Re: Remarks on gtk docs
Ed James wrote: Joost, and others, I tried learning to use gtk, gdk, cairo, pango, etc several years ago and was frustrated by the difficulty in getting good docs, sample code, etc. Even worse was finding that "constant change" meant me having to rewrite code fairly often. Note that I'm an "old guy" who has written code for a living since I was a "young guy". But this has been the most difficult venue in which I've tried to work. I feel and share your pain in producing something of quality and lasting value. I know it can be done, but I pretty much work alone, and it's not easy. I switched to writing my own set of "widgets" in C++ which more or less look and act somewhat like Java's AWT, but nowhere near as powerful yet. I've got simple projects like a telnet client working, but I feel like I'm mining gold with a fork. My one big question to this list is (and no disrespect is meant), is there a elist similar to this one dedicated to Xlib programming? This list does have many very talented people, some of whom I'm in awe of. But I'm veering into a different direction and just need pointer towards that direction. ... xorg would be as close as any: xorg mailing list x...@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg ___ gtk-list mailing list gtk-list@gnome.org http://mail.gnome.org/mailman/listinfo/gtk-list
Re: so, is this claim about pango still true? or does nobody actually care?
Russell Shaw wrote: Daniel Kasak wrote: Russell Shaw wrote: ... I'm using a pentium 166 as a dedicated mozilla machine among a few other things. Correction: 266MHz (with 64MB ram) But that is very sluggish because of ram swapping when running mozilla. In a typical gtk window with a dozen widgets in it, dragging the lower right corner to resize the window shows a very sluggish iterative loop of repositioning/resizing all the widgets, on a 2.4GHz AMD athlon. I think the video card is unaccelerated in dual-head mode that i have. However, even if unaccelerated, a well designed system should still be lightning fast. You'd expect this if every individual pixel was retrieved from the X server, alpha-composited, then written back. ___ gtk-list mailing list gtk-list@gnome.org http://mail.gnome.org/mailman/listinfo/gtk-list
Re: so, is this claim about pango still true? or does nobody actually care?
Daniel Kasak wrote: Russell Shaw wrote: I don't care how much tedious rendering pango does; if a gui isn't lightning fast on a 100MHz pentium (or 25MHz 386 for that matter), it's fundamentally broken in either or both design and implementation. Oh come on! A pentium 100? That's what you're using, right? Sure? Really sure? I don't believe you. Computers from that time are starting to pack it in. In fact, computers from that time have probably *finished* packing it in. If anyone has a pentium 100 ( or a 25MGz 386 for that matter ) still around, it's *NOT* to run Gtk2, it's to run a firewall, DNS, etc, and it will not have X at all. I'm using a pentium 166 as a dedicated mozilla machine among a few other things. Everyone else ( bar the occasional minimal-freak ) uses widget toolkits designed to run on today's hardware. God help you when XGL / Glitz arrive :) Well, i write a lot of stuff for embedded systems, and a full windowing graphical widget kit for a 12MHz 8bit cpu using scaleable fonts is entirely feasible. There is absolutely no reason for the sluggishness of graphical toolkits on anything faster than an old 386. Excessive layers of convoluted bloat and poor architectural design is the problem. Rendering of arbitrary unicode glyphs is fast and easy (unlike pango). Old laptops with 1MB video ram should be entirely useable for web browsing with mozilla and gui word processing. ___ gtk-list mailing list gtk-list@gnome.org http://mail.gnome.org/mailman/listinfo/gtk-list
Re: so, is this claim about pango still true? or does nobody actually care?
[EMAIL PROTECTED] wrote: is this still true? does anybody care? is there a way to avoid pango entirely and still get AA fonts inside GTK2? will this ever be fixed before everyone is using h/w acceleration to print button labels? the issue raised here will *kill* ardour dead, and would force us to also have to abandon GTK for Qt (a move I would really, really not want to make). some clarification would help Everyone has something to complain about. Yes, Gtk+ 2.x is much slower than Gtk+ 1.x, but you also have to take into consideration two things. First, there are a ton of new features in 2.x that are very useful such as new widgets. Second, computers are a lot faster now. Look at what he is saying: 40%-60% of a 1.2 GHz processor. Well, first of all, that is a very slow processor, but I don't think that is constant. I am running an Athlon 2500+ and a Sempron 2800+ in my laptop and have never seen any problems with the speed of Gtk+ that were significant. Of course there is always room for improvement, but from what I have been seeing in discussions across the net, there is a big focus on making Gtk+ more efficient. Lastly, I would like to say one thing. Gtk+ 1.x may be faster, but I refuse to use any applications that still use it. Compared to Gtk+ 2.x applications, they look horrible. I can't stand it because, yes they may be using X fonts which are faster, but it also has the blocky/amateur look that comes with that. In conclusion, if speed is a concern, I would recommend just maintaining both the Gtk+ 1.x & 2.x trees because your team has already put a lot of work into both. On the other hand, if you realize that most people are running machines capable of handling a larger library, 2.x is very worth using. (Don't use Qt. You will alienate every Gnome user because Qt looks terrible on Gnome. Take this from someone that cannot even look at KDE because it tears me apart. It's an aesthetic thing...) I hope this helps. And just to note, I'm not trying to start a flame about Gnome vs. KDE ... to each his own even though this is the Gtk list... I don't care how much tedious rendering pango does; if a gui isn't lightning fast on a 100MHz pentium (or 25MHz 386 for that matter), it's fundamentally broken in either or both design and implementation. ___ gtk-list mailing list gtk-list@gnome.org http://mail.gnome.org/mailman/listinfo/gtk-list
Re: How will cairo influence GTK's performance?
Daniel Campos wrote: Pango seems to be quite slow, I agree with that... Clemens Eisserer escribió: Wouldn't it have something to do with client-side rendering the fonts then shipping all those pixels to the server? ___ gtk-list mailing list gtk-list@gnome.org http://mail.gnome.org/mailman/listinfo/gtk-list
Re: newbie question
Globe Trotter wrote: Hi, I have a program which was written in gtk+1.2. What steps do I need to take to convert it to gtk-2? Are there any tools available for such an enterprise? Not really. It's a good excuse and exercise to learn the gtk2 api tho;) Install the devhelp help browser and you can read all the api docs much easier and search with keywords. ___ gtk-list mailing list gtk-list@gnome.org http://mail.gnome.org/mailman/listinfo/gtk-list
Re: gtk-config with GTK+-2.0
Glen Engels wrote: Hello, I'm a new linux user trying to configure/install a pre-written program. I'm getting this error: checking for gtk-config... no checking for GTK - version >= 1.2.0... no *** The gtk-config script installed by GTK could not be found. From this archived post: http://mail.gnome.org/archives/gtk-list/2002-August/msg00202.html It says that gtk-config is no longer used with GTK+-2.0 (which is what I have installed). You'll need to install gtk-1.2. Has someone written a gtk-config script that can be used with GTK+-2.0 for the sake of previously written programs, to be able to install them? Or should I assume that the GTK API has changed between 1.2 and 2.0 such that a program expecting 1.2 would not work anyway, and just give up even attempting to install it? Correct. You can still install it if you have gtk-1.2 installed, which has gtk-config. ___ gtk-list mailing list gtk-list@gnome.org http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Help - Unsubscribe
Bobby King wrote: Help. I need to stop subscribing to the GTK list. That is; I need to be removed from the GTK distribution list. I sent an email with "unsubscribe" as the subject, which works for many groups, but obviously does not for this one. Will someone be kind enough to inform me how to "unsubscribe" from the GTK list? Thanks, Follow the instructions on the link below: ___ gtk-list mailing list gtk-list@gnome.org http://mail.gnome.org/mailman/listinfo/gtk-list ___ gtk-list mailing list gtk-list@gnome.org http://mail.gnome.org/mailman/listinfo/gtk-list
Re: dereferencing
Mustafa Sanver wrote: Hi, Any idea how to solve this dereferencing problem? Thank you in advanced. = ERROR GtkNotebookPagegtsrc.c: In function `set_notebook_tab': gtsrc.c:144: error: dereferencing pointer to incomplete type CODE /* This is an internally used function to set notebook tab widgets. */ void set_notebook_tab(GtkWidget *notebook, gint page_num, GtkWidget *widget) { GtkNotebookPage *page; GtkWidget *notebook_page; page = (GtkNotebookPage*) g_list_nth (GTK_NOTEBOOK (notebook)->children, page_num)->data; notebook_page = page->child; // LINE 144 gtk_widget_ref (notebook_page); gtk_notebook_remove_page (GTK_NOTEBOOK (notebook), page_num); gtk_notebook_insert_page (GTK_NOTEBOOK (notebook), notebook_page, widget, page_num); gtk_widget_unref (notebook_page); } You're probably passing in a bad pointer. ___ gtk-list mailing list gtk-list@gnome.org http://mail.gnome.org/mailman/listinfo/gtk-list
Re: i know is off of topic, but...
Mario Motta wrote: hi all, someone can tell me why this piece of sw compile and works even if it should not ? $ gcc --version gcc (GCC) 3.3.1 (mingw special 20030804-1) $ gcc impossible.c -o impossible == #include int main(int argc, char *argv[]) { int n,lv; printf("enter lv:"); scanf("%d",&lv); int impossible[lv]; // declaring an array without a constant size ? This is a C89 GNU extension and standard in C99. for(n = 0; n < lv; n++) { impossible[n] = n+10; printf("\nimpossible[%d] = %d",n,impossble[n]); fflush(stdout); } return 0; } TIA /mario ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Printing in GTK+
Roger Leigh wrote: "Carl B. Constantine" <[EMAIL PROTECTED]> writes: I have some data I want to print out in my application. How do I do it? To make things a little different, I *might* print to one of those POS type printers (USB) not a regular printer. Pointers? Which classes do I need to use? GTK+ itself doesn't currently offer any printing support, so the question doesn't really involve GTK+ directly. You need to either 1) Open the printer device (e.g. /dev/usb/lp0) and write data to it. 2) Open a pipe to the printer spooler (e.g. lpr -P queuename) and then write the job data and close the connection. By a POS printer, I take that to mean a 40 col receipt printer sort of thing. These are basic devices which are basically ASCII thermal printers plus some additional control codes and possibly some extra bitmap raster graphics mode. For these you just send the control codes inline with the text you're printing to either (1) or (2) above. Your program is entirely responsible for the formatting, layout and cutting etc. For the work I do, I send jobs to a receipt printer via CUPS. After completion of a transaction, the receipt gets printed after a barely noticeable delay (< 0.5s). Your best bet is to use a spooler. This also makes networked operation totally transparent, and is far more flexible. You can use libcups, lpc or lpstat etc. to get a list of all available printers if the user needs to choose one. It's really quite easy to generate and output postscript, which can then be printed to any postscript printer or non postscript printer via ghostscript. Look for the postscript "blue book" and "red book". HP PCL 5 might be an option too. IIRC, there's some new way of printing from X windows directly (Xprt). Stay away from win gdi printers. http://linuxfinances.info/info/printing.html http://members.tripod.com/rpragana/gdiprinters.html http://www-cdf.fnal.gov/offline/PostScript/AdobePS.html http://www-cdf.fnal.gov/offline/PostScript/AdobePS.html#SamplePostScriptfiles ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: learning gtk - callbacks
greg wolski wrote: Hi there, I am trying to learn gtk and I have a hard time to understand signals. This comes from helloworld in tutorial g_signal_connect (G_OBJECT (window), "delete_event", G_CALLBACK (delete_event), NULL); g_signal_connect (G_OBJECT (window), "destroy", G_CALLBACK (destroy), NULL); Question: How does gtk know what functions it should call? I mean for me it basically says: call this or this. Is it about the arguments that are declared in callback functions? This is a little bit confusing because no arguments are indicated in those connections. Thanks for any help, The best way to learn is to google for gobject. A lot of detail on gobject is here: http://www.le-hacker.org/papers/gobject/ There's also previous posts on the gtk lists on tutorials being written. There's also: http://www.gtk.org/tutorial/ It will take a long time to get the hang of it, but it's much easier after you do. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: MDI (not winMDI) frameworks ... again
Tim wrote: Firstly, I agree with all prior posts that M$ has destroyed the concept of MDI by having the window in window approach using a window menu and having maximized sub windows, tiled or cascaded. This type of MDI is what another user called winMDI and I totally agree that it sucks big time. However, MDI is still used (very well I might add) by other applications such as IDE's (KDevelop, Netbeans, Eclipse and even Apples X-Code). Many have suggested that MDI should be the job of the window manager. This cannot work for the time being since no *nix app should ever be tied into one window manager. Maybe someday in the long future when the WM world stabilizes sufficiently, this would work but for the foreseeable future this is an unrealistic ideal. In the meantime, developers are left hanging with putting out apps that have just way too many windows. I've used Gimp on my OSX box and it totally sucks having to first clicking the window to get focus before being able to click the item. If the app ran on a MDI framework, this would not be necessary. I have ton's of apps open at one time and if I did not have MDI for those apps, I'd have too many windows to manage. If you need the effect of lots of sub-windows in a larger window, then each "sub-window" can be implemented as an object derived from a GtkWidget or GtkContainer. The titlebar of each subwindow can have a close/maximize/minimize button. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Creating printable reports
Daniel Kasak wrote: Hi all. I'm a newbie Perl developer, and I'm part-way through porting our database front-ends from Access to Perl / Gtk2 ( running on Linux ). We need a few printable reports that I have to rebuild. I've started out doing them in a Perl CGI script, exporting an HTML 'report' and firing up Mozilla to view / print them. This is less than perfect. Mozilla doesn't support the @page directive, so controlling pagination and headers / footers is a nightmare. What options do I have with GTK / Gnome? Requirements are basically headers & footers and pagination, and of course Perl bindings, but I'll check this out. Any takers? Haven't dealt with text processing for a while but there's a few ways: - generate the reports directly as TEX or postscript (TEX is hard to learn), - generate the reports in a word-processor and embed keywords in it, and save in TEX or postscript format as a canned file, then run sed or C regex lib commands to replace the keywords with your own end-user text. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: passing 3 widgets as parameters to g_signal_connect
Sven Neumann wrote: Hi, Russell Shaw <[EMAIL PROTECTED]> writes: G_CONST_RETURN gchar *oper1,*oper2; oper1 = gtk_entry_get_text(GTK_ENTRY(elem->op1)); oper2 = gtk_entry_get_text(GTK_ENTRY(elem->op2)); Why would you use G_CONST_RETURN here? The right thing to do is to use the const qualifier directly. Yes, not thinking again (too much pasting). ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: passing 3 widgets as parameters to g_signal_connect
Leslie Harlley Watter wrote: Hi Andrei, Thanks for the answer, but it only changes the error. Now I get: warning: assignment discards qualifiers from pointer target type Ah, BTW I'm using gcc -o calc main.c `pkg-config --cflags --libs gtk+-2.0` main.c: In function `on_vezes_clicked': main.c:217: warning: assignment discards qualifiers from pointer target type main.c:218: warning: assignment discards qualifiers from pointer target type - void on_vezes_clicked (GtkButton *button, gpointer user_data) { CALC *elem; (CALC *) elem = user_data; elem=(CALC*)user_data; G_CONST_RETURN gchar *oper1,*oper2; oper1 = gtk_entry_get_text(GTK_ENTRY(elem->op1)); oper2 = gtk_entry_get_text(GTK_ENTRY(elem->op2)); ... ires = iope1 * iope2; gchar result[80]; sprintf(result,"%4d", ires); g_snprintf(result,80,"%4d",ires); /* Here I get A Segfault :( */ gtk_entry_set_text(GTK_ENTRY(elem->res), result); } ) oper1 = gtk_wentry_get_text (GTK_ENTRY (elem->op1)); ) oper2 = gtk_wentry_get_text (GTK_ENTRY (elem->op2)); Cheers, LEslie ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: GLIB / Pango install
Andreas Hagele wrote: Russell, running rpm -i glib3-2.4.2-1.i386.rpm produces: file /usr/lib/libglib-2.0.so.0 from install of glib2-2.4.2-1 conflicts with file from package glib2-2.2.1-1 file /usr/lib/libgmodule-2.0.so.0 from install of glib2-2.4.2-1 conflicts with file from package glib2-2.2.1-1 file /usr/lib/libgobject-2.0.so.0 from install of glib2-2.4.2-1 conflicts with file from package glib2-2.2.1-1 file /usr/lib/libgthread-2.0.so.0 from install of glib2-2.4.2-1 conflicts with file from package glib2-2.2.1-1 running rpm -U glib3-2.4.2-1.i386.rpm produces: error: Failed dependencies: glib2 = 2.2.1 is needed by (installed) glib2-devel-2.2.1-1 so either way does not work. Why not? Drives me spares this thing. I think I'll give up on gimp 2 Andreas I haven't tackled rpms for a long time. Maybe the old packages should be uninstalled first. Doesn't rpm have a force option? ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: GLIB / Pango install
Andreas Hagele wrote: Sven, thanks for your reply. I must say I read the INSTALLs for glib,pango, gtk and gimp several times over. And I went so far that I started completely from scratch. Re-installed RedHat 9 again (with all options on to make sure all the required tools are handy). And I found rpms for glib 2.4.2 and pango. However the problem is still the same. When installing glib 2.4.2 it complains about glib 2.2.1 is required (which by itself is rather weired) but glib 2.2.1 is installed. Running rpm does not create any logfiles to show more details on what is going wrong. But in essance it's the same fault as when running the ./configure for compiling the source for glib. So I got glib2-2.2.1 rpm and tried an 'update' and a 'freshen' option when running rpm, but it refused to run saying glib 2.2.1 is already installed. What is wrong here? Check that pkg-config returns the right results: pkg-config --modversion glib-2.0 ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: GLIB / Pango install
Andreas Hagele wrote: Russell, thanks for that. I just grabed pkgconfig-0.15.0.tar.gz. I'll compile and istall it. How will the Glib/pango use that. Is that just part of the ./configure script? yes. You can use it manually to see what version of library it detects: pkg-config --modversion glib-2.0 ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: GLIB / Pango install
Andreas Hagele wrote: Hi, I'm new to Linux and try to install Gimp 2.0 onto a RedHat 9 installation with Gnome desktop. Gimp needs a new GTK which needs Pango which in turn needs GLIB >= 2.4.0 I can get GLIB installed without any errors but the Pango ./configure can't find it. The configure script fails with "glib.h" not found. I have glib.h in /usr/include/glib-2.0 I also ran /sbin/ldconfig to 'activate' the library. But no avail. Where is glib.h meant to be? And why does the install process not do it right? Below a snippet from the pango config.log file Maybe someone can help me here. Install latest pkg-config and test. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Install GTK+2.4.3 - error required pango and xft
David Stevens wrote: Hi. I have this same problem what had Ori Rosen. During configure GTK+2.4.3 I receive massage: configure: error: Pango 1.2.0 and Xft backend is required for x11 target I check pango and xft: [EMAIL PROTECTED] gtk+-2.4.3]# pkg-config --modversion pango 1.4.0 [EMAIL PROTECTED] gtk+-2.4.3]# pkg-config --modversion xft 2.1.2 That looks like I have this packages but steal there are not recognize. I install glib-2.4.0, atk-1.6.0, pango-1.4.0 with -prefix=/usr Thanks for any suggestion, how solve this problem. How about: pkg-config --modversion pangoxft The error messages can be related to what's in configure.in. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Making an empty TreeView sub-column not take up any space
ERDI Gergo wrote: [Please CC replies to [EMAIL PROTECTED] Hi, I'd like to create a pixbuf sub-column inside a GtkTreeView column that doesn't take up any space when the pixbuf is empty. Currently I'm setting up the view like this: column = gtk_tree_view_column_new (); renderer = gtk_cell_renderer_pixbuf_new (); gtk_tree_view_column_pack_start (column, renderer, false); gtk_tree_view_column_add_attribute (column, renderer, "pixbuf", COLUMN_ICON); renderer = gtk_cell_renderer_text_new (); gtk_tree_view_column_pack_start (column, renderer, true); gtk_tree_view_column_add_attribute (column, renderer, "text", COLUMN_LABEL); gtk_tree_view_append_column (treeview, column); However, as you can see in this screenshot: http://cactus.rulez.org/files/treeview-align.png the same width is allocated for empty pixbufs. How can I avoid that? Maybe you could modify the data before it's displayed with gtk_tree_model_filter_set_modify_func(). ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: GTK+-2.4.3 installation
Ori Rosen wrote: --- Russell Shaw <[EMAIL PROTECTED]> wrote: Ori Rosen wrote: --- Russell Shaw <[EMAIL PROTECTED]> wrote: Ori Rosen wrote: I have installed the most recent versions of glib, pango and atk. When configuring gtk2.4.3, I am getting the error: Pango 1.2.0 and Xft backend is required for X11 target. Just above that, it was able to find pango 1.4.0, so why does it need pango 1.2.0? Make sure pkg-config returns the versions of the libraries you just built. It finds /usr/local/bin/pkg-config, which is the only one I have. See if pkg-config is reading the right .pc files: pkg-config --modversion pango 1.4.0 pkg-config --modversion xft 2.1.2 It found pango 1.4.0 but didn't find xft. Try installing libxft2-dev or something. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: GTK+-2.4.3 installation
Ori Rosen wrote: --- Russell Shaw <[EMAIL PROTECTED]> wrote: Ori Rosen wrote: I have installed the most recent versions of glib, pango and atk. When configuring gtk2.4.3, I am getting the error: Pango 1.2.0 and Xft backend is required for X11 target. Just above that, it was able to find pango 1.4.0, so why does it need pango 1.2.0? Make sure pkg-config returns the versions of the libraries you just built. It finds /usr/local/bin/pkg-config, which is the only one I have. See if pkg-config is reading the right .pc files: pkg-config --modversion pango 1.4.0 pkg-config --modversion xft 2.1.2 ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: GTK+-2.4.3 installation
Ori Rosen wrote: I have installed the most recent versions of glib, pango and atk. When configuring gtk2.4.3, I am getting the error: Pango 1.2.0 and Xft backend is required for X11 target. Just above that, it was able to find pango 1.4.0, so why does it need pango 1.2.0? Make sure pkg-config returns the versions of the libraries you just built. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Building gtk docs
Hi, I upgraded all the xml/xsl/sgml packages i could find on my system. I get an error when building the gtk docs (from cvs): ~/SRC/gtk+/docs: make Making all in tutorial make[1]: Entering directory `/home/russell/SRC/gtk+/docs/tutorial' make[1]: Nothing to be done for `all'. make[1]: Leaving directory `/home/russell/SRC/gtk+/docs/tutorial' Making all in faq make[1]: Entering directory `/home/russell/SRC/gtk+/docs/faq' make[1]: Nothing to be done for `all'. make[1]: Leaving directory `/home/russell/SRC/gtk+/docs/faq' Making all in reference make[1]: Entering directory `/home/russell/SRC/gtk+/docs/reference' Making all in gdk-pixbuf make[2]: Entering directory `/home/russell/SRC/gtk+/docs/reference/gdk-pixbuf' make[2]: Leaving directory `/home/russell/SRC/gtk+/docs/reference/gdk-pixbuf' Making all in gdk make[2]: Entering directory `/home/russell/SRC/gtk+/docs/reference/gdk' make[2]: Leaving directory `/home/russell/SRC/gtk+/docs/reference/gdk' Making all in gtk make[2]: Entering directory `/home/russell/SRC/gtk+/docs/reference/gtk' *** Building HTML *** rm -rf ./html mkdir ./html cd ./html && gtkdoc-mkhtml gtk ../gtk-docs.sgml /usr/bin/gtkdoc-mkhtml: line 45: 5605 Killed /usr/bin/xsltproc --nonet --xinclude --stringparam gtkdoc.bookname $module --stringparam gtkdoc.version "1.2" $gtkdocdir/gtk-doc.xsl $document make[2]: *** [html-build.stamp] Error 137 make[2]: Leaving directory `/home/russell/SRC/gtk+/docs/reference/gtk' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/home/russell/SRC/gtk+/docs/reference' make: *** [all-recursive] Error 1 ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: GLists
zze-DEPREZ Olivier FTRD/DMI/SOP wrote: Hello again, Do we need to create a new Glist (or GSList) each time it is called in a function ? Not if it is global. It seems that a call to g_list_nth_data(), even without changing indexes with g_list_next or else, into separate functions causes a segmentation fault. There should be at least n elements in the list. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Building glib docs
Sven Neumann wrote: Hi, Russell Shaw <[EMAIL PROTECTED]> writes: When building glib docs from cvs, i get: ... ../glib-docs.sgml:69: warning: failed to load external entity "http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd"; If your Docbook XML setup was complete (this includes properly setup XML catalogs), then nothing would have to be loaded over the network. From what I heard quite some modern distros finally come with a proper setup for processing XML documents. Debian testing and unstable work fine and I've been told that Mandrake 10 and the latest Fedora releases got it right as well. Ok, it works now after upgrading various xml packages. Is there something wrong with some versions of /bin/sed? Each of these steps takes 5-10+mins: ... Writing glib-Arrays.html for refentry(glib-Arrays) Writing glib-Pointer-Arrays.html for refentry(glib-Pointer-Arrays) Writing glib-Byte-Arrays.html for refentry(glib-Byte-Arrays) ... ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Building glib docs
Hi, When building glib docs from cvs, i get: touch tmpl-build.stamp *** Building XML *** cd . && \ gtkdoc-mkdb --module=glib --source-dir=../../.. --output-format=xml --sgml-mode --output-format=xml --ignore-files=trio 100% symbol docs coverage (1264 symbols documented, 6 not documented) See glib-undocumented.txt for a list of missing docs. The doc coverage percentage doesn't include intro sections. touch sgml-build.stamp *** Building HTML *** rm -rf ./html mkdir ./html cd ./html && gtkdoc-mkhtml glib ../glib-docs.sgml I/O error : Attempt to load network entity http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd ../glib-docs.sgml:69: warning: failed to load external entity "http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd"; I can get "http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd"; in a web browser. After that, i get tons of errors like: ../xml/macros.xml:219: parser error : Entity 'nbsp' not defined a : ^ ../xml/macros.xml:222: parser error : Entity 'nbsp' not defined b : ^ ../xml/macros.xml:249: parser error : Entity 'nbsp' not defined a : ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: custom widget embedding stock widget
mj s wrote: I am attempting to build a custom widget based off of the GtkDial example in the 2.0 tutorial. I wanted to add a vertical progressbar which would float above the dial widget until the user button_releases the mouse. I cannot find any infomation about embedding this stock Progress widget inside of my own. I thought that all I really need to do is create a progressbar instantiation within the custom widget and find a way to paint the progressbar's output on top of mine. Are there any examples of something like this in the CVS that I could look at or perhaps some documentation? Thank you. What do you mean "float" the widget? If you want it packed vertically, derive your custom widget from GtkVbox. Then you can pack that extra widget into your own widget with a vbox cast. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: On Click event for an image
Chris W wrote: I am thinking that the best way to get the look I want for the calculator simulator I want to create is to have one graphic for keyboard. So I need to respond to clicks and then know where on the image they clicked so I can then figure out what button is at that point.How do you do that? Read the button-press coordinates from a GtkDrawingArea that has a picture of the keypad on it. You'll need to manually define where the hot areas for each button lie. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: glib-mkenums: my: command not found
Erik Lorimor wrote: I've tried to build several things and am having this error: /usr/local/bin/glib-mkenums: my: command not found Current program I am trying to build is ATK 1.7 I've built glib 2.4.0, 2.4.1, 2.4.2 on Solaris 9, sparc all with the same results. For ATK my configure results and build output: ... /usr/local/bin/glib-mkenums: my: command not found /usr/local/bin/glib-mkenums: sub: command not found /usr/local/bin/glib-mkenums: my: command not found /usr/local/bin/glib-mkenums: my: command not found /usr/local/bin/glib-mkenums: line 20: syntax error near unexpected token `(s' /usr/local/bin/glib-mkenums: line 20: `for $opt (split /\s*,\s*/, $opts) {' gmake[2]: *** [s-enum-types-h] Error 2 gmake[2]: Leaving directory `/skipps/downloads/atk-1.7.0/atk' gmake[1]: *** [all-recursive] Error 1 gmake[1]: Leaving directory `/skipps/downloads/atk-1.7.0' gmake: *** [all] Error 2 If you need any additional information or could provide any insite, I would appreciate it! Are you using the latest pkg-config? Is perl installed? glib-mkenums is a perl script and it seems like something non-perl is reading it. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Installing GTK/Glib for CHEOPS
Jon Langel wrote: > Guys, > > I'm trying to install CHEOPS and am installing GTK to > support it. > > When I launch the CHEOPS install I get an "unsatisfied > libgtk.so.1 and libglib.so.1 dependency message. > > I've installed GTK 1.2.0 and 2.0. CHEOPS says it is > 'tested' with the 1.1.x versions. > > I'm a LINUX newby, so compared to you guys, will miss > the obvious. Can you tell me what I need to get > going? If necessary, can the 1.1.x versions of GTK > still be obtained? ls /usr/lib: ... /usr/lib/libgtk-1.2.so.0 /usr/lib/libgtk-1.2.so.0.9.1 /usr/lib/libgtk-x11-2.0.a /usr/lib/libgtk-x11-2.0.la /usr/lib/libgtk-x11-2.0.so /usr/lib/libgtk-x11-2.0.so.0 /usr/lib/libgtk-x11-2.0.so.0.400.2 /usr/lib/libgtk.a /usr/lib/libgtk.la /usr/lib/libgtk.so ... If you don't have them, try reinstalling. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Problems with building GTK+ 2.4.3
Michael B. Trausch wrote: On Tue, 15 Jun 2004, Russell Shaw wrote: pkg-config --modversion xft (i have 2.1.2 on debian) You should have something like libfreetype6-dev on the system. Ugh. The problem this time was that I didn't have /usr/X11R6/lib/pkgconfig in $PKG_CONFIG_PATH... My head is exploding with all the variables I hafta set to make this stuff work... I don't really use packages on this system, by the way, I am (mostly) building everything from scratch, so all the headers are already on the system. FT2 I had to download to successfully build the sources from X.org, so it's here... *shrugs*. I'm attempting another build here, now that 'make' is running... Okay. So it finally built successfully... *sigh* That took way too long to do. However, I had to change one file to make this work I don't recall exactly how to use 'diff' for sure at the moment, and I don't know how many files would need changed to make things work with the latest FreeType, however, on this system, it wouldn't build unless the line that read: #include ... was commented out and replaced with: #include #include FT_FREETYPE_H Seems fairly stupid to me, because I don't see why you'd want to do that. Anyway, the file that I had to do that in was pango-ot.h from the Pango header files that were installed in /usr/include/pango-1.0/pango when I re-installed that. Someone who has access to Pango's centralized source base may want to make that change in their code and retar the distro... Just an idea. I was doing the same thing today putting those lines in to try building gtk 2.2.4 with the newer libfreetype6-dev (2.1.7 instead of an older one like 2.0.9). Are you sure you were building gtk 2.4.3? With the newer libfreetype6-dev (2.1.7), gtk 2.4.3 build doesn't need those header file mods. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Problems with building GTK+ 2.4.3
Michael B. Trausch wrote: On Tue, 15 Jun 2004, Russell Shaw wrote: Make sure pkg-config can find its .pc files in /usr/local/lib/pkgconfig. Add this path in the shell: PKG_CONFIG_PATH="/usr/local/lib/pkgconfig: /usr/lib/pkgconfig" That had been done already... *sigh* I probably should've mentioned that. That was one of the first things that I'd come across reading someone else's other scenerio. I've since then decided to re-try this whole thing yet again. So I rebuilt everything, thinking that maybe there was a problem using /usr/local as my prefix -- so I changed that to /usr and rebuilt everything. Now, GTK+ 2.4.3 doesn't get past ./configure, complaining: checking For sufficiently new FreeType (at least 2.0.1)... yes configure: error: Xft version 2 is required for x11 target [EMAIL PROTECTED]:~/gtk-stuff/gtk+-2.4.3$ Xft version 2 is on the system - and in fact, it didn't die before from this error! What in the world is wrong with this? It seems that the X Windowing System should've been the hardest thing on this system to compile, and GTK+ is quickly proving that wrong... pkg-config --modversion xft (i have 2.1.2 on debian) You should have something like libfreetype6-dev on the system. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Problems with building GTK+ 2.4.3
Michael B. Trausch wrote: Hello there, I've recently obtained the GTK+ library as well as Pango, ATK, and GLib. I was able to successfully build Pango, ATK, and GLib, and they've installed just fine. However, GTK+ absolutely *refuses* to build, and I can't quite make any sense of it, because everything seems to be in order for it *to* build. The system in question has the following specs to it: Linux Kernel 2.4.20 glibc 2.3.1 gcc 3.2.2 X.org X11R6.7.0 FreeType 2.1.8 (IIRC, anyway, can't find the exact thing at the moment) Latest Stable Pango, ATK, GLib. The problem that I seem to be experiencing is this: /bin/sh ../../libtool --mode=link gcc -g -O2 -Wall -o gtk-demo appwindow.o button_box.o changedisplay.o colorsel.o dialog.o drawingarea.o editable_cells.o entry_completion.o expander.o hypertext.o images.o list_store.o menus.o panes.o pixbufs.o sizegroup.o stock_browser.o textview.o tree_store.o ui_manager.o main.o ../../gdk-pixbuf/libgdk_pixbuf-2.0.la ../../gdk/libgdk-x11-2.0.la ../../gtk/libgtk-x11-2.0.la gcc -g -O2 -Wall -o .libs/gtk-demo appwindow.o button_box.o changedisplay.o colorsel.o dialog.o drawingarea.o editable_cells.o entry_completion.o expander.o hypertext.o images.o list_store.o menus.o panes.o pixbufs.o sizegroup.o stock_browser.o textview.o tree_store.o ui_manager.o main.o ../../gdk-pixbuf/.libs/libgdk_pixbuf-2.0.so -L/usr/local/lib ../../gdk/.libs/libgdk-x11-2.0.so -L/usr/X11R6/lib ../../gtk/.libs/libgtk-x11-2.0.so /home/fd0man/gtk-stuff/gtk+-2.4.3/gdk/.libs/libgdk-x11-2.0.so -lXrandr -lXinerama -lXext -lXft /usr/lib/libfreetype.so -lz -lXrender -lfontconfig /usr/lib/libpangoxft-1.0.so /usr/lib/libpangox-1.0.so /usr/lib/libpango-1.0.so /home/fd0man/gtk-stuff/gtk+-2.4.3/gdk-pixbuf/.libs/libgdk_pixbuf-2.0.so -lX11 /usr/local/lib/libpangoxft-1.0.so /usr/local/lib/libpangox-1.0.so /usr/local/lib/libpango-1.0.so /usr/local/lib/libatk-1.0.so /usr/local/lib/libgobject-2.0.so /usr/local/lib/libgmodule-2.0.so -ldl /usr/local/lib/libglib-2.0.so -lm -Wl,--rpath -Wl,/usr/local/lib ../../gtk/.libs/libgtk-x11-2.0.so: undefined reference to `pango_find_base_dir' collect2: ld returned 1 exit status What I've done with attempting to make this build is as follows: The first time I ran 'make' after './configure', I was greeted with an issue stating that I didn't have libgobjects.2.somethingoranother.la in /usr/lib. Well, I knew that much, because everything had been installed in /usr/local/lib. The solution that seemed to permit that to work was symlinking from the original files in /usr/local/lib to /usr/lib, and that seemed to permit the build to go on, until... Now. I have this crazy message stating that there is an undefined reference to 'pango_find_base_dir', however, that's kinda mind-boggling. I checked /usr/local/lib/libpango-1.0.so and saw that it did indeed have the function requested: [EMAIL PROTECTED]:~/gtk-stuff/gtk+-2.4.3$ nm /usr/local/lib/libpango-1.0.so |grep find_base 0001d4ec T pango_find_base_dir So I'm not quite understanding where the compiler sees this as non-existant, when it very plainly exists right there in front of my face. Anyone who can help with this, would be very appreciated, and if there is any more information that is needed to help solve this, please feel free to ask me and I will try to provide it. I just want to get this all working :-). Make sure pkg-config can find its .pc files in /usr/local/lib/pkgconfig. Add this path in the shell: PKG_CONFIG_PATH="/usr/local/lib/pkgconfig: /usr/lib/pkgconfig" ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Howto build Gtk 2.4.2
[EMAIL PROTECTED] wrote: You might find it a lot easier and more productive to simply upgrade to the recently released Fedora Core 2 - that includes glib/gtk 2.4.1 as Thank you for your siggestion. The easy path is not always the best one :-) This is not what do I need. I want to build gtk-2.4.2 from scratch. Can anybody help me ? TIA http://developer.gnome.org/doc/API/2.0/gtk/gtk-building.html ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Tracing errors
Jean BrÃfort wrote: Le dim 13/06/2004 Ã 07:07, Russell Shaw a Ãcrit : Hi, I just compiled gtk and now when i run a gtk prog such as gnumeric, i get: ** (gnumeric:9135): WARNING **: Cannot open font file for font Verdana 10 ** (gnome_segv:9136): WARNING **: Cannot open font file for font Verdana 10 and the prog exits. How do i find the specific missing package without resorting to gdb? fc-list shows i have: Verdana:style=Bold Verdana:style=Bold Italic Verdana:style=Italic Verdana:style=Regular I have gtk/glib/pango/atk installed in /usr/local/lib. I ran ldconfig with: /usr/X11R6/lib /usr/local/lib in /etc/ld.so.conf (OS is debian) Do you get these errors with all gtk apps? Which versions do you use? Hi, I don't get the error with gtkfontsel, or gimp or abiword. I do get the error with gnumeric and devhelp (Gnome devhelp 0.8.1). Did you compile everything yourself? Only gtk/glib/pango/atk. strace devhelp: ... access("/usr/local/lib/pango/1.4.0/modules/pango-basic-fc.so", F_OK) = 0 open("/usr/local/lib/pango/1.4.0/modules/pango-basic-fc.so", O_RDONLY) = 32 read(32, "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0p\f\0\000"..., 512) = 512 fstat64(32, {st_mode=S_IFREG|0755, st_size=70374, ...}) = 0 old_mmap(NULL, 10236, PROT_READ|PROT_EXEC, MAP_PRIVATE, 32, 0) = 0x41006000 old_mmap(0x41008000, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED, 32, 0x1000) = 0x41008000 close(32) = 0 getpid()= 11764 write(2, "\n** (devhelp:11764): WARNING **:"..., 75 ** (devhelp:11764): WARNING **: Cannot open font file for font Verdana 10 ) = 75 open("/usr/share/fonts/truetype/Verdana.ttf", O_RDONLY) = 32 fcntl64(32, F_SETFD, FD_CLOEXEC)= 0 fstat64(32, {st_mode=S_IFREG|0644, st_size=139640, ...}) = 0 old_mmap(NULL, 139640, PROT_READ, MAP_PRIVATE, 32, 0) = 0x418e3000 close(32) = 0 brk(0) = 0x8432000 brk(0x8453000) = 0x8453000 brk(0) = 0x8453000 brk(0) = 0x8453000 brk(0x8451000) = 0x8451000 brk(0) = 0x8451000 --- SIGSEGV (Segmentation fault) --- --- SIGSEGV (Segmentation fault) --- --- SIGSEGV (Segmentation fault) --- ... --- SIGSEGV (Segmentation fault) --- +++ killed by SIGSEGV +++ strace gnumeric: ... open("/proc/meminfo", O_RDONLY) = 15 fstat64(15, {st_mode=S_IFREG|0444, st_size=0, ...}) = 0 old_mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x4001c000 read(15, "total:used:free:"..., 4096) = 522 close(15) = 0 munmap(0x4001c000, 4096)= 0 getpid()= 11769 write(2, "\n** (gnumeric:11769): WARNING **"..., 76 ** (gnumeric:11769): WARNING **: Cannot open font file for font Verdana 10 ) = 76 brk(0) = 0x8469000 brk(0x848a000) = 0x848a000 open("/usr/share/fonts/truetype/Verdana.ttf", O_RDONLY) = 15 fcntl64(15, F_SETFD, FD_CLOEXEC)= 0 fstat64(15, {st_mode=S_IFREG|0644, st_size=139640, ...}) = 0 old_mmap(NULL, 139640, PROT_READ, MAP_PRIVATE, 15, 0) = 0x411f1000 close(15) = 0 --- SIGSEGV (Segmentation fault) --- write(3, "\232\21\3\0\6\0\200\1,\0\0\0\33\1\2\0\0\0\0\0", 20) = 20 write(3, " \21\2\0\0\0\0\0", 8) = 8 write(3, "+\21\1\0", 4) = 4 read(3, "\1\0P\0\0\0\0\0\23\0\0\1P1)\0\0\0\0\0\1\0\0\0\37\0\0\0"..., 32) = 32 fork() = 11770 waitpid(11770, ** (gnome_segv:11770): WARNING **: Cannot open font file for font Verdana 10 [WIFSIGNALED(s) && WTERMSIG(s) == SIGSEGV], 0) = 11770 --- SIGCHLD (Child exited) --- sigreturn() = ? (mask now [SEGV RTMIN]) _exit(1)= ? ls -l /usr/share/fonts/truetype/Verdana.ttf -rw-r--r--1 root root 139640 Dec 12 1998 /usr/share/fonts/truetype/Verdana.ttf I have fontconfig installed, and cat /etc/fonts/fonts.conf: ... /usr/share/fonts /usr/X11R6/lib/X11/fonts/Type1 /usr/local/share/fonts ~/.fonts ... ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Tracing errors
Hi, I just compiled gtk and now when i run a gtk prog such as gnumeric, i get: ** (gnumeric:9135): WARNING **: Cannot open font file for font Verdana 10 ** (gnome_segv:9136): WARNING **: Cannot open font file for font Verdana 10 and the prog exits. How do i find the specific missing package without resorting to gdb? fc-list shows i have: Verdana:style=Bold Verdana:style=Bold Italic Verdana:style=Italic Verdana:style=Regular I have gtk/glib/pango/atk installed in /usr/local/lib. I ran ldconfig with: /usr/X11R6/lib /usr/local/lib in /etc/ld.so.conf (OS is debian) ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: static linking of gtk+ 2.4.1 and programs
Jean BrÃfort wrote: Le ven 11/06/2004 Ã 06:39, Mirco Mueller a Ãcrit : Hello there everybody! After reading the web and one straight day/night of trying out stuff, I cannot figure out why static linking doesn't work. According to the DirectFB-project this seem to be possible (http://www.directfb.org/documentation/GTK_Embedded/). I got those libs: atk-1.6.0 fontconfig-2.2.2 freetype-2.1.8 glib-2.4.0 gtk+-2.4.1 jpeg-6b libpng-1.2.5 pango-1.4.0 tiff-v3.6.1 zlib-1.2.1 All (if possible) where compiled with --enable-static. After "make install" I really have all those *.a files in /lib Now trying to compile test-static.c (see bottom of eMail) with the command... gcc `pkg-config --cflags gtk+-2.0` -L/home/mirco/test-system/lib -lgtk-x11-2.0 test-static.c -o test-static ... works as expected. The resulting binary test-static compiles and runs. If I try the same with static linking like... gcc `pkg-config --cflags gtk+-2.0` -static -L/home/mirco/test-system/lib -Wl,-Bstatic -lgtk-x11-2.0 test-static.c -o test-static ... I get these errors: >> >>/tmp/ccuoq9YZ.o(.text+0xe): In function `delete_event': >>: undefined reference to `g_print' You need to statically link many other libs. See my posts around: http://mail.gnome.org/archives/gtk-app-devel-list/2003-October/msg2.html I did a mixture of static and dynamic libs with: gcc -o shapegen main.o gui.o xmalloc.o object.o menu.o dialog.o -Wl,-Bstatic, -lgtk-x11-2.0,-lgdk-x11-2.0,-latk-1.0,-lgdk_pixbuf-2.0,-ltiff,-lpng, -lpangoxft-1.0,-lpangox-1.0,-lpango-1.0,-lgmodule-2.0,-lgobject-2.0, -lglib-2.0, -Bdynamic,-lfreetype,-lfontconfig,-ljpeg,-ldl, -L/usr/X11R6/lib,-lXinerama,-lXi,-lXft,-lm,-lc Whenever you get an undefined reference, run a small script using nm to find which system lib has the symbol, then include it in your next link. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Notebook and Scrolled windows: bug?
Bob McCallister wrote: If I run this program, I can only scroll on the first page of the notebook. In this example I used textviews, but the same problem occurs with treeviews. I've only been working with gtk (pygtk) for a few months, so I may be missing something, but this seems pretty straightforward. Version of gtk is gtk+-2.4.0 on Solaris 8. Compiled with: gcc notebook.c -o notebook `pkg-config --cflags --libs gtk+-2.0` ... I ran the program and can scroll both pages (gtk 2.2.4 on debian). ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Calendar app
Hi, Is there a gtk calendar app where you can enter times and dates of meetings etc in to a graphical display of days of the month, and sound an alarm or send an email when the date/time is reached? Any other gtk apps for keeping track of appointments and things? ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: dead gtk.org?
Sjoerd Linders wrote: I cann't reach it also, so I don't no. I've had no luck getting to http://www.gtk.org for the last couple of days. Is it dead? Thanks, Alan Blount I can get it ok. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Gtk command line console widget
MAIDA, JAMES C. (JIM) (JSC-SF) (NASA) wrote: I am looking for a command line console for gtk. To do what commands? An xterm clone as a widget that can be embedded into a gtk app? > Something that will work on both Linux and Windows. At the very least, I would like to have a text input window (not just a single line). I am very pleased with gtk so far. It is conceptually and implementation-wise much cleaner than MFC etc. Jim Maida ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: How to set background/foreground on button, entry, or label?
Tony Denault wrote: On Thu, 4 Mar 2004, Owen Taylor wrote: On Wed, 2004-03-03 at 20:39, Tony Denault wrote: Hi All, After doing some research and hacking, I wrote some code to set the foreground / backgroud color of a button, entry, & label. This is still a confusing part of GTK 2.x., an internet search on this question resulted in lots of partial answers, vague description on how it should work, but no clear answer that work across all widgets. Have you seen: http://ometer.com/gtk-colors.html Yes, it was one of the references I came acoss while researching this subject. It indicated using gtk_widget_modify_fg/bg(), but I only got the fg to change for button&label. Could not get the entry text, or background colors for label, entry, button to change. As I stated before, if there is a better technique for gtk 2.0, please modify the code and post. gtk_label directly renders onto the parent (it doesn't have its own window). You can change the text foreground and background colour with pango, but the whole background of the label can be changed by holding it in a gtk_event_box, and changing the fg colour of that (it derives from gtk_widget). Looking in the header file may tell you if a widget has its own window. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: gtk+ compliation problem
adam wrote: Hi, I am trying to compile a CVS version of the gnome tree for development alongside my Debian copy. I want to install the tree in /opt I have set these environment variables: BASE="/opt" GCCOPTS="-g" export CC="gcc $GCCOPTS" export CXX="g++ $GCCOPTS" export PATH="$BASE/bin:$PATH" export LD_LIBRARY_PATH="$BASE/lib:$LD_LIBRARY_PATH" export LDFLAGS="-L$BASE/lib $LDFLAGS" export LD_WARN=yes export LIBS="-L$BASE/lib $LIBS" export CPPFLAGS="-I$BASE/include $CPPFLAGS" export CFLAGS="-I$BASE/include $CFLAGS" export CXXFLAGS="-I$BASE/include $CXXFLAGS" export PKG_CONFIG_PATH="$BASE/lib/pkgconfig:$PKG_CONFIG_PATH" export LIBS="-L$BASE/lib $LIBS" For the glib, atk and pango modules this has worked. ./autogen.sh --prefix=/opt --enable-gtk-doc; make; make install; When doing the same for gtk+ there is a linking error: /bin/sh ../libtool --mode=link gcc -g -I/opt/include -Wall -L/opt/lib -o gtk-query-immodules-2.0 queryimmodules.o libgtk-x11-2.0.la ../gdk-pixbuf/libgdk_pixbuf-2.0.la ../gdk/libgdk-x11-2.0.la -L/opt/lib -L/opt/lib gcc -g -I/opt/include -Wall -o .libs/gtk-query-immodules-2.0 queryimmodules.o -L/opt/lib ./.libs/libgtk-x11-2.0.so /home/adam/devel/cvs/gtk+/gdk/.libs/libgdk-x11-2.0.so -L/usr/X11R6/lib /opt/lib/libatk-1.0.so /usr/lib/libgobject-2.0.so /usr/lib/libgmodule-2.0.so /usr/lib/libglib-2.0.so ../gdk-pixbuf/.libs/libgdk_pixbuf-2.0.so ../gdk/.libs/libgdk-x11-2.0.so -lXinerama -lXext /usr/lib/libXft.so /usr/lib/libfreetype.so -lz /usr/lib/libXrender.so -lfontconfig -lX11 /opt/lib/libpangoxft-1.0.so /opt/lib/libpangox-1.0.so /opt/lib/libpango-1.0.so /home/adam/devel/cvs/gtk+/gdk-pixbuf/.libs/libgdk_pixbuf-2.0.so /opt/lib/libgmodule-2.0.so -ldl /opt/lib/libgobject-2.0.so /opt/lib/libglib-2.0.so -lm -Wl,--rpath -Wl,/opt/lib /home/adam/devel/cvs/gtk+/gdk/.libs/libgdk-x11-2.0.so: undefined reference to `g_unsetenv' ./.libs/libgtk-x11-2.0.so: undefined reference to `g_completion_complete_utf8' ./.libs/libgtk-x11-2.0.so: undefined reference to `g_markup_vprintf_escaped' /opt/lib/libpango-1.0.so: undefined reference to `g_unichar_get_mirror_char' In this command there are both my debian glib libraries (/usr/lib/libgobject-2.0.so /usr/lib/libgmodule-2.0.so /usr/lib/libglib-2.0.so) and the new CVS libraries (/opt/lib/libgmodule-2.0.so -ldl /opt/lib/libgobject-2.0.so /opt/lib/libglib-2.0.so). If i remove the debian libraries and run this command it compiles fine - but since i there are a load of modules to compile this wont do. Manually edit out references to existing library objects that are referred to in any of the newly generated .la (libtool) files. It seems to be the bugginess of libtool. Anyone know a decent bash-script debugger? ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: restricting data entry (filters?)
Carl B. Constantine wrote: I posted a note a while back about adding an "input filter" to a GtkEntry widget. I recieved a couple of responces, but I don't think they are what I need. I want the user to only be able to type numbers in a GtkEntry field. I don't want to have to check the value after the fact. One person came up with declaring the field cast as an INT, that may work, but I have no idea how to hook that into Glade (which is generating my source at present) and thus into my app. Does anyone have other ideas on how to do what I want? Thanks. You could connect a callback to the gtk_entry key-press-event that monitors the typing and blocks any invalid characters with gtk_entry_set_text(); (haven't tried it) ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: gtk+-2.2.4 make fails
[EMAIL PROTECTED] wrote: Is freetype.h on the system and in the search path? The other errors are probably caused by the following implicit declarations. I _knew_ I was asking for a smack in the chops with the bleeding obvious. Thanks for wielding the cluebat :) The final solution was to copy the whole of freetype2's include/freetype into X11R6's include directory. Apparently linking it into /usr/local/include just isn't good enough. It's now complaining about conflicts between 2 versions of libexpat (it wants libexpat.so.0, while Xft used .so.1), but I should have that fixed with a little time and some judicious recompiling... When you have libs already on your system and you compile a new setup that has other versions of the same libraries, libtool can get confused. Beware. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: gtk+-2.2.4 make fails
[EMAIL PROTECTED] wrote: Apologies if this has already been covered, but I've not been able to find it in the docs, the FAQ or the list archives. I'm trying to build gtk+ on a customised Debian 3.0r2 host, and getting the problem described below. It happens regardless of whether I'm using the default version of gcc, or whether I'm trying gtk+ version 2.2.3, 2.2.4 or 2.3.2. I'm assuming here that I've borked some dependency on the host, given the lack of references to this problem (and given my habit of compiling from scratch wherever viable). Any hints as to where I've gone wrong, or even where to concentrate my attention, would be gratefully received. I've compiled glib-2.3.2 from source, as well as pango-1.2.5 and atk-1.2.4. The config line is: ./configure --prefix=/opt/pkgs/gtk+-2.2.4 --enable-fbmanager --with-xinput=yes --with-x The result is always a failure in gdk/x11, with a long stream of complaints about syntax errors. The actual output is: make[2]: Entering directory `/opt/build/gtk+-2.2.4/gdk' Making all in x11 make[3]: Entering directory `/opt/build/gtk+-2.2.4/gdk/x11' /bin/sh ../../libtool --mode=compile gcc -DHAVE_CONFIG_H -I. -I. -I../.. -DG_LOG _DOMAIN=\"Gdk\" -DGDK_COMPILATION -I../.. -I../../gdk -I../../gdk -DG_DISABLE_DE PRECATED -DGDK_PIXBUF_DISABLE_DEPRECATED -DGDK_DISABLE_DEPRECATED -DG_DISABLE_CA ST_CHECKS -pthread -I/opt/pkgs/glib-2.3.2/include/glib-2.0 -I/opt/pkgs/glib-2.3. 2/lib/glib-2.0/include -I/opt/pkgs/pango-1.2.5/include/pango-1.0 -I/usr/X11R6/in clude -I/usr/X11R6/include/freetype2-g -O2 -Wall -c gdkcolor-x11.c mkdir .libs gcc -DHAVE_CONFIG_H -I. -I. -I../.. -DG_LOG_DOMAIN=\"Gdk\" -DGDK_COMPILATION -I ../.. -I../../gdk -I../../gdk -DG_DISABLE_DEPRECATED -DGDK_PIXBUF_DISABLE_DEPREC ATED -DGDK_DISABLE_DEPRECATED -DG_DISABLE_CAST_CHECKS -pthread -I/opt/pkgs/glib- 2.3.2/include/glib-2.0 -I/opt/pkgs/glib-2.3.2/lib/glib-2.0/include -I/opt/pkgs/p ango-1.2.5/include/pango-1.0 -I/usr/X11R6/include -I/usr/X11R6/include/freetype2 -g -O2 -Wall -c gdkcolor-x11.c -fPIC -DPIC -o .libs/gdkcolor-x11.o In file included from ../../gdk/x11/gdkdrawable-x11.h:37, from ../../gdk/x11/gdkwindow-x11.h:30, from gdkprivate-x11.h:36, from gdkx.h:78, from gdkcolor-x11.c:31: /usr/X11R6/include/X11/Xft/Xft.h:35:31: freetype/freetype.h: No such file or directory Is freetype.h on the system and in the search path? The other errors are probably caused by the following implicit declarations. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: g_object_new segmentation fault
divya seshadri wrote: Hi all, I've cross compiled GTK for arm linux (little endian) I'm using glib-2.1.4 gtk-2.2.0 pango-1.0.4 atk-1.0.3 libc-2.3.2 and I have the following problem. GTK applications like calender crash. A backtrace points to g_object_new. Is this a problem with glib or glibc?? If not either where does the problem lie? I've been stuck for the past week. I'm at my wits end.. please help. If it's on custom hardware, is there enough memory? Is it unfragmented? Has a memory check been done? Compile gtk with CFLAGS="-g -O0" and gdb thru the library code. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Show and hide window help(modified source code)
george carter wrote: You can probably tell by the source code I would like to show a window, have a loop count and hide that window. It shows the window but doesn't hide it. Any ideas? Eventually I would like to have this happen and then call another function, but little by little. Any help would be appreciated. Thanks. Make main(): show the window, add a function that calls gtk_widget_destroy(window) ( g_timeout_add() ), then waits in gtk_main(). ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: plotting question
[EMAIL PROTECTED] wrote: Hi, I must insert in my application 2 graphic types: 1) X-Y Axis to plot points in realtime (or with delay). 2) A timeline scrolling (seconds) from left to right that plots points. Could be sufficient using Canvas/Drawing in Gnome/GTK (do you know examples?) or is it necessary/advisable use external libraries? Eventually what are this libraries? If you implement some panning functions that transform window coordinates to sensible graphing coordinates, you can plot on to a gtk drawing area and have a timer thread to pan the graphics like a chart recorder, by scheduling 'expose' events. You can do it all in gtk. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: gtk frame buffer problems
divya seshadri wrote: Hi, The permissions are OK.. and the frame buffer is working. If I dump something to /dev/fb0 the penguin on the LCD gets ver written. But gtk demo says cannot open display device. Do you have any other suggestions? Divya. Step thru it with gdb/ddd. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: gtk frame buffer problems
divya seshadri wrote: Hi, I have GTK with linux-fb cross compiled for arm linux with an LCD display. My problem is that even though the frame buffer is working fine, when I run the gtk demo it says cannot open device '/dev/fb0'. The device is present and working. Any ideas?? Check permissions. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: New to GTK; some beginner questions.
Amit BHATNAGAR wrote: Hi all. I am considering using GTK+ to develop an app for Solaris 8, and I could use some help in answering some of my questions that would allow me to make a decision on whether or not to use GTK+. I have searched the FAQ & online docs, however if any of my questions have documented answers that I failed to find, please point me in the right direction. The app that I would be creating would largely deal with viewing radar images, and I would require to do some image manipulation, (zoom, pan, filtering, contrast, brightness, display multiple images, edge detection, mirror, etc). I would like to use GTKmm (and Glademm) as C++ would be my primary development language of choice. #1 Does GTK provide an API that I can use to perform the above operations? Gtk is more primitive compared to win32 GDI, but it mostly means there's less things to get in your way. win32 GDI has real-world and screen coordinate transformations for doing things like pan and zoom. I've done the same thing in gtk by writing my own transform functions. There's functions to render an array of pixels, so you can do all your edge enhancements with the array before rendering it. I haven't figured out how a pixmap can be got back into an array buffer. Multiple images can be displayed just be opening multiple windows. Since GTK is closely connected with GIMP, is it possible to use GIMP to achieve this? (assuming that GIMP has an API). I have read the documentation for GtkImage and even GdkPixmap, but nothing would suggest that image manipulation is possible from these. #2 Is my only other alternative to use a library like ImageMagick to perform these operations? Is there a GTK(mm) interface for this? I have found GDKMagick but this seems to be very old (obsolete?). There is also Magic++, but again I don't know how this integrates well within a GTK app. Is there any better alternative than ImageMagick? Thanks in advance for your assistance. amit. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Widget Update before gtkmain()
Pisanu Prasertnopakun wrote: Hello list, I write a long gtk code and can't wait till "gtkmain()" to update my widget. How can I update my widgets immeadiately before "gtkmain()"? I read from FAQ of GTK.ORG.. they told about "g_main_context_iteration()", but I don't really understand how. Thanks in advance. Something like: gtk_widget_queue_draw(drawing_area); gdk_window_process_updates(drawing_area->window,TRUE); ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Destroying a Dialog Box....
Hari Prasad Nadig wrote: within a dialog box called 'popup', I've this cancel button in the action area.. button = gtk_button_new_with_label ("Cancel"); g_signal_connect (G_OBJECT (button), "clicked", G_CALLBACK (close_dialog), (gpointer) popup); gtk_box_pack_start (GTK_BOX (GTK_DIALOG (popup)->action_area), button, TRUE, TRUE, 0); gtk_widget_show (button); and in the callback, I'm doing this.. static void close_dialog( GtkWidget *widget, GtkWidget *parent ){ gtk_widget_destroy(parent); } Basically, I'm trying to put a cancel button which closes the dialog.. Is this a proper method or is there any 'proper' way of doing this? It depends if you want it to be modal or non-modal, and if you want to extract any user-entered data from the popup. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: problem with pangox - and also with pkg-config !
Alain D'eurveilher wrote: Hi all, I'm trying to install GTK2.2.4 on my woody, and i already install the demendencies requiered, and glib, atk, pango. Now, running ./configure of gtk, I've got : <- configure: error: pangox Pango backend is required for x11 target -> I've made a google search on my problem, and i've found a post with : "(...) Try running 'pkg-config --list-all' and see what pango packages it lists." but i have : <- % pkg-config --list-all Package gtk+-2.0 was not found in the pkg-config search path. Perhaps you should add the directory containing `gtk+-2.0.pc' to the PKG_CONFIG_PATH environment variable Package 'gtk+-2.0', required by 'gnome-window-settings-2.0', not found -> ok, I'm trying to install gtk2.2.4, and i don't have the 2.0 verison. And where could I find the PKG_CONFIG_PATH ??? apt-get install -t unstable libpango1.0-dev locate pango*.pc /usr/lib/pkgconfig/pango.pc /usr/lib/pkgconfig/pangoft2.pc /usr/lib/pkgconfig/pangox.pc /usr/lib/pkgconfig/pangoxft.pc ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Messed up my GTK really, really bad
Bruce Ferrell wrote: OK gang, go ahead and laugh. In the long and torturous path to getting an application going on my system, I have meesed up my GTK tool chain something feirce. Somewhere along the line, I'managed to create a situation where gtk+2.09 from source thinks glib-2.2.3 is installed in /usr/local/glib-2.2.3 as evidenced by this: /bin/sh ../libtool --mode=link gcc -O2 -march=i386 -mcpu=i686 -Wall -version-info 0:9:0 -export-dynamic -rpath /usr/lib -export-symbols-regex "^[^_].*" ../gdk-pixbuf/libgdk_pixbuf-2.0.la -L/usr/X11R6/lib -lXft -lXrender -lXext -lX11 -lfreetype -lz -Wl,--export-dynamic -lpangoxft-1.0 -lpangox-1.0 -lpango-1.0 -lgobject-2.0 -lgmodule-2.0 -ldl -lglib-2.0 -lm -o libgdk-x11-2.0.la gdk.lo gdkcolor.lo gdkcursor.lo gdkdraw.lo gdkevents.lo gdkfont.lo gdkgc.lo gdkglobals.lo gdkkeys.lo gdkkeyuni.lo gdkimage.lo gdkpango.lo gdkpixbuf-drawable.lo gdkpixbuf-render.lo gdkpixmap.lo gdkpolyreg-generic.lo gdkrgb.lo gdkrectangle.lo gdkregion-generic.lo gdkwindow.lo gdkenumtypes.lo x11/libgdk-x11.la grep: /usr/local/glib-2.2.3/lib/libgobject-2.0.la: No such file or directory /bin/sed: can't read /usr/local/glib-2.2.3/lib/libgobject-2.0.la: No such file or directory libtool: link: `/usr/local/glib-2.2.3/lib/libgobject-2.0.la' is not a valid libtool archive make[3]: *** [libgdk-x11-2.0.la] Error 1 make[3]: Leaving directory `/usr/src/redhat/BUILD/gtk+-2.0.9/gdk' make[2]: *** [all-recursive] Error 1 make[2]: Leaving directory `/usr/src/redhat/BUILD/gtk+-2.0.9/gdk' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/usr/src/redhat/BUILD/gtk+-2.0.9' make: *** [all-recursive-am] Error 2 suggestions on how to fix this other than re-formatting would be very much appreciated... I already did do a re-install If you have old and new libraries installed, libtool can get confused. Edit any of the above 2.09 .la files to remove dependency references to 2.2.3, or else remove 2.2.3. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Intalling GTK-+ on Debian woody / Installer GTK-+ sur Debian Woody
Alain D'eurveilher wrote: Hi all, I just want to know if it is possible to install in an easy way GTK-+ on my woody ? using for example : # apt-get install gtk-+ (etc..) or other installation tools. Or if i absolutely need to follow the instruction on the gtk website ? apt-get install libgtk2.0-0 Apt-get any dependencies if it complains. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: stopping signals
dave walters wrote: Greetings: I have a spin button with a callback function invoked by the "value_changed" signal. The callback func has a gtk_spin_button_set_value(). When this executes it causes the spinner to emit the "value_changed' signal which invokes the callback which executes set_value() etc. It's recursive. I tried using g_signal_stop_emission_by_name() put in before the set_value() but it didn't work. Any ideas?? Why? If the next spin value is wrong, then you should have set the right values when the button was created. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: first creation evet of a window
halil agin wrote: When a window is firstly constructed, does it generate a signal or event? if so, what is it? I want to make some initialization on my components in the window when it first created... I thought that when it firstly created, it will generate a signal or event and i want to learn what it is... May be i am wrong, if so how can i do that? Maybe with the GtkWidget "realize" signal. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Problems building gtk+-2.2.4
Aitor Garcia wrote: Hello everybody: I am having problems building gtk+-2.2.4 Installation order followed: 1 pkgconfig-0.15.0 2 glib-2.2.3 3 pango-1.2.5 4 atk-1.2.4 5 gtk+-2.2.4.tar.gz All the programs have been built using configure, make and make install with no options. This is the error message I get when building gtk. make[3]: Entering directory `/root/gtk/gtk+-2.2.4/demos/gtk-demo' /bin/sh ../../libtool --mode=link gcc -g -O2 -Wall -o gtk-demo appwindow.o button_box.o changedisplay.o colorsel.o dialog.o drawingarea.o editable_cells.o images.o item_factory.o list_store.o menus.o panes.o pixbufs.o sizegroup.o stock_browser.o textview.o tree_store.o main.o ../../gdk-pixbuf/libgdk_pixbuf-2.0.la ../../gdk/libgdk-x11-2.0.la ../../gtk/libgtk-x11-2.0.la gcc -g -O2 -Wall -o .libs/gtk-demo appwindow.o button_box.o changedisplay.o colorsel.o dialog.o drawingarea.o editable_cells.o images.o item_factory.o list_store.o menus.o panes.o pixbufs.o sizegroup.o stock_browser.o textview.o tree_store.o main.o ../../gdk-pixbuf/.libs/libgdk_pixbuf-2.0.so -L/usr/local/lib ../../gdk/.libs/libgdk-x11-2.0.so -L/usr/X11R6/lib ../../gtk/.libs/libgtk-x11-2.0.so /root/gtk/gtk+-2.2.4/gdk/.libs/libgdk-x11-2.0.so /root/gtk/gtk+-2.2.4/gdk-pixbuf/.libs/libgdk_pixbuf-2.0.so -lXrandr -lXinerama -lXft -lXrender -lXext /usr/lib/libfreetype.so /usr/lib/libpangox-1.0.so /usr/lib/libpango-1.0.so /usr/lib/libgobject-2.0.so /usr/lib/libgmodule-2.0.so /usr/lib/libglib-2.0.so -lX11 /usr/lib/libpangoxft-1.0.so /usr/local/lib/libpangox-1.0.so /usr/local/lib/libpango-1.0.so /usr/local/lib/libatk-1.0.so /usr/local/lib/libgobject-2.0.so /usr/local/lib/libgmodule-2.0.so -ldl /usr/local/lib/libglib-2.0.so -lm -Wl,--rpath -Wl,/usr/local/lib /usr/bin/ld: warning: libXft.so.1, needed by /usr/lib/libpangoxft-1.0.so, may conflict with libXft.so.2 ../../gdk/.libs/libgdk-x11-2.0.so: undefined reference to `g_fprintf' ../../gdk/.libs/libgdk-x11-2.0.so: undefined reference to `g_get_application_name' ../../gtk/.libs/libgtk-x11-2.0.so: undefined reference to `g_sprintf' collect2: ld returned 1 exit status make[3]: *** [gtk-demo] Error 1 make[3]: Leaving directory `/root/gtk/gtk+-2.2.4/demos/gtk-demo' make[2]: *** [all-recursive] Error 1 make[2]: Leaving directory `/root/gtk/gtk+-2.2.4/demos' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/root/gtk/gtk+-2.2.4' make: *** [all-recursive-am] Error 2 It looks like the script is taking the library files both from /usr/lib and /usr/local/lib. In the /usr/lib directory seems that there is an old version of the libraries and this causes the display of the error message. Buggy libtool. How can I tell the script to take the library files only from one directory ? Go into the relevant libtool (.la) archives and delete references to /usr/lib objects that already have a /usr/local/lib equivalent. When installing, glib how can I tell the configure script to install the library files in /usr/lib in order to avoid this error message ? Set the environment variables as it says in the online installation manual. (prefix etc) ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Getting gtk+ to find my private copy of libgobject-2.0.so
Michael Terry wrote: Hello, list. I am trying to compile GTK+ CVS, and have run into a problem with it linking against my system's version of GLib instead of my private copy of a compiled GLib CVS. That fault's exactly the same as mine (just posted on [EMAIL PROTECTED]) ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: why is gtk install so difficult?
Steve Grimaud wrote: Hi, I'm sorry, but the statement below by Reuben hits the nail on the head for the "user." ... > I tried building a program from source (jpilot) on RH9 (shrike). I > encountered a "problem" with not finding GTK+ 2.0.0 or higher (although > I believe it's somewhere on my computer). I then downloaded GTK+; then > glib, pango, and atk+, since they had to be updated; then pkg-config, > since it wasn't the most uptodate (according to the other > configure/make/make install error messages). I know I installed atk+, > but GKT+ tells me it can't find it. > I give up. You have to have more than a basic understanding of the operating system and general installation procedures for programs to get these all to build and install. The biggest problem is that an obscure build error means reading and understanding a heap of things about autoconf/configure, linkers, versions, pkg-config, and various other things that normal users don't need for installing binaries. Users just have to accept that until all the variations in parameters of a system are covered, they'll have to have *some* detailed system know-how to fix errors on their own. If you have library objects and headers in a non-standard location, then ./configure LDFLAGS= CPPFLAGS= To find the path to atk headers: locate *atk*.h /usr/include/atk-1.0/atk/atk-enum-types.h /usr/include/atk-1.0/atk/atk.h /usr/include/atk-1.0/atk/atkaction.h ... To find the path to atk objects: locate libatk* /usr/lib/libatk-1.0.a /usr/lib/libatk-1.0.la /usr/lib/libatk-1.0.so /usr/lib/libatk-1.0.so.0 /usr/lib/libatk-1.0.so.0.200.4 ... ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: why is gtk install so difficult?
Tara Milana wrote: On 2003.10.19 06:23 Sven Neumann wrote: As outlined in http://bugzilla.gnome.org/show_bug.cgi?id=103708 your problem to compile gdk-pixbuf is a version incompatibility with your binutils and you can hardly blame the GTK+ developers for this. Hi, I just upgraded to the version I was told to use to compile for GTK: ldd (GNU libc) 2.1.3 Did that fix the problem? From http://sources.redhat.com/ml/binutils/2001-12/msg00182.html The following patch adds anonymous version nodes (per Ulrich's request anonymous version nodes are allowed only if there are no other nodes in the version script). gcc -shared .libs/gdk-pixbuf.o .libs/gdk-pixbuf-animation.o .libs/gdk-pixbuf-data.o .libs/gdk-pixbuf-io.o .libs/gdk-pixbuf-loader.o .libs/gdk-pixbuf-scale.o .libs/gdk-pixbuf-util.o .libs/gdk-pixdata.o .libs/gdk-pixbuf-enum-types.o -Wl,--whole-archive pixops/.libs/libpixops.a -Wl,--no-whole-archive -L/usr/lib /usr/lib/libgmodule-2.0.so -ldl /usr/lib/libgobject-2.0.so /usr/lib/libglib-2.0.so -lm -Wl,--export-dynamic -Wl,-soname -Wl,libgdk_pixbuf-2.0.so.0 -Wl,-version-script -Wl,.libs/libgdk_pixbuf-2.0.ver -o .libs/libgdk_pixbuf-2.0.so.0.200.4 /usr/bin/ld: anonymous version tag cannot be combined with other version tags Notice the above link involves -Wl,-version-script The error appears to be creating a new library by linking other library objects. If some of these libraries had anonymous version nodes and others did not, maybe that is what causes the error. Maybe you could try passing --disable-new-dtags, or else recompile any old libraries in the above link. `--enable-new-dtags' `--disable-new-dtags' This linker can create the new dynamic tags in ELF. But the older ELF systems may not understand them. If you specify `--enable-new-dtags', the dynamic tags will be created as needed. If you specify `--disable-new-dtags', no new dynamic tags will be created. By default, the new dynamic tags are not created. Note that those options are only available for ELF systems. Versioning is in this manual: http://www.gnu.org/manual/ld-2.9.1/ld.html My guess is there really needs to be some way of autoconf reading the ld versioning info from each library to ensure compatibility between them. Maybe there should be an extra ld utility for doing the dump. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: why is gtk install so difficult?
Paul Davis wrote: I don't recall the specific error, but to test that libraries work, an autoconf macro can compile and run a small program using a specific library to see if it works. Autoconf macros can be specific to each platform and/or test for all known implementations of ld to get a version number. that assumes that the problem shows itself in a simple test case. there are linker/library incompatibilities that will not reveal themselves unless a program does specific things to reveal them. and besides, what would your proposal result in? configure would say: "installed linker (ld.so vN.n) and the installed version of libfoo (vM.m) are not compatible" > and a new message would sally forth to gtk-list with the subject "why is gtk install so difficult?" I looked over the previous posts: http://mail.gnome.org/archives/gtk-list/2003-October/msg00200.html http://bugzilla.gnome.org/show_bug.cgi?id=103708 I assume some pre-existing libraries were linked with an older linker, then a newer linker was installed? The only solution in that case is for users to know the meaning of: /usr/i386-pc-linux/bin/ld: anonymous version tag cannot be combined with other version tags Maybe ld should have a more meaningful error message on object/linker versions. It would be tedious to write an autoconf sanity check macro for every library to be linked. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: why is gtk install so difficult?
Sven Neumann wrote: Hi, Russell Shaw <[EMAIL PROTECTED]> writes: You could make an autoconf macro to check the ld.so version from ldconfig -V. And how would I know that the user has libraries installed that are linked with a different and incompatible linker? The problem is not that GTK+ would depend on a specific version of the linker; that would be easy to check indeed. The problem is that the user messed up his system by mixing libraries compiled with incompatible linker versions. (At least that's what I believe to be the culprit). If you want to investigate this further, I suggest you look at the ld source code and grep for the particular error message. I don't recall the specific error, but to test that libraries work, an autoconf macro can compile and run a small program using a specific library to see if it works. Autoconf macros can be specific to each platform and/or test for all known implementations of ld to get a version number. If the user messed up their libraries *after* running ./configure, they have only themselves to blame. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: why is gtk install so difficult?
Sven Neumann wrote: Hi, Russell Shaw <[EMAIL PROTECTED]> writes: Any version incompatibilities should be handled in the relevant configure.ac. We are talking about incompatibilies between the linker and installed libraries. This is completely out of the realm of GTK+. You could make an autoconf macro to check the ld.so version from ldconfig -V. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: why is gtk install so difficult?
Sven Neumann wrote: Hi, Michael Torrie <[EMAIL PROTECTED]> writes: As outlined in http://bugzilla.gnome.org/show_bug.cgi?id=103708 your problem to compile gdk-pixbuf is a version incompatibility with your binutils and you can hardly blame the GTK+ developers for this. Perhaps this should be in the permanent FAQ. I had never heard about this before now. I saw one reference to it on the mailing list archives (the same problem as she had) but no one responded. That was about six months ago. I guess it's in the permanent record now, as it's now in the list archive. :) Well, it shouldn't be a problem that hits a lot of people. If I understand this correctly, it means that you messed up your system badly. This should not happen under normal circumstances. Any version incompatibilities should be handled in the relevant configure.ac. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Glade 2 & passing data to signal handlers
Dave Reed wrote: On Saturday 18 October 2003 11:24, Russell Shaw wrote: Joe Scaduto wrote: On Sat, 2003-10-18 at 06:01, Sven Neumann wrote: Russell Shaw <[EMAIL PROTECTED]> writes: Maybe you need: g_signal_connect((gpointer) okLoginButton, "clicked", G_CALLBACK (on_okLoginButton_clicked), GTK_OBJECT (loginDialog) ); Yes. Or simpler (w/o all the redundant casts): g_signal_connect (okLoginButton, "clicked", G_CALLBACK (on_okLoginButton_clicked), loginDialog); or even without writing a dedicated callback function: g_signal_connect_swapped (okLoginButton, "clicked", G_CALLBACK (gtk_widget_destroy), loginDialog); Both of your suggestions seems logical to me and I will try them. The only problem I have is that Glade 2 created that siganl connection not me. And at the beginning of the file where the signal connections are it says "DO NOT EDIT THIS FILE - generated by Glade" Can i still go in and change the signal connections as you suggested? Yes, but it will be over-written next time you generate the code. Most users don't use this generated code and use gladelib xml which is less intrusive on your program. Which leads me to the question: how do you pass extra parameters when you're using libglade along with glade? voidglade_xml_signal_connect_data (GladeXML *self, const char *handlername, GCallback func, gpointer user_data); ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Glade 2 & passing data to signal handlers
Joe Scaduto wrote: On Sat, 2003-10-18 at 11:24, Russell Shaw wrote: Yes, but it will be over-written next time you generate the code. Most users don't use this generated code and use gladelib xml which is less intrusive on your program. Thanks I appreciate the help.I found what the problem was at least I think. When Glade created the g_signal_connect_swapped((gpointer) okLoginButton, "clicked", G_CALLBACK (on_okLoginButton_clicked), GTK_OBJECT (loginDialog) ); connection the first parameter of on_okLoginButton_clicked is loginDialog not okLoginButton. Which would make my callback function: void on_okLoginButton_clicked(gpointer user_data, GtkButton *button) { } not(which Glade had orginally created for me): void on_okLoginButton_clicked(GtkButton *button, gpointer user_data) { } Correct me if I am wrong but I think the difference between g_signal_connect and g_signal_connect_swapped is the order the parameters are recieve by your callback function. Yes. The connect functions are usually macros involving g_signal_connect_object(): #define g_signal_connect(instance, detailed_signal, c_handler, data) #define g_signal_connect_after (instance, detailed_signal, c_handler, data) #define g_signal_connect_swapped(instance, detailed_signal, c_handler, data) gulong g_signal_connect_object (gpointer instance, const gchar *detailed_signal, GCallback c_handler, gpointer gobject, GConnectFlags connect_flags); http://developer.gnome.org/doc/API/2.0/gobject/gobject-Signals.html Some theory: http://www.gtk.org/tutorial/sec-theoryofsignalsandcallbacks.html ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Glade 2 & passing data to signal handlers
Joe Scaduto wrote: On Sat, 2003-10-18 at 06:01, Sven Neumann wrote: Hi, Russell Shaw <[EMAIL PROTECTED]> writes: Maybe you need: g_signal_connect((gpointer) okLoginButton, "clicked", G_CALLBACK (on_okLoginButton_clicked), GTK_OBJECT (loginDialog) ); Yes. Or simpler (w/o all the redundant casts): g_signal_connect (okLoginButton, "clicked", G_CALLBACK (on_okLoginButton_clicked), loginDialog); or even without writing a dedicated callback function: g_signal_connect_swapped (okLoginButton, "clicked", G_CALLBACK (gtk_widget_destroy), loginDialog); Sven Both of your suggestions seems logical to me and I will try them. The only problem I have is that Glade 2 created that siganl connection not me. And at the beginning of the file where the signal connections are it says "DO NOT EDIT THIS FILE - generated by Glade" Can i still go in and change the signal connections as you suggested? Yes, but it will be over-written next time you generate the code. Most users don't use this generated code and use gladelib xml which is less intrusive on your program. ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Glade 2 & passing data to signal handlers
Joe Scaduto wrote: Hey I am a newbie to GTK+ 2 and Glade 2. I have been playing around with GTK and I understand the concept behind signals and events. My problem arised when I decided to use Glade 2 and wanted to connect a 'clicked' signal for a 'modal' 'dialog' window 'button'. When connecting any signal handler to a signal I know you can pass extra data to the handler if you wish. In Glade 2 when connecting a signal you can also set the extra data and it casts it as a GTK_OBJECT. In my callbacks.c file my handler is there: code snippet void on_okLoginButton_clicked(GtkButton *button, gpointer user_data) { gtk_widget_destroy( GTK_WIDGET(user_data) ); } My call to destroy the dialog window only destroys the button. However when I change the widget to destroy to: GTK_WIDGET(button), everything works fine. Seems to me something is reversed. Also I went into my interface.c file and found the actual code to call the handler and found this: g_signal_connect_swapped((gpointer) okLoginButton, "clicked", G_CALLBACK (on_okLoginButton_clicked), GTK_OBJECT (loginDialog) ); Where loginDialog is my dialog window I want to close. If you actually read the whole thing and are willing to help a newbie out I would greatly appreciate any explanation of this if anyone has ran into this problem. Maybe you need: g_signal_connect((gpointer) okLoginButton, "clicked", G_CALLBACK (on_okLoginButton_clicked), GTK_OBJECT (loginDialog) ); ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Key pressed and modifier
gtk gtk wrote: Hi! I'm a new user of gtk/gdk and I can't cope with the following problem. I need to handle the event m. Here is the way I connect this signal to my widget: SIGNAL_CONNECT(GTK_OBJECT(my_widget), "key-press-event", GTK_SIGNAL_FUNC(key_pressed_handler), NULL); and the associated handler: void key_pressed_handler(GtkWidget *widget, GdkEventKey *event, gpointer func_data _U_) { /* Ctrl M on the line */ if (event->type == GDK_KEY_PRESS) { if ((event->state == GDK_MOD2_MASK) && if ((event->state == GDK_CONTROL_MASK) && Correction: if ((event->state & GDK_CONTROL_MASK) && ((event->keyval == GDK_m) || (event->keyval == GDK_M))) { } } } The problem is that I get nothing when I run this... I tried with "m" only (without ) and it's ok. What can I do so as the modifier is taken into account? ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Key pressed and modifier
gtk gtk wrote: Hi! I'm a new user of gtk/gdk and I can't cope with the following problem. I need to handle the event m. Here is the way I connect this signal to my widget: SIGNAL_CONNECT(GTK_OBJECT(my_widget), "key-press-event", GTK_SIGNAL_FUNC(key_pressed_handler), NULL); and the associated handler: void key_pressed_handler(GtkWidget *widget, GdkEventKey *event, gpointer func_data _U_) { /* Ctrl M on the line */ if (event->type == GDK_KEY_PRESS) { if ((event->state == GDK_MOD2_MASK) && if ((event->state == GDK_CONTROL_MASK) && ((event->keyval == GDK_m) || (event->keyval == GDK_M))) { } } } The problem is that I get nothing when I run this... I tried with "m" only (without ) and it's ok. What can I do so as the modifier is taken into account? ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Help on Glib and pango installations.
nkb wrote: Hi. Beginner's question. I'm trying to install gtk2.2.3. I downloaded and installed glib-2.2.3. When installing pango-1.2.5, it reports an error that says Glib 2.1.3 or better is required. But I just installed successfully [without error] my glib 2.2.3! So, how do I check and confirm that my glib was installed successfully? Or how do I tell pango that I've glib 2.2.3 installed? Any pointers will be helpful. locate *.pc|sort /usr/lib/pkgconfig/glib-2.0.pc /usr/lib/pkgconfig/glib.pc pkg-config --modversion glib-2.0 2.2.2 pkg-config --modversion glib 1.2.10 ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: new bie
Chorn Sokun wrote: Sjoerd Site is on strike where can I get the source packages? http://www.gtk.org/download/ http://www.gtk.org/tutorial/ ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: text looks different on different displays
[EMAIL PROTECTED] wrote: On Sun, 10 Aug 2003 11:53:49 EDT, Jeff Abrahamson <[EMAIL PROTECTED]> said: Thanks, I was thinking something like this, but it seems the metric information is wrong on one. That is, buttons displayed on one machine look like normal, +--+ | button | +--+ on the other they get displayed all spread out like this: ++ | b u t t o n | ++ Well, they could still be different fonts.. remember that the XLFD spec has a 'width' field (third from the end)... s: % xlsfonts -fn "-*-*-*-r-normal-sans-*-140-100-100-*-*-iso8859-15" -b&h-lucida-bold-r-normal-sans-19-140-100-100-p-0-iso8859-15 -b&h-lucida-bold-r-normal-sans-20-140-100-100-p-127-iso8859-15 -b&h-lucida-medium-r-normal-sans-19-140-100-100-p-0-iso8859-15 -b&h-lucida-medium-r-normal-sans-20-140-100-100-p-114-iso8859-15 -b&h-lucidatypewriter-bold-r-normal-sans-19-140-100-100-m-0-iso8859-15 -b&h-lucidatypewriter-bold-r-normal-sans-20-140-100-100-m-120-iso8859-15 -b&h-lucidatypewriter-medium-r-normal-sans-19-140-100-100-m-0-iso8859-15 -b&h-lucidatypewriter-medium-r-normal-sans-20-140-100-100-m-120-iso8859-15 ... Is there any way to find out which file on the system contains a given font? ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Re: Autoconf hello world
Russell Shaw wrote: Hi, I'm looking for a GTK+ hello-world example that uses configure.in and automake.am (autoconf and automake). Ok, i found a faq: http://www.gtk.org/faq/#AEN412 ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Autoconf hello world
Hi, I'm looking for a GTK+ hello-world example that uses configure.in and automake.am (autoconf and automake). ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Printer friendly docs
Hi, I downloaded the gtk+ and glib 2.2.2 packages. The documentation gets built as html (from sgml/xml) to be viewed online. Because i'm learning the package, it's too tedious to read all the documentation on the monitor. What's the best way to get a printable output preferably with a table of contents and page numbers? (ps, pdf, dvi etc) ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list
Printer friendly docs
Hi, I downloaded the gtk+ and glib 2.2.2 packages. The documentation gets built as html (from sgml/xml) to be viewed online. Because i'm learning the package, it's too tedious to read all the documentation on the monitor. What's the best way to get a printable output preferably with a table of contents and page numbers? (ps, pdf, dvi etc) ___ gtk-list mailing list [EMAIL PROTECTED] http://mail.gnome.org/mailman/listinfo/gtk-list