RE: passing arguments to hildon applications
ext Andrew Daviel mailto:[EMAIL PROTECTED] wrote: On Thu, 5 Jun 2008, [EMAIL PROTECTED] wrote: It's actually been poorly supported/unsupported for years. How hard can it be .. download a file to a temporary directory, spawn an application and tell it the filename. It worked in Mosaic, it works in Lynx. (never mind, I'm not sure I really want to know ...) Go read the spec. some of the handlers want urls not files. Now here's a simple problem: How do you know which handler to pick? Oh, you check the mime type. How do you find the mime type? Oh, you download the file. What if the file is behind a one time token? Oh, you're in trouble. What if the file requires a cookie? Oh, good luck passing that url to a handler. Now, if you only care about the simple file cases, then yeah, that's almost supported by most things. But the spec supports all sorts of fun bits. Someone should write a dbus wrapper thing I found one (Eero told me how to make it work). Yeah, I was hoping someone would include the name (I know it's in the archives), as someone else asked me about it for some other purpose yesterday. I'm not sure if I can pass an argument in a Dbus service record (probably not; in which case I'd have to create a separate wrapper for each MIME type, or get a name from a symbolic link or something). But in reality I'm not that likely to want to port Xdvi or Xfig to the tablet; most documents are saved in PDF. Mplayer on the other hand would be nice to have - preferably with a robust plugin that would play those Hollywood trailer embedded WMV movies or streaming media from CBC.ca Personally, I've tried to suggest that people get totem or a similar proper NPAPI plugin configured. The browser's default plugin can be disabled from the mangify button in the toolbar, which means it really shouldn't be too hard. The only trick is deciding how much integration you want w/ the system components (I'd suggest none). What I've suggested internally is that the idea media plugin would basically support streaming and pass the data to an out of [browser] process which would handle the media playback. Whether it sticks ui controls into the browser or just has a button which brings the media player forward is not my problem :). From playing w/ flash movies on my mac w/ two displays, I'm actually fairly annoyed w/ them... Because they have this full screen button, which will full screen on the proper display (aux), but when I click on another application, the full screen mode is discarded. So I can't actually watch a full screen video while using my computer's primary display to do work I mention this because I think wanting to full screen videos is a fairly common task, and bad implementations, especially for embeddings will result in unhappy users. Of late, I can't really think of many cases where I actually need to be able to see a video and do something else on my tablet at the same time, and in the cases when I do, I'm fairly certain that having them fixed in the browser's frame will not agree w/ me (iow, I'd rather the video float above the browser, so that I can scroll around the page while watching the video). (they dropped RealPlayer from a lot of the site) :( - on my desktop I end up doing things in Firefox like Tools/PageInfo/Media, clip the media URL and give it to mplayer on the command line - ugly, but usually works. Heh. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: maemo Bug Jar #7
Stephen Gadsby wrote: 61 bugs were closed: Luca Olivetti wrote: I noticed that some of the bugs are marked as fixed in diablo, and that's not really useful unless you release diablo rsn or provide updated packages for chinook. I suppose you think that black mail or threats is useful? As it happens, you can actually update your device to diablo prerelease images using application manager, as has been mentioned both on this list and on the web. We're trying to be open and indicate which versions will have fixes. This is a lot better than the past where bugs.maemo was mostly a black hole. The target milestone (which people should be using instead of simply writing fixed in diablo) will enable people who look at bugs.maemo.org after fremantle to determine which release had a fix. This is a valuable feature. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: maemo Bug Jar #7
Eero wrote: I would like the close bug jar section to separate things that were marked as wontfix from things that were fixed (or worksforme). seconded ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: maemo Bug Jar #7
Stephen Gadsby wrote: A Quick Look at maemo Bugzilla 2008.05.29 through 2008.06.04 As of 2008.06.04 maemo Bugzilla contains 3166 (+22 this week) items, including 1152 open issues (-40 this week): * 783 open bugs (-43 this week) * 9 critical/blocker (-3 this week) * 22 moreinfo (+4 this week) * 25 crash (-2 this week) * 21 patch (-2 this week) * 293 unconfirmed (+1 this week) * 369 open enhancements (+3 this week) * 7 moreinfo (+1 this week) * 5 patch (no change this week) * 104 unconfirmed (no change this week) I'm finding that I want to load all the bugs you're listing each week. And since I'm using outlook (yeah, I know, but Outlook-QuoteFix makes it almost bareable) clicking individual links is painful. Could you include links in the summary sections? The easiest way is: https://bugs.maemo.org/buglist.cgi?bug_id=1,2 The alternative is to include the queries you used w/ date pinnings, however those are less useful and won't really survive, although they could help people get into triaging: https://bugs.maemo.org/buglist.cgi?bug_status=UNCONFIRMEDchfield=[Bug+c reation]chfieldfrom=1w https://bugs.maemo.org/buglist.cgi?resolution=---bug_severity=enhanceme ntchfieldfrom=1wchfieldto=Nowchfield=bug_severity ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: maemo Bug Jar #7
Frederic Crozat wrote: Remember that most people are not used to Nokia policy to embargo any date (even estimate) regarding software (or hardware) release. For simplicity's sake, I don't know when Diablo will be released. I also didn't know when Chinook was going to be released (in fact, I believe a number of end users got Chinook before engineers at Nokia discovered/learned it was released). Releasing is a management decision, whether that's an open decision or a closed decision, and whether it's for an open, closed, or semi open project, it's still that. I also have absolutely no idea when Firefox 3 will be released. And that's the open source project that I work on. I don't have any idea when Safari 3.1.2 will be available, or IE8. The only products that I know release dates for are date based products (I can predict Ubuntu 8.04). Nokia's products are generally seasonal. And they have some wiggle room. The Mozilla view which I would hope people are familiar with is: It'll ship when it's ready. http://www.mozillazine.org/talkback.html?article=1179 http://www.pcpro.co.uk/news/124630/firefox-we-caught-microsoft-asleep-a t-the-wheel.html http://blog.johnath.com/index.php/2007/09/07/gaming/#comment-1 Kairo (one of the driving people behind SeaMonkey) writes: Only, ever, ship when it's ready to be shipped, never before that point. Except Quim has specifically requested those instructions to not be disclosed on this list or on the web and the hole in the update repository was closed for non insiders (or people who did the update before it was closed). Regardless, it was discussed both here and there that it is possible to retrieve the updates and that they're not simply vaporware. OpenSolaris also lists fixes at bugs.opensolaris.org before they're available anywhere (as it happens, they're date based, so a fix should be in the next binary drop w/in 2 weeks unless it fails QA which happens from time to time). Similar behavior can be found with other open projects (netbeans comes to mind, although I haven't worked w/ them in nearly a decade). As for Apple and Google, you will find that they may or may not announce a date for a product, and there can easily be a lag time between when it's announced and when you can get it. As for Kernel. If I'm using a Linux distribution, I know that just because mm or someone has committed a fix to an upstream kernel doesn't mean I'll instantly get that fix in my Linux distribution, in fact, for maintenance distributions (centos 3, rhel), I may be forced to wait for the time when I'm willing and able to upgrade to something newer. If I'm using a non-stock Linux or a non stock compiler )embedded-arm, codesourcery) then I'll probably have to wait for them to integrate the fix into a new release, and then pick a time to upgrade. If you aren't used to any of these things, then hrm... I dunno... I'm certainly used to waiting. And whenever people tell me about improvements, I'm happy, because it means that instead of bugging them about X, I know that they're responsive and I can choose to bug them about Y that they didn't know about. This applies to Qt (I filed a bug against Qt, they confirmed it, I'm still waiting for them to fix it) and Opera (I'm not sure how many bugs I've filed their way) and Apple (I need to file a bug about iLife's iWeb generating bad content). Note that with Trolltech, if I'm told that they've fixed a bug in Qt, and I'm using QtEmbedded (which I was at the time), then I have to wait for them to backport it to QtE, or I have to wait for a new version of QtE (and those lag a bit). I really don't know of many groups who deliver software instantly... At least, not good software. FWIW, in Mozilla if someone says X is fixed, that just means there's a code fix, it doesn't mean it has passed through all verification steps. It could cause a regression and be backed out. As was written elsewhere (by the original complainant), there's a risk in fixing things, and perhaps it'll be backed out, or regress something else or Now I'm not going to discourage people from using bleeding edge software, bug reports based on the latest sources and binaries are always more welcome than old stale versions. But I understand the reason people don't want to upgrade (I rarely upgrade my software) and I would hope that people understand that if they're afraid of using the latest prerelease, then asking that the latest prerelease be released is at least a bit ridiculous, changing the name from alpha, may destroy your computer to official doesn't mean it won't destroy your computer. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: maemo Bug Jar #7
Marius Gedminas wrote: Personally, I'm disappointed that when a bug is closed as fixed in diablo, as a user I've no idea how to get the updated package on my N810. Please excuse my disbelief ... you guys are the strangest users I've ever met. I've got scratchbox handy, In fact, I've got scratchbox handy is one of many definitions of not a user. Another is actually using Bugzilla at all (* see long paragraph below) If you're using Scratchbox, or if you're using Bugzilla, you're a developer. As a developer, there should be a much higher bar[ier] (reading a couple of pages) to entry. If you're using bugzilla, there are a few pages you should read Here's a short list of links that you can get to from enter_bug and show_bug: https://bugs.maemo.org/page.cgi?id=bug-writing.html Oddly, you can only reach bug writing if you have CANCONFIRM (people w/o CANCONFIRM should get guided which doesn't offer it) the link is bug writing guidelines https://bugs.maemo.org/page.cgi?id=fields.html (this is auto generated), there are links from status, severity, resolution on any bug. http://maemo.org/community/wiki/codenames/ This link should be present for most bugs (excluding bugs which are in video converter and misdirected) as a link from target milestone Note that this page indicates what 4.1 is and that it's equivalent to Diablo (thanks to whomever maintained this wiki page) http://browser.garage.maemo.org/news/8/ is sorta the only other page I know of with numbers, but it's not really a page that's likely to be maintained. I can't find any pages on maemo.org that talk about dates (in fact, the roadmap pages are utterly useless, typically even the utterly useless Mozilla roadmaps which are perennially out of date at least try to include guesses about seasons next year). I think it'd be nice if someone actually made page with release dates. Something like: http://developer.mozilla.org/en/docs/CVS_Tags (which is sadly incomplete) would be nice Probably an interim fix would be for http://maemo.org/community/wiki/codenames/ To be updated to indicate that Diablo will not be available for download until after the N810 WiMax Edition is available for sale. If it happens that this statement is wrong, someone can correct the wiki when a better statement is available. Most people in the interim would be able to follow the links read that, and realize that they'll just have to wait like the rest of us. (I'm actually waiting for Diablo, and I'm an internal!) just point me at the new source package and I'll backport it -- but repository.maemo.org has no 'diablo' in dists/. Note that bugzilla now only talks about versions (4.0, 4.1). And anytime someone says a bug is fixed in Diablo, they should make sure it has target milestone 4.1 When 4.1 (Diablo) is released, then everyone can upgrade to it using whichever upgrade methods are available (I'll probably flash since I'm running stuff from 3.x or pre 4.0). Btw, for people who said they can just download Firefox nightlies, try doing it for/from your tablet, Mozilla nightlies are available for a few platforms but that doesn't include tablets. Yes, as a convenience for testing they make available binaries for a limited set of common platforms, but for the rest you often get to wait for a release cycle. Oh, and if you were actually around when Netscape 6 was in development, you could see the same process that Nokia has now in place for bugs * (long paragraph from above) n.b. I work on Bugzilla, I'm a developer. When I claim to be a user, I read the user documentation; follow the instructions for reporting problems, sit back and wait until I'm asked for more input I don't ask questions about development process. As a user, I of course look in the official channels to see if there's a new version of software available for me. As a user I have: OS X 10.3.9 (the 9 is an auto update thing, the 3 is because I like the rest of you don't buy software, I buy hardware). Safari 1.x (dunno what x is, but it should be whatever it was when apple end of life-d it), Camino 1.6.1 multilingual (it's the latest and greatest!) Firefox 2.0.0.x Python 2.3 - I can't run: Firefox 3 Safari 2/3 Mercurial (because it requires Python 2.4) XP SP2 (users don't get prerelease service packs, and in business units SPs are always delayed) IE6 (no one offered me IE7 - I think my IT disabled that update), Firefox 2.0.0.x (generally not the latest because IT disables updates and delivers them urgently late), Office 2003 (I'm not responsible for purchasing software, so this is what's on my box) CentOS 3 (it came in a pretty vm - it's also the latest available for the host in question) Python2.2 Ubuntu 7.10 (it came in a pretty vm) ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Having trouble installing the scratchbox development environment.
Hendrik Boom wrote: I'm following the instructions in http://inz.fi/blog/2007/10/22/multi-target-development-for-maemo/ as advised by Graham Cobb, and have run into a snag. When I get to the point of creating the bora rootstrap, it tells me: [EMAIL PROTECTED]:~$ sb-conf rs mistral-armel /scratchbox/packages/mistral-armel-rootstrap.tgz ERROR: You don't have a Scratchbox user account! [EMAIL PROTECTED]:~$ Now doing adduser hendrik sbox was easy, but it didn't help. Evidently something else is also required. Did you log out? ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Having trouble installing the scratchbox development environment.
Hendrik Boom wrote: Did that; I suspect I am really, truly, don't have a Scratchbox account, and that something more is involved than just having an entry in /etc/group: Wow. Well, one thing to do is check /scratchbox/users/hendrik I believe that directory needs to exist. The alternatives are sh -x `which sb-conf` ... and strace -f `which sb-conf` ... ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: passing arguments to hildon applications
Andrew Daviel wrote: I'd like mailcap to work as a fallback; it's simple, Kinda, it's not so simple (see below) I understand it, and it's been standard in Unix for years. It's actually been poorly supported/unsupported for years. https://bugzilla.mozilla.org/buglist.cgi?query_format=advancedshort_des c_type=allwordssubstrshort_desc=mailcaplong_desc_type=substringlong_d esc=bug_file_loc_type=allwordssubstrbug_file_loc=status_whiteboard_ty pe=allwordssubstrstatus_whiteboard=keywords_type=allwordskeywords=em ailassigned_to1=1emailtype1=exactemail1=emailassigned_to2=1emailrepo rter2=1emailqa_contact2=1emailtype2=exactemail2=bugidtype=includebu g_id=votes=chfieldfrom=chfieldto=Nowchfieldvalue=cmdtype=doitorder =Reuse+same+sort+as+last+timefield0-0-0=nooptype0-0-0=noopvalue0-0-0= https://bugzilla.mozilla.org/buglist.cgi?bug_id=83305,115041,120380,1627 42,173106 https://bugzilla.mozilla.org/show_bug.cgi?id=83305 Add correct parsing/spawning of handlers from mailcap entries https://bugzilla.mozilla.org/show_bug.cgi?id=115041 support %u in mailcap helpers https://bugzilla.mozilla.org/show_bug.cgi?id=120380 needsterminal flag in mailcap must be respected https://bugzilla.mozilla.org/show_bug.cgi?id=162742 Helper Applications doesn't prepopulate based on mailcap https://bugzilla.mozilla.org/show_bug.cgi?id=173106 copiousoutput ignored in mailcap I think this doesn't include the fairly important bug about file handles not being closed before helper applications are spawned. Apart from the streaming media playlists, I have entries on my desktop for things like Xfig, LaTeX DVI, PostScript, Excel etc. while installers for things like Java Web Start assume they can add mailcap entries. I'm not knocking mailcap. I happen to think it's nice in theory too. But in practice, there are a lot of problems, a number of which are hard if not impossible to solve properly. Any application specifically written for the tablet would use DBUS, but we might still port desktop applications which run from the command line. Someone should write a dbus wrapper thing But yes, gecko kinda supports mailcap and microb or tablet-browser breaks it. Sorry about that. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: [gtkmm] Gdk::Pixbuf::create_from_file() segmentation fault inmaemo
Dinh Khac Thanh wrote:
Thank you for your fast reply, that deferencing one was
correct. However the error is still there after I changed it.
I am suspecting that it is because of thread conflict ( I
tested the create_from_file without threading and it worked
fine). I am not sure if gtkmm is thread-safe or not?
Google?
http://library.gnome.org/devel/gtk-faq/stable/x482.html
http://library.gnome.org/devel/gdk/stable/gdk-Threads.html
GTK+ is thread aware but not thread safe - it provides a global lock
controlled by gdk_threads_enter()/gdk_threads_leave() which protects all
use of GTK+. That is, only one thread can use GTK+ at any given time.
http://www.gtkmm.org/docs/gtkmm-2.4/docs/FAQ/html/index.html
Neither X, nor GDK nor GTK+ nor gtkmm are thread safe by themselves. You
must use either the gdk_threads_{enter,leave}() functions to protect any
and every
Note that you really should not send email to more than one mailing
list.
I don't know how to set a followup-to from outlook. But please only
reply to gtkmm
___
maemo-developers mailing list
maemo-developers@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: D-BUS service name confusion
Kimmo wrote: libosso.h says: * @param application The name of the application. * This name forms the last part of the default (D-BUS) service name of the * application. This is wrong. Last part implies that foo.bar will become either com.nokia.foo.bar or com.nokia.bar neither of which are correct, As it will actually become foo.bar Note that the D-BUS service name will be * 'com.nokia.application', where 'application' is the value you gave as the * parameter. Note also that this argument must be identical to the * X-Osso-Service value in the desktop file, or the D-BUS daemon will kill * your application. This is wrong. As application is not identical to com.nokia.application It needs to say the resolved value. And in fact, this special casing of Nokia is just plain stupid. But that's a different story. Vendors should not make it easy for everyone else to claim to be them. If a vendor wants to make it easy for the vendor's own apps to save characters, it should use a vendor specific wrapper. Note that this shortcut is probably a perf penalty since it has to involve extra checks and copies/allocs. The byte savings on flash are really not worth any of this. If you do not want to use the default 'com.nokia' prefix, * you must pass the whole D-Bus service name, e.g. 'org.maemo.my_app'. This should not be embedded into the same paragraph. * The only valid characters that the name may contain are letters a-z and * the underscore '_'. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Segmentation fault with the latest microb-engine
ext Zhihai Wang mailto:[EMAIL PROTECTED] wrote: (gdb) run Program received signal SIGSEGV, Segmentation fault. 0xb64e8aaa in nsFrame::BoxReflow (this=0x839c854, [EMAIL PROTECTED], aPresContext=0x82800b8, [EMAIL PROTECTED], aRenderingContext=0x83b2d28, aX=0, aY=0, aWidth=0, aHeight=0, aMoveFrame=1) at /home/wf/browser/microb-trunk/trunk/microb-engine/microb-engin e/build-tree/mozilla/layout/generic/nsFrame.cpp:6291 6291if (metrics-mLastSize.width != aWidth) (gdb) bt Congratulations you got symbols. Now you can go read the unix debugging faq (as referenced earlier). You could also practice using bugzilla. http://bugzilla.mozilla.org/query.cgi Answers (use this to check your work, do try to read the faqs and use the search tools before you read below the line): (gdb) info locals (bugzilla) 436232 [there are about 10 other resolved answers, but you said head, and afaik the others are all resolved] (code) metrics ?= 0 ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: passing arguments to hildon applications
On Tue, 27 May 2008, [EMAIL PROTECTED] wrote: Fwiw, at the present time, video/x-ms-wxv is among the hacks. These hacks were requirements. We're hoping that some future version may lose some or all of these hacks. But don't hold your breath, I've been holding my breath for two years and only just now got a sign that maybe something might happen eventually. Andrew Daviel [EMAIL PROTECTED] wrote: As I recall, these playlist content-types such as WXV, WXA and Realmedia RAM are required because the browsers do not understand RTSP URLs. I don't want to try to dig through my memory for why they were created. Our browser has *different* hacks for rtsp and similar urls. Actually, I can say for certain that feed: was created exactly for the opposite reason. Web servers wanted to ensure that some other application safely got a url. So there's a short text file containing the URL of the stream. Sadly from reading bugs, it seems that one microsoft mime type can mean either a stream or a playlist. I can't remember if this was by design or simply because the microsoft plug-in was tolerant and web sites were lazy. For a video file available via HTTP, using the playlist will let Mplayer stream it via HTTP, otherwise you have to wait for the entire file to download. Our hacks are basically designed to address this unfortunate standard behavior for the non playlist case if you're using the normal plugin path. It of course doesn't work correctly for mplayer (although in theory it could work as well as it does for the standard player). and it doesn't always work properly in general... It may be a hack, but I'm not sure how you'd get around it. Well, atm the browser has logic that should be in a plugin. How the plugin should be written should not be the browser team's problem :). I explained that when I first arrived, and maybe someday that might be the way it is. For people curious about NPAPI, the browser will give you a url and a chance to receive the data. You can choose to let the browser stream it to a file, feed it to you piecemeal, or not download it. At one time I thought there was a mailcap variant that caused Netscape to pass a URL to the helper application rather than downloading the file and passing the filename. There is (%u), support for this has been poor for years. I can't remember how well n4 supported it, but generally speaking Mozilla support didn't exist until fairly late (iirc bz tried to add some support for it). http://www.spinnaker.de/mutt/mailcap has an example The problem is that by the time you have a mime type, the browser also already has a stream of the file, and if it's a one time stream or a stream guarded by authentication, passing the url to someone else means paying twice ($+$) and possibly not getting anything at all ($ - /dev/null; 0 - /dev/audio). Or I guess you write a browser plugin. We have a strange browser plugin. Given the choice of this or a normal browser plugin, I'm hoping to get a normal one. I prefer an external helper that doesn't take the browser with it when it crashes... There are a number of strategies, but in the end, you'll lose. audio and video are coming to beta browsers near you by the year's end. * This is *not* a commitment by the microb team to include such a feature, merely a statement that prereleases of Safari, Firefox, and Opera will all include this feature in the near future. I can't predict when MicroB might get such a feature (we obviously need it for compatibility reasons). Audio processing can be done out of process, a plugin only really needs to accept a stream and set it to a different process for handling. It shouldn't even need to do much in the way of parsing. It *should* be possible for a couple of teams to write safe plugins which perform this sort of out of process implementation. There are a number of projects to do things like this, if only to enable 64bit browsers to host 32bit plugins, or for Qt based browsers to host Gtk based plugins. Actually, as far as browser crashes, diablo includes a feature such that the ui doesn't die when the engine crashes. It's not full sessionstore ala firefox, but we hope you'll like it. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Segmentation fault with the latest microb-engine
Zhihai Wang wrote: I have checked out the latest source code from the repository https://garage.maemo.org/svn/browser/mozilla/trunk/microb-engine and compiled in scratchbox under CHINOOK_X86, everything went smoothly. After installing the built debian packages, I started run-standalone.sh /usr/lib/microb-engine/TestGtkEmbed, and try to access some website such as www.google.com or about:blank, I got a Segmentation fault. ... /usr/bin/run-standalone.sh: line 11: 21205 Segmentation fault (core dumped) $@ Anyone knows what might have happened? Anything. Before sending messages like this about *any* software to any list you should: 1. build everything with debugging symbols For gecko, this means adding: --disable-strip --enable-debugger-info-modules (or --enable-debug) to your mozconfig (it also means doing something to prevent debugging symbols from being stripped by debian) 2. get a stack trace For gecko, this means: ./run-mozilla.sh -g -d gdb ./TestGtkEmbed In maemo this means: /usr/bin/run-standalone.sh /bin/sh cd /usr/lib/microb-engine/ ./run-mozilla.sh -g -d gdb ./TestGtkEmbed If you were actually dealing with a device instead of a scratchbox then gdbserver would probably be your friend (-d gdbserver). Note that you want to build an install the latest version of gdb (6.8 or something) instead of using whichever version your system has. You should also read any debugging guides for your platform. Google: debugging maemo 4 http://maemo.org/development/documentation/how-tos/4-x/maemo_debugging_g uide.html Google: debugging unix faq http://developer.mozilla.org/en/docs/Debugging_Mozilla_on_Linux_FAQ Google: maemo-debug http://maemo.org/development/tools/doc/chinook/maemo-debug-scripts/ You'll want to use debug-dep-install to grab symbols for the system libraries. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Segmentation fault with the latest microb-engine
Zhihai Wang wrote: I'm not good at gdb, and I'm just learning how to debug. Following is the trace I got, if this is not enough, I would like to collect more trace for you: Program received signal SIGSEGV, Segmentation fault. 0xb64d42ea in XRE_GetFileFromPath () from /targets/CHINOOK_X86/usr/lib/microb-engine/libxul.so (gdb) bt #0 0xb64d42ea in XRE_GetFileFromPath () from /targets/CHINOOK_X86/usr/lib/microb-engine/libxul.so You don't have symbols. You need to install symbols. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: D-BUS service name confusion
Axel Sommerfeldt wrote: So in total I find the current situation very bad: - The Maemo 4.0 Tutorial is wrong, it must be osso_context = osso_initialize( org.maemo.example_libosso, 0.0.1, TRUE, NULL); instead of osso_context = osso_initialize( example_libosso, 0.0.1, TRUE, NULL);, otherwise this example will simply not work. - The documentation of osso_initialize() is wrong/misleading. - The maemopad example is misleading: It works, but only as long as you don't change the com.nokia to something different, otherwise it stopps working and you simply don't know why. So, did you https://bugs.maemo.org/enter_bug.cgi?product=Development%20platform file a bug? ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: certificate dialog in microb
Zhihai Wang wrote: When accessing some websites like https://www.gmail.com;, a certificate dialog will show to the user. I guess that the one of the following signals should be emited: G_WEB_SIGNAL_CERTIFICATE_DIALOG (defined in gweb.h) G_WEB_SIGNAL_CERTIFICATE_PASSWORD_DIALOG (defined in gweb.h) G_WEBWIDGET_SIGNAL_MODAL_CERTIFICATE_DIALOG (defined in gwebwidget.h) I searched over the source code, but I failed to find where this signal be emited. You didn't search very hard. But instead of trying to search hard over the source code, please learn to use the cross reference. http://timeless.justdave.net/mxr-test/garage/search?string=certificate-d ialogfind=browser%2F If you want to skip the documentation, you can use a smarter search: http://timeless.justdave.net/mxr-test/garage/search?string=certificate-d ialogfind=browser%2F.*\.c It must have been emited since my N800 show me a certificate dialog ... ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Let's do something together in LinuxTag
Simon wrote: We'll also have to think about a better way of organising IRC meetings - it must have been rather painful for Quim writing answers to questions asked ~10min previously, and then missing the comments people made as he answered (and also confusing for people asking questions, wondering why they weren't being answered straight away). Someone suggested a separate chat channel and a channel simply for questions. Niels wrote: I think we can try out this approach in the next meeting. Use #maemo for the general chatting and ask questions and moderate #maemo-meeting. We can then give people voice when they have asked a question on #maemo and it is their turn to get their question answered? You should look at how undernet.org runs its #class channel http://www.user-com.undernet.org/class/ Specifically the after portion. People send messages to identified users who queue questions and then present them. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Scratchbox cs2007q3-glibc2.5-arm7 toolchain support?
Ben wrote: Trying to run this on an n810 gets [EMAIL PROTECTED] hello-world $ ./hello ./hello: /usr/lib/libstdc++.so.6: version `GLIBCXX_3.4.9' not found (required by ./hello) Welcome to the wonderful world of c++ and glib versioning. FWIW the normal C only hello-world compiled with the new toolchain runs fine on the n810. It doesn't require any newer libraries it would seem. What is the recommended solution for folks wanting to use the more recent g++ and toolchain? The new linking method is supported out of the box for the more recent ld in this toolchain ;) I haven't tried it, but personally I'd try to downgrade the scratchbox libstdc++.so so that it's the same one used by the older scratchbox. That's ideal, since what you want is the new compiler and the new linker, but the same libc as your target. If you can't do that, you should be able to point the build system to the old libc++ and say don't link against the compiler's libc++. I could imagine doing similar things to the things that were mentioned on a recent thread about glib. Pretty much. In the end I think it's probably best for the community to maintain a /usr/maemo/lib which when created is added to /etc/ld.so.conf and when files are added here ldconfig gets run. The quick and dirty version being having a personal lib directory and forcing its use with LD_LIBRARY_PATH in a little wrapper script for any tools I compile like: Yep, this is the standard mozilla approach for libraries. The theoretical downside for embedded devices is that you will run out of space if everyone ships their own copy of each library c++ (it's 740k today). $ gcc4-launch foo Or forcing the scratchbox to link to libstdc++.so.6.0.9 instead of libstdc++.so.6 and just shoving the libstdc++.so.6.0.9 into /usr/lib and being done with it. This becomes problematic, if you're going to do it, please coordinate w/ the community to make sure that there's an actual package for it and that you depend on it, otherwise when a second provider has it, they'll conflict with you, which will suck. I thought I'd throw it out there in case someone is using this newer toolchain and has a solution. Fwiw the mozilla mobile - fennec people intend to use the newer toolchain for their release (if they ever make one?), so you could wait and see what they do. Offhand I'd expect them to go the LD_LIBRARY_PATH route since that is what gecko traditionally does. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Who's who in maemo
Dave Neary wrote: I've been looking for a Who's who in maemo - a list of active developers (Nokia and otherwise) and what they specialise in. I haven't found one. I wrote one for the browser team: http://browser.garage.maemo.org/news/5/ I think that such a list would be very useful. In particular, I'd like to know when I see a nokia.com email address replying to a subject about the roadmap whether the person is a decision maker, a developer, or somewhere in between. There was a crusade for this a while ago. Oddly I think I defended one side of the fight and am now writing from the wrong side. For which I feel quite embarrassed because in addition to it making me look hypocritical, I fully agree w/ my original stance and am sure that the fact that I'm writing this email now indicates I am in the wrong writing from a mokia.com address (and was right in my original stance). https://bugs.maemo.org/show_bug.cgi?id=2035 Bugzilla Bug 2035 Bugzilla users who are Nokia employees should be noted It would also help, I think, to allow people to personalise and better direct any concerns and complaints, rather than just saying It's Nokia's fault. Speaking of which. I was just having lunch w/ Quim and I lamented not recognizing that you work for Nokia as a maemo representative. (Yes, I read the notice that there were new maemo representatives, but see below.) What do people think about this? Should I just get started and see how far I get? That's always fun. I tried making a map showing where we worked and to whom we reported. It was valid for at most three months (technically, I believe it was out of date before it was printed), and usable for another two years. The last reorganization meant that it's 100% useless (and then they tore down our walls and removed my chart, so now it's truly gone). And no, I won't and can't publish it anywhere. Dave Neary GNOME Foundation member [EMAIL PROTECTED] It'd be nice if your address was @maemo.org, it sounds like Quim could help make that happen. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Diablo, do we need a separate repository?
Graham cobb wrote: If there are *any* library changes (you mentioned libssl but I *really* hope there will be an up to date version of glib!) Perhaps we did a really bad job explaining what not changing the platform means. I'm 99.99% certain that glib2.0 will be based on 2.12.12, just as it was in the previous release. Assuming I'm correctly reading the changelog there was only one change to glib2.0, and that was to fix shlibs in debian/rules. Sadly it looks like the changes were all done in some private repository because my snapshots from both projects show the 4 version bumps all happening in one week even though the datestamps imply it would have happened weeks earlier. My guess is that this was because their versions failed integration (hopefully this is explained below). There was one attempted change, however it was reverted because it would have broken the ABI. Diablo is not a new OS. Most applications haven't changed significantly** According to the marketing material it's Internet Tablet OS: maemo Linux based OS 2008 feature upgrade The only major changes are feature updates to the browser (not actually a new browser, it's still based on the same old gecko as 2008), a new mail client, and the ssl change to support WiMax. I'll actually be working on browser release notes starting this week (it takes a long time). I might actually try to grab the highlights for the other apps if I manage to do the browser notes in fewer than 2 weeks. ** many applications probably haven't changed at all, I can do a diff at some point to get more details, but in general the maemo platform people actually provide pretty colored tables of this, so I don't need to. then apps built for Diablo will not work on Chinook. So, you have the problem that users still running chinook will find that apps in the chinook repository will not install! No. The only case where this should happen is an app that uses libssl. And libssl 0.9.8 *should* be in the repository. And fwiw, diablo includes the libssl 0.9.7 library (and package), so apps built against it from chinook would still work in diablo (this actually scares me, but I don't want to read the changelog to figure it out). (presumably Nokia does not allow a chinook package to upgrade libssl) Wrong, as explained above. Both 0.9.7 and 0.9.8 are installed and owned by their own independent packages. Note that there is no /usr/lib/libssl.so, /usr/lib/libssl.so.0, nor /usr/lib/libssl.so.0.9 only /usr/lib/libssl.so.0.9.7 and /usr/lib/libssl.so.0.9.8 - so there is no problem. An app that needs 0.9.8 simply writes: Build-Depends: libssl-dev (= 0.9.8) Depends: libssl-dev An app that doesn't care writes: Build-Depends: libssl-dev Depends: libssl-dev The former will force the system to install libssl0.9.8 into chinook as part of the install. And the latter will just work in either place (as long as the builder doesn't start with 0.9.8). It does mean that the autobuilders should actually use chinook with repository access to diablo, otherwise the results will be things that aren't the most pleasant of experiences. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Keeping Glib up to date (was RE: Diablo, do we need a separate repository?)
That said, to some extent people obviously do want to use later versions of libraries when/where possible. No one loves the idea of using code that's many years out of date with its ever growing set of known bugs. However sometimes bug-wise compatibility triumphs. Graham Cobb wrote: During the life of an installed release, I agree. Between Nokia-issued firmware releases, I disagree. I suspect the original hope was that diablo would have been delivered by SSU. If you keep that in mind, does it help change your view? Also note that the merge cost for hundreds of packages exceeds the small window for a project like diablo (which really really was a dot release). My view (I realise you disagree) Actually, in this case, I have no particular opinion. I understand why Nokia did it, and I can understand why you're upset. From a technical perspective, the browser team did not have enough time/resources to merge to trunk (nor was there a stable trunk of any value until long after we were frozen) and get any work done for diablo. We therefore had to choose not to merge to trunk and plan to do it for a future release. Most other projects (excluding wimax) probably had much fewer resources than the browser (in some cases they probably had no resources at all). is that at each new release Nokia should update all shared libraries. I think the key is that you're ascribing this to be an OS release. It isn't. it's a dot release. We never claimed it was a new OS release, the marketing information on this is quite clear, and I can't imagine anyone from Nokia would have claimed otherwise. http://www.backports.org/dokuwiki/doku.php You are running Debian stable, because you prefer the stable Debian tree. It runs great, there is just one problem: the software is a little bit outdated compared to other distributions. That is where backports come in. Think of chinook as a debian stable. Diablo is basically a collection of libraries provided by Nokia for chinook. You still have old libraries, and because it isn't actually newer software, the more you use it the more little bit outdated your software will become until an actual new distribution is released. As for how you manage to get a backports.org up and running, obviously that the package manager makes it harder is well... Unfortunate. But such is life. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Diablo, do we need a separate repository?
I wrote: I'm 99.99% certain that glib2.0 will be based on 2.12.12, just as it was in the previous release. Assuming I'm correctly reading the changelog there was only one change to glib2.0, and that was to fix shlibs in debian/rules. I should have written to _Nokia's_ glib2.0 for diablo. Marcin Juszkiewicz wrote: glib2 keeps source and binary compability between releases so upgrading from 2.12.12 to 2.16.x should not even require recompiling of applications. It might, however, please see the changelog I referenced in one of my other replies. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: autoconf upgrade problem
Kwan Hong Lee wrote:
For some reason, even if I have the new autoconf directory
added to the beginning of the directory, when I call
pulseaudio/autogen.sh it uses the older autoconf, currently
2.59. I can't modify sb_autoconf_wrapper because it's read only.
What can I do?
Please don't reply to individuals. This is a list for a reason.
[EMAIL PROTECTED]:/scratchbox/tools/bin$ echo *2.50*
autoconf2.50 autoheader2.50 autoreconf2.50
[EMAIL PROTECTED]:/scratchbox/tools/bin$ ls -l autoconf2.50
lrwxrwxrwx 1 root root 38 2008-03-31 18:52 autoconf2.50 -
../autotools/autoconf2.59/bin/autoconf
[EMAIL PROTECTED]:/scratchbox/tools/bin$ cd ../autotools/
[EMAIL PROTECTED]:/scratchbox/tools/autotools$ ls
autoconf2.13 autoconf2.59 automake-1.4 automake-1.7 automake-1.8
automake-1.9
[EMAIL PROTECTED]:/scratchbox/tools/autotools$ mkdir ~/ac262
[EMAIL PROTECTED]:/scratchbox/tools/autotools$ pushd ~/ac262/
~/ac262 /scratchbox/tools/autotools
[EMAIL PROTECTED]:~/ac262$ wget
http://ftp.gnu.org/gnu/autoconf/autoconf-2.62.tar.bz2 2 /dev/null
/dev/null
[EMAIL PROTECTED]:~/ac262$ tar jxf autoconf-2.62.tar.bz2
[EMAIL PROTECTED]:~/ac262$ cd autoconf-2.62/
[EMAIL PROTECTED]:~/ac262$ ./configure
--prefix=/scratchbox/tools/autotools/autoconf2.62
[EMAIL PROTECTED]:~/ac262/autoconf-2.62$ make -s
[EMAIL PROTECTED]:~# cd /home/timeless/ac262/autoconf-2.62
[EMAIL PROTECTED]:/home/timeless/ac262/autoconf-2.62# make install 2 /dev/null
/dev/null
[EMAIL PROTECTED]:/home/timeless/ac262/autoconf-2.62# cd /scratchbox/tools/bin
[EMAIL PROTECTED]:/scratchbox/tools/bin# ls -l autoconf2.50
lrwxrwxrwx 1 root root 38 2008-03-31 18:52 autoconf2.50 -
../autotools/autoconf2.59/bin/autoconf
[EMAIL PROTECTED]:/scratchbox/tools/bin# ln -s
../autotools/autoconf2.62/bin/autoconf autoconf2.60
[EMAIL PROTECTED]:/scratchbox/tools/bin# ln -s
../autotools/autoconf2.62/bin/autoheader autoheader2.60
[EMAIL PROTECTED]:/scratchbox/tools/bin# ln -s
../autotools/autoconf2.62/bin/autoreconf autoreconf2.60
[EMAIL PROTECTED]:/scratchbox/tools/bin# patch /dev/stdin
--- sb_autoconf_wrapper 2008-05-05 09:40:45.0 +0300
+++ sb_autoconf_wrapper 2008-05-05 09:42:47.0 +0300
@@ -26,2 +26,4 @@
ac250 ();
+ } elsif ($force eq '2.60' || $force eq '2.62') {
+ ac260 ();
} else {
@@ -158,2 +160,9 @@
-# Default to 2.13.
-ac213 ();
+my $force = $ENV{'SBOX_DEFAULT_AUTOCONF'};
+if ($force eq '2.50' || $force eq '2.59') {
+ac250 ();
+} elsif ($force eq '2.60' || $force eq '2.62') {
+ac260 ();
+} else {
+# Default to 2.13.
+ac213 ();
+}
@@ -222 +223,5 @@
+sub ac260 {
+run_autoconf (/scratchbox/tools/bin/${mode}2.60);
+}
+
sub run_autoconf {
^D
patching file sb_autoconf_wrapper
Hunk #2 succeeded at 161 (offset 1 line).
Hunk #3 succeeded at 224 with fuzz 1 (offset 1 line).
[EMAIL PROTECTED]:~$ scratchbox
[sbox-test-arm: ~] SBOX_DEFAULT_AUTOCONF=2.60
[sbox-test-arm: ~] export SBOX_DEFAULT_AUTOCONF
[sbox-test-arm: ~] autoconf
autoconf2.60: no input file
___
maemo-developers mailing list
maemo-developers@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: autoconf upgrade problem
Kwan Hong Lee wrote:
What do I have to do for autogen to use the latest autoconf I
installed?
Probably depends on your environment.
lrwxrwxrwx 1 root root 19 Mar 31 08:52 /scratchbox/tools/bin/autoconf
- sb_autoconf_wrapper
--
if ($force eq '2.13') {
ac213 ();
} elsif ($force eq '2.50' || $force eq '2.59') {
ac250 ();
if ($mode eq 'autoreconf' $arg eq '--install') {
ac250 ();
# Default to 2.13.
ac213 ();
sub ac213 {
run_autoconf (/scratchbox/tools/bin/${mode}2.13);
sub ac250 {
run_autoconf (/scratchbox/tools/bin/${mode}2.50);
--
[sbox-test-arm: ~] ls /scratchbox/tools/bin/autoconf
autoconf autoconf-2.13 autoconf2.13 autoconf2.50
Personally, I'd probably just manually run autconf2.62 or however it's
spelled.
Instead of installing in /usr, you could install in ~/ and have your
path be of the form:
~/bin:~/usr/bin:$PATH
If ~/usr/bin/autoconf2.62 exists, you could ln -s ~/usr/bin/autoconf2.62
~/bin
And then move on with life.
Installing things to / are imo one of the most foolish things people can
do. For fun just look at all the problems people have w/ /usr/bin/python
(which might be /usr/bin/python: symbolic link to `python2.5', or
/usr/bin/python: symbolic link to `python2.4' or /usr/bin/python:
symbolic link to `python2.3' or ..., where different versions of python
are fairly incompatible, and you never know which things will break if
you change the symlink).
___
maemo-developers mailing list
maemo-developers@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Reboot with charger plugged in
Juuso wrote: I just observed that the reboot of N810 or N800 is not a perfect reboot if you have charger plugged in during the switch off - switch on - sequence. At least manually loaded (insmod) kernel modules remain loaded and the ssh connections to the device does not result in connection lost if the charger is plugged in. The reboot time seems to be shorter as well.. This behavior was a bit confusing at first... :| It's a design feature. It's actually quite useful. If you're ssh'd in, use [/sbin/]reboot instead, that will kill your device completely. The fact is that the charger screen is actually managed by X, so if you turn off your device, the system essentially switches to the special run level (ignore the detail in case it isn't a true run level) where X is running w/ the charging screen, and when you turn on your device, the system essentially switches to the normal run level (*). As a user, this is usually fine. As a power user (anyone using an ssh session can not claim they are not a power user), this isn't a problem, and again is useful (the one time I recently executed reboot instead of using the power switch, I was annoyed because I lost my ssh session!). And as mentioned above, as a power user, if you want the other behavior, you can execute a true reboot from your session. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: the shared libraries used by /usr/lib/browser..
Vinod Hegde wrote: I am trying to get the shared library dependencies that browser in n810 has. (the ld.so trick would let you do it if you could find the right binary which may be browser.launch or something similarly strange.) Why are you trying to do this? Browsers generally speaking are dynamic creatures and will load other modules at run time. Debian package management has a concept of dependencies which you could look at to get a general understanding of which libraries are required by another binary. In fact, generally speaking the debian build system is designed to answer this exact question Note that neither of these techniques will tell you that the flash player plugin is used by the browser. For that, you're better served by strace or something that lets you look at loaded libraries at runtime. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: RE : Re: Fullscreen and Flash
Frédéric Charrier wrote: I can't find this file, there is no tutorial-applet on my n810. Is there a way to install it ? It shipped w/ the os update, you should run the latest version of os2008, not the earlier ones. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Using the scrollbar on the N810
[snip observations that the web widget is catching mouse events outside the thumb but inside the scrollbar] I wrote: Unfortunately, this isn't really how things work. Either a widget asks to capture input, in which case it can get events outside its boundaries, or it doesn't. If a widget does ask to capture events outside its boundaries, it then needs to have some idea about what to do with them. And that's hard (and each group will implement it differently, but poorly) Kalle wrote: I don't understand this, why would the browser widget be interested in events that happen on the scrollbar widget? Pictures required: Browser Widget +---+ |This is a web page. It's a bit wide..|^| |This is web content, it isn't interes| | |Sometimes there might be /^%$TRER^Y\ | | |Other times you mightHSDfkjl435H |X| |have a link or buttonDSFGDFGDR$T | | |there are other objects SDFDSFDSFDS | | |but most people forget SDFDSFSDFSD | | |about them. sDFSDFDSFDS | | |_/v| |XXX x| +---+ We have two thumbs (labeled X and XXX for vertical and horizontal scrolling in their respective tracks). If you click on X, and then drag over the button or link, or any other part of the page, then X gets the events and has to decide what that means. If you happen to reach all the way to the link in my picture, something has /probably/ gone wrong. That said, the exact algorithm that Gtk uses is to my knowledge not exposed (unlike Gtk Widget theming which is exposed). The windows algorithm seems to involve the thumb halting movement when the distance from the track is ~60px, offhand, I don't believe there's an API available to find out this information either. +---+ | |^| | | | | | | | 1 |2| | | | | | | | | | | | | |_/v| | 3 4| +---+ 1 - browser content 2 - browser widget emulating vertical scrollbar 3 - browser widget emulating horizontal scrollbar 4 - browser unhappy More detailed pictures are of course available, but usually the interesting ones involve much more complicated pages. For example, pages that are 200,000 pixels tall (historically some widgets especially related to X toolkits had issues with 16bit limits). if you think that your toolkit can really handle such widgets correctly, I have a number of crashing bugs involving X and images that are e.g. 1x200,000 in dimension [you're of course welcome to volunteer to fix them]. You will not find a properly functioning css3 capable web browser which actually uses native widgets. They *all* synthesize them (some do more synthesis, and some less, but all widgets are synthetic). That's what happens anyway, I start a drag over the scrollbar outside the thumb and the page is panned as if I dragged on the page. Unless of course (and I guess this is the case) it's not really two widgets but one widget doing funny things. It's not not two widgets, but it isn't a browser widget and a gtk scrollbar widget. It's a browser content area and a browser scrollbar widget emulating a gtk scrollbar widget, both inside a browser widget. I'm glad you figured out the general problem. I don't see this with any other application, the scrollbar will jump to the direction of the stylus as expected if I start a drag outside the thumb. So I suppose this particular bug can and should be fixed in the browser UI / browser engine... Lots of things can be done in lots of places, some things can't be done correctly in any one place. Some things shouldn't be done. And some things might never be done. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Using the scrollbar on the N810
Kalle wrote: I don't understand this, why would the browser widget be interested in events that happen on the scrollbar widget? That's what happens anyway, I start a drag over the scrollbar outside the thumb and the page is panned as if I dragged on the page. Eero wrote: AFAIK the scrollbar is part of the HTML widget. Bah, eero's answer was way to short (accurate, but short). ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Using the scrollbar on the N810
Yes, the general problem seems to be that there is a bug in the emulation that makes it work differently and, more importantly, in a way that is error prone and confuses the user. Let's take that pretty picture here: Glad it came in handy (and I do appreciate the explanation, because w/o yours, I indeed had a different understanding of which problem was being discussed) +---+ | |^| | |-| | |-| | 1 |2| | |-| | |-| | |-| | |-| |_/v| | 3 4| +---+ Now let's consider the two scenarios: 1) user pushes stylus down on 2 (the thumb) and starts dragging 2) user misses 2, hitting the dashed area instead and starts dragging 1) is fine, works reliably (as long as the touchscreen does) and so on. 2) is where normally nothing but a jump in the location towards the cursor happens (for the record, works like this on this ff3b3 I'm writing this on). On ITOS2008/microb, the page is not jumping but instead starts to pan. Gah, I misread this twice. (it's late) Here's what I originally wrote (it's irrelevant): Some of these problems stem from bogus behaviors in the kernel/x-server. And some of them stem from hacks to work around the aforementioned. I believe that some of these problems are resolved in internal builds. Capture to me is only interesting when the user accidentally strays from 2 into 1. If you're only worried about 2 and 2's track, ok. We have some changes for it (but they're useless w/o a new xserver). Right, this is purely a bug introduced by the panning code, and basically it's probably capturing too much (capturing is dangerous) http://timeless.justdave.net/mxr-test/os2008/source/microb-engine-1.0.3/ debian/patches/gtkmozembed/gtkmozembed.diff?raw=1 Is the relevant file as shipped with 2008. The evil code can be seen here: http://timeless.justdave.net/mxr-test/garage/source/browser/mozilla/trun k/microb-engine/microb-engine/debian/patches/gtkmozembed/gtkmozembed.dif f#2043 From what I can tell, it doesn't think about scrollbar (tracks) and is very brutish. If you can figure out a way to make it behave sanely, I'll commit it tomorrow :). Offhand, it should probably find out the clientRect (?) bounds for the elemnt it's trying to scroll and if the point is outside those bounds it should fall through to the normal behavior. Perhaps dolske (irc.mozilla.org) or someone else there w/ a device would be able to help (I'm reviewing about 600 bugs right now, and while I'd love to fix this one, I unfortunately need to push those around instead), it's probably ~10 lines of code limited to that one function. Now, it's normally not an issue (nobody scrolls from the shaft are anyway) but when it is difficult to hit the thumb due to it's size (those huge pages you talked about) Indeed, please feel free to create some nice testcases and file a user interface feedback bug against the ui designers explaining that tiny thumbs don't work for big thumbed users. (sorry, I've given up, I'm hoping they listen to customers.) Actually, w/ about:config, whether I'm zoomed to 80% or 240%, the scrollbar thumb size doesn't change, it's about 1/4th the size of any of my fingers (which upon reflection is clearly fairly unreasonable as targets go...). and the fact that the touchscreen is a shotgun instead of a sniper rifle, it confuses the user when scrolling actually goes to the opposite direction than what was intended. Sure, personally I blame the ui designers for this too. They introduced panning which is 180 degrees reversed from scrolling and wow, users get confused when bad things happen. I'm sorry. I'd love to fix this (and I have a fix: disabling panning), and my fix happens to solve a number of other problems too. But really, in order to do this, we'd need to make the scrollbars much more finger friendly (by we, I mean you+ui designers, I want no part in it, as long as whatever is done doesn't make web browser interaction worse than it is in firefox or microb today). Usually you end up multiple pages in the wrong direction, Hrm, panning should never let you go more than 1.5 pages away from where you are ... (clicking repeatedly in the track otoh...) and *poof* went your mental context of what you were doing (something considered harmful to the user experience). This was the issue raised originally, AFAICT. If the browser gets sent a mouse up event and then a mouse down event in the track, you'd have this problem. We're pretty much at the mercy of the x-server. Second guessing it leads to bugs (some of which you're suffering from in the track,
RE: Developing applications in Flash
Alfie wrote: I notice that the Getting Started applet is a full screen Flash application. yep Does anyone know how this is achieved, specifically: - How to add to start menu - How to add to desktop This is basically a distinct application/applet. - How to launch the swf It creates a browser eal instance and then asks the eal to load a url, which happens to be flash. As it happens, the application doesn't depend on flash, and it doesn't ensure that flash is available. If you're going to implement something like this, please ensure the following: 1. your application has a depends: adobe-flashplayer | macromedia-flashplayer 2. you have your own profile directory with a plugins directory containing a symlink to the flash player library 3. provide a desktop file that claims to handle .swf and the mime type Here's a fairly short overview of what it does (written in C++ coding style so I can declare variables as I need them; note that proper release of modules and similar is not implemented in this quick description): /* XXX there is *no* function which lets you get the user's prefered browser, * and in fact there's virtually no way to figure out which ones are available. * we hard code microb-eal. */ #define EAL_BROWSER microb-eal GModule* module = g_module_open(/usr/lib/lib EAL_BROWSER .so.0, G_MODULE_BIND_LAZY); if (g_module_error()) return 0; setenv(MOZ_COMP_PATH_OVERRIDE, .swfplayer, 1); setenv(MOZ_PROFILE_OVERRIDE, FlashPlayer, 1); if (!g_module_symbol(module, g_web_set_env, (gpointer *) web_set_env)) return 0; if (!g_module_symbol(module, g_web_new, (gpointer*) web_get_new)) return 0; web_set_env(); GWeb* web = web_get_new(); if (!web) return 0; /* this code is a copy of code copied from code copied from svn which you can read at * http://timeless.justdave.net/mxr-test/garage/source/browser/browser-ui/t runk/eal-test/src/main.c * however, do note the / between the VendorName and the Version. * This is required by the UserAgent specification, * which is unfortunately not followed by the example above, nor by the people who copied it. */ g_web_set_ua_string(web, SWFPlayer/0.1); GWebEngine* engine = g_web_new_web_engine_window_with_context(web, NULL, FALSE, FALSE); if (!engine) return 0; GObject* widget = g_web_engine_get_engine_widget(engine); gtk_fixed_put(GTK_FIXED(container), GTK_WIDGET(widget), 0, 0); gtk_widget_set_size_request(GTK_WIDGET(widget), 0, 0); gtk_widget_show(GTK_WIDGET(widget)); GObject* notify = g_web_engine_get_engine_notifier(engine); g_signal_connect(notify, G_WEBWIDGET_SIGNAL_FINISHED_LOADING, G_CALLBACK(finished_loading), NULL); g_web_engine_load_url(web, file:///tmp/foo.swf); Note: none of the code to do shutdown is included here. The plugin directory should be /home/user/.swfplayer/plugins - Does it have to be full screen? No, the player happens to hard code this. I'm interested in how one might write applications for OS2008 using Flash alone, in order to make development easier for the new breed of Flex developers! There's actually a browser bug that's catching people unhappy. And given the existence of the tutorial, I'm going to see if I can solve it. However if you get a working standalone player sooner, all the better. Note: it'd be great if you could use rewrite the browser .desktop file in a postinst. For an example, see http://timeless.justdave.net/maemo/microb-pageupdown-0.1/DEBIAN/postinst http://timeless.justdave.net/maemo/microb-pageupdown-0.1/DEBIAN/postrm It should be something like this: PREFFILE=/usr/share/applications/hildon/browser.desktop PREFBACK=$PREFFILE.flash-mimetype OLDPREF='(MimeType.*)application\/x-shockwave-flash;' OLDPREFG='MimeType.*application\/x-shockwave-flash;' NEWPREF='\1application\/x-shockwave-flash-in-browser;' (not tested) Does anyone think it would be possible to develop Flash apps that make use of the Hildon menus, etc? This would presumably need native calls from Flash and I'm not sure this is currently possible. That's a question for people who probably can't comment. Good luck. In the medium/long term it would be great to see Adobe AIR fully supported on OS2008. See http://www.adobe.com/devnet/air. That seems unlikely. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Using the scrollbar on the N810
Richard Booth wrote: I find it particularly awkward at the times when you want to fast scroll down a page. I really wish the N810 had a scroll wheel Eero wrote: I'm not sure whether this helps, but you can scroll the page with long-press of the down key in keyboard. Short press moves just to next link. And if you don't like the link navigation, you can disable it ... http://timeless.justdave.net/maemo/microb-pageupdown-0.1.deb ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Using the scrollbar on the N810
Neil wrote: Bug 2961[1] now opened for what it's worth... I've inadvertently assigned it to the incorrect component (input method framework) when it should be Window Manager (I think) - perhaps someone would be kind enough to reassign? After further investigation this problem is not due to the thumbtrack - simply dragging within the scrollbar region OUTSIDE the thumbtrack will result in the drag events being passed to the main application. I suspect I've often been missing the thumbtrack in the past and thus hitting this bug - ideally the window manager would allow for a few extra pixels above and below the thumbtrack widget when determining if it's been clicked upon! 1. https://bugs.maemo.org/show_bug.cgi?id=2961 Unfortunately, this isn't really how things work. Either a widget asks to capture input, in which case it can get events outside its boundaries, or it doesn't. If a widget does ask to capture events outside its boundaries, it then needs to have some idea about what to do with them. And that's hard (and each group will implement it differently, but poorly) ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Using the scrollbar on the N810
Neil wrote: Hi Josh - I'm guessing your reply concerns my comment about the thumbtrack capturing events slightly above/below and not the bug being discussed (2961)? If so I can understand what you're saying - it would just be nice if the scrollbar were able to allow for slightly mis-directed tocuhscreen hits! Ultimately I guess scrollbars will disappear to be replaced by more finger friendly drag/kinetic scrolling I'm not sure how practical this is for full web browsers. I know everyone says ooh look shiny when they look at iPhone/iPod touch, but the fact is that if you need to be able to allow full integration w/ Google Maps and the rest of Google's applications (note: Apple cheats by shipping their own Google Maps UI), then trying to trap user interactions is a disaster (which is quite visible from the bad compromises we've made for the tablet browser). For any other application that doesn't have to host full applications (i.e. flash embedded into an application has the same problem as a web browser), then finger friendly whatever may be better, but those aren't my problem spaces, and I don't own ui design, so I'm staying away from them. so improving thumbtrack hit detection isn't such a big deal, but fixing the broken click drag event processing certainly is something which needs fixing. Some of our problems relate to trying to counteract some bad kernel/x behaviors (they're fixed in newer internal builds, which means we can remove some of our bad hacks), and some of them stem from trying to implement bad design decissions from the platform (dragging a web application with a finger, see earlier complaints). If you're looking for another case where things don't work correctly, try typing into google reader's main screen using the n810's keyboard. I have tested bug 2961 on a 770 running OS 2007HE and Opera, and it is not present. And on a N810, bug 2961 doesn't appear to be present with other applications such as the RSS Feed Reader (which also supports page drag scrolling) so perhaps bug 2961 is in fact an EAL/MicroB issue? I was pretty sure that the bad dragging/scrolling hacks were in microb-eal, but I can't find them... ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: maemopad exits suddenly when clicking on Save icon
Jakov wrote: I'm using Maemo 4.0 Chinook. After I load maemopad, click on Save icon, this application suddenly exits. The msg in the console is: [sbox-CHINOOK_ARMEL: ~/maemopad-2.1] run-standalone.sh src/maemopad qemu: Unsupported syscall: 264 This is the problem, you need a version of qemu w/ support for more of the arm cpu. Please keep in mind that the ARMEL environment is designed for cross compiling, not for hosting/testing. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: SDL, tearing, X overhead and direct framebuffer gfx
Tobias wrote: Q: Where can I find the sources to the OS2008 SDL? Eero wrote: Where all the other sources are i.e. from the maemo.org repository: http://repository.maemo.org/pool/maemo4.0/free/source/libs/ Or you could search for them: http://timeless.justdave.net/mxr-test/os2008/find?string=sdl or http://mxr.maemo.org/os2008/find?string=sdl if you have http://timeless.justdave.net/maemo/mxr-maemo-org-dns-0.1.deb installed ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: 810 not always obeying redirects from websites
Simon wrote: Just done some more tests and the problem seems to go away on G wireless network and only be present on a B wireless network - so maybe it is related to wireless rather than browser. Has anyone else had these symptoms? I don't have a site to test with setup it is on an internal network. You're going to need to get some tcpdump or wireshark style logs. But it sounds like your scenario is way too flaky for someone else (like me) to want to go off and investigate it. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Fullscreen and flash, new question
Jussi wrote: Josh mentioned .browser already. I just wanted to add a piece of advice: Don't even think about modifying browser settings from your application. Your users will hate you for that. Embedding the mozilla engine is a better solution if you really want total control of the UI. Oh yes. I should have said that, and I'd like to thank Jussi for noting it. Your programs should *never* make changes to these files (a package which clearly only changes a single setting which a user must manually install individually is different). Users absolutely do not expect it, and the results are always bad. I'm usually fairly good about yelling at developers internally who indicate some plan to change browser files (or bookmarks files), and I think I did tell people on this list not to do it in other threads. I'm sorry if anyone got the idea that changing these files was somehow an endorsed or reasonable idea. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Fullscreen and Flash
Carlos wrote: i published the newest issue of my maemo ezine yesterday and had two ideas, wishes and questions :) 1st: Is there a way to force the browser to go Fullscreen without the address field at the bottom of the screen? No. this is a violation of the user interface (and security) model we offer to our users. You can tell users how to turn off the full screen toolbar [app menu]viewshow toolbar[ ] Full screen 2nd: Is it possible to display Flash on the today-screen aka desktop? Yes, theoretically, but it's really not recommended. In order to do this, you'd need to successfully embed the browser (there are a couple of apps that do this, including get started), but keep in mind that if flash is disabled and you try to embed flash, your applet will be on average fairly unhappy (the get started applet sure is). ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: WebKit Incremental Layout
-Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of ext Alp Toker Sent: February 13, 2008 5:01 PM To: David Carson Cc: maemo-developers; [EMAIL PROTECTED] Subject: Re: WebKit Incremental Layout David Carson wrote: the main issue that it tries to address is that on really really slow networks, the user goes to a site and takes a long time to load, and they think the browser has stopped working or is not doing anything and give up and cancel the load. By showing some content, some progress, the user can see that there is activity and will continue to wait for the page to load. Again, this is for slow network connections (ie low bandwidth) If this is the actual problem you want to solve, I would hope there are better ways that we could solve it. I used this on a real mobile device with a slow connection for a couple of weeks to get a feel for it. As a browser developer, the feature at first seemed neat. But as an end user I found that the appearance of styling up to ~10/20 seconds later was annoying, since by that point I'd already started reading the unstyled content and was getting the information I needed. The subsequent CSS load often meant I had to find the newly-styled paragraph and start reading it all over again. Maybe it's subjective, and maybe there's a technical solution to make the transition from unstyled to styled content less distracting. I've pinged some other mobile browser developers to get their thoughts on this. I've heard a lot of people complain about the style jump problem, and I experience it too. It'd be less of an issue if: 1. the user was prevented from accidentally clicking as things moved (this requires disabling input for a couple of seconds before/after a rendering) - and this is almost certainly going to annoy someone, but perhaps not as much as accidentally clicking on the wrong thing. On mobile devices, accidental clicks are very expensive as you lose the current page, have to wait for a new page, and going back can result in you waiting again. 2. somehow enable the user to indicate exactly which element they're looking at (this is impossible with most input methods today - eye scanning being the only exception), plus distinguishing between overlapping elements :) Given that both 1 and 2 are expensive, 1 is hard and 2 is impossible, I think the blocking approach on CSS is better. I think basically that Opera got a press win for this behavior, but overall we'd be better off with everyone refusing to do this form of incremental rendering, as it'd encourage sites to design content where the css is not expensive and is designed to be preloaded+cached for future pages. Note: I'm replying to a cross posted entry to maemo-developers. I would not normally comment to webkit-dev and suspect that they might find this offensive, I'm sorry and do not intend to send any further comments. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Fullscreen and flash, new question
Frédéric wrote: I tried the --full-screen option of the command line, but I still have problems : 1) when I launch the browser with the --full-screen option, during a few seconds, I can see the browser in a normal display mode. After 3-4 seconds, it goes to fullscreen mode. Is there a way to display the browser ONLY when it is in fullscreen mode ? I'm fairly certain this is just a bug (I'm not certain there's actually a report for this in bugs.maemo.org, there should be if there isn't). 2) About the address field at the bottom of the screen, is there a way to force it hidden without any user action ? By example, by changing an option in a file ? ~/.browser Full_Screen=false Norm_Screen=false The browser should not be running while you try to change these (or any other) settings. About the first question, I'm thinking for instance about a C program which run the browser, hide it, wait 4 seconds, and display it again... ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: RE : RE: Fullscreen and flash, new question
Frédéric wrote: ~/.browser Where can I find a documentation of this config file ? The settings are either explanatory or ducible by tweaking browser settings and diffing. On average, the latter. I haven't actually tried those settings, and am merely guessing based on their strange names that they're the right ones. Is there other config files for the Maemo browser ? The browser ui is controlled by that one file. Bookmarks is not part of browser and mostly has its single data file (MyBookmarks.xml ? - which should be owned by user:users, not root:root - in case anyone else decides to be evil and use a package to attempt to do evil things to it). There is a history file for browser (somewhere, strace would find it) The engine (opera, gecko) of course has its own files, and these are documented by the respective engine providers. For examples of evil things that you can do to the browser, there are some packages at: http://timeless.justdave.net/maemo/ Specifically: http://timeless.justdave.net/maemo/microb-pageupdown-0.1.deb http://timeless.justdave.net/maemo/microb-googlesearch-0.1.deb These were generally built by blackbox engineering and google searches and not based on any internal knowledge of any Nokia (or debian packaging!) code. Note that of course any debs I may provide are unsupported, subject to breaking randomly and well open source, so read before using :). specifically the sources to the debs are available as directories that are siblings of the .debs themselves, although you might want to use dpkg-deb to extract the deb to make sure that the version of the deb has the same scripts as the sources, sometimes the debs may be out of date if I'm hacking them. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
Repository corruption checker
I mentioned to ferenc before he left that I had a script that I used to check for corruption of the repositories. The perl script is here: http://timeless.justdave.net/maemo/repository-check.txt And syntax for using it is here: http://timeless.justdave.net/maemo/repository-check.input I kinda forgot where I had put it, and of course, ferenc is gone, so this is more of a general hey, in case people want it. It's obviously quick and dirty, but it did work once or twice for me :). Of course, it assumes you do have a local copy of the repository (timeless.justdave.net does - http://timeless.justdave.net/maemo/repository-maemo-org-dns-0.1.deb ) but that's not strictly necessary, you could w/ very little effort adapt it to operate against the live repository (or create a script which only downloaded the indexes and directory lists and parsed those instead of using stat). ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Maemo's glib version, lobbying for GIO
Philip wrote: Therefore I'm pleased that at the level of glib a team has started working on GIO. http://svn.gnome.org/viewvc/glib/trunk/gio/ GIO not only comes with a streaming API, but also with a standard way for making cancellable asynchronous APIs. http://neutronic.spaces.live.com/blog/cns!6CAF0CCA79C2F340!145.entry The wonderful thing about standards... ... is that there are so many to choose from I can't speak for anyone, however, I don't expect mozilla to instantly add gio. It took a while before someone added gnomevfs, and now that's going to go the way of the dodo (three cheers!) It uses the Asynchronous pattern for this which results in types like GAsyncResult and a type GCancellable. As Tinymail's main developer I would like to start using these types as soon as possible (that's instantly after my first release). I would like to standardise on how future glib GObject based libraries will do async APIs that are cancellable and I would like to standardise on my stream related APIs too. http://arstechnica.com/news.ars/post/20080211-samsung-sued-over-defectiv e-first-gen-blu-ray-players.html Samsung sued over defective first-gen Blu-ray players Jumping to new standards can get you in trouble. To sooner I can do this, the fewer times my API will need to be broken, I don't see how this follows. Someone jumping to maemo1.0 has probably had their impl broken at least yearly (so 4 times?). the fewer major releases I have to do. The less headaches development teams depending on Tinymail, like Modest's team, will have. I'm sure they appreciate fewer headaches (I sure would like to have fewer headaches). If after this current distro Maemo would not get a glib for half a year that includes a GIO, people like me could be blocked for the entire time from delivering APIs like E-mail ones I'm not sure this follows. I'm not a representative of maemo, but if you look at the history, OS releases have been named after years (2005, 2006, 2007, 2008). As a normal developer, I wouldn't expect something faster than that. Looking at product announcements from Nokia, http://www.linuxdevices.com/news/NS8069179684.html Sprint will offer a Mobile WiMAX-enabled version of Nokia's N800 Internet Tablet to North American customers So far most Nokia product releases relating to maemo are either at the beginning of the year (n800) or at the end of the year (n810). Given an announcement, and the fact that the device clearly hasn't surfaced yet, I think it's safe to assume it'd come out closer to the end of the year. This of course doesn't imply that you get a new OS, it might run 2008 (I have no idea). unless I hack-in a standalone version of GIO (sounds ugly to me, but if necessary ...). My questions ... Therefore I wonder about what the plan is for Maemo's glib version. Upgrading glib generally speaking breaks the entire toolchain and scares everyone (for some releases it's attempted and rejected because too many things break). So it should never be done in the middle of a release cycle. The same applies to the kernel, X, gcc, ld, and a number of other core components - this of course excludes bug fixes [gio is not a bug fix]. How soon will Maemo upgrade? (I don't speak for maemo.) NPTL was added for 4.0 http://maemo.org/development/sdks/api_changes_between_maemo_3_2_and_maem o_4_0.html But it was actually a risk item. It could very well have missed even though most people wanted it. What can we do to help speeding up GIO's inclusion into Maemo? My personal believe is that it's unlikely that there's anything you can do here. Would a standalone version be acceptable? I think you should try to find a way to get gio to work on the existing glib and then provide a package which could be installed into existing OS releases. Or yes. Just package it properly and support it. Add it to maemo-extras and version it well My opinion ... .. is that the sooner we have this API in Maemo, the sooner application and- library developers will be synchronised with each other's APIs and IO sharing. I don't think it'll happen that fast anyway. The sooner people develop and test w/ gio, the sooner they can start working out some of the bugs. For example HTML components need to share IO with the E-mail framework (inline attached images in base64 encoding). I don't think this is truly necessary. Are you planning on loading hundreds of images at a time? The VFS needs to share IO with the E-mail framework (Attachment-Save As). I'd rather see fuse implemented. http://arstechnica.com/journals/linux.ars/2007/09/28/gnome-2-22-planning -gio-and-gvfs-proposed-for-inclusion The media player (hi there Valagume developer!) could share IO with the E-mail framework (streaming an attached MP3), or GStreamer could share the IO (a source element that knows about GIO's streams). Streaming an attached mp3 sounds like a really odd thing to do. Are you streaming from a
RE: TestGtkEmbed on N800
Andrew wrote: Maybe i'm going about this wrong. All i really want is to write an application that has the browser shipped with the device embedded in a widget inside it. I thought getting TestGtkEmbed would best the quickest starting point, but maybe not. Does anyone know of any examples that i can compile in scratchbox that does this? Officially we want to encourage people to use the eal apis instead. http://browser.garage.maemo.org/docs/eal/ I'm not sure there are any good or useful eal demos out there (and the one that people might mention should not be mentioned, as it has some major non technical issues). Using the EAL should enable your browser consumer to use whichever browser may be on the maemo tablet. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: TestGtkEmbed on N800
Andrew wrote: You suggest i need a symlink, but i'm still unclear where. The libgtkembedmoz.so is at /usr/lib/ on the device, but ldd says that TestGtkEmbed doesn't depend on it anyway. Using nm on TestGtkEmbed in the armel environment gives B addresses next to all the gtk_moz_embed functions, which suggests to me that the symbols are embedded in the program? Can you tell me where the symlink would be needed? TestGtkEmbed is probably linked against libgtkmozembed.so, a symlink from libgtkembedmoz.so to libgtkmozembed.so would probably make things happy. ./run-mozilla.sh `which ldd` ./TestGtkEmbed ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: TestGtkEmbed on N800
This too can be answered w/ google (as mentioned in one of the links in my previous email, which sorta cross paths with the one to which I'm now replying) or do i need to copy all these files somewhere to get the TestGtkEmbed working? There is one (ibid) https://garage.maemo.org/svn/browser/mozilla/trunk/microb-engi ne/microb- engine/debian/patches/125_MICROB_autoset_grehome_display.diff +export GRE_HOME=/usr/lib/microb-engine But again, the question is what's your goal. Do you want to use the one that ~worked so well~ in scratchbox, or do you want ot use the one that is shipped w/ the device. The choice is yours :). I would like to use the one shipped with the device, to keep any installed files to a minimum. I've patched run-mozilla.sh with the diff file linked above. This sets the GRE_HOME for you so my command line becomes ./run-mozilla.sh ./TestGtkEmbed. I couldn't find the ibid file / llibrary you mention. ??? Ibid is a latin reference. It means something like the same as the previous reference The closest match is a libid3-3.8.so.3 which is already installed on the device. With the above command line i now get Couldn't find GTKMozEmbed symbols, but i guess this has something to do with the ibid file? GtkMozEmbed has a different name in microb, just to be difficult, a symlink would be needed. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: TestGtkEmbed on N800
Does anyone know what tools i can use to track down where these symbols are coming from inside the scratchbox environment? If i can do that then i should be able to work out whats missing on the device. Typically nm and ldd are your friends. That and a debugger which will tell you what's loaded. I'm still confused. Libidl should only be needed by xpidl, which you didn't indicate you were going to run ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: TestGtkEmbed on N800
Andrew Gatt wrote: I've been playing around with the TestGtkEmbed program under the VMWare device downloaded from http://wiki.mozilla.org/Mobile/Build/Maemo_Build_Instructions and i'd like to test it out on my N800. I've copied over TestGtkEmbed and run-mozilla.sh and as before run; Sorry, you copied them over *where*? GRE_HOME=. ./run-mozilla.sh ./TestGtkEmbed But it says Couldn't start XPCOM. Now i'm not really sure what the run-mozilla script is doing but maybe i need to adjust these runtime configurations to get the TestGtkEmbed program to run on the actual device? Anyone had any success with this? ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: extras: promotion interface
Niels wrote: - Packages can not only be promoted, but also demoted if they are breaking things. (With the ability to tell the author why it was demoted) How do you deal with packages that are already installed on people's devices when they're demoted? Personally, if I have bad software installed and it's managed by some central system, I'd like to be notified. Preferably given the option to uninstall it. - All promotions and demotions should be logged to some mailing list to keep the process as transparent as possible. I'm biased, but I think you should consider either a bugzilla.mozilla.org like system or an addons.mozilla.org like system :) We would also need to provide an easier way for developers to upload packages. If we can make a webinterface where developers can add their package to the upload/promotion queue, I think we will make uploading to extras a lot less complicated. (Regular dput uploads would also end up in this same queue) We can add this as a first step, auto-builders would be a next step to add. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: TestGtkEmbed on N800
Andrew wrote: OK so i should have tried harder. :) GRE is the Gecko Runtime Environment, which under scratchbox is setup in the /dist/bin folder of the VMWare device. I now suppose the question is, is there a GRE already in the rootfs somewhere that i need to specify to the GRE_HOME= , This too can be answered w/ google (as mentioned in one of the links in my previous email, which sorta cross paths with the one to which I'm now replying) or do i need to copy all these files somewhere to get the TestGtkEmbed working? There is one (ibid) https://garage.maemo.org/svn/browser/mozilla/trunk/microb-engine/microb- engine/debian/patches/125_MICROB_autoset_grehome_display.diff +export GRE_HOME=/usr/lib/microb-engine But again, the question is what's your goal. Do you want to use the one that ~worked so well~ in scratchbox, or do you want ot use the one that is shipped w/ the device. The choice is yours :). ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: TestGtkEmbed on N800
Andrew wrote: I copied them to /home/user/MyDocs/.documents and chmod'ed them both to 777 and ran them after gaining root access (sudo gainroot). GRE_HOME=. ./run-mozilla.sh ./TestGtkEmbed And you thought this would work because you also copied a GRE to the ~/MyDocs/.documents directory? You'll have to forgive my ignorance, but i suppose this is half the question. I don't know what a GRE is or how to copy one or what it does. I couldn't find any reference to it anywhere and so just tried what was working under scratchbox. If you don't mind, I'd rather you read about it. Running random commands from random web pages on random devices is not a good idea. GRE_HOME is of course an environment variable, and . is of course a reference to the current directory, which is of course relatively empty, because you've copied two files (run-mozilla.sh, and TestGtkEmbed). In short, GRE_HOME specifies *where* the GRE you *want* to use lives. A simple google query for GRE_HOME (or possibly GRE) would do wonders. Here are a few queries thrown in at no extra charge. http://www.google.com/search?q=gre_homeie=utf-8oe=utf-8aq=trls=org.m ozilla:en-US:officialclient=firefox-a . https://garage.maemo.org/svn/browser/mozilla/trunk/microb-engine/microb- engine/debian/patches/125_MICROB_autoset_grehome_display.diff . http://groups.google.com/group/mozilla.dev.builds/tree/browse_frm/month/ 2006-04/1c6a0feb22268562?rnum=1_done=%2Fgroup%2Fmozilla.dev.builds%2Fbr owse_frm%2Fmonth%2F2006-04%3F http://www.google.com/search?q=gre+mozillaie=utf-8oe=utf-8aq=trls=or g.mozilla:en-US:officialclient=firefox-a . http://developer.mozilla.org/en/docs/GRE Note that typically the right thing to do w/ a mozilla build is to rsync dist/bin to whereever you want. I think that's something like: builder:~/mozilla/obj-firefox-i386-pc-linux/dist/bin% rsync -azL . victim:firefox_build builder:~/mozilla/obj-firefox-i386-pc-linux/dist/bin% ssh victim victim:~% cd firefox_build victim:~/firefox_build% #do random stuff here. But, I'm not really sure what your goal is. Do you want to use microb's own GRE, or the one you built? (GRE_HOME of course lets you pick which one to use, but picking a non existent one is not a good idea, as I hope you now understand.) ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: flashplayer plugin doesn't effect in microb
I just installed microb browser from .deb files reside on http://repository.maemo.org/extras/pool/bora one by one. The installion procedure went smoothly and the browser could be properly launched and mozilla based browser for maemo has been set to default homepage. But I found 2 questions: 1. Set engine is missing from menu. You can manually force this by changing a line in ~/.browser to say: hidden=true 2. Couldn't render flashs (.swf), just displayed Flash not supported or similar stuff in the Flash area. I viewed the about:plugins, which shows that Shockwave Flash 9.0 r31 has been installed and the status is Enabled. Are you using ARM or X86? because we don't have flash for x86... ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: explanations of the recent download problems
Ferenc wrote: When apt-get update talks to our repository it actually fetches a package database [1] first. From that file apt-get knows that it supposed to receive 12380 bytes for the pidgin-extprefs_0.7-0nix2_armel.deb package. If you go to the pool's index [2] you can see that the file is actually 12382 bytes long. This could explain the size mismatch error. Wget will not match the file size with a package db, so it will just fetch it without a problem. Question: why the Packages have the wrong size. Well, frankly I don't know. It can only be because our repo manager tool had a hiccup, though it has been working quite well in the past years. Anyways, I try to rebuild that package database and see if that will help. Fwiw there are a number of mismatched files but most of them are pidgin related, I found them while trying to mirror and trouble shooting people's complaints. I ended up writing a script that compared my repo against my index. I think I decided part of the problem might relate to multiple uploads (of the same exact filename) or multiple arches, but that's pure speculation. I can run the script this weekend, but I'm sure ferenc can make an equivalent script... ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Where can I find source codes of following MicroB Browser Packages
On 16/01/2008, Huang Gao [EMAIL PROTECTED] wrote: I am studying MicroB for Bora, and can not find source codes of following packages: tablet-browser-controls_0.1.16-1_armel.deb tablet-browser-default-plugin_0.1.2-1_armel.deb tablet-browser-dialogs_0.1.9-1_armel.deb tablet-browser-ui_0.2.9-1_armel.deb tablet-browser-widgets_0.13.11-1_armel.deb Is there anybody can tell me where I can find them? Daniel Bainton wrote: I believe those are all closed source by Nokia. They're not really a part of MicroB, just the browser UI. Daniel is correct, and this should actually be clear from the package names. A pity really that it isn't a completely free browser... You should be able to build and install microb-refui which should give you TestGtkMozEmbed ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Where can I find source codes of following MicroB Browser Packages
Huang Gao wrote: I have checked out the codes from the SVN several days ago, however its version is too old, and the install package is 0.0.1 and can not be used in my N800 device. I am not sure if this SVN has been updated during these days, in that I can not connect to it any more. It's indeed not for use, the actual browser ui work is done in a private repository. This was a management decision beyond the control of the people who wanted to open source it. BTW, I can not find these packages, even binaries in Chinook repository, and therefore does it mean that Chinook has a different architecture and does not use these packages for UI? Chinook uses tablet-browser-* which is a much newer version of browser-ui (as you noticed). Calling it a different architecture isn't reasonable. The architecture is effectively the same, the package names are slightly different and the code is slightly different, but the general structure / design / interfacing hasn't changed that much. Antonio should not have recommended that directory to you (or anyone). ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Compiling Kryptpad ...
Massimo Mund wrote: I am trying to compile kryptpad for OS2008. Running ./autogen.sh works fine, but ./configure gives me: ... checking for ANSI C header files... (cached) yes checking for pkg-config... /scratchbox/tools/bin/pkg-config checking pkg-config is at least version 0.9.0... yes checking for HILDON... configure: error: Package requirements (hildon-libs = 0.9.50 hildon-fm libossohelp) were not met: No package 'hildon-libs' found No package 'hildon-fm' found No package 'libossohelp' found Consider adjusting the PKG_CONFIG_PATH environment variable if you installed software in a non-standard prefix. Alternatively, you may set the environment variables HILDON_CFLAGS and HILDON_LIBS to avoid the need to call pkg-config. See the pkg-config man page for more details. So, any suggestions? http://timeless.justdave.net/mxr-test/os2008/find?string=hildon.*pc Shows that the library got a new name. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: n810: Light Sensor (?)
Luca De Cicco wrote: Sorry if ask something which has been already discussed... I was wondering if there is some specification about the small thing near the camera which appears to be a Light Sensor. Andrew wrote: Your best bet is to hook up SSH or an xterm or something and poke around the filesystem. I imagine Nokia are going to be closed-mouthed about it for the time being like they were with the FM radio in the N800. Alternatively looking at kernel sources would be a good idea. http://timeless.justdave.net/mxr-test/os2008/search?string=light%20senso rfind=kernel- http://timeless.justdave.net/mxr-test/os2008/search?string=light%20senso rfind=hal- Or if you have this installed: http://timeless.justdave.net/maemo/repository-maemo-org-dns-0.1.deb Then these urls http://mxr.maemo.org/os2008/search?string=light%20sensorfind=kernel-; http://mxr.maemo.org/os2008/search?string=light%20sensorfind=hal-; Of interest is probably this: http://timeless.justdave.net/mxr-test/os2008/source/kernel-source-rx-34- 2.6.21.0/debian/changelog#361 361 * I2C: TSL2563 Light sensor integration time fix 365 -- Yauheni Kaliuta [EMAIL PROTECTED] Tue, 04 Sep 2007 19:34:02 +0300 Which shows that the i2c TSL2563 thing is Nokia related Googling for i2c and TSL2563 should probably get you the rest of the way. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Microb browser and GTK configuration
Tomàs Jiménez Lozano wrote: This means I should build microb browser from scratch but I've read in this list that it's not possible to do it: http://www.gossamer-threads.com/lists/maemo/developers/31766 You're misinterpretting. microb-engine is essentially a library, and libgtkembedmoz or libgtkmozembed whichever is also a library. As long as you have the tablet-browser-ui (which anyone running os2008 has), you can simply package your replacement libraries and drop them in place. The note is more for people who think that they can use microb-engine standalone on an empty scratchbox that doesn't include the Nokia portions, e.g., an intel scratchbox (which as I recall does not include the Nokia ui). As it happens, microb-eal is an even smaller library, so as long as you have enough prerequisites, you should be able to skip building microb-engine. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Creating a simple installer
Michael Stepanov wrote: Thanks for your suggestion, Rafael. I already created the package and specified there all dependencies. But the problem is that Application Manager cannot install dependencies automatically. It just displays exception. Application manager will only handle dependencies if you install from a repository. If you try to install the same .deb using install from file, it will not follow dependencies. In theory, if you create: 1. a deb with proper dependencies 2. a repository and insert the deb 3. a .install file that ensures the necessary repository(ies) are available and instructs application manager to install the package (1) Then when the user loads the .install file (3) it will get the repository (2), look for the deb (1), and get the dependencies. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Build the entire Mozilla based browser for maemo
Yong Zhang wrote: Can someone give some guidelines or point me some links on how to build the Mozilla based browser for maemo from scratch? Which packages from garage.maemo.org/svn/browser/ need to be included to make the final application 'browser'? Any feedback would be greatly appreciated. https://bugs.maemo.org/show_bug.cgi?id=2152#c0 Has better steps. However, you can't build the browser ui, the sources are not published (they got stuck somewhere between management and legal). The best result you can build from scratch is a TestGtkEmbed application. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Build the entire Mozilla based browser for maemo
Brian wrote: Hopefully this is just a temporary issue right? Don't hold your breath. It is not as if someone has killed publication I hope. I'm not sure if I can say. I am actually looking forward to the browser UI. I'm not sure why. Personally, I'm hoping to redesign the ui, as is the ui designer, and it would seem just about everyone else. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Microb browser and GTK configuration
Ross Burton wrote: As far as I know, microb ... can render forms using GTK+ widgets It doesn't use gtk widgets, it emulates them. That said, there should be an xsetting you can change, instead of rebuilding. Xsetting sounds a bit arcane... ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Apache + PHP5 + MySQL/SQLite
Joshua Layne wrote: Graham Cobb has made some SQLite packages available repo: Graham's (cobb.uk.net) repository http://www.cobb.uk.net/apt chinook user package: sqllite This is confusing. http://timeless.justdave.net/mxr-test/chinook/find?string=sql Shows that sqlite3 is a package included as part of chinook itself. Is there some reason to have a second non Nokia version available? http://repository.maemo.org/pool/chinook/free/s/sqlite3/ Index of /pool/chinook/free/s/sqlite3/ Name Last modified Size Description Parent Directory 22.11.2007 13:52 - sqlite3_3.4.1-1osso3.dsc 08.11.2007 15:26 470 sqlite3_3.4.1.orig.tar.gz 08.11.2007 15:26 2217587 sqlite3_3.4.1-1osso3.diff.gz 08.11.2007 15:262231 libsqlite3-0_3.4.1-1osso3_armel.deb 08.11.2007 15:26 177292 libsqlite3-dev_3.4.1-1osso3_armel.deb 08.11.2007 15:26 242354 libsqlite3-0_3.4.1-1osso3_i386.deb 08.11.2007 15:26 208174 libsqlite3-dev_3.4.1-1osso3_i386.deb 08.11.2007 15:26 271178 ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Do you need a maemo LXR-type service?
Juha wrote: there used to be a source code cross-referencing [1] service in maemo.org for maemo sources. For reasons unknown (to me), it's no longer there. maemo.org has too few people and resources :). I've made plans to maintain a replacement, I'm hoping to get it up with branding and a maemo.org dns entry this year (I'm currently syncing garage-all - the index reduced view and index update push will probably have to wait for next week). Would you find such a service useful? What would it need to index? Is it important for you that the service is under maemo.org? I'm just charting out the feeling of the community on this issue and I don't want to belittle any (one?) of the cross referencing services for maemo that are out there already :-) Indexed Debian sources packages: http://timeless.justdave.net/mxr-test/chinook/ http://timeless.justdave.net/mxr-test/sardine/ Index svn repositories: http://timeless.justdave.net/mxr-test/maemo/ http://timeless.justdave.net/mxr-test/garage/ ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: Embed microb
Andrew Gatt wrote: OK so i followed this http://scratchbox.org/wiki/Apophis-r4 under Crocodile heading which allowed me to compile libidl and libidl-dev which are now both installed using dpkg -i. Going back to the libgtkembedmoz svn checkout and running dpkg-buildpackage -rfakeroot throws up another dependency issue with bc. apt-get install bc doesn't work. Anyone know where i can get this from? Alternatively, you could search bugzilla https://bugs.maemo.org/query.cgi?product=Browserresolution=--- Summary /contains/: doc https://bugs.maemo.org/buglist.cgi?product=Browsershort_desc=docshort_ desc_type=allwordssubstrresolution=--- 2152 [EMAIL PROTECTED] ASSIBrowser MicroB e Documentation on how to build MicroB is not correct https://bugs.maemo.org/show_bug.cgi?id=2152 And read the bug. Sorry, the document in question is currently in such a position that I can't figure out how to fix it. A replacement or patch attached to the bug would speed up the process. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
RE: [Fwd: Looking for npapi microb-engine header files]
Jean-Sébastien Légaré wrote on November 28, 2007 10:34 AM To: maemo-developers@maemo.org Subject: Looking for npapi microb-engine header files I would like to port an existing mozilla plugin to the microb browser on the N800 running the latest OS2007. The whitepaper on the microb-engine available at browser.garage.maemo.org doesn't say much about their implementation of NPAPI and NPRuntime. It's standard gecko from a certain time period, but you shouldn't need to ask, NPAPI has functions you can call to ask if features. http://timeless.justdave.net/mxr-test/chinook/find?string=npruntime Where could I get the set of header files, specific to the microb-engine on the n800 that I need to include in my plugin code? NPAPI is designed such that you should not need to get some specific version. Use the version you're using, and the run it. If you write your plugin correctly so that it asks for features then it will just work. I am referring to the header files that we normally find in the gecko-sdk, such as npapi.h and npupp.h. http://timeless.justdave.net/mxr-test/chinook/find?string=npapi Shows that chinook has a package whose source includes the files of interest. Reading through: http://timeless.justdave.net/mxr-test/chinook/source/microb-engine-1.0.3/debian/control It seems that the only likely candidate is 68 Package: microb-engine-dev Also, once I manage to compile the plugin as a shared object, how can I register it in the browser ? e.g. Can I install it with xpi from the browser? http://www.google.com/search?num=100hl=enclient=firefox-arls=org.mozilla%3Aen-US%3Aofficialhs=fr6q=site%3Abrowser.garage.maemo.org+extensionsbtnG=Search http://browser.garage.maemo.org/docs/extension.html How to create an extension package for Mozilla based browser for maemo Clearly shows conversion from xpi to debian package http://browser.garage.maemo.org/news/1/ clearly says: How can I extend the browser? Thanks to the open standards support and a code base shared with Firefox, developers are able to easily port add-ons and extensions available from addons.mozilla.org. Firefox extensions can not be installed directly. http://browser.garage.maemo.org/news/9/ Says: There is a project to help enhance MicroB, called browser-extras which is busy porting and packaging content and extensions to improve your browsing experience as well as reduce the pressure on our in house developers. Thanks to their contributors there for settings it up and working on it. You can examine the packages they produce: http://browser-extras.garage.maemo.org/ At the present .deb is the only official way to install things, and this is a platform requirement (xpinstall was explicitly disabled per platform integration requirements) Can I place it in user's home directory under .mozilla/plugins/ or something similar ? There are actually a couple of choices System plugins are installed into: /usr/lib/browser/plugins with symlinks pointing to them in ~/.mozilla/plugins This is probably the best behavior as it enables hypothetical users to choose which plugins to have enabled w/o forcing multiple installations of the same files. Please please please keep in mind that there can easily be more than one browser on the system. My systems tend to have at least 3: MicroB Opera (I have a 770...) MiniMo Your plug-ins should be written according to NPAPI which means runtime interogation and proper error handling. You should not be crashing just because the user has a different browser (newer, older, or just plain different) hosting your plugin. NPAPI requires that plugins be browser agnostic. ___ maemo-developers mailing list maemo-developers@maemo.org https://lists.maemo.org/mailman/listinfo/maemo-developers
