plucker-build --foo vs. -foo
I recall a time when the Spider.py script would bomb on invalid arguments, and now... well, it doesn't, it just crashes or continues on, ignoring invalid options. I was trying to figure out the one for forcing an icon (8bpp icons on IIIc images consistantly disappear after launching the db at least once) $ plucker-build -f /tmp/CreateRSS -H "http://www.perl.com/lpt/a/2001/11/15/creatingrss.html"; --zlib-compression --bpp=4 --no-urlinfo --maxdepth=1 --foo Traceback (most recent call last): File "/usr/local/bin/plucker-build", line 1023, in ? usage (text) File "/usr/local/bin/plucker-build", line 804, in usage print 'Error: ' + reason TypeError: cannot add type "instance" to string $ plucker-build -f /tmp/CreateRSS -H "http://www.perl.com/lpt/a/2001/11/15/creatingrss.html"; --zlib-compression --bpp=4 --no-urlinfo --maxdepth=1 -foo Working for pluckerdir /home/hacker/.plucker ZLib compression turned on [...] Note the first one with --foo bombs, but -foo happily continues on. Any idea on the icon issue? /d
Re: Plucker Feedback
> Hi. I'm really enjoying and appreciating Plucker. However, I've found > what seems to be one small bug. It doesn't seem to work to use 4 bpp > for your images, because ppmtoTbmp doesn't have a 4bpp switch, only a > 2bpp switch -- unless there's a better version of it out there somewhere > I can't find? How old is your version of Plucker? We haven't used ppmtoTbmp in close to a year now. It was deprecated long ago in favor of the new PIL and ImageMagick stuff. /d
Re: PDF converter? (David again :-)
> David, is the pdf converter on the site working? Yes, it has always worked. Well, ok it broke when I moved everything from plucker.gnu-designs.com to plkr.org, but it's working now. From the logs, 1,239 different PDF files have been converted using the tool since I created it. > I put this url in: http://www.therationaledge.com/content/nov_01/PDF/The > Rational Edge - November 01 Issue.pdf and hit the button. A little > while later it wants me to download pdf2pl.pl, but it doesn't have > anything in it. Is this the actual perl code to parse it? Do I need > Perl installed locally? I was expecting to get a pdb file back. You must be using some form of IE. It's been a known bug in IE for quite some time, and still exists in IE6 (with the .NET client lib infrastructure underneath it). http://support.microsoft.com/support/kb/articles/Q267/9/91.asp http://support.microsoft.com/support/kb/articles/Q281/1/19.asp http://support.microsoft.com/support/kb/articles/Q279/6/67.ASP However, you did uncover two problems I had, which I managed to fix. In fact, I am going to gut and rewrite that so it can handle Word as well as PDF, with a file-upload function so users can upload, then delete the file after conversion (uploading it for conversion is the only way to handle local documents, server-side, but a URI reference to the doc will work also). 1.) I was using a very stupid insecure manner of parsing scalars. 2.) Filenames with spaces in them must be urlencoded first before conversion. I really can't stand when people (improperly) use a space as a character in a filename. I blame Windows for propagating that gross misconception that a space is a delimiter as well as a syntactical character in a filename. Ugh. In any case, a space in a filename in a URI is an illegal character, and it must be escaped to be %20 and so on. Your filename in this case, escaped, turns out to be: http%3A%2F%2Fwww.therationaledge.com%2Fcontent%2Fnov_01%2FPDF%2FThe%20Rational%20Edge%20-%20November%2001%20Issue.pdf This is now being escaped automagically (URI::Escape, three lines of code total to do this part of the chop, escape, convert, gotta love perl) > Also, it was kind of hard to find the link to do the pdf conversion (I > found it in the old news articles). Perhaps it should have a link or > section on the test drive or samples page? Perhaps also the samples > page should have a link to a site that collects e-texts in plkr format? I will take a look. I'm thinking of beefing up the "Build It" page a bit, making the site "flow" a bit better. Under the hood, it's running well, but could be optimized a bit more, more interactivity. I have a few post-it notes up for some lingering issues. The PDF/Word conversion is one of them, and you're right... it needs to be more prominently figured on the website. > I know you're busy this week & weekend. Just some thoughts, no rush. Busy? BAH! I'll have lots of spare time soon. Incidentally, I'm flying out to CT for two weeks to hang out with my girlfriend and her parents over Turkey Day, so I may be in and out of email for a bit, depending on the dialup situation there. After I return from this "Sabbatical", I will have a lot of time to devote to building, fixing, solving these issues. /d
Re: Plucker 1.1.13 source (David)
> David, can we get a source zip on the downloads page for the 1.1.13 > stuff like there is for 1.1.12. I'd like to build the viewer and I > don't know how to use CVS (probably have firewall problems too). Sure, let me take care of that this weekend. I'm cleaning up some last-minute business at my (now former) job, so I'm a bit time-limited until this weekend. /d
Re: Plucker Desktop
> ...and that doesn't help Chris, since it only gives you the main trunk > ;-) Ok, my bad. I thought he wanted the cvs snap itself, not the Plucker Desktop branch. Blech. Well, there's always the 'Download' link for each file in the cvs view =) /d
Re: Plucker Desktop
> No, the problem here is that the sys admin doesn't want to allow CVS > thru the firewall. I'm working on it tho. (Let's see if I stand on my > tippie-toes and lean way over...) ...which is why the "Snapshot" link exists.
RE: Plucker Desktop
> Please help me unsubscribe the address [EMAIL PROTECTED] from this > list. The initial email you received when subscribing had confirmation instructions which detail how to unsubscribe from the list. You will need to locate that email, or send a message to [EMAIL PROTECTED] to be removed from the list. Good luck. /d
Re: Plucker Desktop
> Still can't access CVS? I have put a package with the code from CVS at > http://www.sslug.dk/~micke/plucker/desktop/ Are people having trouble accessing CVS? It's been up without incident (except for a slight downtime due to kernel upgrades). Let me know what machines you're connecting from. I'm blocking most of 64.x.x.x because of repeated Nimda.e attacks. Every machine that hits me with a Nimda.e request gets blocked with iptables. That may be the cause of some of it. Let me know, and I'll do what I can to address any issues. /d
Re: ADV: Palm QuickWrite
On Thu, 8 Nov 2001 [EMAIL PROTECTED] wrote: > no text body I am reminded of a Monty Python song... o/~ ...SPAM SPAM SPAM SPAM... o/~
Color images when zoomed, appear in greyscale
I've been working with a user trying to figure this out. He started by asking me to decipher this "Bikini of the Day" site he wanted, and has uncovered a very unusual "bug", though I hesitate to call it that yet. When he gathers the content, using his Windows-style home.html file under Win98SE (file shown below), he gets his "Bikini of the Day" page, and when the image is tapped, it zooms to show him a greyscale image, not a color version. When I tried this using plucker-build, I get a color image which is too large to display (Alert message is shown). Is there something different in the Windows parser that perhaps is dithering down the image from color to greyscale instead of showing an alert message? Dirk? The home.html file is below, and the db is compressed and attached: Plucker Home Page Plucker Home Links Weather http://www.yourdailybikini.com"; STAYONHOST MAXDEPTH="1" MAXWIDTH="150" ALT-MAXWIDTH="500" BPP="8">Daily I told him tables weren't allowed, and he changed it, but this is his uncorrected home.html file. Attached is also his original db, apologies for the size (the original was 143k though, I had him cut it down and rebuild it for this bug report). I also noticed that in a IIIc POSE session (I didn't have a Handspring ROM to test with, he has a Prism), after leaving Plucker, the icon is erased for the one I created (bikini.pdb in the attached tarball) using his home.html file under linux (with launchable = yes in ~/.pluckerrc). I just see a title called 'bikini' under where there should be an icon, but is a blank space. The syntax I used for this was 'plucker-build -f /tmp/bikini --alt-maxwidth=500 --alt-maxheight=500'. Some memory related bug taking out the icon? Just launching it and then immediately exiting it causes this to happen. Resetting POSE seems to cure it and the icon comes back until launched again. Weird. /d BikiniOfTheDay.tar.gz Description: Binary data
Re: Error page on PDA schedule (fwd)
> In any case, forging my Referer with Junkbuster solved it. Another volley of emails and now the site is open, sans Referer. The previous restriction on passing the "AvantGo" headers has been removed. YAY! /d
Re: Error page on PDA schedule (fwd)
I love watching fish out of water. *sigh* Apparently my "PC" is causing their server to barf. That's one I've never heard of before. Apparently using a different PC is supposed to fix their server-side issues. [spits milk out nose laughing] I only bring this up because from time to time, I go re-scrape the database of URLs here to make sure they're valid, and some turn up turnips, so I email the admins to see what the problem may be. In any case, forging my Referer with Junkbuster solved it. -- Forwarded message -- Date: 12 Nov 2001 18:06:32 -0800 From: [EMAIL PROTECTED] To: [EMAIL PROTECTED] Subject: Re: Error page on PDA schedule David, The URL is valid. Try again using a different PC. On Mon, 12 November 2001, "David A. Desrosiers" wrote: > I get an error trying to access the following URL on your site: > http://www.pdaschedule.com/avantgo/content/index.asp > > The error I get is: > An error occurred on the server when processing the URL. Please > contact the system administrator. >
Re: So about that 6000 dollars
> basic requirements: > 306K for application OWCH! What are they putting in these applications?! > Looks like they've improved matters slightly now that version 3.0 is out > of beta, but I still have no desire to waste 128kB. This smells of the "Why improve or optimize the speed of our code, let's just bump up the requirements of the hardware needed to run it!". Sound familiar? /d
Re: So about that 6000 dollars
> If I hammer the server hard enough, can I make it catch fire? (OK, poor > joke, I'm sorry... I've just seen enough bad php to last me many > lifetimes.) The language doesn't make bad code, programmers make bad code. Some languages make it easier to write bad code, however, but I think that's also dependant on the programmer's level of understanding of the actual code and the objective. TMTOWTDI. > I don't see that your comments make a decisive case. Libraries exist > for many other languages, including perl, to do what you want and most > of those languages are more powerful and elegant than PHP. Nobody said it was decisive. Someone offered, it was one less thing I had to do myself, and I have some local talent I can prod for help if I get stuck looking at/fixing/auditing the code. Shortest distance between two points.. Seriously though, if I can do something better in perl, I will. > Yes, that's good, but are they going to help maintain it? I made sure he understood this was not just a "throw it over the wall" type of project, but even if it is, and he walks away, I want to make sure I can understand the code enough to be able to fix/update it as necessary (note bugs.plkr.org was my first foray into the PHP menage, and it's doing ok with little babysitting). I know php enough to be probably at the Amateur Lethal status right now. Level 4 Sorceror. Nothing more. > > I make it a point never to turn down a contribution or contributor. > > That is not a healthy statement. Again, I know that's politically > incorrect, but that's the way it is. I meant 'turn down' from a morale and future contribution perspective, not from a "Hey, we'll just toss this code in there and see what happens" perspective. If you turn them down now, they won't be inclined to learn more, help you further in the future, or contribute. If you teach them to fish... /d
Re: So about that 6000 dollars
> OK, I'll bite: php? Dear God, nooo! Is plucker going for a > record number of languages used, or something? ;-) You forget whose house is right up the road, and whose house the Plucker server is currently colocated within... Rasmus Lerdorf, author of php. Not that he'll help me, but I can be annoying enough to get some of his unique criticism when required, I think. Seriously though, php is ideal for this over perl as much as I hate to admit it. It does lend itself well to session handling and form processing and very fast database access. All of which will be required attributes for this system to work. Perl in this case would be much more work for me to implement in the short term. Whatever makes the job easier. Besides, I have a volunteer that knows php, and that's less work I have to do actually coding it. I can get onto other things taking up my time and burn through them. I make it a point never to turn down a contribution or contributor. /d
Re: So about that 6000 dollars
> So where are the people promoting Plucker in that thread? I only read > my local palmtops group, as I don't have that much time or a > particularly big stake in palms at the moment, but I often promote it > there. I have sent a message to the proprietor of GoChannels.net to see if he is willing to work with us. I'm sure PODS could infuse his collection quite extensively. He has 32 "channels" now, and currently PODS has 543 records. I should audit it to make sure they're all still valid. I'm sure 20-30% have moved or been closed down by now anyway. We'll see what happens. It's important to note that AvantGo restricts content from being shared. How is this fostering the spread of information? How does this benefit content providers? /d
Re: So about that 6000 dollars
> From the palm newsgroup archives the past few days, for those that don't > read it full time. I didn't see one single active plug for Plucker there. Hrm. I should really do a head-to-head and compare/contrast AvantGo with Plucker with iSilo with Sitescooper, etc. I was at ALS on Saturday and spoke with Betsy Waliszewski (Product Marketing Manager for O'Reilly) and she promised to get me the name of someone who would be interested in my articles regarding PDAs and the work we're doing in this space. The conversation started when I showed her the O'Reilly Linux Device Drivers book and Samba books online in Plucker format. I offered to help them get their content "PDA-ready" on their online books download section. She seemed interested in the idea. Perhaps it would be a good entry point for Plucker and for doing a comparison of the alternatives. In the meantime, there's always this message I posted from awhile back: http://gnu-designs.com/code/plucker-vs-avantgo.txt Someone want to memorize those features, and offer them when the newsgroups rile up like this and think that AvantGo is the only alternative out there? /d
Re: So about that 6000 dollars
> From the palm newsgroup archives the past few days, for those that don't > read it full time. . This is a nice example of what David was discussing > in regards to the heavy tolls being tossed onto the small independent > content providers. I've got a guy volunteering already to help with the php-based front end for "Wucker" (ok, I couldn't think of a name, Web Plucker), which is going to expose the innards of PODS to the users. We're working out a schema now, which will make it easier to do some things AvantGo and other similar services don't yet do. /d
Fling-It (duplicate CreatorID warning)
As you probably know, deleting Fling-It via the Palm-provided launcher facilities will also delete all of your Plucker data and applications as well, because they share the same CreatorID. I just did this and tapped [Ok] when prompted before I remembered that gotcha. Oof! Deleting it via Z'Catalog, Filez, McFile, etc. will not have that smae effect, and will preserve your Plucker data and application. My bad. /d
Look familiar?
Look very closely at that toolbar: http://freewarepalm.net/communication/flingit.shtml I haven't seen Ben post in awhile. I just installed it, and there's a very small sentence there about the GPL on the installer "License" window. No biggie, except.. he didn't change the CreatorID, so Plkr still exists. Ben, if you're here, can you check a few things in there, and make sure you've got your own CreatorID for Fling-It? People who install both will be (un)pleasantly surprised when their application icon changes and the Fling-It application is nowhere to be found. Duplicate CreatorIDs are bad. Also, the app seems to crash quite a bit in general. Load it up with no content, then tap on Open and you'll get a NilPtr fatal on like 4450 of MemoryMgr.c. I tapped Last and it brought me to the last Plucker document I had read inside Plucker, but launched Fling-It to view it. Crash. Just an FYI.. /d
Re: How to get source of Plucker 1.1.14
> How can I get the complete source code of 1.1.14 as a tarball? Download it from the cvs. The branch is plucker-1_1_14, and you can browse it here: http://cvs.sourcefubar.net/cgi-bin/vcvs/viewcvs.cgi/?only_with_tag=plucker-1_1_14 You could also use the Snapshot link from the website, but that will pull HEAD from the cvs with the date of the time you click the link (It's pulled live). /d
Re: Struggling windows user
> I have double clicked the Plucker\PyPlucker\Spider.py script, and it ran > though in a Dos window quite quickly and then closed. How do I know if > it has done anything? Have you downloaded the Windows tool and installed it? I don't think you need to bypass the tool to go running the parser scripts directly. What do you expect clicking on Spider.py to do for you? > If there is a "Plucker starter guide for windows non programers" I would > like to read it. If there is not I would like to write one. We would be glad to get a "Starter Guide" from you. That would be a welcome addition to the project. > I have deleted avantgo, and am missing the bbc news site, which is all I > can fit after the avantgo program on my 2Mb palm III. Plucker is a much > nicer size and takes heaps less time loading and unloading, I am looking > forward to using it. These might help get you back on track: http://news.bbc.co.uk/hi/english/static/avantgo/default.htm http://news.bbc.co.uk/text_only.htm Funny thing is the second one, labeled 'text_only', has graphics, and the first one does not. Weird Brits. =) /d
Re: Change the Palm OS requirement
> Adding #ifdefs would not cause a lot of pain; to support and test an > unlimited amount of different versions will be a real PITA, though. $ make [lite, full] $ Is your handheld shown in the list below? [display a list of compatible handheld units here] It may be more pain for those that compile, but they're expected to know what they're doing for the most part. I've talked to people that say that they compile all their stuff with the same three commands in a script: './configure; make; sudo make install', and it makes me laugh. In any case, perhaps with a gui tool like Roberts, you pick which handheld you use, and the appropriate viewer is selected, and the same can be modeled in a script as well. /d
Re: Change the Palm OS requirement
> I guess the number of devices running 2.0 are less than 1% and it would > make more sense to strip out all the 3.x stuff from the viewer to create > a "Lite" version for those devices (they are already low on memory). > Anyone using 3.0 can upgrade to 3.3. My only concern would be that we are treading close to alienating the Handspring users with the 3.1 or greater requirements. Can we make sure we test those capabilities on the Handspring units? The only units which do not support a multibyte ROM right now and are OS3.1 or later are the IIIx, IIIe (not flashable, but running 3.3), IIIxe, V, VII, VIIx, ALL of the Handsprings (none are flashable), the original IBM Workpad (which is running 3.0, but is flashable), the two Qualcomm units, the Symbol units (1500, 1700, and 1740, runs OS3.2 anyway though), and the original TRGPro (but it runs 3.3 or 3.5.1, and is flashable). We shouldn't be in the game of forcing or requiring users to join the Palm Developer Pavilion to get access to PalmOS ROM images to flash their own devices to allow Plucker to work if we move beyond OS3.1, but I think as long as we hang at 3.1 for awhile, or provide "lite" versions, I would be fine with that. /d
Re: This one's weird..
> But it seems to have a content-type: > > > $ HEAD http://www.netlet.net/interact/Babe.jpg > > 406 No acceptable objects were found > > Content-Type: text/html > > Client-Date: Sat, 10 Nov 2001 00:58:20 GMT > > Client-Peer: 206.58.152.238:80 A jpeg most-certainly does not have a Content-Type of text/html. Yes, that's probably the Content-type of the response/error page spitting back at me, but as you can see, it can't determine charset, language, or Content-type. That's why I said force/forge (note the latter) to try to munge it through. /d
This one's weird..
Some guys in irc wanted a way to get this "Bikini of the Day" site
on their Palm, and they were using AvantGo, which I weaned them off of, but
Plucker doesn't like the url they're passing to it. What's weird is...
neither does anything else I point at it, except Netscape/Mozilla and wget.
All of my other tools barf trying to retrieve it. Here's the url, and here's
what happens:
$ HEAD http://www.netlet.net/interact/Babe.jpg
406 No acceptable objects were found
Content-Type: text/html
Client-Date: Sat, 10 Nov 2001 00:58:20 GMT
Client-Peer: 206.58.152.238:80
And when plucker-build is pointed at the image directly:
$ plucker-build -f /tmp/bikini -H "http://www.netlet.net/interact/Babe.jpg";
--maxdepth=1 --bpp=8 --zlib-compression --no-urlinfo
Working for pluckerdir /root/.plucker
ZLib compression turned on
Using exclusion list /root/.plucker/exclusionlist.txt
Using exclusion list /root/.plucker/exclusionlist.txt
Processing http://www.netlet.net/interact/Babe.jpg.
0 collected, 0 still to do
Retrieved failed: 406 -- No acceptable objects were found
Fetching the home document failed. Aborting all!
Fatal error while processing.
And when pointed to the html page that holds that image:
plucker-build -f /tmp/bikini -H "http://www.netlet.net/interact/bikini.htm";
--maxdepth=1 --bpp=8 --zlib-compression --no-urlinfo
Working for pluckerdir /root/.plucker
ZLib compression turned on
Using exclusion list /root/.plucker/exclusionlist.txt
Using exclusion list /root/.plucker/exclusionlist.txt
Processing http://www.netlet.net/interact/bikini.htm.
0 collected, 0 still to do
Retrieved ok
Writing out collected data...
Writing document 'bikini' to file /tmp/bikini.pdb
http://www.netlet.net/interact/bikini.htm has charset 4 (ISO-8859-1)
Converted http://www.netlet.net/interact/bikini.htm
Default charset is MIBenum 4 (ISO-8859-1)
Converted plucker:/~special~/index
Converted plucker:/~special~/metadata
Wrote 1 <= plucker:/~special~/index
Wrote 2 <= http://www.netlet.net/interact/bikini.htm
Wrote 5 <= plucker:/~special~/metadata
Mapping:
2 => http://www.netlet.net/interact/bikini.htm
Done!
It smells like a missing Content-type header, and I thought there
was a fix in cvs for this. I recall someone mentioning this before, but I
can't find the reference in my archives.
Bill, can we add a feature to the parser to pass (force/forge) a
Content-type header to the parser? --content-type={$type}.
/d
My bad, ignore that last one
I forgot to run autogen.sh from the toplevel dir first.
Kaboom! (install-plucker crash)
Simple errors: medusa:/tmp/plucker/unix# ./install-plucker This script will guide you through the installation of Plucker @@VERSION@@ If you run this script as root the default values should be OK, if you want to install Plucker as a normal user you have to change the paths. Installation path prefix [/usr/local]: In which directory do you want to install the binary files [/usr/local/bin]: Install Python parser in default location for site packages, /usr/lib/python2.1/site-packages? [n] y In which directory do you want to install the Plucker data [/usr/local/share/plucker]: In which directory do you want to install the documentation [/usr/local/share/plucker/doc]: Do you want to install pluck-comics (used to gathers comics from selected websites)? [y] n Do you want to build the documentation? [y] n Do you want to build the viewer and ZLib binaries? [y] n Running ./configure --with-pypluckerdir=/usr/lib/python2.1/site-packages/PyPlucker --disable-docbuild --disable-palmosbuild sh: ./configure: No such file or directory Continue? [y] y Running make... make: *** No targets specified and no makefile found. Stop. Traceback (most recent call last): File "./install-plucker", line 126, in ? os.symlink(os.path.join(MOD_DIR, "PyPlucker", "Spider.py"), os.path.join(BIN_DIR, "plucker-build")) OSError: [Errno 17] File exists /d
Re: Plucker 1.1 Release Article
> But how will I use an IDE like KDevelop with prc-Tools,compile and debug > plucker or other proejcts? Can someone guide me please? This may be something more suited for the pilot-unix list for the early term, using Plucker as a model for your design and development. You can find the necessary information here: http://hcirisc.cs.binghamton.edu/pipermail/pilot-unix/ /d
Re: Handheld directories
> Is there any dmoz.org similar organizations for organizing/maintaining a > reviewed directory of handheld-friendly sites? Yes, PODS, 500+ URLs now. I maintain the db, but it's not clear about liability and legalities if I make them all public. I don't want to put content providers in violation of their restrictive agreements with "Those Other Companies(tm)". Hit me privately. /d
wxPerl! YEAH!
Alright, you say po-tay-to, I say wxPerl! http://www.perl.com/pub/a/2001/09/12/wxtutorial1.html I'll see if I can get Robert's gui or parts of it working in this, just something to tinker with. You all know how I feel about Python. /d
RE: I cannot delete one article
> I wish there was a delete function from within the open document. Ever climb a tree and cut the branch you were standing on? It's generally not a good idea in the Palm environment, to go deleting resources from within your event loop while memory is locked by the resource currently in use. I'm sure there's a way, but... you don't do this on the desktop, why would you do it here? Do you delete your files by opening them first in their respective applications, then deleting them from within the app? Personally, I use a shell or if in Windows, a file manager, to handle those kinds of operations. /d
Possible idea for "No Docs Installed"
One of my coworkers gave me a Kyocera 6035 the other day (gratis) to keep, and I notice that the voice mail messager application has an interesting screen that is similar to something we're talking about with the "No DOCs Installed" idea in the DocMgr. Check out the attached screenshot. When that entry is tapped, a form is brought up that says "You have no new Voice Mail". Perhaps we could use something similar when no docs are installed in the viewer? It's quick, it's small, and it only consumes one form resource. /d Kyocera_Messager.gif
Re: Plucking Slate.com, a Python example
> The MSN change has affected Slate.com, an online magazine owned by MS. Even their own page has broken links: http://slate.msn.com/?id=111843 (many of these are broken now) I used to use this link: http://slate.msn.com/Code/TodaysPapers/TodaysPapersHHF.asp That's also broken. I'll keep digging though. It's there. Personally, I'm all for not supporting this at all, and letting their readership drop like a rock. I don't like supporting a company that consistantly lies to it's readers, then slaps those of us supporting standards in the face like they've done recently. Let them hemmorage on their own. They're doing a good job of that. /d
Re: While we're gutting the python parser...
> I'm still not sure how this would work. The user already sets the font > preferred to read documents in via the Preferences panel. No, they don't "set" the font, they *CHANGE* the font via the Preferences panel. What I'm suggesting is that we have a way to set a default font for that pdb at pluck time. From there, the user can change the font to be whatever they want, but if they beam it to someone, restore it from a backup, whatever, the default font is "narrow" for example (if that's what they used at pluck time). > Presumably this setting would always override any "default viewing font" > specified in the document. So what's the point of putting it in? Because it's stored in the main pdb, not in the secondary pdb, which nobody beams across when sending content or restoring from backups, or distributing to users. /d
Re: behavior of missing tag?
> This is extremely counter-intuitive. The user is jumped to some place > that may be wildly out of context (for a long page). I'm going to > change the parser to treat such URLs as excluded, just as with any other > URL which doesn't exist. I have seen pages do this *INTENTIONALLY* by design. Yes, lazy html authors who use #top and don't include an anywhere in the top of the content, but they know it'll jam the user back up to the top of the page. Are we sure changing this by default is right? Shouldn't we act like browsers would in this situation, not like an AI would? /d
Plucker-1.1.13.exe mirror, 1.1.14 update
I've mirrored the Windows Plucker-1.1.13.exe package onto plkr.org, based on about a dozen emails I've gotten from people who have tried to download it from prohosting. Dirk, I'll work on a more secure way that you can upload the package to the proper directory automagically, and not have to worry about prohosting. Just have to clean up a bit of the scripts around the web stuff first. Did you manage to solve the bugtracker issue on your end? I also dropped the 1.1.14 packages in there, and plopped them on the download page. I'll make a front-page announcement later on today. I slashed my hand pretty bad in a fall yesterday, so it's hard to type. EEK! Let me know if I typo'd something. I also noticed that Mozilla completely breaks the page, viewing it, so I'll put some time in this weekend and try to find out what happened there. Probably my nasty nested table rendering stuff. Incidentally, I got a message from the Beyond2000 guys about promoting their PDA-sized page on the Plucker site, so I'm going to start a new section there "Plucks of the Week" or something, and have some of those sites highlighted. Anyone graphically inclined want to make a small "Plucker Friendly Site" banner that we can pass around to these "partner" sites? I made one awhile ago, but I'm sure we could update it a bit more. http://plkr.org/images/pf.gif /d
Re: While we're gutting the python parser...
> I don't think I understand this suggestion, David. The HTML specifies > the various fonts; I'm not sure what overriding it on the command line > would add. I am suggesting we store the default font that Plucker *DISPLAYS* the content in, not the font that the original HTML presented at gather time. > Are you suggesting this just for text files? In that case, I think it > would be better to keep them as they are, and allow the user to adjust > the UI in the viewer (via the preferences) to select an appropriate > font. My goal was to set it at pluck time, to set a default font, which can be overridden via Preferences. If no font is specified at pluck time, the generic Standard font is used. /d
Re: Bug tracking website login?
> David do you have changes something with the bug tracker scripts? Nope, nothing has changed. I upgraded the version several months ago, but nothing has changed since. > http://gnu-designs.org/bugs/login.php?f_username=heiserd&f_password=XX This works here, I just tried it: http://gnu-designs.org/bugs/login.php?f_username=guest&f_password=guest&f_perm_login > If yes, can we please get this feature back? (yes i'm to lazy to type > the Name/Pass every time :-) Nothing changed. Perhaps the cookie was confused? Try deleting the cookie, close your browsers and try again. /d
Re: Python Parser Retrofit
> Keep in mind that the parser ask for user input if an link need > authentication. In this case you get an problem. Sure sure, that's not a problem in my particular case though, since I'm pulling publically available (static, controlled) content from various places and need to jam it in as standard input. /d
Python Parser Retrofit
While we're on the subject of using standard redirection in the parser, so we can do plucker-build [...] > foo.pdb, I wonder if we can also use the opposite for input. I have a need now to do things like: plucker-build [...] < $content_stream ..which is raw html itself, not a url reference or a file. Possible? /d
Re: debug.c
> I don't think any of the top-level capitalized docs should be turned > into XML. They are for people to bring up in an editor (emacs, of > course :-), not a web browser. ...and before a release is made, they are run through the proper style sheet to turn them into 7-bit ascii format suitable for loading into a plain editor. However, I need to use them and manage them so they are digestable from the web, with foo links intact (and clickable), which in an editor, would become rapidly annoying. > But isn't it easy to bring up text directly from the CVS, too? I mean, > for the README, I can type > http://www.plkr.org/index.pl/cvs/README?rev=HEAD, and get a perfectly > readable version. The point is to make the content maintainable, while separating it from the presentation layer, much like I've done with the news articles on the site right now. The next step to that, of course, is storing content (and not presentation) in a database of URLs, which users can then choose from for their own needs, and arrange accordingly. /d
Re: debug.c
> Unless you're in the MS world, which seems to have completely junked use > of the In-Reply-To and References headers. Outlook in particular is a > bad offender in this area. ...and as Microsoft would have it's users believe, market share dominance seems to be what determines standards these days. Witness their blocking of browsers from the MSN.com and partner websites (thus denying their paying customers and paying advertisers of their guaranteed revenue streams), claiming "standards compliance" as a motive. "Oh, you mean you block browsers which don't adhere to the Microsoft standards... not the w3c standards... I see..." More FUD from our friends in Redmond. If they convince 80% of their users not to worry about using any other browser than IE, and other users who are using non-IE browsers begin to forge the UserAgent string to "smell" like IE to the servers, the Microsoft userbase increases (even if the number of actual IE instances goes down), because people will be pseudo-IE users. Personally, and I'm sure Bill sides with me on this one, given his charge at Xerox/PARC, I can't stand this abuse of the standards by a company who has already been found guilty of monopolistic behavior. Market share dominance does not determine standards. In any case, I think that you can change the behavior of how Outlook threads email anyway, but if not, Outlook Express and IE both contain rudimentary methods of threading as well.. and there are dozens of Windows MUAs available to replace the broken, non-compliant, Microsoft versions. Incidentally, I'm thinking of converting the basic README/FAQ/etc. text files in the CVS into an easily parsable format (XML, likely), so that I can pull them directly from the cvs to the website for viewing (i.e. index.pl/faq) and maintain one copy. I'm still chewing on some other interesting modulations for the website, much more interactivity will be coming soon. /d
Re: debug.c
> Really sorry about that. I was just being lazy so I could avoid typing > the To: address. I did, however, change the subject line. I am not sure > what underlaying mechanism is used to signal a new thread. I thought it > was the suject line. Typically, threading is handled by the MUA, and relies on the Reply-To and In-Reply-To headers. Some mail readers (mutt for example) do add additional granularity if the Reply-To/In-Reply-To headers match, but subject differs, but subject is definately not the mechanism used to determine threading. Since you're using Yahoo's webmail client, you may want to build an address book and use that to create addresses, or use some templates. Perhaps keep a "dummy" message to the list in your drafts folder, and always "reply" to that one? /d
Google Groups search for Plucker
Anyone poke around here lately? Some interesting things I see there that have never made the list or the bug tracker: http://groups.google.com/groups?q=%2B%22plucker%22++-%22pheasant%22+-%22byron%22+-%22all+source%22+-%22freshmeat.net%22+-%22duck+plucker%22+-%223D%22&hl=en&btnG=Google+Search I tried to narrow the search down quite a bit as you can see from the search box, to eliminate the things that are non-Plucker (the PalmOS app) related. /d
Re: Support for Japanese characters.
> I think I'm going to make my own web-page which has patches for Plucker > original sources. Why not integrate your useful changes into the main Plucker tree? /d
Re: move to Python 2.x?
> 2.0 is available in debian-unstable, but for some reason isn't in > debian-testing. 2.0 seems to be the default for SuSe. I'm not sure how > to figure out what the default is for RH 7.x. The Redhat 7.2 that I've been running for about a month has 1.5.2 by default, but python 2.0.1 was also installed. We should have a python "grace" period, like Mike suggested. Nothing we have in there really is such a new whizbang feature as to require the 2.x series (yet). We should continue to support a "legacy mode" if 1.x is found, and if 2.x is found, use the newer features, even if it's for 3-4 months, until we're sure that 2.x of some version is in the stock distros. Personally, I compile those critical system things myself from source, not packages. That includes perl, apache, python, gcc, yadda yadda. /d
Re: Feature-Wunsch
> In English FYI: > > Hans-Peter asked two feature wishes: Thanks for translating this. I'm sure many of us were wondering what exactly he had asked. Babelfish and other translators don't quite do well. My Deutsch is very rusty. /d
While we're gutting the python parser...
I suggest a small block of new options: --font=std,bold,large,largebold,narrow If we're moving towards storing the actual font value in the database itself, why not put a pointer there to specify it upon creation? /d
Re: Should 'separate' page images always be large?
> I'm looking at the support for alternate sizes of images, and wondering > if something should change. As you know from a previous message, there's a definate mathematical scaling factor issue with the current image parsers in the Python code. I haven't been able to locate where this happens (or if it's just ImageMagic, in which case, we may be able to work around the flaw with Python). A quick example is the following pseudocode: plucker-build [...] --alt-maxwidth=320 (a double-normal image, the original was a 160x160 gif) The parser reports: ppmquant: 16 colors found in colormap ppmquant: mapping image to new colors... pnmtopalm: input is grayscale 160x160, maxval 255 -- using 4 bits-per-pixel plucker-build [...] --alt-maxheight=320 (the same image, original 160x160 gif) ppmquant: 16 colors found in colormap ppmquant: mapping image to new colors... pnmtopalm: input is grayscale 150x150, maxval 255 -- using 4 bits-per-pixel Somehow I just lost 10 pixels on the height and the width. Why? In both cases, you'll notice that I did not get an image that was double the size (when viewed on the Palm, though I can't find a way to find out exactly what the size of the converted image actually was. The parser only reports the input size, but not the output size, currently). If I take that 160x160.gif image and make it 160x320 (height being the larger value), and run the following: plucker-build [...] --alt-maxwidth=1000 I get: ppmquant: 16 colors found in colormap ppmquant: mapping image to new colors... pnmtopalm: input is grayscale 125x250, maxval 255 -- using 4 bits-per-pixel If I change it so that I ask for an image with exactly double the height (640), it then tells me the input file is much different than actual: ppmquant: 16 colors found in colormap ppmquant: mapping image to new colors... pnmtopalm: input is grayscale 150x300, maxval 255 -- using 4 bits-per-pixel The reason I want exactly double the image size, is because fonts that are not aliased in the original 160x160 image (a graph) look sharp and crisp at 320x320. Anything less or more than an multiple of those (2x 3x, etc.) results in blurry, completely unreadable fonts. I have yet to find a way to get exactly double the image size in any image that I pull into the parser. The fact that I get a different size of "input" images (using the same unmodified source image) depending on whether I use --alt-maxwidth vs. the same multiplier with --alt-maxheight, baffles me. This should not change the output. > To refresh everyone's memory: There are two kinds of image pages > included in a Plucked document, inline and separate. There's a slight issue here as well, in that if you use --alt-maxfoo on any plucked document, *EVERY SINGLE IMAGE* then applies to those values. What this means is that even a document where an image leads to a separate page (), will then take even the smallest "bullet" image and make it a nice huge image when tapped. Even simple images on a page that are linked to nothing are applied this state. Not quite sure that's the desired action. I do, however, recommend an option we should support inside the href element: > I'm proposing to change this so that if 'alt_max*' is specified, and a > separate image page is produced as the result of a link, it will be > scaled to the 'alt_max*' sizes instead of the regular 'max*' sizes, and > no additional separate page will be created. If 'alt_max*' is not > specified, the regular 'max*' restrictions will be honored. This sounds like a good idea, but can we fix the scaling first =) It may complicate things to inherit that scaling "fracture" in the new design for the parser. /d
Re: Plucker 1.1 Release Article
> I tried to download the following item, but always fails. Windows > ME/NT/98/95 version (offsite link) Try this (and upgrade your browser to something more compliant): Permanent URLs to the latest Version (1.1.13) of the Plucker Windows installer - For the Webpage: http://www.dirk-heiser.de/plucker - Direct Download: http://www.dirk-heiser.de/plucker/plucker.exe [2.08MB]
Re: Plucker conduit
> How does it handle "promises" which are identical from different > sites? Pavuk handles this very well, you may want to look at the source. ..and also, how does it handle: http://slashdot.org";> vs. The content retrieved will be the same, athough the URI may vary. /d
Re: Plucker conduit
> Adding multi-process fetching of Web pages would indeed speed things up, > but it would also increase the likelihood of bugs significantly. We talked about this for awhile on the perl parser with the LWP::Parallel (which is blindingly fast, the same fetch under python takes 4 minutes longer, roughly 700%, than the perl fetch with parallelism). The issue isn't bugs, per se, it's tracking which pages have already been fetched, so you don't re-fetch them. You have to build several arrays which construct silos to fetch and parse from. You should have a "seen" "dead" or "down", "fetch", "duplicate" and "completed" arrays (or however Python keeps it straight). Every url stripped off of a page is plopped into "seen", a second (parallel) pass then hits those with HEAD to make sure they're actually alive and responding (good ones go into "fetch", bad ones go into "down" [using a callback of course to terminate after 'n' seconds] and are then either written to disk and read back in at the next parse (no sense in testing a dead or down url every time you parse) and then a separate pass (parallel?) can be used to skim off urls found in "dead" to see when they come back up), then the "fetch" silo will be used for the REAL gather process, because those urls are guaranteed to be valid, up, and responding. You get the idea. If you fetch Slashdot to a depth of 2, and also fetch Freshmeat, Newforge, and Linux.com, you'll get a lot of overlapping duplicate links because they all link to each other. Adding the environment loosely described above eliminates this. > So you'd go through and issue GET requests for each page, and get back a > 'promise' for each. Then you'd loop through the list of promises, > reaping each as data arrives. How does it handle "promises" which are identical from different sites? Pavuk handles this very well, you may want to look at the source. /d
Re: newsrc files
> I've written it mostly in python because all systems using plucker > should have python installed, right? I don't think anyone has won that thumb-wrestling match yet =)
Re: Regd. plkr source
> Thanks for your mail. I dont know unix so i am worried in going thru the > document. Not knowing unix is going to make the learning curve steeper, but not impossible. This issue has been tackled before, try reading up on the thread starting here: http://plkr.org/list/2Q2001/0052.html and here: http://plkr.org/list/2Q2000/0690.html Or use the search engine on the Plucker homepage to search the various other people who have played around with CodeWarrior and this project. > Is there any other easy way to do this task. Or can u send me the latest > .rcp file after converting it from .rcp.in file. I have plucker 1.1.12 > source with me. Easy? I doubt it. CodeWarrior is a confusing environment, riddled with bugs, but for some people, it works well. Porting Plucker over to using it is going to be more than just renaming a .rcp.in file to .rcp. Good luck. /d
Re: Regd. plkr source
> I have renamed your .rcp.in file to .rcp and compiling. Not exactly the right way of doing things. The .in is a file which is used to create viewer.rcp as an output file. If you search the file for every location with an '@' sign, you'll see lots of replacable parameters that get filled in when the ./configure or ./autogen.sh commands are run. We didn't just "spell it wrong", there is a reason for that file having the name it does. It's all part of the build process. Use it. > I have added pilRC plugin to the project to convert .rcp file to .r > file. Am i doing any mistake here. It is showing 2 errors 1)MENU ITEMS > short cuts 2)Duplicate resource tFBM of ID 3002. Why it is showing like > this. Yes, see above. > lang.h is missing from your source, as of now i have commented that > line. If i compile this program successfully i will be grateful to u. > Any help regd. this is appreciated. It's not missing at all, it's just not created yet. You didn't use the proper automake/autoconf procedure to build it, so those files aren't going to be there. The build process is quite dynamic, and several steps have to be taken first, before you get those files. /d
Re: newsrc files
> I'm writing a tool to pluck news. Do any non-Unix newsreaders hold > articles in a spool-like structure and use newsrc files? There was a perl script out there awhile back that someone specifically wrote to put news directly into Sitescooper and Plucker. I know I posted it here in the list, but I can't recall when or what the tool's name was. DOH! 10lbs of potatoes in a 5lb bucket (my brain). It was on a .nl site IIRC. I'll search around and see if I can find something. /d
Russian Translation of Plucker (complete?)
Since this looks like complete garble on my Palm (wrong codepage,
I'm sure), but it is functional and all operations do as they appear, I'm
going to say that this is complete. ru.{h|rcp} are in the cvs, and the
ChangeLog and configure.in have been udpated. I will work with Oleg to make
sure this works before we put it into 1.2's release.
/d
SDK4.0 changes
Another quick one: TxtCharWidth was replaced by FntWCharWidth in the 4.0 SDK. /d
Russian viewer (almost complete)
I was working on the Russian Plucker viewer tonight (Oleg's translation) and ran into some weird anomalies with the compilation, related to (once again) the TRG headers. After fumbling around with my environment (I use both prc 0.6 and prc 2.0 for building various things), noticed distclean in viewer/Makefile (line 85) removes the AutoCF.h, ffslib.h, and Trg.h headers upon build. Should we really be removing these with distclean? The other thing I found was that the version is improperly reported by autogen.sh now (this used to work): "Plucker @@VERSION@@ was configured as follows:" I'll keep posting some more gotchas as I find them.. I'll see if I can't roll up a quick patch to hit some of these issues. /d
RE: (Plucker) User's Guide Slashdot Link
> To be fair, the Dread Competitor allows you to both add a 'Custom > Channel' (with any URL) and schedule update times for channels. Except that custom channel is then stored on their server. /d
Re: (Plucker) User's Guide Slashdot Link
> I am sure that you have already been informed of this, but the "first" > link to Slashdot is out of date. The new link should be: Fixed in CVS. > One of my biggest gripes with AvantGo is that pages must be under a > certain size (i.e. "Thinking in Java" chapters are too big) and they > must be online. Limited by their contractual obligations. Yet another reason Plucker exceeds where AvantGo fails. Here are 13 others: http://www.plkr.org/list/plucker-list/0565.html > P.S.: Why choose DOC over zLIB? Speed? Yes, it is faster to render, yet larger in memory. > Why default to DOC? Is DOC decompression supported by the OS or is it > just included with the viewer? Can I get rid of it? Yes. > Do I ask too many questions? Yes. Just kidding, some of them are valid. We should be moving to zlib as a standard though, I think, but defaulting to DOC means that it's handled natively by the viewer, and using zlib requires the users to install SysZLib.prc onto their palm as well. There are tradeoffs in size over speed. /d
Re: SOT: Referer header and the http GET (was: Re: Plucker 1.1Release Article)
> David> The header passed to the server is, in fact, HTTP_REFRERER. > No that the name of the environment variable passed to the CGI I should have said: "..in fact, in HTTP_REFERER" ...bad verbage on my part. I'm aware of the difference between Referer and HTTP_REFERER. Remember, it was I who discovered this was how to get around "Their(tm)" protection of website content using forged Referers. > So the Web browser need send the "Referer" header to the web server > (this are the task between the browser and the server) and not the > "HTTP_REFRERER" (this are the task between server and CGI). Exactly. /d
Re: Plucker 1.1 Release Article
> David> That would be IE and AOL and WebTV browsers. None of them send > David> HTTP_REFERER headers in the requests. I managed to get around it by > > not HTTP_REFERER. The header passed to the server is, in fact, HTTP_REFRERER. Here's an example of the entire environment requested in a GET from a local page on http://localhost/index.pl on my laptop. In perl, you can get that by doing the following: use Env; use CGI qw(:all); print "\n"; /d
Re: Plucker 1.1 Release Article
> But as i say these are basic problems in "How to surf the Web" and not > related to plucker. So its really off topic here. I'll mirror the releases on plkr.org as they appear. > And if the user disable cookies you get the same problem :-) Yep. However, I don't want people with non-standard browsers poking around my filesystem and playing with URLs to get into protected areas. I don't use banners or ads, so my design choice has a different justification. /d
Re: Plucker 1.1 Release Article
> I guess that's because you do not send the referer header with your > request. That would be IE and AOL and WebTV browsers. None of them send HTTP_REFERER headers in the requests. I managed to get around it by requesting a cookie from the browser, after sending one. /d
Re: Plucker 1.1 Release Article
> I try to download 1.1.13 installer for win without success. it always > refreshing the same page. I wrote two letters to the dirk. No answer. You're the second person this week to email about that. The first one was a private email to me. What browser are you using? What platform? I went to the link and it brings me to the right file, and I'm able to download it without any trouble. /d
Re: CopyToMemo feature
HOLY Patch Batman! Can you gzip it and attach it as a compressed file next time? Please =) > When "Copy To Memo" in the menu is selected, it forces to redraw screen > and copies the text in the viewport to the memo. URLs of the links in > the viewport is also added after the tag strings. Isn't there a way to do this without redrawing? We already know what percentage down in the document we are, so can't we just take the equivalent of that value (x,y) and stuff the contents into the clipboard? The other thing is, we should definately put some error trapping in here so we don't inadvertantly wipe out anything that may already be in the clipboard (a Plucker clipboard?). Example: I clip an address or phone number, or "cut" it to the clipboard (removing the original entry in the process) and then go to Plucker, and (accidentally or legitimately) tap on the option or shortcut to Copy to the clipboard, it will wipe out my unsaved record already present in the clipboard. Remember also, the Palm clipboard is persistant over powering off. If I cut an address in a meeting, power off, an hour later launch Plucker and see some content I want to Copy, it should not stomp on my clipboard contents. Mike will say this is a user problem, and I agree, but we should still try to take some precautions to protect the user from themselves. /d
Re: character sets in HTML files?
> One of the advantages of using Python 2 for parsing is that it can work > with a complete 32-bit Unicode charset encoding (UTF-8), rather than > just a locale-specific subset, and includes support for transforming > many (most) subsets into UTF-8. My understanding is that you need the catalogs and NLS support built into Python to take advantage of that, and that means ensuring that the package maintainer (or if you do source builds on your own) did not use the --disable-nls switch when compiling. Many do (and there's good reason to). /d
Re: character sets in HTML files?
> Should plucker just parse XML and feed non-xml stuff to tidy to > reformat? Just an idea to simplify things. I think it simplifies > things, at least. Remember, implementing an XML parser is no trivial matter. If the XML page or application fails validation, the page is bitbucketed. In the current scheme, Plucker tries to make sense of what's left of the broken HTML, but with XML, that's not allowed. > Indeed, but I thought XML was in unicode? Or did I dream that? Probably > did, as I'm sure I've seen encoding="iso-8859-1" in some files, > actually. It is indeed unicode, however, you can override it. /d
Re: Anthrax Plucker Doc - thank you and a quick question
> I was wondering if you've considered adding a few Q&A's on the > Biological warfare aspects of Anthrax? > Again, well done, Look forward to similar releases in the future, I'm accepting any and all suggestions (and have gotten quite a few so far. For example, "Anthrax" is a different disease in France, so I have to correct that in 1.3). Do you have any particular questions you'd like me to cover? /d
Re: plucker and WAP?
> I notice that http://www.ananova.com/ advertises pages for WAP phones. > Anyone here making use of those pages via plucker for their Palm? If they are truly using WAP and WML in their implementation, I doubt that the current Python parser that Holger and Bill Janssen have written will correctly grok that content. If, however, it's just "light HTML", then the chances are pretty high. This one will work though: http://www.ananova.com/pocket/ (and yes, it was from PODS) /d
Re: making plucker-build easier to use?
> Well, to be more accurate, under the windows MacOS has an environment > very much familar to a Unix / Linux developer. Whether that OS is > technically a Unix or Unix-like is probably not the issue for this > mailing list. Right, religious OS turf wars aside, here's the count of operating systems that have either successfully, moderately-successfully, or with some hacks, been able to use Plucker currently: * Linux (all flavors) - repeated problems with Redhat 7.x Python builds * OS/2 Warp 3 and 4 and Aurora * Solaris * HP-UX * AIX * Windows98/98SE/NT4/2000/ME (no testers for XP yet) * BSD (both FreeBSD and NetBSD) * Macintosh (prior to MacOSX) - any OSX testers want to help? I know of only one so far That's a pretty good segment of operating systems and environments. I think we're doing good so far. One of the questions I have though, is what does Palm plan to do to support their Palm Desktop offering and Hotsync conduit stuff for the ongoing migration of Macintosh/Apple users to the new OSX environment? The rules change slightly.. Then again, Palm did spin off their OS business, perhaps they're going to focus on more hardware and tools than the OS itself.. /d
Re: Where to get Robert O'Connor's betas?
> > I couldn't find Robert O'Connor's email address on his website > > neither on this list... > > Maybe you should look in the AUTHORS file included in the Plucker > package ;-) Or on the Plucker website, on the download page... Or on his original message about the german viewer... Or on the files in the cvs, both directly and on the web... We're all pretty easy to find =) /d
Re: making plucker-build easier to use?
> I'm going to pursue this a bit further on Windows, first. There must be > some way to write binary output. We should also have Alexander test this on OS/2, though in my recollection when I ran it, it was more Unix-like than Windows-like. Macintosh will still be a problem, since there is no shell. /d
Re: Attempt to report broken link fails...
> When i attempted to follow the advice of the plkr.org web site in > reporting a broken link, I got this error msg... Fixed. anti-SirCam measures. /d
Re: Plucker conduit
> Although I've done this for my company, the code should be GPL still > since it's just a port (spelling mistakes and all :-) You mentioned there's some BSD code in there. Is that code under the BSD license? Or is it *nix code, developed under BSD, using the GPL license? /d
Re: mailto links update
> If you change "Subject" to "subject" it will work just fine, i.e. it is > a minor "case sensitive" problem in the parser. Take a look at the > 'parse' function in PluckerDocs.py if you want to fix it. Hrm, thanks Mike, I'll try it. /d
Re: mailto links update
> I'm mixed about this. Those modifiers are distinctly non-standard and > I'm unsure that they're used in any consistent manner by the world at > large. By non-standard you must mean "not used often". They're used quite a bit in academia, and are also in rfc2368. http://www.ics.uci.edu/pub/ietf/uri/rfc2368.txt I also found a few other random references: http://www.sightspecific.com/~mosh/WWW_FAQ/mailsubj.html http://lists.w3.org/Archives/Public/www-talk/1998JulAug/0002.html > On an semi-related point, is anyone else seeing plucker occasionally > goof when presented with a long URL to copy to memo? How long? /d
mailto links update
This may be in Chris' court, and if he doesn't have time to look at it, I'll poke around in emailform.c, but.. subject and mailto modifiers are ignored when they're parsed into Plucker. Example: mailto:[EMAIL PROTECTED]?Subject=help";>Help When clicked in Plucker, brings up the proper "To:" address, pulled from the record, but does not populate either the Cc: or Subject: of the form displayed. A feature add? With the addition of this 'feature', we could start populating some "online" bug reports for sites, and content providers could "load" mailto links with useful information. /d
Once again, we revisit the missing --stay{$FOO} options
I've been banging on the python code again trying to make sense of the currently broken (in my opinion, no flamewar please) --stayonhost and --staybelow options. I assert that --stayonhost should stay on a particular FQDN, which it currently does, but leaves a gaping hole in the implementation. A perfect example is http://slashdot.org/palm. The image that shows up on that main page comes from images.slashdot.org. If I want to gather this site to a depth which allows me to get comments (--maxdepth=3 or greater), it also traverses offsite and to other linked sites. That's bad, since it gets very large, and we're out of the "prepared for Palm" format of the site. Right now, I either get the site without comments (useless) or I include comments, and manually have to add all of the offsite links to exclusionlist.txt. Very ugly. --staybelow=slashdot.org also does not work as you would expect, as it requires an actual URL there. We're getting into the misconception of the "up" and "down" of URLs (as you all know, there is no such thing as "up" or "down" in any web content. Everything is exactly one hop from everything else). This IMHO, should be adjusted to take a domain as an argument, not a URL (complete with protocol). Using --staybelow=slashdot.org and --staybelow="http://slashdot.org"; has very different results. In any case, there's a missing option here (and always has been missing); --stayondomain. With --stayondomain=slashdot.org, for example, images.slashdot.org, www.slashdot.org, banjo.slashdot.org, and slashdot.org can be maintained, and you can "package up" the content so that it never leaves this domain. I could spider it to a maxdepth of 100, and be assured that it would never get out of hand and go offsite (yes, the file would be large, but it would be very self-contained). Here's what I propose: --stayonhost: Will not ever leave the FQDN you specify in your -H syntax. --staybelow: (should take a URI as an argument, not a URL) Will restrict ascention to the supplied URL as a parent. --stayondomain: Will never leave the network you specify, so that www.foo.com, images.foo.com, and foo.com will all be assumed to be included in the same "pluck". Content from all "member domains" will be included. Sound feasable? /d
RE: Plucker Desktop GUI Manager
> I disagree with you on this. This is exactly what Plucker is doing, and > what is supposed to be doing. I tell it the source of the information, > then it is auditing out the javascript, the HTML comments, and the other > crap I don't want, and formatting it to suit my medium which is a Palm > pilot with a small screen and low processing power. I think you misunderstood my explanation, so let me try to come up with a better analogy. When you tune in with a television, you have a guide, a schedule that tells you which shows are playing, and when. These shows have been "prepared" for you, sprinkled with commercials, to fit within a given timeslot. You sit and watch from the beginning of that timeslot to the end, and you get the "programming schedule" for the night. At no point were you involved in the creation of this programming schedule. It was prepared for you beforehand. It came from one central source, and was guaranteed to fit within the timeslot allowed, and was guaranteed to be broadcasted and available from that channel. AvantGo, by example, uses this same mentality. They aggregate their collections of content on a schedule, store them on the AvantGo servers. This content is prepared *FOR* AvantGo by the content providers, formatted for the Palm, tweaked, optimized, and customized for that screen and viewing area. When you connect via AvantGo, either wireless or serially, you are being delivered the "programming schedule" for your Palm, based on what you decided to "tune into". Plucker, in contrast, is 100% user driven *CONTENT*, vs. user-driven *PROGRAMMING*. The difference here is that the content may break, be too large for them, of the wrong type, badly written, may not be available, or access to it may be denied. It is certainly not "prepared" for them beforehand (think content, not parser, that's a different beast). The difference with a channel vs. a Plucker "stream|pluck|document" is that a channel is a known entity, controlled, customized beforehand, knowing the end-user device, credentials, and so on. Within the Plucker design, it is completely 100% unknown, can change, may break, I'm not 100% opposed to any name: channel, pluck, document, zorgoblif or whatever. I'm simply saying that we may end up confusing the users into believing that there's some central repository where this content can be obtained from, some sort of "content aggregation service", which implies a channel, which implies some sort of subscription. Sure they can download an XML file which described their content. I can easily see the "I subscribed with the XML file, but the content doesn't appear when I hit sync" type of problems. > Document isn't bad, connotation is somewhat static though. We've misused it before, and I'm sort of opposed to propagating the term, since there are something like 23+ doc readers now. At some point, it may come that Plucker, via a plugin, supports reading DOC files directly, which then means Plucker content (created *BY* plucker) and Documents (*NOT* create by Plucker) may be available within the Plucker UI. If we globally use the term "Document" now, that's going to get much more confusing later on if that feature is added. > I think "Pluck" isn't bad either, and the word can be extended into > whatever meaning desired. The drawback is that mmediately people don't > know what one is talking about. It reminds me of "Flooz". One has to > always define and remind what one is talking about. One also makes the > user feel stupid by giving them terms they don't understand. An easy-to-explain analogy: pluck == fetch How do you describe a channel in one word? > A question on this: I believe the license is just a restricted > directory, not a restriction of exclusive providership to handhelds, is > this the case? Does the content provider just have to have a separate > contentprovider.com/avantgo directory and a contentprovider.com/handheld > directory to get their info on all the handhelds? It's a restriction on the number of users allowed to use that prepared content on the provider's site. Take a peek at this message: http://plkr.org/list/plucker-list/0567.html Specifically: "[anon] for instance, www.pocketpcthoughts.com are now shutting down their official AvantGo channel because they're being charged with $6000 USD since they have more than 5000 registered AvantGo subscribers" The content providers are, by contract, told to restrict the access to the content with any measures that will only allow the content to be retrieved by AvantGo's browser, and also restrict access to any means of getting access to that same content. As an example, in my local database here, I have: mysql> select rowid, url from pods where url like '%avant
RE: Plucker Desktop GUI Manager
> Since that was an item included in the draft (under channel manager and > the main>advanced options), I should ask why you recommend not packaging > software with any sample Plucker-compatible content, for new users who > don't know any URLs, or don't feel like always keeping up to date with > them as they change? I'm sort of against packaging actual *CONTENT* with the distribution, and it's also the reason I haven't dumped the urls I've scraped up over the past few months on the masses. Redistribution of modified content is verboten in 99% of all cases, unless explicitly granted by the content provider (and we should be able to prove that approval). Transcoding and redistribution is a sticky mess, and I'd like to make sure the Plucker team doesn't get dragged down in it. > I find it frustrating to have to hunt down that the site I want is under > either some deep directory as a /low /lite /pda /handheld /palm /mobile > /whatever in order to get it into Plucker. I've done this for hundreds of urls already, but in some cases, a simple email to the content provider, explaining your need, can work wonders. > A button on the manager pulls the updated XML file down from plkr.org > whenever the user feels like updating it to get the new/changed URLs. Who keeps the XML file updated? /d
Re: Thax and 2 wishes...
> Right now I'm reading the plucker converted manual for a palm basic > interpreter and would like to copy and paste one of the examples. Don't > work !!!() Is there any chance to see that feature in the plucker > viewer?? This may help: http://www.plkr.org/list/plucker-list/0352.html /d
1.1.13 fall down, go boom
Can someone verify this for me: plucker-build -H "http://home.pacbell.net/bullnose/pyth.gif"; \ -f /tmp/Pythagorean --zlib-compression --no-urlinfo --bpp=8 \ --launchable --maxwidth=740 --maxheight=820 When I specify both height and width, the parser scales it to 370x410, and when launched, I get: Plucker (1.1.13) called SysFatalAlert with the message: "DataMgr.c, Line:7025, DmWrite: DmWriteCheck failed". Removing maxwidth, but leaving maxheight, the parser scales the image down to 150x166, and it seems to work ok. Ideally, I'd like an image 2x the size of the original. I scaled it locally with gimp, and pointed it to file:/tmp/pyth-LG.gif, which was detected as a 740x820 image. It crashes with the same exact error. What's also weird is that if I leave the --maxheight=820 value, and omit the --maxwidth=740 value, the file is scaled to 150x166. If I leave the --maxwidth=740 and omit the --maxheight=820, I get an image which is scaled to 226x250. If the image is being detected as the same, but one is defined by height (and no statically defined width), and the other is defined by width (no statically defined height), shouldn't they scale to be the same physical size? Using both height and width together crashes the Palm, hard. The reason I bring this up is because I was working with a guy in irc in another window who is trying to convert a bunch of his blueprints, diagrams, etc. over from FireViewer (non-free) to something better. He tried the Acrobat reader, but it munches the fonts to death. I suggested Plucker, and we went down this path. We managed to get one working, but it's not 2x, and it's not quite 1x, but the fonts went from sharp to blurry in the conversion process. /d
RE: Plucker Desktop GUI Manager
> I am certainly open to what to call the things other than channels; it > is easier to specify the term now than more time spent rewriting things. I understand that. It would be harder to have to "undo" anything we implement now, using a term that doesn't fully explain our architecture and design that we've built around Plucker and supporting scripts. > All the other GUI subscribed portable Palm web systems I have ever seen > use the term channel, probably for easier first user/migration, and the > fact that new people are familiar with the concept of a 'channel' in the > real world, as being some content that is delivered onto your device by > specifing where to dial into the originating source. The only one I know of that does this is AvantGo. Sitescooper uses "scoops" or "nightly scoops", for example. I'm not aware of any others that actually ACTIVELY pull content from sites and parse it for a digestable view on the Palm. In fact, I was looking to see if they even had a trademark on this, and it seems, though not specific, they have some interesting open trademarks: http://www.nameprotect.com/cgi-bin/FREESearch/search.cgi?action=SEARCH&db=PTO&ss=avantgo&imageField.x=10&imageField.y=11 What's interesting is their "Avantgo Dot Com" one: 071107 - Dwellings, buildings, monuments, stadiums, fountains, structural works, and building materials 260302 - Geometric figures and solids 260312 - Geometric figures and solids 260321 - Geometric figures and solids However, I have a bit more of a fundamental issue with calling these things a "channel", and I've talked to a couple of people about this, so let me try to explain/aggregate my/our thoughts. A "channel" in the current paradigm (i.e. AvantGo) is generally thought of as a "push" service, where content is aggregated at the server, pre-prepared, converted, and then pushed down to the end-user's Palm device. They "subscribe" to "channels", and then sync, and the content magically appears on their palm. To the users, they think they're "pulling" content, but in fact, they're being pushed prepared content from one server to their Palm device. No problems, makes some sense. With Plucker, the user initiates a script/gui/whatever, which goes out on the LIVE internet, gathers content from live sites, which may be up, down, or slow, and then parses it locally, which creates a file they have to sync to their Palm. At no point were *WE* (Plucker) involved in the access, structure, or availability of this content. In fact, we only provide a client-side parser, the rest is completely out of our control. It comes from anywhere from 1 to thousands of live sites. To contrast the two, we're hitting live content provider sites, which present unprepared content to Plucker (i.e. not necessarily formatted to fit on the Palm screen), where the users using AvantGo have to sign in, and are pushed content that was batched on a schedule, and parsed for them. Calling what we do a "channel" is misleading in this aspect. Channels in the vein of telecommunications or television, indicates that you have something "canned" for your viewing pleasure (or displeasure of late), which has been audited/edited to suit your medium. This doesn't apply in our case, for a majority (most) of our users who use Plucker. > Nevertheless, terms can be created that aren't always the most > intuitive. But for myself, any other term that I could think of, pretty > well would have in the help file this phrase: "Well, it's like a > channel" to describe what the term actually meant. I get this every day too, when people see me using Plucker, running Plucker, from usenet posts, and people emailing me from links on the website or whatever. Bill Janssen is starting to get some emails now as well. Describing what Plucker *IS* is not hard, but describing what it *DOES* and *HOW* it does that, can be difficult for some general users to grasp. If I introduce the channel aspect, to speed the explanation process, they understand the "what", but they have a completely incorrect idea as to the "how". > Some of the possibilities: > --Document: (ties with viewer's Document manager well). Documements can be stored in Plucker, so this fits well. Is a document a "channel"? Likely not. Can you "pluck" a document? Yep. Here we introduce the nature of "plucks" (suggested by another user, just today online. I think I've tossed the term around before also) > --Database: the former name of Library manager. Content is stored in a database or databases. This also fits, but one does not make a channel of "databases", however one can have "database of channels". One can also pluck content into a database. > --Ebook, Site, File, Plucker File, Plucker Document, Plucker Channel, > Plucker Database. Still don't know of a suggestion that would co
Re: Plucker Desktop GUI Manager
> I may not understand what "That Other Product" is. Just a few examples: http://avantgo.com/channels/index.html http://avantgo.com/support/mobile_support/palm/device/index.html http://avantgo.com/support/mobile_support/palm/desktop/dksubscribe.html I'm only suggesting that we strive to be unique, since our entire architecture is fundamentally different. We're going to get into a mess of support questions with "Why are their channels incompatible with your channels?" "How come Plucker can't sync with AvantGo channels?" "How do I add a channel from your website?" /d
RE: Plucker Desktop GUI Manager
> Now that's something to think about--adding some sort of command line > gateway to the manager also, to let people still update their channels > without a windows manager. Certainly have to consider how to build a GUI > manager so that it is still accessible with a command line. Smells like a daemon with a GUI front end that can issue callbacks into the daemon, ala gnome-pilot and friends. Eventually, I've envisoned that pilot-link will do the same, with GUI apps like JPilot and friends issuing callbacks... (well, not JPilot, because he may be dropping pilot-link support soon in favor of ColdSync, but you get my idea) The code should be in a small daemon, the gui should be secondary, and have hooks to talk to the daemon... ideally. /d
RE: Plucker Desktop GUI Manager
> The overall polish of the programs way also quite good. Mahogany really > has a lot of options for email, but somehow still couldn't prevent me > from paying internet time by the minute to download spam and Sircam > attachment files. :) I personally don't like gui applications for mail, since it restricts my availability to mail, a very critical part of my daily job function. Regarding your SirCam issue, you can 'defang' those with fetchmail, procmail, or some intelligent sendmail filters. There's hundreds of sources on the net for those. > Good questions. I don't know the answers on these. Will spend some time > investigating, as dropping from these would be very useful. Isn't there prior work here from another contributor that was working on a Macintosh "toolbar" plugin or some such for Netscape? Perhaps he has some ideas (can't recall his name at this point) > Yourself and the rest of Plucker Team always has good ideas for building > software well--can there be a cvs branch for a Plucker Desktop on the > Plucker cvs at gnu-designs.org for the interested Plucker coders and > fans to collaborate on? Sure, give it a name, and I'll create a repository for it on sourcefubar.net (physically the same box as plkr.org) and we can reference it there. I'm not sure that this should be inside the Plucker cvs directly, even though it's a supporting tool, it's not part of the same codebase. /d
RE: Plucker Desktop GUI Manager
> Was considering the concept of making a channel 'Due'. When the program > is running, or is started, it looks at the time when the Channel was [...] Stop stop stop... let's not call this a "Channel" please. We do not want to continue the incorrect association (legally, morally, ethically) with "That Other Product", do we? Can we think of another name for this? /d
Re: making plucker-build easier to use?
> > DO NOT EVER SEND BINARY DATA TO STDOUT, BAD! > It's easy to check whether stdout is a terminal, lookup isatty(3C). I > agree that sending binary data to the terminal is a bad thing. (GNU zip > does this check when you feed it data from stdin.) I wasn't implying that precautions couldn't be taken, simply that before implementing a solution involving the possibility of sending binary data to STDOUT, that these precautions should be implemented. /d
Re: Hi and short bug report
> Is that problem know and is there any support for Handera planned. I > voluntere to test on that device :) This may help: http://plkr.org/list/4Q2001/0076.html /d
Re: Plucker Desktop GUI Manager
> I believe it uses GTK+ widgets on Unix, doesn't it? There is a wxWindows for GTK+ http://www.wxwindows.org/faqgtk.htm > Does it have to be running to do timed plucks correctly, or will it > notice when it's next started that it's time to do some plucking? ..and more importantly, what if there's no GUI running when it's time to spawn a "pluck"? /d
Re: Plucker Desktop GUI Manager
> It is written in C++ in wxWindows (which I believe is included in > Debian). The resources are an XML file (which I was waiting for, and are > now available in wxWindows 2.3.1). Since a GUI app, most of the work is > widgets and a separation of code from an easily hand-editable resource > file is helpful. I've used wxWindows very briefly with some Wine work I was doing for a customer about a year ago. I'm familiar with Glade, somewhat, and the more recent builds allow interface changes via the XML UI definition file. I rather like that approach. I'm curious. I started a very (very) loose skeleton for this about two years ago in GTK+ (which at the time, was.. uhm, "rough" with toolkits constantly in flux). I've posted the link here to the png with it. It's orphaned off now with the new GNOME 2.0 stuff. Incidentally, Debian has 1.67c7 of wxWindows, via libraries. I could not find the proper headers, but a source build would definately include them. > Examples of existing programs are Audacity (wav editor), and Mahogany > (high-powered email client). wxWindows (www.wxwindows.org) is Free. These must be Windows tools, I've never heard of them, and Mandrake, Redhat, Debian haven't either... or at least they don't include them (Debian had audacity in unstable though, searches around..., AH!) http://mahogany.sourceforge.net/ http://audacity.sourceforge.net/ ..both on my favorite open source aggregation site! (grumble) Sourceforce would be nice if every project had files. At last parse, 40% of their projects are empty, no files, and quite old. People put these "mental projects" there as placeholders, hoping that the code will just magically write itself. Ugh. Ok, back on track.. > I finished the initial draft this week. The 10 screenshots are here: > http://www.rob.md/projects/plucker/2001_10_13/ (windows widgets shown). I like it, though I rather like the "tabs on the side" approach vs. the "tabs along the top" design. A bit more intuitive for the way I personally work. YMMV though. > Things like correctly dealing with linefeeds, preference saving, and (I > believe) drag n' drop are supported across the 3 platforms with the same > C++ code. Does this include dnd from Netscape and standard GNOME/KDE widgets also? This leverages the "drag a webpage onto the canvas" approach I was shooting for with PDBuddy. Code anywhere to bang on yet? > The Plucker Desktop is designed to have the full power of Plucker, but > also organized into basic and advanced sections, so that a first-time > user can still pick it up and use it. Just be aware, wxWindows is LGPL, not GPL. http://www.wxwindows.org/newlicen.htm Your work on Plucker to date has been nothing short of exemplary. Keep up the good work! /d
Re: making plucker-build easier to use?
> Both interesting ideas, though perhaps not for Plucker. Incidentally, I > deleted his note, or I'd forward it along. I keep "most" of my relevant mail, so if necessary, I'll forward it along.. if Norm does not object. > However, it got me thinking about the plucker-build command line. > (Those of you who don't use command lines can stop reading right now :-). You mean... there's *ANOTHER* way to do it? >=) > plucker-build foo.txt >foo.pdb I'm all for simplifying the obvious, however.. doesn't that complicate your parsing routines? If I omit the redirection (which obviously is not parsed as a command passed to the parser) your suggestion would spit the (binary) data to STDOUT. *DEFINATELY* not something you want to enable by default. If, however, we could somehow let the parser know that we were going to use STDOUT and redirect it, and *IF* that redirection was omitted, it should spit out "human readable" text or HTML to STDOUT. Let me reiterate: DO NOT EVER SEND BINARY DATA TO STDOUT, BAD! > I like it! I'd like to check this in, but thought I'd see what people > thought before doing so. Another point though, please do not remove the -f argument to the parser. Your redirection should be in addition to it (and a feature I personally cannot find a use for), but it should not replace it. Also, what do you do when there is a combination of both? Right now I can see that using -f /tmp/csets and > csets.pdb will put the textual "dump" of errors and reporting into csets.pdb (as ascii text) and the real Plucker database would be in /tmp/csets.pdb, and if these were both pointed to the same location, the latter would trash the creation of the former. Fault tolerance is always a good design criteria. > If we write the document to stdout AND no doc_name has been > specified or found, the home URL is used as the doc name. Writing the document to STDOUT is bad. For one it sends potentially bad binary control sequences to the terminal, to the session handling the terminal, and to the system interpreting them. Over remote, ssh, screened, dumb-terminals, thin-clients, this would cause *VERY* unstable and undesired results. /d
Re: PalmOS 3.5
> I got a lot of static from the list a while back 'cause I didn't upgrade > my IIIx to 3.5. I'm thinking about doing that now, but, since I use > Linux exclusively how can I? (My Palm CDrom has never been out of it's > package!) Romeo is your friend. ..Or osflash. Both can be found linked from here: http://romeo.sourceforge.net/ Beware, osflash *DOES NOT* work with OS4, so if you try it on your Palm to flash it to OS4, you will create a nice expensive brick from your Palm. I just unbricked my PalmVx a few days ago with Romeo. Our wonderful "rolling blackouts" in California hit one afternoon when I was flashing the Palm from OS3.1 to OS4, and I turned the Vx into a paperweight. Blecch. /d
Re: Support for HandEra 330? [workaround for refresh problem]
> Thanks to all for your work on Plucker - let me know if there's any > testing I can do on my Handera that would help out. Mike Perry has loaned his Handera to Micke for the purposes of getting the "Scale to Fit" working, and the side benefit is that there may be some very cool VFS features coming out of his testing of it. Once I get some time, I may hack in support for JogDial/ScrollWheel support in the viewer, unless Mike gets to it first =) /d
Microsoft Word to Plucker convertor, 1.0
I have sucessfully hacked together an MSWord to Plucker conversion script last night, and would like to slap it up on the website for people to use. There's a slight glitch, however.. it's going to have to behave differently than the PDF-to-Plucker one does now, in that most people have Word documents that are not located at remote URLs which are publically accessible (for the server to grab it). This means implementing a public "File Upload" facility to send the doc to the server for conversion. Being an advocate of privacy and anonymity, I don't want to have to write any files to disk. I can do this, but it's harder to "prove" that the files weren't written to disk during the conversion (they're held in a scalar in memory, as is the output .pdb file until it's sent to the user). Any ideas on how I should go about this? Or should I just dump the idea and move onto other things? The output from Word to Plucker is fairly impressive. Much more impressive than Abiword's output or even the Microsoft native "Save As HTML" output. Ideas? Comments? /d
RE: Emdash
> Sounds very reasonable. Just want to design in a way that the same > document is portable among different devices platforms like a Linux PDA, > etc. *cough* XML... it was made for exactly that (well, that and to get rid of the complexity of the non-standard, oddly used SGML tagsets) /d
Re: Emdash
> I am very much not up to date on the different character sets and coding > systems available, but I was wondering about some advice for emdash. > AFAIK, there is no official '&symbol;' for emdash in HTML. Some HTML/XML > type documents require it, and it seems to be '&emdash;' as the common > usage. I have seen three usages, depending on browser and parser agent: — — and — > Alot of text makes liberal use of emdash. What is the best way for > Plucker? They do?! I've never seen it used in a normal text yet. > 1) Replace &emdash; with two hypens in parser, so outputs as -- in > Plucker document instead of nothing (Easy. Works, though not as pretty > as a real emdash. What I initially did in a test parser hack a while > back). Would this fail if one was calculated at the end of a line, i.e. splitting the two "characters" so one - was at the end of a line, another - at the beginning of the next? > 2) A special character function code in viewer. Function code for > character followed by a number of the special character to output (in > this case: Palm's emdash character (Harder. More of a pain for > international sets and to calculate line widths too perhaps?) Well, similar to the way you draw your now, we could easily do a 5-pixel wide (just a guess) line, in response to the parser seeing any of the three implementations of emdash in the document parsed. We don't care to represent the HTML "validly" on the Palm, only represent it "visually". /d
