG
-PNG
-XPM

PNG, GIF and JPG libraries were put in so that can preview those images in the editor.

Probably over the longer haul, there will be a switch over to exclusively XPM for 
icons, as it 
eliminates the requirement for the PNG library, which is an extra dependency to worry 
about on 
POSIX. The xpms would be kept in a zip file for compression, and read from the zip as 
a zip 
filesystem.

There is no requirement for 22x22. Feel free to use whatever size you fancy. The only 
thing to 
change is that if you open up the editor_dialog.xrc file, you will see parameters in 
the editor 
toolbar XML for the size of the bimaps to use. (Doing a search for 22 should bring you 
to the 
right spot). Just change those parameters, put in your different sized bimaps in the 
appropriate resource subdirectory and restart the desktop. (The nice part of XML 
resources, in 
action).

Best wishes,
Robert
___
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev