Page 28 of 46 FirstFirst ... 18232425262728293031323338 ... LastLast
Results 541 to 560 of 917
Like Tree43Likes
  1. alan sh's Avatar
    Posts
    766 Posts
    Global Posts
    913 Global Posts
    #541  
    Ah, I see. I didn't realise the pReader was so dependant on some WebOS internals.

    Cheers

    Alan
  2. #542  
    I'm sorry if that sounded unclear. Preader isn't dependent on something from webOS Internals. This is a bug with webOS itself. It just happens that it also effects preware & apparently any app that accesses databases in this manner. It only effects physical TouchPads (not emulator) running webOS 3.0.4.

    According to jappus' post above, HP is aware of, and acknowledges, this bug. There's no ETA for a fix.
  3. #543  
    After installing pReader Native, it will not always load....especially after loading the normal pReader first. I have to re-install and reboot each time, to fix.
  4. #544  
    I just deleted the Native preader...it was completely buggy. It skipped ahead dozens of pages at a time and went from 13% in a book to 34% in one page. The next time I loaded it, it loaded completely blank. A shame...cause when it was working at first, it was very quick.

    Just FYI, I did a total erase of my TP and started all over and the same issues with Native preader happened again.

    Also, the Native reader does not play well with the regular preader. When you alternate back and forth between them, they will stop opening and you have to reboot again to get either of them to open.

    I am on 3.0.4
    Last edited by TouchMe79; 11/19/2011 at 08:20 PM.
  5.    #545  
    Quote Originally Posted by TouchMe79 View Post
    I just deleted the Native preader...it was completely buggy. It skipped ahead dozens of pages at a time and went from 13% in a book to 34% in one page. The next time I loaded it, it loaded completely blank. A shame...cause when it was working at first, it was very quick.

    Just FYI, I did a total erase of my TP and started all over and the same issues with Native preader happened again.

    Also, the Native reader does not play well with the regular preader. When you alternate back and forth between them, they will stop opening and you have to reboot again to get either of them to open.

    I am on 3.0.4
    As I mentioned above, all these bugs are due to WebOS 3.0.4 on TouchPad devices. HP introduced a very massive change in 3.0.4 (moving the LunaSysManager to a different CPU Core than the applications you ruin) which exposed huge numbers of bugs and race conditions in the LunaSysManager.

    And since this Manager is responsible for many aspects of the WebOS operating system accessible through the SDK, lots of programs are affected. Furthermore, if only one application triggers the bug, it causes this manager to completely space out and break many other applications.

    For example, the way the pReader (both native and classic) interacts with the accelerometer causes the WebOS Depot interface to crash for all applications, which causes PreWare, Feedreader, the other pReader version and lots of other application to crash, stall, hang or load up empty. This bug is not just caused by the pReader, but also by several other applications, as it seems that all you have to do is to listen to accelerometer events and use WebOS Depot in some way to cause the bug.


    As such, yes, at the moment your only recourse is to not use any pReader version until a WebOS version greater than 3.0.4 is released (you don't need to uninstall the pReader, though, as not running it obviously does not trigger the issue). Ben Combee from HP Developer Relations has stated that the HP engineers have identified the root causes for these bugs and are busy fixing them. Unfortunately, he could not tell me a timing for when the new version is released.

    Here's the thread on the HP Developer Forums, by the way:

    https://developer.palm.com/distribut...p?f=70&t=18926
  6. #546  
    Jappus,

    Could you disable accelerometer access within your app as a temporary fix?
  7.    #547  
    Quote Originally Posted by Toasters View Post
    Jappus,

    Could you disable accelerometer access within your app as a temporary fix?
    Yes, but that's only way to trigger the bug, and thus it's not sure whether not registering for Accelerometer events helps in any way. After all, as far as I understand it, what crashes the LunaSysManager is not the Accelerometer itself, but rather a race condition in the WebSQL code used for the Depot interface (a simple database that JavaScript applications can use for data storage). Basically, the Accelerometer only triggers the bug, but it does not contain the bug itself. As such, that would be just a stopgap measure.

    Of course, if HP takes too long to fix this bug, I might decide to simply stop using the Depot-API altogether. Of course, as Govotsos has correctly pointed out "simply" is unfortunately not the same as "simple" here.

    But I'll nonetheless take a look at it later and see if disabling the accelerometer registration works. In that case, affected TouchPad owners can use it while everyone else does not suffer from reduced functionality. But as I said, I'm not sure whether it will avoid the issue in all circumstances...
  8. #548  
    Quote Originally Posted by Jappus View Post
    As I mentioned above, all these bugs are due to WebOS 3.0.4 on TouchPad devices. HP introduced a very massive change in 3.0.4 (moving the LunaSysManager to a different CPU Core than the applications you ruin) which exposed huge numbers of bugs and race conditions in the LunaSysManager.

    And since this Manager is responsible for many aspects of the WebOS operating system accessible through the SDK, lots of programs are affected. Furthermore, if only one application triggers the bug, it causes this manager to completely space out and break many other applications.

    For example, the way the pReader (both native and classic) interacts with the accelerometer causes the WebOS Depot interface to crash for all applications, which causes PreWare, Feedreader, the other pReader version and lots of other application to crash, stall, hang or load up empty. This bug is not just caused by the pReader, but also by several other applications, as it seems that all you have to do is to listen to accelerometer events and use WebOS Depot in some way to cause the bug.


    As such, yes, at the moment your only recourse is to not use any pReader version until a WebOS version greater than 3.0.4 is released (you don't need to uninstall the pReader, though, as not running it obviously does not trigger the issue). Ben Combee from HP Developer Relations has stated that the HP engineers have identified the root causes for these bugs and are busy fixing them. Unfortunately, he could not tell me a timing for when the new version is released.

    Here's the thread on the HP Developer Forums, by the way:

    https://developer.palm.com/distribut...p?f=70&t=18926
    Thanks for the info. I wish I had known that before I used WebOS Doctor and started over from scratch. I should have read more in this forum before using the Doc. For whatever reason, my web browser is now having issues since the WebOS Doctor and Preware install. I should have just done Preware and not the Doc.

    I think, in general, 3.0.4 is a mess and I can't wait for the next update which will hopefully clear up a lot of the bugginess affecting the TP right now.
  9. #549  
    Dear pReader developers,
    I love your product! I understand you are having problems with the new webos update, but I still believe in your product! I am sure you are receiving a lot of criticism regarding that but i just wanted to thank you for creating this free program for me and for everyone. I am really looking forward to your new update to fix these problems and your new features. Do your best! We are all waiting! Some more patiently than others.

    tigerboy211
    Los Angeles, CA.
    lar3ry likes this.
  10. #550  
    @Tigerboy211 BTW its not developers, its one guy (jappus) doing all this in his copious spare time while at University and now at a new job at Intel Germany. He's done a great job and dedicated a lot to the project / community - even to the point of having to track down & buy a TouchPad to verify & diagnose this issue because the emulator doesn't emulate this behavior.

    Jappus is a tremendous (and very patient lately) asset to webOS users. His work should definitely be appreciated by all. If you choose to, there are links to PayPal in the app & the first threads if you'd like to support the project financially. No, it's not required, or even solicited (he had to be convinced to add them), but it's nice anyway
    Zippich likes this.
  11.    #551  
    Good news, everyone!

    I've just released v0.9.14. I wanted to release this update quite a bit sooner, but I was quite swamped with work the last few weeks. Anyway, here's the changelog. Enjoy!

    1. Tracked down the reason for crashes on TouchPads. The reason is a bug in WebOS 3.0.4. HP engineers have promised to fix this issue with high-priority for the next WebOS release.
    2. Implemented support for user-added bookmarks.
    3. Selecting a default encoding now actually does something.
      PLEASE NOTE: If you upgraded, you might need to select one or else your new books might not open.
    4. Added support for many more encodings with an option to select a custom one for specific books.
      PLEASE NOTE: Your currently imported books might have had their encoding reset by this.
    5. Fixed a minor bug concerning images in the two-page-layouter.
    6. Fixed issue where reordering metadata fields caused the book name to default back to "[author] - [title]".



    Happy reading!
    MikeN68 likes this.
  12. alan sh's Avatar
    Posts
    766 Posts
    Global Posts
    913 Global Posts
    #552  
    So, a quick question. I know you have found bug no. 1, but have you done anything in your code to stop it happening?

    Alan
  13.    #553  
    Quote Originally Posted by alan sh View Post
    So, a quick question. I know you have found bug no. 1, but have you done anything in your code to stop it happening?
    Yes, I have looked into it, and to prevent the bug from affecting the pReader, I'd have to remove all references to both the Cookie and Depot API -- which mean rewriting the entire options-storage (and some other minor thingies). To additionally not trigger the bug for other applications, though, I'd also have to excise all references to the accelerometer and orientation APIs, as an overactive sensor call deep inside WebOS is one culprit of the entire malaise.

    So basically, I'd need to spend ~a week on porting the options storage reliably to the C++ back-end and then have to live with the limitation of locking the entire app into one screen orientation. And at the end, I still can't be sure that I don't accidentally trip over another way to trigger the bug inside WebOS. It wouldn't hurt the pReader anymore, but it'd still cause 100% CPU usage on one of the cores and preclude the use of PreWare and some other applications until a Luna restart.

    All in all, I think it's currently the best option to wait for HP to fix the root cause of the bug. Ben Combee has implied that the fix is already done and it just depends on when the next WebOS update is pushed out.

    Of course, if HP -- in its infinite wisdom -- decides to drop / sell-off / cancel WebOS, such an update might never materialize, in which case I'll have to do the things mentioned above and hope for the best. But I'm not yet at that grade of desperation.
    Last edited by Jappus; 11/27/2011 at 05:59 PM.
  14. alan sh's Avatar
    Posts
    766 Posts
    Global Posts
    913 Global Posts
    #554  
    ok thanks. I wasn't trying to denigrate your efforts - which are herculean and very well received.

    As it turned out, I installed the app and ran it. No issues until I then ran Preware when it did the classic rotating circle on me. A quick luna restart and back to normal.

    I really hope that HP fix this.

    Alan
  15.    #555  
    Quote Originally Posted by alan sh View Post
    ok thanks. I wasn't trying to denigrate your efforts - which are herculean and very well received.
    Ahh, no worries, mate.

    If my previous posting seemed a bit ... blunt; well, that may be just my German nature showing through. If I have learnt one thing while working for Intel, it is that the way Germans approach problems and communicate them to others is quite strikingly different from the US-American way.

    Basically, if you give most Germans a task which is needlessly hard (or impossible), they'll very bluntly tell you that it is needlessly hard or impossible and then usually propose to do it in a better way. I have found that US-Americans (no clue about Canadians, though ) react quite differently. Usually they'll tell you how incredibly awesome and great your idea is and say that they'll implement it with extreme enthusiasm right away. Then, over the coming weeks they'll step by step tell you that some things were done a bit differently than originally planned, but are now much, much better.

    In the end, both approaches achieve the same end-result of implementing the actually reasonable ... but do so in strikingly different way.
  16. #556  
    What you're trying to say in a nice way is we're good rear kissers while you're good at telling someone what to do with themselves more directly

    That passive aggressive is partly why the mental "health" business is booming over here. Your way might hurt someone's feelings (boo hoo) but it saves a lot of bs time.
  17.    #557  
    Quote Originally Posted by govotsos View Post
    What you're trying to say in a nice way is we're good rear kissers while you're good at telling someone what to do with themselves more directly
    Heh, sounds like you too have some Germanic/Nordic blood in you.

    Of course, mix both, and you get the modern Anglo-Saxon/Norman again; seems like bluntness works just like negative numbers. Multiply them, and they turn out extremely positive.
  18. #558  
    Hey I'm an avid pReader fan who uses both versions daily, so I definitely appreciate all your hard work as a developer, especially when HP is so up in the air about what they're going to do w/WebOS! So I was excited to grab your newest pReader release off Preware yesterday, but on WebOS 1.4.5 on my Pre Plus it just sticks on the "loading backend..." step. Although it doesn't cause a restart as occasionally w/ the previous version, now it just freezes on that screem & won't load. I did try an uninstall, restart device, then re-installed, just in case, but no luck.
  19. #559  
    Quote Originally Posted by Jappus View Post
    Heh, sounds like you too have some Germanic/Nordic blood in you.

    Of course, mix both, and you get the modern Anglo-Saxon/Norman again; seems like bluntness works just like negative numbers. Multiply them, and they turn out extremely positive.
    Yup, half Greek (the names a giveaway ) and half German.
  20.    #560  
    Quote Originally Posted by wakegirl40 View Post
    Hey I'm an avid pReader fan who uses both versions daily, so I definitely appreciate all your hard work as a developer, especially when HP is so up in the air about what they're going to do w/WebOS! So I was excited to grab your newest pReader release off Preware yesterday, but on WebOS 1.4.5 on my Pre Plus it just sticks on the "loading backend..." step. Although it doesn't cause a restart as occasionally w/ the previous version, now it just freezes on that screem & won't load. I did try an uninstall, restart device, then re-installed, just in case, but no luck.
    The pReader should generate quite a lot of log data while running. It'd be great if you could send me that log file.

    What you need to do that is just this:
    1. Download and install "InternalZ" with PreWare
    2. Retry starting the pReader
    3. Close it if it hangs, after waiting a few seconds (~20s should be more than enough)
    4. Open InternalZ and navigate to "/var/log"
    5. Copy the "messages" file to your USB disk ("/media/internal")
    6. Connect your Pre Plus to your PC, download and compress the file.
    7. Share it!


    If you don't want to upload the file for all to see, just drop me a PM, and I'll give you my E-Mail address.


    In other news: I found a way to at least don't use the Depot interface for the global Options. While this does not solve the bug, it might circumvent it enough to at least allow some semblance of pReader usage on WebOS 3.0.4 -- of course, you'll still have the problem of causing 100% CPU usage on one core and having to restart to use PreWare and some other apps, but it's better than nothing, I guess.

    If it works, you can expect a v0.9.15 shortly, as I can restrict this workaround quite easily to just affect WebOS 3.0.4 Touchpads and nothing else. The workaround will even auto-deactivate the moment you upgrade WebOS.

Posting Permissions