[maemo-developers] VKB problems
I am trying to put together package of AbiWord 2.4.6 for Maemo 2.1, but run into difficulties with the VKB -- the kbd never comes up when the window is clicked on, although the same code used to work with OS2005. I see in the Maemo 2.1 docs that "the input methods work automatically ... whenever text widgets (entry, text view) are pressed ..." -- is there a way to make the VKB to work automatically for a widget of a different type, or do I have to manually call hildon_gtk_im_context_show() on focus_in events ? Tomas ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] .doc and .odf for OS2006?
Eduardo de Barros Lima wrote: > On 8/3/06, Marc McGuinness <[EMAIL PROTECTED]> wrote: >> Hello, >> > Hi Marc, > >> is anyone working on a port of a text processing software for OS2006? >> > > We used to have Abiword that can handle both .doc and .odt for the > OS2005 version, but I don't know if there is someone working on it for > the OS2006. I am currently working on the port of Gnumeric and hope to > release it real soon. > > I also would like to know if someone has started to work on Abiword > for OS2006. I could help if needed. > I have; there was a problem with the updated cs3.2 C++ compiler initially, but I finally got the CVS head building just before I went on holiday, for now still using the HildonApp api. Once I get the debian repo set up for it, there could be at least an unstable package available. Tomas ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] library updates
Hi, Neal H. Walfield wrote: > At Thu, 20 Jul 2006 18:06:31 +0100, > Tomas Frydrych wrote: >>> how are updates of library packages which are not visible in the >>> application manager expected to work? We just noticed that when updating >>> gpe-calendar the updated libeventdb does not get installed. libeventdb >>> is visible because its section currently is "user/libs", so in this case >>> i can install the update but what if the package is not visible? >> I think you just need to make sure that the main package includes >> appropriate version number in the dependency declaration, not just the >> library name; at least that seems to work for me. > > That's a hack and a poor one at that. Versioned dependencies are used > when the applications depends on a particular ABI version of a > library. I think it is a bad idea to abuse this. > What florian wants is `apt-get upgrade' functionality, i.e. install > updates for any installed package. If the application needs a particular version of the library, or higher than (for whatever reasons, not just ABI compatibility), then the package should declare that -- that is not a hack; if the application does not need a newer version than is on the device already, then this question becomes entirely moot. (Why should you, the application distributor, care about it getting upgraded?) > The application manager could use > the following predicate to determine if an application is up to date: > if and application and all of its dependencies (direct and indirect) > are the latest version available (respecting APT::Default-Release and > the pin settings, of course), it is up to date. This would cause a > bit of confusion when a new version of a commonly used library, e.g., > GTK+, becomes available as then pretty much all applications > indirectly depend on it. These can be exclude by having a base system > container. Yes, to mark applications as not being up to date because there is a newer version of a library they use, even though they do not need the newer version, would seem to me rather unintuitive from the user point of view. Perhaps the manager should simply offer the option to upgrade packages that are not normally visible in the UI, but that it installed with the applications (it should not be hard to do; it already knows to to remove the 'invisible' dependency packages when you remove the main package). Tomas ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] library updates
> how are updates of library packages which are not visible in the > application manager expected to work? We just noticed that when updating > gpe-calendar the updated libeventdb does not get installed. libeventdb > is visible because its section currently is "user/libs", so in this case > i can install the update but what if the package is not visible? I think you just need to make sure that the main package includes appropriate version number in the dependency declaration, not just the library name; at least that seems to work for me. Tomas ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] Application catalog from repositories
Tommi Komulainen wrote: > I see "libsqlite0 2.8.13-2" in two repositories / sections: > * under user/libs from Kernel Concepts > * under libs from Maemo repository > I think library packages that applications might want to pull in should not be using the user/whatever section convention, but the normal libs section as the Maemo package here does. Putting a library under user/libs clutters the application list with packages that are not applications and the user does not have a clue what they are (and should not need to know about at all). If you declare the section as libs instead, the installer will still pull that dependency from the repository, but without it showing in the installer UI. Tomas ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] IT2006 library installation
Steven Hill wrote:
> Is there a way to write a .deb package that the Application Installer in
> IT2006 will understand that only installs a library? Because the only
> way to install missing libraries into the filesystem is to install
> xterm, become root somehow (like use "becomeroot") and use "apt-get
> install" Most general users won't want to do this.
You make a standard package for the library ('standard' as regular
debian package, not one using the user/something section declaration)
and you declare it as a dependency in the main, user-installable
application package (in the control file). You then put that package
into your repository alongside your application package; it will not
show in the installer, but the installer will automatically pull it when
installing the application.
Tomas
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers
[maemo-developers] problems compiling C++ code on Maemo 2
I am trying to build armel AbiWord package for Maemo2, but the linker bails out for reasons I do not understand. The exact error message is as follows: ../../../../src/text/fmt/xp/libFmt.a(fv_View.o): In function `~AV_Listener':../../../../src/af/util/xp/ut_string_class.h:55: multiple definition of `UT_GenericStringMap::_key(UT_GenericStringMap::UT_Cursor&) const::bad' ../../../../src/wp/impexp/libImpExp.a(ie_imp_PasteListener.o):../../../../src/af/util/xp/ut_string_class.h:55: first defined here The odd thing is that ~AV_Listerner() is a NOP virtual destructor, so it makes no use of UT_GenericStringMap, and ut_string_class.h:55 contains yet another NOP virtual destructor, but the class is unrelated to UT_GenericStringMap either. The code builds fine both with gcc and MSVC, and used to with the Maemo1 toolchain, so I suspect this is a bug in the toolchain, but hoping there might be some workaround -- if any of the C++ gurus here would have any suggestions, it would be much appreciated. Tomas ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] desktop integration
Danny, been there and this works for me: Application Desktop file * The X-Osso-Service line must not contain a fully qualified name, you want something like X-Osso-Service=myapp * Must include at least one of the Exec or StartupWMClass lines, otherwise there will be no icon in the Task Navigator panel for your application. Application Service file * Call your service com.nokia.myapp where myapp is whatever you put into the X-Osso-Service line in the desktop file: Name=com.nokia.myapp Tomas Danny Milosavljevic wrote: > Hi, > > Oookay... > > I just spent four hours trying to find out how > to integrate an application into the maemo > environment. > > Now that was unneccesary and I'm giving up > for now, hoping for help. > > I ported a nice little program to maemo > (see http://www.scratchpost.org/nokia-770/packages/ ), > and it worked fine, when started from xterm. > > When I tried to start by using the application > menu, however, my osso_initialize call > failed. > > By now I tried every single one of the ~ 43000 (j/k) > combinations that could be valid in the > - desktop file > - service file content > - service file name > - osso_initialize call name > - osso_initialize call version > - osso_initialize call activation > > And it just doesn't work. > > For now I commented out the osso parts and now it > can be launched from the menu. :) > > However, stangely enough, now the application icon > of the menu entry is gone. It was there before I > commented out the service stuff. I didn't comment > out "Icon=", it's still there. > > Helppp :) > > And before I don't forget, a few nitpickings about the > tutorial: > > - > On > http://www.maemo.org/platform/docs/tutorials/Maemo_tutorial.html#LibOSSO-library > there is no mention how the include file needed > for libosso is called (until wy down). > > - the osso_initialize example uses a partial > dbus service name only, and isn't the > third parameter deprecated by now? > > - I sure hope the weird voodoo done by > osso_initialize (magically prefixing > "com.nokia.") is already gone? If not, it's > not mentioned. > > - And the example doesn't use "com.nokia", > but "org.maemo". Which would make it break in > mysterious ways. > > - the valid chars in the dbus service name are not > mentioned (and there is no NO DASHES warning) > > - "Icon – Name of icon that represents this application" > (in tutorial) says: > "Format of the file should be PNG and the prefix (.png) > should not be defined." > > That is probably a suffix, and my voodoo detector goes > off again. > Don't ever do that. The file is called how it is called, > no adding/removing stuff to *cough* "simplify" it please. > > Phiew... now I need sleep. Good night :) > > cheers, > Danny > ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] Optimized memory copying functions for Nokia 770
> Like Dirk already replied, the implementation is in macros in the .h > file. I see. That makes the comparison with memcpy somewhat unfair, since you are not actually providing replacement functions, so this would only make difference for -O3 type optimatisation (where you trade speed for size); it would be interesting to see what the performance difference is if you add the C prologue and epilogue.# BTW, you can instruct gcc to use inlined assembler version of its memcpy and friends as well, I think -O3 includes this, but if I read bits/string.h correctly in my sbox, there are no such inlined functions on the arm though, so there is certainly value in doing this. Tomas ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] Optimized memory copying functions for Nokia 770
There seems to be no source for the functions in the tarball. Tomas Siarhei Siamashka wrote: > Hello All, > > Here are the optimized memory copying functions for Nokia 770 (memset is > more than twice faster, memcpy improves about 10-40% depending on > relative data blocks alignment). > > http://ufo2000.sourceforge.net/files/fastmem-arm-20060312.tar.gz > > These functions were created as an attempt to experiment with getting > maximum memory bandwith on Nokia 770 (powered by TI OMAP1710) and also > learning ARM assembler in process. Getting maximum memory bandwidth > utilization is needed for 2D games and probably other applications which > need to process a lot of multimedia data. I'm particularly interested in > getting the best performance for Allegro game programming library > (http://alleg.sourceforge.net) on Nokia 770 and that was the motivation > for writing this code. > > After a few experiments with reading/writing memory using different data > size for each memory access operation, appears that writing in a bigger > chunks is much more important for reading, that means writing 16-bits > per memory access is usually twice faster than writing using 8-bit, > 32-bit memory access is also twice faster than 16-bit access. There is > no such significant performance degradation for reading with smaller > chunks, so optimizing reading seems to be less important. After trying > some orher half empirical experiments with writing to memory even more > seems like the most efficient memory bandwidth is achieved by using > 16-byte burst writes aligned on 16-byte boundary using STM instruction. > And this seems to provide at least twice better memory bandwidth > utilization than the standard 'memset' function on Nokia 770. Having > such fantastic results, I decided to try making some optimized functions > that can serve as a replacement for standard memset/memcpy functions. > Aligned 16-byte write with STM instruction is a core part of all these > functions, all the rest of code deals with leading/trailing unaligned > data chunks. > > It implements the following functions (see more detailed comments in the > code): > memset8, memset16, memset32 - replacements for memset, optimized > for different alignment > memcpy16, memset32 - replacements for memcpy, optimized > for different alignment > > Testing framework is included, which allows to ensure that this code > provides valid results and is also really fast. In order to run the > tests, this file should be compiled as c-source with > FASTMEM_ARM_TEST_FRAMEWORK macro defined. > > Requirements for running this code: little endian ARM v4 compatible cpu > > Results from my Nokia 770 are the following: > >--- running correctness tests --- >all the correctness tests passed >--- running performance tests (memory bandwidth benchmark) ---: >memset() memory bandwidth: 121.22MB/s >memset8() memory bandwidth: 275.94MB/s >memcpy() memory bandwidth (perfectly aligned): 104.86MB/s >memcpy16() memory bandwidth (perfectly aligned): 113.98MB/s >memcpy() memory bandwidth (16-bit aligned): 70.37MB/s >memcpy16() memory bandwidth (16-bit aligned): 101.31MB/s >--- testing performance for random blocks (size 0-15 bytes) --- >memset time: 0.410 >memset8 time: 0.260 >--- testing performance for random blocks (size 0-511 bytes) --- >memset time: 2.360 >memset8 time: 1.140 > > TODO: >1. implement memcpy8 function (direct replacement for memcpy) >2. provide big endian support (currently the code is little endian) >3. investigate possibilities for getting the best performance > on short buffer sizes >4. better testing in real world and on different ARM based devices > > I'm especially interested in getting feedback from running this code on > different devices. It is quite possible that these functions are only > optimal for OMAP1710, but do not provide any benefit on other devices. > > Currently this code improves Allegro game programming library > performance quite a lot (in my not yet finished patch), but it might be > also used for SDL. It is interesting if using these functions can > improve GTK performance as well. In that case we could have a nice user > interface responsivety improvement. > > As soon as a complete replacement for memcpy (memcpy8) is done, it can > be probably also used as a patch for glibc to improve performance of all > the programs automagically. > > Waiting for feedback, suggestions and test results on other ARM devices > (not only Nokia 770). > > > ___ > maemo-developers mailing list > maemo-developers@maemo.org > https://maemo.org/mailman/listinfo/maemo-developers ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] Re: problem in Maemo installation
Either DISPLAY is not set (try 'echo $DISPLAY'), but that is not very
likely. More likely you do not have sufficient priviliges to open the
diaplay -- run Xephyr as root to see if that makes any difference.
Tomas
sudarsana narashiman wrote:
>
> Hi ,
> I have installed scratch box 0.9.8.6 and maemo 1.1 version.
>
> but when tried to run the following script outside the scratch box
>
> #!/bin/sh -e
> prefix=/scratchbox/users/${LOGNAME}/targets/SDK_PC/usr
> export LD_LIBRARY_PATH=${prefix}/lib; export LD_LIBRARY_PATH
> exec ${prefix}/bin/Xephyr :2 -host-cursor -screen 800x480x16 -dpi 96 -ac
>
> I am getting the error
>
> Xephyr cannot open host display.Is the display set?
> could you please tell me the reason.
>
> regards
> sudarsan
>
>
>
> PROTECTED]>
>
>
>
>
>
> ___
> maemo-developers mailing list
> maemo-developers@maemo.org
> https://maemo.org/mailman/listinfo/maemo-developers
>
>
> !DSPAM:440e94ec103151430218324!
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] Problems with installing scratchbox
Andrey Khurri wrote: > Hi all, > > Most probably this question has already been pointed here but I did not > find answer... May be some of you could help me. > When installing scratchbox and running script ./run_me_first.sh there is > an error after the last script question: > --- > Would you like me to create the group 'sbox' for you? [yes/no] (yes): > No such group! > Creating group "sbox"... > /scratchbox/sbin/sbox_configure: line 20: groupadd: command not found > ERROR creating group: sbox > --- > How can I manage it? > Thanks, > A. Just add the group manually in whatever wat that is done on your distro (your distro might use some other command than groupadd, like newgrp). Tomas > ___ > maemo-developers mailing list > maemo-developers@maemo.org > https://maemo.org/mailman/listinfo/maemo-developers > > !DSPAM:440816df14629328119275! > ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] Icon to task navigator... again
Tomi Ollila wrote: > Hi > > I'm not sure if this was briefly discussed in a while ago, but > I just can not find anything much related to it > > I have 2 programs that has pretty much identical .desktop files, > but one launched from TN (or using run-standalone.sh) gets > an icon shown on TN after launched but the other doesn't. > > So, if I iconify the second program there is no way to bring > the GUI window back to the screen. > > Needless to say I've now spent 3 hours trying all kinds of tricks > to get it working, but no go (guess how log period is is constantly > deinstalling, copying new version, installing and rebooting the machine...) > > Latest configuration is: > > I dropped program.service file and removed X-Osso-Service=prog from > .desktop file > > There is file called usr/share/applications/hildon/bar.foo.prog.desktop. > > symlink to that is located at > etc/others-menu/extra_applications/1229_progname.desktop > > (note: prog != progname, but I think I've tried with same string > there... can not remember, tried so many cobbinations... I just tried by > renaming that symlink to match -- no help... as expected) > > latest content of the desktop file: > > [Desktop Entry] > Encoding=UTF-8 > Version=1.0 > Type=Application > Name=Cool Program > Exec=/var/lib/install/usr/lib/progdir/prog.sh > X-Osso-Type=application/x-executable > X-HildonDesk-ShowInToolbar=true > Icon=qgn_list_home adding StartupWMClass = prog.sh Should do the trick. Tomas ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
[maemo-developers] AbiWord package
Done some work on the 770 port of AbiWord in the past two weeks, in particular: * build without printing stuff (reduces the installed footprint by nearly MB, * run as a proper mutliview application, * some basic hibernation support. A deb package based on today's cvs's is at http://www.abisource.com/~tf/abiword_2.5+cvs20060302_1maemo1_arm.deb With the usual disclaimers for unstable CVS snapshot. Tomas ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-users] [maemo-developers] Gnumeric 1.6.2 and goffice 0.2 Released
Armin M. Warda wrote: > (Cross post to maemo-developers@maemo.org from [EMAIL PROTECTED]) > > Hi, > > unfortunately this release of gnumeric lacks the ability to put > an icon into the taskbar of the Nokia770, thus if you 'accidentally' > minimize the running gnumeric app, you cannont maximize it and > thus have to kill it from the xterm. > > This is a known restriction - Eduardo describes it on his webpage. > > Who can assist Eduardo to add the 'icon in taskbar' feature? You probably need to add the StartupWMClass line into the desktop file, as described in the maemo tutorial. Tomas ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] osso_initialize()
Ilkka Poutanen wrote: > Hi, > >>* The full name of your service has to be com.nokia.revsh > > That is not actually true. I'm using eg. a fi.tut.cs prefix and my app > works fine both in Scratchbox and on the device. Can you start your application from the desktop menus? From what you are saying I have done exactly the same, but without success (there are other ways to start the app that work, like from xterm or in SB using run-standalone.sh, or by removing the X-Ossso-Service line from the desktop file). If you can start it from the menus, would you mind pasting in the complete content of your desktop and service files? Tomas ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] osso_initialize()
Tomi Ollila wrote: > ... as my prog still gets killed... should I also give 3rd arg to > osso_initialize() as true instead of false. Yes, all the mamemo examples pass in TRUE, but according to the docs this parameter is deprecated and should be FALSE. I ended up passing FALSE at the end, it works for me. Also make sure that the version string you pass to osso_initialize() is identical to that in your desktop file. I spent about 4 hours last w/k trying to get this work, so some observations on what the whole thing does: * When you select your app from Task Navigator menu, the TN usesses some lib osso function to start your app based on the service name in the desktop file. * osso starts your application, using the Exec line in the service file (so if your app gets started at all, it is working up to here). * the system waits for the app to register the dbus handler; this happens when you call osso_initialize(), * the system gives the app a few seconds to start up the dbus service, if it does not happen, it kills (I think it just sends it sigkill). The waiting period is quite long, I have managed even to attach gbd during it few times (not that it does much good once the app is killed). There are two main reasons why the kill signal is sent: a) your call to to osso_initialize() fails; this would invarriably happen to me when I tried to start the app in scratchbox from gdb to see what is happening; at the end, the only way to start it inside gdb was to run gdb like this: run-standalone.sh gdb Then you can start it, but, it does not do much good, because now it is not attached to the TN, so it will not get killed. Best probably to print lot of dbg stuff to the console. b) your call to osso_initialize() succeeds, but the name of the service does not match the name of the service the system is looking for -- if that happens, double check the desktop and service files for typos: whatever is found the desktop file and as a parameter to osso_initialize() will endup prefixed with 'com.nokia.' The other thing that has occurred to me as possible (but this is just a hunch), is that you not only have to create the service, but have to register some of the callbacks for it; my app registers 3, the HW event call back, the rpc default callback and the application exit callback. It is also possible to run you application without dbus service at all -- you simply remove the X-Osso-Service line from you desktop file, in which case the TN will start your appl just by using the Exec line, and it will not get killed. The major problem with doing that is that some of the significant events get only notified over dbus, and I think without having a correctly registered service you would not receive those. Hope some of this helps. Tomas ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] osso_initialize()
The osso debus service name handling is badly designed and cannot handle
properly quialified names; to get your app working:
* The full name of your service has to be com.nokia.revsh
* the desktop file *must not* contain fully qualified name:
X-Osso-Serivce=revsh
* the service file *must* contain fully qualified name
Name=com.nokia.revsh
* you *must* pass the short name to osso_initialize:
osso_initialize("revsh" ...)
It is silly and ignores the dbus spec on naming, but basically, whatever
you pass into osso_initialize() will always end up prefixed with
'com.nokia.'.
Tomas
Tomi Ollila wrote:
> Hi
>
>
> I've tried to get my "menu-launched" program to work. System keeps
> kill-9:ing my prog since it was not communicating properly using
> D-BUS messages
>
> Currently I am using the "dbus escape hax" like:
>
> if (fork()) { sleep(4); _exit(0); }
> close(0); close(1); close(2); setsid();
>
> to get the program working. With that the (escaped) prog doesn't get
> killed but ...
>
> I've heard that calling osso_initialize() should somehow do the work,
> and I have been trying to search how that actually should be used.
> Thew only source code I've read is from maemopad and that seems not
> to do much more than calling osso_initialize() and saving the return
> value (but not using it elsewhere). The available documentation
> and usage tips I've found so far have not helped too much -- and there
> are so much uncertainty there that explodes my testing options...
>
> So far I've tried the following:
>
> In code:
>
> (void)osso_initialize("revsh", PRJ_VER, false, null); /* called after
> gtk_init() */
>
> (what should I do with return value ???)
>
>
> In revsh.desktop
>
> [Desktop Entry]
> Encoding=UTF-8
> Version=1.0
> Type=Application
> Name=RevSH Tunnel
> Exec=/var/lib/install/usr/bin/revsh-tunnel-gui
> X-Osso-Service=invalid.too.revsh
> # FIXME Icon
> Icon=qgn_list_home
>
>
> In revsh.service
>
> [D-BUS Service]
> Name=invalid.too.revsh
> Exec=/var/lib/install/usr/bin/revsh-tunnel-gui
>
>
>
> For maximum portability with minimum #ifdefs the program (currently)
> uses standard gtk widgets, (with the exception of setting "autocap" in text
> entries)...
>
> The flashed system is Nokia770-51
>
>
> Any help is appreciated
>
> Tomi
>
>
>
>
> ___
> maemo-developers mailing list
> maemo-developers@maemo.org
> https://maemo.org/mailman/listinfo/maemo-developers
>
> !DSPAM:43e0a44d211872090367755!
>
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] Custom backup locations
Erik Karlsson wrote: Currently you can use the /etc/osso-backup/applications/ directory for backup configuration files when you want to add some new backup locations. Except, of course, the Application Installer does not let you install anything outside of /var/lib/install/ ;) Tomas However the backup should also work when configuration files are put into the /var/lib/install/etc... directory so I think we have a bug here. ___ How much free photo storage do you get? Store your holiday snaps for FREE with Yahoo! Photos http://uk.photos.yahoo.com ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] ipkg vs dpkg
Kalle Vahlman wrote: Build environment as in the Maemo and packages as in the existing dbus/ dir contents (bad wording, sorry). If you need to port all of Maemo to a new pkg management just to use ipkg, well... Granted, you need to recreate the packages for AI currently too so perhaps it's not an issue. In fact, there is very little involved in making a deb package work with the AI -- if your app is well designed, you basically just need to pass --prefix=/var/lib/install/wherever_you_want_it to configure, and make sure to replace any standard system dependencies with the 'maemo' dependency; there are useful scripts in ApplicationCatalog to help with this. So far in this discussion nobody has been able to demonstrate any genuine advantage to using ipkg over dpkg on the 770. More significantly, the dpgk is already on the system and maintains a database of installed applications. It really escapes me what the issue here is, other than Nokia wanting to keep the system (under /) completely separate from the applications installed by the AI (under /var/lib/install) -- unless Nokia changes this policy, there is not much to be discussed; your packages need work with the AI. Tomas ___ Yahoo! Model Search 2005 - Find the next catwalk superstars - http://uk.news.yahoo.com/hot/model-search/ ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] Still problems with Linux flasher tool...
[EMAIL PROTECTED] wrote: I do it always so that I put the flasher to waiting and then turn on the device. Works always. I don't know about the Windows-flasher, I have never tried it. And by the way the first time I heard about the home key thing. Perhaps if you are confident that there is a such issue with the Windows-flasher (that it does not have the wait mode), you may file a bug against it. Or just use Linux and be happy, e.g. I wouldn't do software development in anywhere else anyway :). The 'home key' technique is the one (the only one) the win32 flashing wizard guides you through (and IIRC also described on the 770 homepage); as such not having a waiting mode is almost certainly a feature, not a bug. Personally I might share your 'use Linux and be happy' sentiment, but considering that the win32 flasher is the tool (only tool?) made available to the normal users by Nokia for software updates, and that undoubtedly most of the 770 users will be coming from the win32 desktop environment, it might not do any harm for Nokia's developers to get some first hand win32 experience with the 770? ;) Tomas ___ Yahoo! Model Search 2005 - Find the next catwalk superstars - http://uk.news.yahoo.com/hot/model-search/ ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] Still problems with Linux flasher tool...
Roberto Resoli wrote: 2005/11/21, [EMAIL PROTECTED] <[EMAIL PROTECTED]>: To my knowledge (please correct me if this is not true in all conditions) there is no need to start the update BEFORE turning on the device. I found that with the R&D mode enabled the 'hold the home key' technique does not work; the device does a normal boot regardless -- starting the flasher before connecting the device to the usb is the only way in which I could get it into update mode (incidentaly, this means that the windows flasher is useless when you have R&D enabled, because it does not seem to have a waiting mode). Tomas ___ To help you stay safe and secure online, we've developed the all new Yahoo! Security Centre. http://uk.security.yahoo.com ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] [maemo-users] The application Catalog page
[EMAIL PROTECTED] wrote: At this point, A WIKI based page will propably do. In the future, moving to a more flexible solution like a database would ofcourse be a great idea!:) First we just have to have enough programs for it to be worth it. My only comment on that would be would be that databases are easier to set up when there is not a large amount of data to input :). I do not think it would be a huge amount of work to do this (i.e., 1 mysql table, and a one or two php scripts). Tomas ___ To help you stay safe and secure online, we've developed the all new Yahoo! Security Centre. http://uk.security.yahoo.com ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] [maemo-users] The application Catalog page
[EMAIL PROTECTED] wrote: True Eero. The question propably is that what is more important for the community. We have 2 ways to go about this. Of course, you could easily have both views available if the document catalog is stored in a database. Tomas ___ How much free photo storage do you get? Store your holiday snaps for FREE with Yahoo! Photos http://uk.photos.yahoo.com ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] TAbiWord
Yes, I have wanted to have ink support in AbiWord for a while and reviewed the TAbiWord code when it first appeared; at that time the TAbiWord code was Win32 specific (fair enough) but unfortunately was done in a manner that broke the cross-platform framework of AbiWord. The AbiWord team did not have the resources to clean it up, and the orginal developer did not either AFAIK (I have not looked at the code for a couple of years now). With the arrival of the n770, ink support has moved high up on my TODO list and I am in planing stages of a new project to run alongside AbiWord with a provisional name libStylus, the aim of which is to provide generic ink support that would allow any gtk-based application to add such functionality with minimum code; I was planing to make an announcement on the maemo-dev list next week :) . If any of you are interested, please get in touch with me. Tomas P.S. AbiWord, of course, is available for the n770, see the application page on maemo.org. Jose Manrique Lopez de la Fuente wrote: Hello, Any Abiword developer has rescued/asked for this[1] code? An extension of AbiWord (version 2.04 code base) that supports digital markup and annotation on the Tablet PC. [1] http://www.cs.umd.edu/~meesh/kmconroy/ -- J. Manrique López de la Fuente http://www.jsmanrique.net msn: [EMAIL PROTECTED] jabber: [EMAIL PROTECTED] ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers ___ Yahoo! Messenger - NEW crystal clear PC to PC calling worldwide with voicemail http://uk.messenger.yahoo.com ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
Re: [maemo-developers] The future of the Application installer
Marius Vollmer wrote: It don't regard the 770 as an embedded system. Once you talk about installing arbitrary software on a system, it is no longer embedded, regardless of how limited its resources are. It is also quite resourceful actually. I think there is a lot to be said about using the 'real thing' when you can. I think we can use dpkg+apt resourcewise, so we should. The big issue is the screen realestate: even in the File Manager it is difficult to get to your files in a folder if you have more than 3 or 4 in there. The same goes for the installer: I am not sure what kind of UI would make it possible to trace a large number of dependencies without it being major pain, which is why I actually quite like the applet only keeping track of applications, not the system. If you want the installer to be fullblown dpkg frontend, there should be a 'normal user mode', that would be similar to how the intaller works just now, and some kind of an 'experienced user mode', and make the 'normal user' mode the default. Synaptic-like view of the system can be overwhelming even on a big screen with mouse and keyboard -- whatever you do, design it first of all to work for the ordinary user; just because the hackers find it not powerful enough does not mean it is bad. Tomas ___ How much free photo storage do you get? Store your holiday snaps for FREE with Yahoo! Photos http://uk.photos.yahoo.com ___ maemo-developers mailing list maemo-developers@maemo.org https://maemo.org/mailman/listinfo/maemo-developers
