Re: [Barry-devel] bjavaloader works great!
On Fri, Jan 30, 2009 at 04:49:37PM -0800, Dirt Dawg wrote: > Hi, I just wanted to confirm to you all that bjavaloader works great! > I made a fresh build of Barry just today and the new bjavaloader erase > command worked flawlessly! You've made me a happy man. Thanks for the report! > P.S. If it's all the same to the developers, I'm going to start > telling others about this as I think Java uploading has been a long > awaited feature. Of course, all will get fair warning this is under > heavy development :D Please do. The more testers the better, and Barry can always use the free publicity. :-) > P.P.S. I sent this same message from different email by mistake. Would a > moderator please erase the duplicate? Thanks. Done. - Chris -- This SF.net email is sponsored by: SourcForge Community SourceForge wants to tell your story. http://p.sf.net/sfu/sf-spreadtheword ___ Barry-devel mailing list Barry-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/barry-devel
Re: [Barry-devel] stuck trying to implement get eventlog in javaloader
On Sat, Jan 31, 2009 at 02:02:53AM -0500, Josh Kropf wrote:
> To get log entries the command 7400 0200 is used. This time there is
> no JL_ACK response, but instead, another write occurs which just has
> a 16 bit integer value which increments and starts from zero.
> Following this is the JL_ACK and the log entry data.
>
> Now Barry seems to know when to send this 16 bit integer packet since I
> don't do any special handling for this. However the problem is that the
> value 2 is written instead of starting from zero (might not be an
> issue) but never increments.
>
> I use a loop to get the entries and inside the loop I have:
>
> packet.GetEventlogEntry();
> m_socket->Packet(packet);
> m_socket->Receive(response);
> MAKE_JLPACKET(jpack, response);
> uint16_t size = be_btohs(jpack->u.logentry.size);
It's a bit hard to tell without seeing your code, but note that
the packet object can build two kinds of packets: one with just a simple
command, or a packet with command + data. I'm assuming that GetEventlogEntry()
is creating a command + data packet.
If so, then you'll want to pass the argument to GetEventlogEntry() containing
the incrementing count.
For example, from a different function:
int JLPacket::GetSubDir(uint16_t id)
{
SimpleCmd(SB_COMMAND_JL_GET_SUBDIR, 0, 2);
return BigEndianData(id);
}
SimpleCmd() gives the command, and the amount of data to follow in the
next data packet (2). The data is given as a uint16_t big endian value.
The combination of the two, creates a JLPacket object with two chunks
of data to send.
Hopefully I'm answering the right question here. :-)
- Chris
--
This SF.net email is sponsored by:
SourcForge Community
SourceForge wants to tell your story.
http://p.sf.net/sfu/sf-spreadtheword
___
Barry-devel mailing list
Barry-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/barry-devel
Re: [Barry-devel] Linux 2.6.28: error submitting URB: No such file or directory
On Sat, Jan 31, 2009 at 12:55:26PM +1100, Jason Thomas wrote: > Hi Chris, > > The kernel configs are different and below is a diff. Hi Jason, I took a quick look at the diff, but I didn't see anything that rang any bells for me, but I'm not a kernel guru either. The most suspicious change to me was: > @@ -3436,7 +3536,7 @@ > CONFIG_USB_EHCI_ROOT_HUB_TT=y > CONFIG_USB_EHCI_TT_NEWSCHED=y > CONFIG_USB_ISP116X_HCD=m > -CONFIG_USB_ISP1760_HCD=m > +# CONFIG_USB_ISP1760_HCD is not set > CONFIG_USB_OHCI_HCD=m > # CONFIG_USB_OHCI_HCD_SSB is not set > # CONFIG_USB_OHCI_BIG_ENDIAN_DESC is not set I would try doing a 'make oldconfig' on .28 using the .27 config and seeing if that works. Preferrably on the latest kernel if possible. If it still doesn't work, let me know, and I'll have to see if I can move 'upgrading my kernel' higher on my todo list. :-) - Chris -- This SF.net email is sponsored by: SourcForge Community SourceForge wants to tell your story. http://p.sf.net/sfu/sf-spreadtheword ___ Barry-devel mailing list Barry-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/barry-devel
Re: [Barry-devel] Linux 2.6.28: error submitting URB: No such file or directory
On Sat, Jan 31, 2009 at 04:55:22AM -0500, Chris Frey wrote: > On Sat, Jan 31, 2009 at 12:55:26PM +1100, Jason Thomas wrote: > > Hi Chris, > > > > The kernel configs are different and below is a diff. > > Hi Jason, > > I took a quick look at the diff, but I didn't see anything that rang > any bells for me, but I'm not a kernel guru either. Found it. http://marc.info/?l=linux-usb&w=2&r=1&s=libusb+broken+on+2.6.28+x86_64&q=b The patch: http://marc.info/?l=linux-usb&m=123074111001861&w=2 And yes, this fix is in the latest 2.6.28.2. :-) - Chris -- This SF.net email is sponsored by: SourcForge Community SourceForge wants to tell your story. http://p.sf.net/sfu/sf-spreadtheword ___ Barry-devel mailing list Barry-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/barry-devel
Re: [Barry-devel] Barry OpenSync plugin alpha
On Fri, Jan 30, 2009 at 10:06:24AM -0700, Nathanael Noblet wrote: > I would love to test this, however I'm getting a bit lost in what to > download. Could a snapshot system be setup so I could easily grab the 0.40 > alpha version to compile and test? Hi Nathanael, Thanks very much for the offer to test. The 0.4x port of the plugin is still early... I haven't had a chance to fully test it myself yet. We do have a snapshot system (see the OpenSUSE Build Service link under Status at: http://www.netdirect.ca/software/packages/barry/ ) But... There is no binary packaging support for 0.4x yet, nor is there really binary packaging support for opensync itself, that I know of. There aren't even tarballs available for the opensync 0.3x devel tree. So if you want to test, you'll have to compile opensync from SVN, as well as Barry from CVS or git. There is documentation on compiling Barry from CVS/git here: http://www.netdirect.ca/software/packages/barry/cvs.php And if that isn't enough, please ask on the list, and I'll be happy to answer if I can. :-) - Chris -- This SF.net email is sponsored by: SourcForge Community SourceForge wants to tell your story. http://p.sf.net/sfu/sf-spreadtheword ___ Barry-devel mailing list Barry-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/barry-devel
Re: [Barry-devel] [patch] documentation for VSM files
On Thu, Jan 29, 2009 at 10:29:56PM -0600, Robert Yaklin wrote: > Thanks Chris, > I went ahead and installed git since making a patch with it looked > less involved than with cvs. I believe the attached patch will work. > I'll get the hang of this yet :) That was a beautiful patch. :-) Applied. Most git style patches can be included in the mail message inline, and any comments you want to make go after the -- mark in the patch. But either way is fine: inline or attachment. Thanks, - Chris -- This SF.net email is sponsored by: SourcForge Community SourceForge wants to tell your story. http://p.sf.net/sfu/sf-spreadtheword ___ Barry-devel mailing list Barry-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/barry-devel
[Barry-devel] binary package updates and OBS snapshot
Hi folks, I spent a good chunk of time on updating the binary packages, adding manpages, and doing a lot of test compiles, on my own machine, and on the opensuse buildservice. I also worked on getting Barry to compile on OpenSUSE 11.1, which meant getting libtar to compile as well. Things seem much more strict on OBS for opensuse 11.1, and the libtar autoconf files were a bit of a nightmare. :-) The outstanding udev issues on opensuse 11.1 are still outstanding. Anyway, the barry-util now depends on the FUSE library, and bfuse is included, which means fuse-devel is now a dependency. I see Chris Stover has already added fuse to his packages... he's on the ball! :-) This is a heads-up for anyone else who is following my package scripts. Due to the new FUSE dependencies, the new 0.15-20090131 snapshot is not available for the following distros, which don't have FUSE on the OBS servers: CentOS5 RHEL5 I don't know how important these are for users. Version 0.14 is still available, but not the latest 0.15-20090131 snapshot. That's all for now, - Chris -- This SF.net email is sponsored by: SourcForge Community SourceForge wants to tell your story. http://p.sf.net/sfu/sf-spreadtheword ___ Barry-devel mailing list Barry-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/barry-devel
Re: [Barry-devel] Barry OpenSync plugin alpha
Hi, The opensync-plugin 0.4x isn't still usable :( The plugin can be built, but after a first test I can list it : msynctool --listplugins don't see the plugin :( If you checkout the opensync-0.4x branch, you have to apply this patch to solve the compilation issue : diff --git a/opensync-plugin-0.4x/configure.ac b/opensync-plugin-0.4x/configure.ac index 1ff435d..bf26ddc 100644 --- a/opensync-plugin-0.4x/configure.ac +++ b/opensync-plugin-0.4x/configure.ac @@ -36,10 +36,10 @@ PKG_CHECK_MODULES([BARRY], [libbarry-0 libusb]) AC_SUBST(TREE_BUILD_CXXFLAGS) AC_SUBST(TREE_BUILD_LDFLAGS) -OPENSYNC_CONFIGDIR=$(pkg-config --variable=configdir opensync-1.0) -OPENSYNC_PLUGINDIR=$(pkg-config --variable=plugindir opensync-1.0) -OPENSYNC_FORMATSDIR=$(pkg-config --variable=formatsdir opensync-1.0) -OPENSYNC_HEADERDIR=$(pkg-config --variable=headerdir opensync-1.0) +OPENSYNC_CONFIGDIR=$(pkg-config --variable=configdir libopensync) +OPENSYNC_PLUGINDIR=$(pkg-config --variable=plugindir libopensync) +OPENSYNC_FORMATSDIR=$(pkg-config --variable=formatsdir libopensync) +OPENSYNC_HEADERDIR=$(pkg-config --variable=headerdir libopensync) AC_SUBST(OPENSYNC_CONFIGDIR) ## This is the place where you can install the default configuration files AC_SUBST(OPENSYNC_PLUGINDIR) ## This is the dir where you plugin will be installed I'm working on this plugin. I have packaged opensync from SVN snapshot : [nico...@dahlia opensync-plugin-0.4x]$ dpkg -l | grep opensync ii libopensync1exp5 20081121 ii libopensync1exp5-dev 20081121 ii opensync-format-vformat 2008 ii opensync-plugin-evolution2008 ii opensync-plugin-irmc 2008 ii opensync-plugin-syncml 2008 ii opensyncutils20081121 If you help me to write the Barry opensync plugin, I can share my package. Also, Barry opensync plugin requires that you use Barry from GIT. Regards, Nicolas Le vendredi 30 janvier 2009 à 10:06 -0700, Nathanael Noblet a écrit : > I would love to test this, however I'm getting a bit lost in what to > download. Could a snapshot system be setup so I could easily grab the > 0.40 alpha version to compile and test? > -- > This SF.net email is sponsored by: > SourcForge Community > SourceForge wants to tell your story. > http://p.sf.net/sfu/sf-spreadtheword > ___ Barry-devel mailing list > Barry-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/barry-devel -- Nicolas VIVIEN -- This SF.net email is sponsored by: SourcForge Community SourceForge wants to tell your story. http://p.sf.net/sfu/sf-spreadtheword ___ Barry-devel mailing list Barry-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/barry-devel
Re: [Barry-devel] stuck trying to implement get eventlog in javaloader
On Sat, 31 Jan 2009 04:44:12 -0500 Chris Frey wrote: > SimpleCmd() gives the command, and the amount of data to follow in the > next data packet (2). The data is given as a uint16_t big endian value. > The combination of the two, creates a JLPacket object with two chunks > of data to send. > > Hopefully I'm answering the right question here. :-) Perfect... thanks Chris, that's what I needed. Patch coming soon. -- Josh Kropf -- This SF.net email is sponsored by: SourcForge Community SourceForge wants to tell your story. http://p.sf.net/sfu/sf-spreadtheword ___ Barry-devel mailing list Barry-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/barry-devel
Re: [Barry-devel] further testing with fc10/8120/tmobileus...
Dave Booth wrote: Just an update on this issue - tried it again after grabbing the latest 8120 device software, in case there was a bug there. A bit of a forlorn hope since it was already working on windows and so it proved. Results identical with bb device software v4.5.0.124. Dave. Dave, After re-reading your original post and the pppd manual I believe this is what is happening. Your system received a request from your provider to authenticate using pap. rcvd [LCP ConfReq id=0x0 ] Your system say's "No auth is possible" and sends a reject message. No auth is possible sent [LCP ConfRej id=0x0 ] Since no authentication is possible the connection is closed. From the pppd manual: The default behaviour of pppd is to agree to authenticate if requested, and to not require authentication from the peer. However, pppd will not agree to authenticate itself with a par- ticular protocol if it has no secrets which could be used to do so. I believe your /etc/ppp/pap-secrets file doesn't have any entries in it, hence "No auth is possible". Try adding one of these to the pap-secrets file: \"\" * \"\" or "" * "" or * * \"\" or * * "" White space should be a tab character!!! Let me know if you get any further. - Andy . -- This SF.net email is sponsored by: SourcForge Community SourceForge wants to tell your story. http://p.sf.net/sfu/sf-spreadtheword___ Barry-devel mailing list Barry-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/barry-devel
[Barry-devel] Barry OpenSync for developper
Hi, First, I have packaged OpenSync Trunk today (for debian) : http://www.progweb.com/modules/blackberry/opensync/ Then, I have ported opensync barry plugin for this last snapshot : http://www.progweb.com/modules/blackberry/opensync-plugin/ Thank's to apply this patch on opensync-0.4x branch. I have an critical issue. I can build without difficulties :) Then, I run opensync : [nico...@dahlia nicolas]$ osynctool --listplugins ERROR: Could not load plugin "/usr/lib/libopensync1/plugins/barry_sync.so". Not a opensync library? EXIT_ERROR: osync_module_check: No version ERROR: Unable to load module: Could not load plugin "/usr/lib/libopensync1/plugins/barry_sync.so". Not a opensync library? Available plugins: evo2-sync syncml-http-server syncml-http-client syncml-obex-client And in traces : >>> osync_module_check(0x8061d50, 0xbfb3a454) >>> osync_module_get_version(0x8061d50) <<< osync_module_get_version: get_version not found. Not a library? ERROR: Could not load plugin "/usr/lib/libopensync1/plugins/barry_sync.so". Not a opensync library? <--- ERROR --- osync_module_check: No version Module check error for /usr/lib/libopensync1/plugins/barry_sync.so: Could not load plugin "/usr/lib/libopensync1/plugins/barry_sync.so". Not a opensync library? >>> osync_module_unload(0x8061d50) I don't understand ! I have well declared get_version function with extern "C". To enable OpenSync traces : export OSYNC_PRIVACY=1 export SYNCML_TRACE=/tmp/sync/syncml export OSYNC_TRACE=/tmp/sync/opensync mkdir -p /tmp/sync/syncml mkdir -p /tmp/sync/opensync Regards, -- Nicolas VIVIEN -- This SF.net email is sponsored by: SourcForge Community SourceForge wants to tell your story. http://p.sf.net/sfu/sf-spreadtheword ___ Barry-devel mailing list Barry-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/barry-devel
[Barry-devel] [PATCH] get eventlog and clear eventlog for bjavaloader
Here is a patch that implements "eventlog" and "cleareventlog" commands. One thing: I put enum definitions in a class called JLEventlogEntry... is there another more appropriate place for these? -- Josh Kropf eventlog_dump_and_clear.patch Description: Binary data -- This SF.net email is sponsored by: SourcForge Community SourceForge wants to tell your story. http://p.sf.net/sfu/sf-spreadtheword___ Barry-devel mailing list Barry-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/barry-devel
Re: [Barry-devel] javaloader erase
On Thu, 29 Jan 2009 21:08:39 -0500 Chris Frey wrote: > I added javaloader erase support tonight, and an experimental implementation > of force erase. I assumed this would erase an application even while it > was open and running on the Blackberry, but it didn't for me. Perhaps > I'm misunderstanding something. Chris I tested this and there is more to it I think. If I use bjavaloader to forcefully erase a module that is in use then I will receive an error. However if I look at the device, the application icon no longer appears on the ribbon. After a hard reset of the handheld, the module is in fact gone. I think this is because the stop command 0x8d has a response package that contains 0x78 (instead of the usual 0x64 JL_ACK response) which in barry has been recognized as an exception case meaning module already installed/in use, cannot continue. In fact if you look at erase-error-inuse.usb you will see that the handheld responds with 0x6c when attempting to erase an in use module without sending the force erase command. I think what 0x78 response means is that a device reset is required... which is precisely what happens when using javaloader.exe on windows: the device performs a hard reset after the erase. If you look at my USB log erase-inuse.usb you will see a bunch of traffic on socket zero after the stop command. This must be what's required to force the hard reset. BTW: loading a module over top of an in use module causes the same 0x78 response behaviour (see load-inuse.usb). -- Josh Kropf -- This SF.net email is sponsored by: SourcForge Community SourceForge wants to tell your story. http://p.sf.net/sfu/sf-spreadtheword ___ Barry-devel mailing list Barry-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/barry-devel
Re: [Barry-devel] javaloader erase
On Sun, 1 Feb 2009 00:24:23 -0500 Josh Kropf wrote: > In fact if you look at > erase-error-inuse.usb you will see that the handheld responds with 0x6c > when attempting to erase an in use module without sending the force > erase command. Silly me I just noticed you are already handling this response code... sorry about that. -- Josh Kropf -- This SF.net email is sponsored by: SourcForge Community SourceForge wants to tell your story. http://p.sf.net/sfu/sf-spreadtheword ___ Barry-devel mailing list Barry-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/barry-devel
[Barry-devel] Use C++ iostream classes for public methods that receive data?
I was just working on the save module command in bjavaloader and I have a question regarding all the public methods in JavaLoader class that deal with file data (load, screenshot, save). Does it make sense to use the standard C++ iostream classes for these methods instead of the Berry Data class? -- Josh Kropf -- This SF.net email is sponsored by: SourcForge Community SourceForge wants to tell your story. http://p.sf.net/sfu/sf-spreadtheword ___ Barry-devel mailing list Barry-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/barry-devel
