pReader (an eBook Reader for Palm-DOC files)

[neve]

Originally Posted by Jappus

Due to popular demand, background and text colours can now be freely customized in the new version 0.5.1.

Hell yeah yet again! As far as I'm concerned, as it stands right now, if there were not another update, it would blow away any e-reader I've ever used. Granted, I've never used a Kindle or anything.. just Palms.. but this is absolutely workable, as-is.

This should definitely hit Preware and Filecoaster ASAP.

OK, my one wish was so quickly granted, I shouldn't look a gift horse in the mouth, but since you're currently in super-responsive mode, and it sounds like you're busy, so who knows how long that will last, here goes with another..

What about changing the text of the file navigation? Can it wrap to the next line? Or maybe lose the icon to the left? Wrap would be best, I think. Right now I have files that look like:
Dick, Philip K - A Maz...
Dick, Philip K - A Sca...
Dick, Philip K - Count...
Dick, Philip K - Now...
Dick, Philip K - Rauta...
Dick, Philip K - The P...
Dick, Philip K - Time...

As you can see, it's diffifcult to tell which books I'm opening.
Also adding one of those scrolling alphabet index bars like in Music Player Remix might be helpful.

___________________________________________

<<Comments on media center "filepicker" widgit deleted, since it's explained more accruately and succinctly below.>>

[Veritas06]

Thanks so much!! I;ve been waiting for an eBook reader like this, & cannot wait for the html support too. I know this is very early in development, & i read your post & i know you're busy with your thesis, however i just wanted to ask:

I have several eBooks that were broken into parts (Stanza on the iPod Touch) & was wondering, in the future, would it be possible for the application to view a directory of the parts as a whole book? If not i understand, was just curious =].

Again -- thank you so much!!

[pjjohn73]

Looking forward to future versions. Right now not working with my free pdb files, I downloaded one from fictionwize.com. Also using free prc files would be cool. And in the future possibly encrypted ereader .pdb files? Thanks for the application and once again looking forward to the new versions.

[tarirene]

Sweeeet. I was going through my collection of pdb files (my old tungsten finally just died on me, so I thought I was never going to be able to use them again) and converting them one-by-one... this is a much better solution. If I knew this was going to exist back when I bought my pre, I would have been much less hesitant about switching to web os.

Any chance of getting code to autoformat books that are poorly formatted - with carriage returns, too few/too many spaces between paragraphs, em-dashes and other "special" characters into something that will show up?

[lollygags]

Thanks a lot, worked fine when I tested it with a couple of pdb's.

[bulls96]

would like to thank you for this app!

Finally have a way to read my Wordsmith memos with the Font formatting. Awesome!

I tried some of my medical books (.pdb files) which are big files (from 7-60MB) and it cannot render them yet.

Is there a file size limit?

[try0625]

Hi i just tried your software to open a txt file written in Chinese. The file can be loaded but the content shown are all messed up. It seems that the code is not supported. Could you please look into it? I also attach one txt file written in Chinese with this reply. Thanks.

[prubin]

Will this work with ereader (fictionwise) books? If not, whjere can we get books in pdb format?

[Jappus]

Good news everyone!

The pReader version 0.5.2 now supports basic eReader PDB PPrs files.

Of course, since I threw this support together in a day, some caveats apply:

• Decompression doesn't happen on the fly, so for large files you have to wait up to 30 seconds till the pReader displays something else but "Reading file..."
• Zlib and PalmDoc compressed files are supported. DRMed files not, because there is exactly zilch documentation online of how those files are encoded. If you load such a file, the pReader displays junk.
• All rich text elements are removed from the file. I've tested it with a few files and the removal seems to be mostly lossless. But since it too was thrown together in a day, it might remove too much or too few metadata.

Despite that, it seems to work pretty well.


Happy reading!

[Jappus]

Originally Posted by SirWill

Out of curiosity any particular reason you haven't put the EXT: in the code to simply put the DB on the /media partition?

Well, for one, ~25.000 books ought to be enough for everyone.

Joking aside, I wanted to keep the DB hidden, because the app will have enough problems as it is. As soon as I'm sure that the app more or less works and can leave beta status, I'll add an option to move the DB to the media partition.

For now, it is simply one less potential error source. :P

[Jappus]

Originally Posted by Jonathan I Ezor

Question: I've got Hebrew DOC files, which show up, but not with the right font. I have Hebrew TrueType fonts, and could move them over to my Pre, but is there a particular name I should give them to use them in pReader, or will doing so screw everything else up?

By the way, if you can implement Hebrew fonts and right-to-left, you've got a *huge*, open market right now that no one else is filling (TealDoc isn't available other than through Classic). {Jonathan}

Originally Posted by try0625

Hi i just tried your software to open a txt file written in Chinese. The file can be loaded but the content shown are all messed up. It seems that the code is not supported. Could you please look into it? I also attach one txt file written in Chinese with this reply. Thanks.

I agree with you that support for more encodings and fonts would be great. And on the display side, it is incredibly trivial, since WebOS displays everything as a UTF-8 HTML page. Therefore, font and encoding support on the display side is a no-brainer. Just a few CSS-style options, afterall. Even right-to-left is no problem whatsover.


Unfortunately, that is just the display end. On the decoding end, thinks look considerably more grim. To understand that, you only need to know a single thing about Javascript:

Javascript does NOT want you to read binary data.

What I do to circumvent that is to completely abuse the XMLHttpRequest() facility of javascript to load a local file into a UTF-8 string that only uses 8 Bit codes. If you're technically inclined, you already see that this has a major drawback: The file must be loaded completely and directly into RAM. Loading partial ranges is not supported.


In any case, after decoding and decompression I end up with a wholly anonymous 8-bit stream. What I currently do is to simply translate those 8-bit chars directly into the low-byte UTF-8 portion which is, as you might have guessed, identical to plain ASCII. The only thing I currently translate are the German umlauts into HTML escapes, but that's just because I'm German and I used those as "proof-of-concept" test cases.

What they proved to me was that, basically, to support other character encodings, I have to write a symbol translation table that maps those 8-Bit codes into their "correct" Multibyte UTF-8 codes, or alternatively into ASCII HTML escape codes.


You can imagine that that's a rather time consuming task, which needs three things: 1.) Test files, 2.) the specification of the encoding and 3.) an unholy amount of patience.


Basically, you could help me out there. What I need to support more 8-Bit encodings is nothing else but a single file which contains in each line the following data:

8Bit-Hex-Code = HTML-Escape / Unicode Escape

for example:

0xC0 = ю
0xC1 = а

For the KOI-8 code

Supporting multi-byte code is a bit more tricky, since I don't know what the different file formats do with them and how WebOS will rect. To be honest, I haven't yet tested multibyte UTF-8 documents, since, as I said, it is rather time consuming to test it all, and I'm completely lacking on the front of test-files.


But as I said in the OP, supporting more encodings is a definite target for the app. It's just very time consuming, that's all.

And well, getting my thesis done sort of has priority at the moment. :P

[Jappus]

Originally Posted by SirWill

You might want to open the file search to include .prc files. I have some that I changed one to a .pdb file and it read it just fine.

The change will be in the next version v0.5.4. I don't think this issue alone already warrants publishing a new version.

I really wish you supported navigation of some kind. iSilo support would be HUGE. Plucker would be HUGE.

The problem with iSilo and Plucker is that they're Rich Text formats that are rather complex and pretty much useless when reduced to the currently supported plain-text display.

Afterall, there was a reason that I started coding with just TXT and Palm DOC support. They're both exclusively plain text and Palm DOC is a really simple format. Just as a comparison: The decompression of PalmDoc takes 46 lines of code (including comments). The ZLib decompression for PPrs files takes 400 lines of code.


Basically, support for iSilo and Plucker needs to wait till I've implemented robust Rich-Text support. And even then, it will be quite a lot of work, since I need to write a Plucker/iSilo to HTML converter.


The good thing with this program being OSS is that as soon as I've established a public repository, and as soon as I've got the time, I have no problems whatsoever with accepting patches from other developers. My ultimate aim is to modularize the program to the point that adding new formats is entirely independent of the main display code.


No single person alone can write a good eReader. And that's the desired target for this reader: good & free.

[Jappus]

Originally Posted by neve

What about changing the text of the file navigation? Can it wrap to the next line? Or maybe lose the icon to the left? Wrap would be best, I think. Right now I have files that look like:
Dick, Philip K - A Maz...
Dick, Philip K - A Sca...
Dick, Philip K - Count...
Dick, Philip K - Now...
Dick, Philip K - Rauta...
Dick, Philip K - The P...
Dick, Philip K - Time...

As you can see, it's diffifcult to tell which books I'm opening.
Also adding one of those scrolling alphabet index bars like in Music Player Remix might be helpful.

Unfortunately, I can change exactly zilch as far as the file selection code is concerned. That's because I call a WebOS Widget called "FilePicker", which can't be styled at all, since it's an entirely seperate application. That's also why you can select audio files and pictures and stuff in it. That has nothing to do with my app. It just comes with the FilePicker widget.


As it stands, I'd be more than happy to lose this crutch, since it is rather buggy. If you reread the original posting again, I've discovered a problem with the FilePicker that leads to an application crash. That's the reason why you have to click on "Reader" after selecting a file, and why the filename is printed into the main menu.

It is a crude workaround for that problem. Currently the FilePicker is all I've got to list files. Of course, there's the FileMgr project on PreCentral that hopes to address this issue, but it's an external, non-javascript service.

On top of that, it's just as much Beta as the pReader, and you really shouldn't mix Betas. It makes debugging virtually impossible.

So.. in closing.. not only have you given us an awesome text/pdb reader, but there's a whole compliment of media players built in, some rivalling or exceeding performance of the stock, standalone apps. I remain a little mystified as to what they're doing there in the first place, and part of me thinks "what the heck.. I'll use them from time to time and try to figure out how the programmer envisioned using these on a day to day basis.. leave pReader open all the time?", and yet another part of me is saying "Damn!.. some of these are a definite improvement over the stock app, maybe they should really stand on their own".

As I said, the FilePicker Widget is a very strange beast. Capable like a swiss-army knife and you have exactly as much influence over its design as with a real swiss-army knife. Namely none.

Basically, the WebOS API more or less says: Use this, or go play somewhere else. I really, really hope that the v1.3 API will be massively better. But I'm not exactly betting on it.

[Jappus]

Originally Posted by bulls96

I tried some of my medical books (.pdb files) which are big files (from 7-60MB) and it cannot render them yet.

Is there a file size limit?

Intrinsically, no. Unfortunately, as I've outlined above I have to load the entire file into RAM. There's currently no way around that. Believe me, I've checked. Apart from writing a Background Java Service, that is. But Palm keeps the API for that one rather closed and installing that service is a problem all on its own.

Another problem has to do with the performance. On the Palm Pre Device, WebOS monitors each application and checks how long it runs without relieving control to the AppController. That means, if a method runs for more than roughly 10 seconds, WebOS terminates that method.

And as you can imagine, since everything happens in Javascript, processing several megabytes can take a lot more than 10 seconds. Therefore, I need to do those long tasks asynchronously, which complicates the source code quite a bit.

The gist of it: Opening such large files has to wait till I've optimized the readers and till the WebOS API offers a better way to read files or at least parts of files.

[Walhalla2k]

Thnak you for this great application. Its almost what I was looking for :-) What I still need to be able to use it is more encoding as it cant display the czech lenguage corectly... So if it could load my own fonts..

like the ones created with the font creator for the palmfiction
http://translate.google.cz/translate...istory_state0=

and the other function I would love to see is the autoscroll.. again, as used on palmfiction...

but again.. thank you for the first really good ebook reader for the pre!!!

[goldyman]

I'm really not adding anything constructive to the thread other than to say:

Jappus. You. Rock.

[kurtlmyers]

Great job making a reader for PDB format! Now at least I can read my son's new Sword Dreamer book on my Pre. The single touch page turn is so much better than a smooth scroll when actually reading a book. It matches the life long experience of real reading and moving from page to page. It will be nice when the graphics come along as well. But how it is right now is great. Congratulations on a job well done and thank you from making it available.

[dump71]

Maybe I'm missing something, but where can I get the most up-to-date version?

[krid]

Curse you Jappus, I was going to release a GPLv3 ebook reader! I've been digging around in the guts of SimpleBible, and I had a pretty clear plan to generalize it to display html books. Congratulations on beating me to the punch.

I've downloaded your sources, and I'll see if I can make my ideas talk to your ideas.

Thanks for putting out pReader.

[remmysmom]

I have read through this forum and still am not clear if ereader books (password) are supported. When I select a book and hit the reader, I get the reading file and it just sits there. I have the updated version and it still happens.

Could someone let me know if the books they have purchased on ereader.com are supported with this program. If so, how did you get it to load?

Thanks!