Page 6 of 46 FirstFirst 123456789101116 ... LastLast
Results 101 to 120 of 917
Like Tree43Likes
  1.    #101  
    Quote Originally Posted by RiverCity View Post
    What can I try next, other than sending my Pre to you or performing a hard reset on it? I am using WebOS 1.4.5.
    As I said earlier, I've brewed up a new IPK for you. It uses a different application name and id, so the jailer application should construct a completely new jail for it. I've just sent you a link via PM.

    But before you do that, I actually have another suspicion why the application might fail. There was a bug with how the usb-partition was mounted when you updated to WebOS 1.4.5 using the OTA updating process. You can find the details here:
    http://forums.precentral.net/webos-i...ta-update.html

    Basically, the gist of it was, that the bug prevented applications from being able to write to the USB-disk ... which is exactly what is happening to you. So I guess you can try to apply the emergency-fstab-fixer patch and reboot the phone. If it works, that'd be on the one hand great, because I finally have a solution for this pesky problem, but immensely bad in the long run.

    Because, you know, there will likely be a lot of other users out there with that problem. The bug should've been fixed by Palm in later versions than WebOS 1.4.5, but you all know how that went down. So the issue can only be really fixed with PreWare (or any other tool that can apply patches), something which I gather a lot of people won't actually do.


    Of course, putting in a check for whether the back-end can write to disk is a simple matter and I can display a suitable error message if that happens, but you know, the solution is not exactly ... easy.
  2. #102  
    Jappus,
    Congratulations!

    Submitted a couple of bug reports: stuck in bold font and slow scrolling.
    Let me know if it is user error or if you need more info.

    Thanks!
  3. #103  
    Mr. Jappus,

    Thank you, the problem is fixed! The Emergency Fstab Fixer .ipk file repaired the problem.

    I can now get into the Native pReader application, find my library of ebooks and load an epub book. Once loaded, I can bring up the book, page around in it and read the text. I have booted up the application more than once with similar results: No more white screen of death.

    I went to the link you sent to me. That page had a link to the .ipk file that was broken. However, that page also referenced other links about the problem. One of those links contained a link that allowed me to download the .ipk file. That reference is:

    Forums webOS • View topic - Alert: 1.4.5 Bug with Writing to /media/internal

    The file link is in the second post (by rwhitby). The file link is:

    http://ipkg.preware.org/feeds/webos-int ... .0_all.ipk

    I checked the USB root directory. The file pReaderNative.log is now in that directory.

    Thanks greatly for your effort.

    RiverCity
  4.    #104  
    Quote Originally Posted by RiverCity View Post
    Thank you, the problem is fixed! The Emergency Fstab Fixer .ipk file repaired the problem.
    I went to the link you sent to me. That page had a link to the .ipk file that was broken. However, that page also referenced other links about the problem. [...]
    Yeah, that's because you can download the patch easily via PreWare, sou you wouldn't have needed to go to the full length of finding a working IPK. But well, it worked, so I'm happy now.

    Of course, now I need to deal with this bug. There's only one upside to this whole mess: Even though the USB-disk can't be written to if your device is affected by the bug, it's still possible to write to the directory the application is in itself. That's also on the USB-disk, so there are no disk-space problems.

    The only downside is that the user can't see that directory. So, I'll probably implement it this way:
    1. Check if the media-partition is writable, if yes, put the library and optional log-file there, too.
    2. If it's not writable, put the library into the application directory and deactivate logging altogether. Notify the user of this problem at least once.


    I think that should clear up this bug. Thanks again for your invaluable help.
  5.    #105  
    Quote Originally Posted by gsparks2 View Post
    Submitted a couple of bug reports: stuck in bold font and slow scrolling.
    Let me know if it is user error or if you need more info.
    Thanks, I'll take a look at them tomorrow and see what I can do.
  6. #106  
    Quote Originally Posted by Jappus View Post
    Thanks, I'll take a look at them tomorrow and see what I can do.
    Update: After fiddling with font settings (changing away from palm default to verdana) bold went away. Still could not get palm default to change either to or from bold font. Verdana works great and appears identical to the verdana font in the original version.

    Scrolling sped up as well after switching to verdana.

    fyi I put the same book in both versions and maneuvered to the same location in the book for comparison.

    Hope this helps.
  7.    #107  
    Quote Originally Posted by gsparks2 View Post
    Update: After fiddling with font settings (changing away from palm default to verdana) bold went away. Still could not get palm default to change either to or from bold font. Verdana works great and appears identical to the verdana font in the original version.
    Could you send me the file in question, so that I can take a look at it? Maybe there's a stray font-tag in the ePub-file that screws up the layout, as selecting "Palm Default" does not actually force a specific font, so any font defined in the ePub takes precedence.


    The difference between the native pReader and the old pReader is that the latter removes all CSS definitions, whereas the former keeps them -- at the moment at least. Your file might help me to decide whether the native version should drop, or at least filter the CSS-styles, too.

    Thanks!
  8. #108  
    Sorry for the delayed response. I work in the nuclear power industry and we have been very busy the last couple of weeks.

    Yes, I will send you the file.
  9. #109  
    Quote Originally Posted by gsparks2 View Post
    Sorry for the delayed response. I work in the nuclear power industry and we have been very busy the last couple of weeks.

    Yes, I will send you the file.
    Yikes talk about understatement...
  10. #110  
    Quote Originally Posted by Unclevanya View Post
    Yikes talk about understatement...
    Yes indeed!
  11. #111  
    Looks good with the first book I tried. Love how fast it loaded. No problems reading, navigating around the book, etc. Setting up themes was OK, but when I created the second one it changed the settings on the first theme instead (leaving the second theme with the first theme's settings).

    Any idea when support for .mobi files will be added? Most of my newer books are in this format.
    Last edited by Nick31; 04/03/2011 at 10:09 PM.
    Nick
  12.    #112  
    Quote Originally Posted by Nick31 View Post
    Looks good with the first book I tried. Love how fast it loaded. No problems reading, navigating around the book, etc. Setting up themes was OK, but when I created the second one it changed the settings on the first theme instead (leaving the second theme with the first theme's settings).
    Mhhm, I'll take a look at that and see if I can replicate that behaviour.

    Any idea when support for .mobi files will be added? Most of my newer books are in this format.
    That might be some time, as I currently try to implement all the features that are currently non-functional, like the autoscrolling. At the same time, I'm currently fixing every layouting and HTML bug that I can find, since they are actually format-independent.

    After that is stable, I'll first implement B&N's ePUB DRM, then the eReader format and shortly thereafter MobiPocket. As soon as those are stable, the "rarer" formats follow, like plain-text, HTML, FictionBook2 an so on.

    At that point, I'll release that as v1.0.0 into the App Catalog. As soon as the bugs of that release are fixed (as more users will inevitably mean more triggered bugs), I'll start to add some features that are nifty, but not immediately necessary.

    Things like the integration of an Over-the-Air book "shop" (in quotes because it'll use completely free sources like Project Gutenberg), a shinier GUI and so on. And somewhere along that way, I'll have to face the transition from Mojo to Enyo, which will be lots and lots of fun ... for certain masochistic definitions of fun.


    So, to answer your question in a short manner: I currently estimate that to take a few more months. The exact amount depends of how much free time I can squeeze in between my job applications and interviews...


    P.S.:
    Of course, if somebody would volunteer to implement the missing formats, as I consider the back-end interface for that to be stable, support for the other formats would come much sooner. So, if anyone among you has some aptitude for programming in C++, you're more than welcome.
  13.    #113  
    Hello everyone.

    I just want to give you a short heads-up, that I'll release v0.9.4 in the next couple of days. The reason why I announce it early is because this version changes some important things in the back-end like data indexing and metadata handling. Most importantly though, it will change the library path from "/media/internal" to /media/internal/appdata/com.mhwsoft.preader-native".

    This means you will lose access to your current v0.9.x library and get a blank, new library. You should be able to manually move the content of the old library folder to the new one though. I've programmed it so that it'll detect the missing additional data & index files and regenerate them as soon as you open a book for the first time.

    I didn't make this process automatic as that'd just introduce clutter that will not be useful at any later stage outside of this alpha version release.


    And yes, I'm aware that the development speed has come to an almost glacial crawl. But there's a very good reason for that: I'm currently in the middle of several job interviews for companies like Google, Intel, Microsoft and some others.

    And as you can imagine, these currently eat a lot of my spare time. Especially Google. Very, very, very especially Google. To say that their interviews are merely hard is somewhat of an understatement. But well, you know what they say about those who dare not go where angels fear to tread...
  14.    #114  
    Good news, everyone!

    I've just submitted v0.9.4 to the PreCentral feed, so you should be able to download it soon. If you can't wait, you can of course already download it via SourceForge, as outline in the first posting of this thread.


    The new version fixes several bugs, of which the most important is the dreaded "White-Screen" bug, that is caused when the pReader can't write to the media-partition.

    The other important change is, that the library was moved from "/media/internal/pReader-Library" to "/media/internal/appdata/com.mhwsoft.preader/.pReader-Library".

    This has several advantages, the most obvious being that WebOS's image tool now ignores that entire folder and won't list the extracted images next to your own snapshots.

    The downside is that you'll start with an empty library. If you really need your v0.9.0-v0.9.3 library, though, you can try to move the content of the old library folder to the new folder.

    It should work and the pReader should fill in the missing data. But no guarantees there.



    So long and happy reading,
    Jappus.
  15. #115  
    Added two new bug reports: position reverts to previous position when the phone is reset and the app is open, gaps in displayed text (more than one page not displayed)
  16. #116  
    So far, the only bug I'e found that isn't already covered is that on my Pre- running 2.1, it grinds to a halt and eventually fails if I try to add too many books a once. All my books are in epub format, and I've found that I can add up to 4 or 5 at a time with no errors, 10-20 depending on size with errors but the books still being added (despite some errors popping up and telling me that adding failed).

    I tried to add 107 epub files at once (select all, done) and my Pre locked up completely about 1/3 of the way through.
    Quote Originally Posted by VeeDubb65 View Post
    It is as if you said that the result a dog and a cat mating was a giraffe. It's so completely wrong, that it's difficult to argue with someone who believes it to be true.
  17.    #117  
    Quote Originally Posted by gsparks2 View Post
    Added two new bug reports: position reverts to previous position when the phone is reset and the app is open
    Duly noted. The reason for this is because the metadata (which contains the current position) is only written back to disk when you close the book or the entire app. Otherwise, every page change would entail writing a few kB to the disk.

    I think I'll be less conservative in the next release and call the write-back on a page-change when the last write-back is more than a few minutes ago.

    gaps in displayed text (more than one page not displayed)
    I half expected that bug, as I've rewritten the relevant parts of the back-end to be faster and cleaner. I didn't notice it with my test-files, but they're not exactly covering all possible cases. If you could send me a file that exhibits that bug, that'd be great.

    Thanks!
  18.    #118  
    Quote Originally Posted by VeeDubb65 View Post
    So far, the only bug I'e found that isn't already covered is that on my Pre- running 2.1, it grinds to a halt and eventually fails if I try to add too many books a once. All my books are in epub format, and I've found that I can add up to 4 or 5 at a time with no errors, 10-20 depending on size with errors but the books still being added (despite some errors popping up and telling me that adding failed).

    I tried to add 107 epub files at once (select all, done) and my Pre locked up completely about 1/3 of the way through.
    I'll take a look at it, as soon as I can find the time. It really sounds like there's a memory leak in the import code, whichs imply clogs up the RAM on a Pre- much quicker than on my Pre2 test device. Especially because you've installed WebOS-2 on it, which means there's even less free RAM.


    One of the disadvantages when switching from JSJSJS $to$ $C$++: $The$ $latter$ $has$ $no$ $automatic$ $garbage$ $collection$, $which$ $means$ $a$ $single$ $method$ $that$ $does$ $not$ $clean$ $up$ $can$ $cause$ $your$ $entire$ $app$ $to$ $run$ $into$ $a$ $brick$-$wall$, $if$ $it$'$s$ $just$ $called$ $often$ $enough$.

    Time to whip out the trusty Valgrind suite, it seems.
  19. #119  
    Quote Originally Posted by Jappus View Post
    Duly noted. The reason for this is because the metadata (which contains the current position) is only written back to disk when you close the book or the entire app. Otherwise, every page change would entail writing a few kB to the disk.

    I think I'll be less conservative in the next release and call the write-back on a page-change when the last write-back is more than a few minutes ago.


    I half expected that bug, as I've rewritten the relevant parts of the back-end to be faster and cleaner. I didn't notice it with my test-files, but they're not exactly covering all possible cases. If you could send me a file that exhibits that bug, that'd be great.

    Thanks!
    I will send you the file and the locations where the gaps occur. The page change thing is not a big deal, I will just close the app when I am done reading. Thanks!
  20.    #120  
    Quote Originally Posted by gsparks2 View Post
    I will send you the file and the locations where the gaps occur.
    Thanks!

    The page change thing is not a big deal, I will just close the app when I am done reading.
    Eh, don't worry. It's an extremely easy fix (quite literally just 3 extra source code lines) which ensures that the app behaves as most people expect it to be.

    I was never a great believer in fixing bugs by forcing one's users to change their behaviour. If at all possible, the application should enable the user to behave in whatever way he or she desired without spontaneously causing the end of the world.

Posting Permissions