[BETA RELEASE] PhoneME Advanced (Java Mobile) - Page 47 - maemo.org

Active Topics

The N900 is now 15 years old! (7)
to Nokia N900 by Danct12 - 2 days, 21 hrs ago
Sailfish OS for the Motorola Moto G7 Power (XT1955-5) - (ocean) (4)
to SailfishOS by Maemish - 3 days, 17 hrs ago
F(x)tec Pro1x (QWERTY) for sale (0)
to Buy & Sell by aln00000 - 6 days, 2 hrs ago
more...

Page 47 of 85

Thread Tools

freemangordon	2012-03-27 , 20:25
Posts: 3,074 \| Thanked: 12,960 times \| Joined on Mar 2010 @ Sofia,Bulgaria	#461

libcvm.so and cvm build according to the instructions.
libcvim.so build in Ubuntu 10.04.4 LTS with -mfloat-abi=softfp -mfpu=vfp -mcpu=cortex-a8 , cvm build in scratchbox

trying opera-mini:

Nokia-N900:/opt/phoneme# $PWD/bin/cvm -Xmx12m -Dmicroedition.profiles=MIDP-2.0 -Dsun.midp.library.name=midp -Dsun.midp.home.path=$PWD/midp/midp_linux_fb_gcc -Dcom.sun.midp.mainClass.name=com.sun.midp.main.Cdc MIDletSuiteLoader sun.misc.MIDPLauncher -suitepath opera-mini-6.5.26955-advanced-en-us.jar -1 Browser
ERROR! Bad CNI/KNI method return type for com.sun.midp.chameleon.skins.resources.LoadedSkinD ata.finishReadingSkinFile()I
java.lang.NoClassDefFoundError: javax.microedition.io.file.FileSystemListener
at java.lang.Class.loadSuperClasses(Unknown Source)
at java.lang.ClassLoader.defineClass(Unknown Source)
at java.security.SecureClassLoader.defineClass(Unknow n Source)
at java.net.URLClassLoader.defineClass(Unknown Source)
at java.net.URLClassLoader.access$100(Unknown Source)
at java.net.URLClassLoader$1.run(Unknown Source)
at java.security.AccessController.doPrivileged(Unknow n Source)
at java.net.URLClassLoader.findClass(Unknown Source)
at sun.misc.MIDletClassLoader.loadFromUrl(Unknown Source)
at sun.misc.MIDletClassLoader.loadClass(Unknown Source)
at java.lang.ClassLoader.loadClass(Unknown Source)
at java.lang.ClassLoader.loadClassInternal(Unknown Source)
at Browser.startApp(Unknown Source)
at javax.microedition.midlet.MIDletTunnelImpl.callSta rtApp(Unknown Source)
at com.sun.midp.midlet.MIDletPeer.startApp(Unknown Source)
at com.sun.midp.midlet.MIDletStateHandler.startSuite( Unknown Source)
at com.sun.midp.main.AbstractMIDletSuiteLoader.startS uite(Unknown Source)
at com.sun.midp.main.AbstractMIDletSuiteLoader.runMID letSuite(Unknown Source)
at com.sun.midp.main.CdcMIDletSuiteLoader.main(Unknow n Source)
at sun.misc.MIDPLauncher.main(Unknown Source)
at sun.misc.CVM.runMain(Unknown Source)
MIDlet suite has unexpectedly quit.

javax.microedition.io.file.FileSystemListener
REPORT: <level:2> <channel:1000> MIDlet suite has unexpectedly quit.

javax.microedition.io.file.FileSystemListener
QThread: Destroyed while thread is still running

trying microemu-demo:

Nokia-N900:/opt/phoneme# $PWD/bin/cvm -Xmx12m -Dmicroedition.profiles=MIDP-2.0 -Dsun.midp.library.name=midp -Dsun.midp.home.path=$PWD/midp/midp_linux_fb_gcc -Dcom.sun.midp.mainClass.name=com.sun.midp.main.Cdc MIDletSuiteLoader sun.misc.MIDPLauncher -suitepath microemu-demo.jar -1 org.microemu.midp.examples.simpledemo.SimpleDemoMI Dlet
ERROR! Bad CNI/KNI method return type for com.sun.midp.chameleon.skins.resources.LoadedSkinD ata.finishReadingSkinFile()I

The result looks ugly

(see attachment), trying to select menus leads to canvs moving, cvm segfaults on exit. Otherwise seems functional. I can bet cvm source is not the latest

running test.sh gives no errors:

*CONGRATULATIONS: test Test completed with 411 tests passed and 0 failures
*Output lines starting with a * should be checked for correctness
*They can be compared to src/share/javavm/test/TestExpectedResult

Attached Images

Quote & Reply |

The Following User Says Thank You to freemangordon For This Useful Post:
Estel

birefringence	2012-03-27 , 20:53
Posts: 19 \| Thanked: 47 times \| Joined on Nov 2010	#462

Since Germany has switched to summer time (CEST) this weekend, the time in Railnavigator is off by one hour (which would then correspond to CET). This is slightly annoying because all query values use the current time as the default ...

Could there be a bug in the time zone handling? (I've checked Railnavigator with Microemulator, and there it works as expected)

Last edited by birefringence; 2012-03-27 at 21:00.

Quote & Reply |

DavyP	2012-03-27 , 20:57
Posts: 306 \| Thanked: 603 times \| Joined on Jan 2012 @ Belgium	#463

Originally Posted by Arthuro_Adam

Thanks for the latest update.
I don't know, were there any changes in Opera Mini landscape/portrait mode or auto-rotation, but on N9, it's still the same. After start OM is in landscape mode and no auto rotation.

I have not changed anything w.r.t to the N9 implementation.
On the N9 emulator, the midlets also start in landscape mode
when the device is positioned in portrait mode.
I cannot test auto-rotate with the emulator, though if I remember
correctly, some people claimed auto-rotation worked for them (I
might be wrong).

I presume you have the same issue with other midlets beyond
Opera Mini (e.g. microemu-demo)?

The only easy fix I can think of is adding another entry in the
midlet settings that would rotate the midlet in the other direction.

Davy

Quote & Reply |

DavyP	2012-03-27 , 21:20
Posts: 306 \| Thanked: 603 times \| Joined on Jan 2012 @ Belgium	#464

Originally Posted by freemangordon

libcvm.so and cvm build according to the instructions.
libcvim.so build in Ubuntu 10.04.4 LTS with -mfloat-abi=softfp -mfpu=vfp -mcpu=cortex-a8 , cvm build in scratchbox

Thanks for testing. It seems I missed a patch that would solve the
finishReadingSkinFile error. However, this error is not critical.

The FileSystemListener error appears because you do not have
JSR 75 support yet. The build instructions I gave are without
additional JSRs. I will add those details later on.

Regarding the dragging: yes, this is an annoying "feature" and
you actually cannot disable it. So for my builds, when compiling
I actually look for a particular subversion commit in the original
subversion repository that added this feature, and then I undo
those changes. I will see if I can make a patch for that and add it
to the list.

About the ugly results: I forgot the fact that for current high
resolution devices the original background bitmaps (the blue
ones) and fonts are way too small. So I doubled the size of the
bitmaps and use those instead. That is why you only see half of
the text in top and bottom blue bars. That is why I usually have
two builds: one for low resolution devices (e.g. 320x240) and
high resolution devices (e.g. 800x480 and up).

I will probably create another package that you have to unpack
on top of the subversion sources and patches.

Despite all this, I am glad to see you got a build working. The
issues you mention are not unknown to me. I will address them
on my website.

Once again, thanks for testing!

Davy

Quote & Reply |

The Following 3 Users Say Thank You to DavyP For This Useful Post:
Estel, freemangordon, Joseph9560

DavyP	2012-03-27 , 21:30
Posts: 306 \| Thanked: 603 times \| Joined on Jan 2012 @ Belgium	#465

Originally Posted by birefringence

Since Germany has switched to summer time (CEST) this weekend, the time in Railnavigator is off by one hour (which would then correspond to CET). This is slightly annoying because all query values use the current time as the default ...

Could there be a bug in the time zone handling? (I've checked Railnavigator with Microemulator, and there it works as expected)

See what happens if you add "-Duser.timezone=GMT+2:00" to the
cvm command line parameters.

In Android, the front-end application (running in Dalvik VM) takes
the current timezone offset (in msec) constructs the above
GMT based timezone, and passes that along when calling cvm.
This means I have to do something similar in Qt4.

Davy

Quote & Reply |

The Following User Says Thank You to DavyP For This Useful Post:
birefringence

DavyP	2012-03-27 , 22:04
Posts: 306 \| Thanked: 603 times \| Joined on Jan 2012 @ Belgium	#466

I added two more patches on my website to fix the dragging problem, and the CNI/KNI skin error.

I also included 2 additional packages with resized skin bitmaps and
JSR 75 sources (with instructions on how to enable this JSR).

Warning: I have not done any thorough testing.

Regards,
Davy

Quote & Reply |

The Following 7 Users Say Thank You to DavyP For This Useful Post:
don_falcone, Estel, freemangordon, fw190, Joseph9560, wwweagle, [DarkGUNMAN]

freemangordon	2012-03-28 , 13:11
Posts: 3,074 \| Thanked: 12,960 times \| Joined on Mar 2010 @ Sofia,Bulgaria	#467

Originally Posted by DavyP

I added two more patches on my website to fix the dragging problem, and the CNI/KNI skin error.

Thanks, those problems are gone now.

I also included 2 additional packages with resized skin bitmaps and
JSR 75 sources (with instructions on how to enable this JSR).

Warning: I have not done any thorough testing.

Regards,
Davy

Resized bitmaps work ok, no more ugly looking midlets

But as long as I add:

export USE_JSR_75=true
export JSR_75_DIR=$WORK_DIR/jsr75
export JSR_75_PIM_HANDLER_IMPL=java

to my build script, libcvm.so no longer compiles. There are lots of error messages for missing classes, see the attachment

Attached Files

phoneme_build.zip (7.4 KB, 103 views)

Quote & Reply |

The Following User Says Thank You to freemangordon For This Useful Post:
Estel

DavyP	2012-03-28 , 15:22
Posts: 306 \| Thanked: 603 times \| Joined on Jan 2012 @ Belgium	#468

Originally Posted by freemangordon

Thanks, those problems are gone now.

Resized bitmaps work ok, no more ugly looking midlets

But as long as I add:

export USE_JSR_75=true
export JSR_75_DIR=$WORK_DIR/jsr75
export JSR_75_PIM_HANDLER_IMPL=java

to my build script, libcvm.so no longer compiles. There are lots of error messages for missing classes, see the attachment

Nice to hear you have a working base MIDP stack. The errors you get with JSR 75 seem to be related to the abstractions for running midlets on a CDC stack rather than a CLDC stack.

I would suggest to leave out JSR 75 support for the time being. I will try to address this issue when I add instructions for including JSR 179 (Location API) support. I still need to clean up this part.

Davy

Quote & Reply |

The Following 2 Users Say Thank You to DavyP For This Useful Post:
Estel, Joseph9560

[DarkGUNMAN]	2012-03-28 , 15:39
Posts: 309 \| Thanked: 456 times \| Joined on Jan 2010	#469

Hi Davy,

Been testing this without issue since the move to .deb install. Thats why I've been a bit quiet of late

Is it possible to get cvm to request packet data access on demand?
When the connection drops or autodisconnects and I try to reload a page, I have to manuall y go to connections and connect again since I have set it to prompt for connection.

Otherwise it is working fine

Quote & Reply |

The Following User Says Thank You to [DarkGUNMAN] For This Useful Post:
Joseph9560

DavyP	2012-03-28 , 18:14
Posts: 306 \| Thanked: 603 times \| Joined on Jan 2012 @ Belgium	#470

Originally Posted by [DarkGUNMAN]

Hi Davy,

Been testing this without issue since the move to .deb install. Thats why I've been a bit quiet of late

Is it possible to get cvm to request packet data access on demand?
When the connection drops or autodisconnects and I try to reload a page, I have to manuall y go to connections and connect again since I have set it to prompt for connection.

Otherwise it is working fine

Yeah, thanks for testing :-). Having a N900 has been quite helpful for me for testing purposes as well.

I actually had some fun disassembling the device. I bought a new digitizer to replace the scratched screen at Tmart.com for 8 EUR, but unfortunately it is missing the metal frame at the back with the 6 screw holes and I cannot seem to separate the metal frame from the original one. Should anyone have some tips where to buy this instead, they are most welcome :-).

Regarding your problem: I presume you are talking about 3G connections? This is not trivial because there are different ways of network connectivity (WiFi, 3G, USB). On Windows Mobile, there is something called ConnectionManager to take care of this. I have no idea if Maemo has APIs to do just that.

Davy