From jredrejo at ...176... Mon Nov 5 13:47:47 2007 From: jredrejo at ...176... (=?ISO-8859-1?Q?Jos=E9_Luis_Redrejo?=) Date: Mon, 5 Nov 2007 13:47:47 +0100 Subject: [Gambas-devel] Some new ugly bugs using gb.gtk Message-ID: <8eb28a500711050447t5bc7f2f0h30c2cfa9c518a1e1@...178...> I continue testing gtk/qt components trying to get them as similar and working perfectly prior to the 2.0 release. I have attached a small project to illustrate two bugs using gtk (with qt the same project work perfectly). - To test the first bug you only have to click on the listview and see how its content changes to ugly characters. It only happens if you use the selected event of the listview object. - To test the second bug you need to drag and drop anything from the desktop or nautilus over the tabstrip object: Using gnome the desktop freezes until you go to a tty and execute "killall -9 nautilus", then you can drag and drop without any problem over the tabstrip until the project is restarted. Don't hesitate to contact me if you need further explanation about them. Tested on Debian, gnome and using gambas2 1.9.90. Regards. -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: testBugs.tar.gz Type: application/x-gzip Size: 4327 bytes Desc: not available URL: From jredrejo at ...176... Mon Nov 5 14:34:53 2007 From: jredrejo at ...176... (=?ISO-8859-1?Q?Jos=E9_Luis_Redrejo?=) Date: Mon, 5 Nov 2007 14:34:53 +0100 Subject: [Gambas-devel] Some new ugly bugs using gb.gtk In-Reply-To: <8eb28a500711050447t5bc7f2f0h30c2cfa9c518a1e1@...178...> References: <8eb28a500711050447t5bc7f2f0h30c2cfa9c518a1e1@...178...> Message-ID: <8eb28a500711050534n105d96eckbf2e21ff98464541@...178...> Another new bug in gtk: Button objects don't raise MouseDown and MouseUp events, mouse_click event works. Checked with gambas 1.9.90 & 1.9.50, on Debian & Gnome. Regards 2007/11/5, Jos? Luis Redrejo : > > I continue testing gtk/qt components trying to get them as similar and > working perfectly prior to the 2.0 release. > I have attached a small project to illustrate two bugs using gtk (with qt > the same project work perfectly). > - To test the first bug you only have to click on the listview and see how > its content changes to ugly characters. It only happens if you use the > selected event of the listview object. > - To test the second bug you need to drag and drop anything from the > desktop or nautilus over the tabstrip object: Using gnome the desktop > freezes until you go to a tty and execute "killall -9 nautilus", then you > can drag and drop without any problem over the tabstrip until the project is > restarted. > > Don't hesitate to contact me if you need further explanation about them. > Tested on Debian, gnome and using gambas2 1.9.90. > Regards. > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From gambas at ...1... Mon Nov 5 14:42:16 2007 From: gambas at ...1... (Benoit Minisini) Date: Mon, 5 Nov 2007 14:42:16 +0100 Subject: [Gambas-devel] Some new ugly bugs using gb.gtk In-Reply-To: <8eb28a500711050447t5bc7f2f0h30c2cfa9c518a1e1@...178...> References: <8eb28a500711050447t5bc7f2f0h30c2cfa9c518a1e1@...178...> Message-ID: <200711051442.16529.gambas@...1...> On lundi 5 novembre 2007, Jos? Luis Redrejo wrote: > I continue testing gtk/qt components trying to get them as similar and > working perfectly prior to the 2.0 release. > I have attached a small project to illustrate two bugs using gtk (with qt > the same project work perfectly). > - To test the first bug you only have to click on the listview and see how > its content changes to ugly characters. It only happens if you use the > selected event of the listview object. > - To test the second bug you need to drag and drop anything from the > desktop or nautilus over the tabstrip object: Using gnome the desktop > freezes until you go to a tty and execute "killall -9 nautilus", then you > can drag and drop without any problem over the tabstrip until the project > is restarted. > > Don't hesitate to contact me if you need further explanation about them. > Tested on Debian, gnome and using gambas2 1.9.90. > Regards. The first bug has been fixed in the last revision. It was an old Daniel's code that badly interacted with my own new code. As for the second bug, it seems that GtkNoteBook widget is a drag & drop destination by default, for apparently no reason. I disabled that, and so your bug should be fixed. But I think that the problem will remain if you set TabStrip.Drop property manually. I noticed too that dropping from Firefox to a GTK+ Gambas application does not work correctly. For example, only the first character of the drop is got, no idea why. Regard, -- Benoit Minisini From jredrejo at ...176... Mon Nov 5 18:37:08 2007 From: jredrejo at ...176... (=?ISO-8859-1?Q?Jos=E9_Luis_Redrejo?=) Date: Mon, 5 Nov 2007 18:37:08 +0100 Subject: [Gambas-devel] Some new ugly bugs using gb.gtk In-Reply-To: <200711051442.16529.gambas@...1...> References: <8eb28a500711050447t5bc7f2f0h30c2cfa9c518a1e1@...178...> <200711051442.16529.gambas@...1...> Message-ID: <8eb28a500711050937u22f7d2f5jd485b5563b4761c@...178...> 2007/11/5, Benoit Minisini : > > On lundi 5 novembre 2007, Jos? Luis Redrejo wrote: > > I continue testing gtk/qt components trying to get them as similar and > > working perfectly prior to the 2.0 release. > > I have attached a small project to illustrate two bugs using gtk (with > qt > > the same project work perfectly). > > - To test the first bug you only have to click on the listview and see > how > > its content changes to ugly characters. It only happens if you use the > > selected event of the listview object. > > - To test the second bug you need to drag and drop anything from the > > desktop or nautilus over the tabstrip object: Using gnome the desktop > > freezes until you go to a tty and execute "killall -9 nautilus", then > you > > can drag and drop without any problem over the tabstrip until the > project > > is restarted. > > > > Don't hesitate to contact me if you need further explanation about > them. > > Tested on Debian, gnome and using gambas2 1.9.90. > > Regards. > > The first bug has been fixed in the last revision. It was an old Daniel's > code > that badly interacted with my own new code. > > As for the second bug, it seems that GtkNoteBook widget is a drag & drop > destination by default, for apparently no reason. I disabled that, and so > your bug should be fixed. Thanks very much But I think that the problem will remain if you set TabStrip.Drop property > manually. Don't care too much about that, but it's a bug anyway. My impression is that the drag signal is launched in nautilus but it stays waiting forever, maybe there's some broken mechanism in tabstrip D&D procedures. It's funny it doesn't happen with other controls. I noticed too that dropping from Firefox to a GTK+ Gambas application does > not > work correctly. For example, only the first character of the drop is got, > no > idea why. > I know, I know, but that seems to be a firefox bug, not a gambas bug. On the other hand, did you see my message about the lost events in the button (and togglebutton) control? I told you about MouseUp & MouseDown, but later I checked Menu event is also lost. Cheers Jos? L. -------------- next part -------------- An HTML attachment was scrubbed... URL: From gambas at ...1... Mon Nov 5 20:57:39 2007 From: gambas at ...1... (Benoit Minisini) Date: Mon, 5 Nov 2007 20:57:39 +0100 Subject: [Gambas-devel] Some new ugly bugs using gb.gtk In-Reply-To: <8eb28a500711050937u22f7d2f5jd485b5563b4761c@...178...> References: <8eb28a500711050447t5bc7f2f0h30c2cfa9c518a1e1@...178...> <200711051442.16529.gambas@...1...> <8eb28a500711050937u22f7d2f5jd485b5563b4761c@...178...> Message-ID: <200711052057.39708.gambas@...1...> On lundi 5 novembre 2007, Jos? Luis Redrejo wrote: > 2007/11/5, Benoit Minisini : > > On lundi 5 novembre 2007, Jos? Luis Redrejo wrote: > > > I continue testing gtk/qt components trying to get them as similar and > > > working perfectly prior to the 2.0 release. > > > I have attached a small project to illustrate two bugs using gtk (with > > > > qt > > > > > the same project work perfectly). > > > - To test the first bug you only have to click on the listview and see > > > > how > > > > > its content changes to ugly characters. It only happens if you use the > > > selected event of the listview object. > > > - To test the second bug you need to drag and drop anything from the > > > desktop or nautilus over the tabstrip object: Using gnome the desktop > > > freezes until you go to a tty and execute "killall -9 nautilus", then > > > > you > > > > > can drag and drop without any problem over the tabstrip until the > > > > project > > > > > is restarted. > > > > > > Don't hesitate to contact me if you need further explanation about > > > > them. > > > > > Tested on Debian, gnome and using gambas2 1.9.90. > > > Regards. > > > > The first bug has been fixed in the last revision. It was an old Daniel's > > code > > that badly interacted with my own new code. > > > > As for the second bug, it seems that GtkNoteBook widget is a drag & drop > > destination by default, for apparently no reason. I disabled that, and so > > your bug should be fixed. > > Thanks very much > > But I think that the problem will remain if you set TabStrip.Drop property > > > manually. > > Don't care too much about that, but it's a bug anyway. My impression is > that the drag signal is launched in nautilus but it stays waiting forever, > maybe there's some broken mechanism in tabstrip D&D procedures. It's funny > it doesn't happen with other controls. > > I noticed too that dropping from Firefox to a GTK+ Gambas application does > > > not > > work correctly. For example, only the first character of the drop is got, > > no > > idea why. > > I know, I know, but that seems to be a firefox bug, not a gambas bug. > On the other hand, did you see my message about the lost events in the > button (and togglebutton) control? I told you about MouseUp & MouseDown, > but later I checked Menu event is also lost. > > Cheers > Jos? L. This was finally fixed in the last svn revision. Regards, -- Benoit Minisini From jredrejo at ...176... Wed Nov 7 10:51:19 2007 From: jredrejo at ...176... (=?ISO-8859-1?Q?Jos=E9_Luis_Redrejo?=) Date: Wed, 7 Nov 2007 10:51:19 +0100 Subject: [Gambas-devel] Bug with radiobutton and gtk Message-ID: <8eb28a500711070151r4d4b53efme83b4385cafc0584@...178...> Hi again, using a RadioButton control with gtk , it does not accept changes by code in its value property. Clicking with the mouse on it works right. So: RadioButton1.value=false does not work in gtk, but does it with qt. Regards. Jos? L. -------------- next part -------------- An HTML attachment was scrubbed... URL: From gambas at ...1... Thu Nov 8 13:29:47 2007 From: gambas at ...1... (Benoit Minisini) Date: Thu, 8 Nov 2007 13:29:47 +0100 Subject: [Gambas-devel] Bug with radiobutton and gtk In-Reply-To: <8eb28a500711070151r4d4b53efme83b4385cafc0584@...178...> References: <8eb28a500711070151r4d4b53efme83b4385cafc0584@...178...> Message-ID: <200711081329.48208.gambas@...1...> On mercredi 7 novembre 2007, Jos? Luis Redrejo wrote: > Hi again, using a RadioButton control with gtk , it does not accept changes > by code in its value property. Clicking with the mouse on it works right. > > So: > RadioButton1.value=false > does not work in gtk, but does it with qt. > > Regards. > Jos? L. Actually isn't it a bug in gb.qt? I see no reason for manually setting a RadioButton value to FALSE. If you need that, you should use a CheckBox, don't you? -- Benoit Minisini From sourceforge-raindog2 at ...19... Thu Nov 8 13:35:51 2007 From: sourceforge-raindog2 at ...19... (Rob) Date: Thu, 8 Nov 2007 07:35:51 -0500 Subject: [Gambas-devel] Bug with radiobutton and gtk In-Reply-To: <200711081329.48208.gambas@...1...> References: <8eb28a500711070151r4d4b53efme83b4385cafc0584@...178...> <200711081329.48208.gambas@...1...> Message-ID: <200711080735.51804.sourceforge-raindog2@...19...> On Thursday 08 November 2007 07:29, Benoit Minisini wrote: > I see no reason for manually setting a RadioButton value to FALSE. > If you need that, you should use a CheckBox, don't you? Well, if you present the user with a set of choices via RadioButton controls, and don't want there to be a default choice, then when you start entering a new record or something you would want all the radio buttons to be cleared, wouldn't you? Rob From gambas at ...1... Thu Nov 8 13:39:34 2007 From: gambas at ...1... (Benoit Minisini) Date: Thu, 8 Nov 2007 13:39:34 +0100 Subject: [Gambas-devel] Bug with radiobutton and gtk In-Reply-To: <200711080735.51804.sourceforge-raindog2@...19...> References: <8eb28a500711070151r4d4b53efme83b4385cafc0584@...178...> <200711081329.48208.gambas@...1...> <200711080735.51804.sourceforge-raindog2@...19...> Message-ID: <200711081339.35035.gambas@...1...> On jeudi 8 novembre 2007, Rob wrote: > On Thursday 08 November 2007 07:29, Benoit Minisini wrote: > > I see no reason for manually setting a RadioButton value to FALSE. > > If you need that, you should use a CheckBox, don't you? > > Well, if you present the user with a set of choices via RadioButton > controls, and don't want there to be a default choice, then when you > start entering a new record or something you would want all the > radio buttons to be cleared, wouldn't you? > > Rob > This is a bad idea I think. You should better add one more RadioButton for the default choice that is not a default choice :-) -- Benoit Minisini From jredrejo at ...176... Thu Nov 8 14:30:27 2007 From: jredrejo at ...176... (=?ISO-8859-1?Q?Jos=E9_Luis_Redrejo?=) Date: Thu, 8 Nov 2007 14:30:27 +0100 Subject: [Gambas-devel] Bug with radiobutton and gtk In-Reply-To: <200711081329.48208.gambas@...1...> References: <8eb28a500711070151r4d4b53efme83b4385cafc0584@...178...> <200711081329.48208.gambas@...1...> Message-ID: <8eb28a500711080530h6e91d45ew4e797dd0ad3be046@...178...> 2007/11/8, Benoit Minisini : > > On mercredi 7 novembre 2007, Jos? Luis Redrejo wrote: > > Hi again, using a RadioButton control with gtk , it does not accept > changes > > by code in its value property. Clicking with the mouse on it works > right. > > > > So: > > RadioButton1.value=false > > does not work in gtk, but does it with qt. > > > > Regards. > > Jos? L. > > Actually isn't it a bug in gb.qt? > > I see no reason for manually setting a RadioButton value to FALSE. If you > need > that, you should use a CheckBox, don't you? > > Checkbox does not update automatically the others options, as radiobutton does. It's quite easy to find its needs: Just think of some program options: you read the options from a Settings file and present them in a form for the user to be able to change them. For instance: male-female, hair color: brown, red, blonde, black, etc. In these cases it's useful to re-read the Settings files and present to the user the values he chose in a previous action. So you do need something like : optSexMale.value=(mySettings["sex"]="male") and automatically optSexFemale will be updated to the right value to show it in a form. In my case the problem arise trying to do something like optInternetEnabled.value=(mySettings["EnalbledInet"]="1"), and there is no logical default value for things like this. Regards. -------------- next part -------------- An HTML attachment was scrubbed... URL: From gambas at ...1... Thu Nov 8 14:42:45 2007 From: gambas at ...1... (Benoit Minisini) Date: Thu, 8 Nov 2007 14:42:45 +0100 Subject: [Gambas-devel] Bug with radiobutton and gtk In-Reply-To: <8eb28a500711080530h6e91d45ew4e797dd0ad3be046@...178...> References: <8eb28a500711070151r4d4b53efme83b4385cafc0584@...178...> <200711081329.48208.gambas@...1...> <8eb28a500711080530h6e91d45ew4e797dd0ad3be046@...178...> Message-ID: <200711081442.45681.gambas@...1...> On jeudi 8 novembre 2007, Jos? Luis Redrejo wrote: > 2007/11/8, Benoit Minisini : > > On mercredi 7 novembre 2007, Jos? Luis Redrejo wrote: > > > Hi again, using a RadioButton control with gtk , it does not accept > > > > changes > > > > > by code in its value property. Clicking with the mouse on it works > > > > right. > > > > > So: > > > RadioButton1.value=false > > > does not work in gtk, but does it with qt. > > > > > > Regards. > > > Jos? L. > > > > Actually isn't it a bug in gb.qt? > > > > I see no reason for manually setting a RadioButton value to FALSE. If you > > need > > that, you should use a CheckBox, don't you? > > Checkbox does not update automatically the others options, as radiobutton > does. It's quite easy to find its needs: Just think of some program > options: you read the options from a Settings file and present them in a > form for the user to be able to change them. For instance: male-female, > hair color: brown, red, blonde, black, etc. In these cases it's useful to > re-read the Settings files and present to the user the values he chose in a > previous action. So you do need something like : > optSexMale.value=(mySettings["sex"]="male") > and automatically optSexFemale will be updated to the right value to show > it in a form. > In my case the problem arise trying to do something like > optInternetEnabled.value=(mySettings["EnalbledInet"]="1"), and there is no > logical default value for things like this. > > Regards. RadioButtons are mutually exclusive, so you just have to set the only one that should be set. All other RadioButtons having the same parent will be unset automatically. Regards, -- Benoit Minisini From jredrejo at ...176... Thu Nov 8 15:18:28 2007 From: jredrejo at ...176... (=?ISO-8859-1?Q?Jos=E9_Luis_Redrejo?=) Date: Thu, 8 Nov 2007 15:18:28 +0100 Subject: [Gambas-devel] Bug with radiobutton and gtk In-Reply-To: <200711081442.45681.gambas@...1...> References: <8eb28a500711070151r4d4b53efme83b4385cafc0584@...178...> <200711081329.48208.gambas@...1...> <8eb28a500711080530h6e91d45ew4e797dd0ad3be046@...178...> <200711081442.45681.gambas@...1...> Message-ID: <8eb28a500711080618u6f7d0b41t609cfb8855ff436a@...178...> > > > > RadioButtons are mutually exclusive, so you just have to set the only one > that > should be set. All other RadioButtons having the same parent will be unset > automatically. > > I know, that's why they are so useful when filling forms from a Settings file, you only have to take care of one of the buttons to set male/female, enabled/disabled, etc.. For me that's the main reason to take the qt behaviour to gtk, i.e. been able to set them by code. Regards. -------------- next part -------------- An HTML attachment was scrubbed... URL: From rospolosco at ...29... Thu Nov 8 19:01:02 2007 From: rospolosco at ...29... (Stefano Palmeri) Date: Thu, 8 Nov 2007 19:01:02 +0100 Subject: [Gambas-devel] Bug with radiobutton and gtk In-Reply-To: <200711080735.51804.sourceforge-raindog2@...19...> References: <8eb28a500711070151r4d4b53efme83b4385cafc0584@...178...> <200711081329.48208.gambas@...1...> <200711080735.51804.sourceforge-raindog2@...19...> Message-ID: <200711081901.02962.rospolosco@...29...> Alle 13:35, gioved? 8 novembre 2007, Rob ha scritto: > On Thursday 08 November 2007 07:29, Benoit Minisini wrote: > > I see no reason for manually setting a RadioButton value to FALSE. > > If you need that, you should use a CheckBox, don't you? > > Well, if you present the user with a set of choices via RadioButton > controls, and don't want there to be a default choice, then when you > start entering a new record or something you would want all the > radio buttons to be cleared, wouldn't you? > > Rob > I'm not sure I understood, but it seems that you need an easy way to reset all the RadioButtons of a Form. Easy workaround: put an invisible RadioButton control in the form. When you want to reset all the other visible RadioButton controls, set the invisible = TRUE. Sorry, if I didn't understand. Stefano > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Splunk Inc. > Still grepping through log files to find problems? Stop. > Now Search log events and configuration files using AJAX and a browser. > Download your FREE copy of Splunk now >> http://get.splunk.com/ > _______________________________________________ > Gambas-devel mailing list > Gambas-devel at lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/gambas-devel From sourceforge-raindog2 at ...19... Thu Nov 8 20:47:31 2007 From: sourceforge-raindog2 at ...19... (Rob) Date: Thu, 8 Nov 2007 14:47:31 -0500 Subject: [Gambas-devel] Bug with radiobutton and gtk In-Reply-To: <200711081901.02962.rospolosco@...29...> References: <8eb28a500711070151r4d4b53efme83b4385cafc0584@...178...> <200711080735.51804.sourceforge-raindog2@...19...> <200711081901.02962.rospolosco@...29...> Message-ID: <200711081447.31159.sourceforge-raindog2@...19...> On Thursday 08 November 2007 13:01, Stefano Palmeri wrote: > Easy workaround: put an invisible RadioButton control in the form. > When you want to reset all the other visible RadioButton controls, > set the invisible = TRUE. > Sorry, if I didn't understand. No, you're right. I came to the same conclusion while driving to work today :) Rob From ronstk at ...124... Fri Nov 9 04:31:06 2007 From: ronstk at ...124... (ron) Date: Fri, 9 Nov 2007 05:31:06 +0200 Subject: [Gambas-devel] Bug with radiobutton and gtk In-Reply-To: <200711081901.02962.rospolosco@...29...> References: <8eb28a500711070151r4d4b53efme83b4385cafc0584@...178...> <200711080735.51804.sourceforge-raindog2@...19...> <200711081901.02962.rospolosco@...29...> Message-ID: <200711090431.06786.ronstk@...124...> On Thursday 08 November 2007 19:01, Stefano Palmeri wrote: > Alle 13:35, gioved? 8 novembre 2007, Rob ha scritto: > > On Thursday 08 November 2007 07:29, Benoit Minisini wrote: > > > I see no reason for manually setting a RadioButton value to FALSE. > > > If you need that, you should use a CheckBox, don't you? > > > > Well, if you present the user with a set of choices via RadioButton > > controls, and don't want there to be a default choice, then when you > > start entering a new record or something you would want all the > > radio buttons to be cleared, wouldn't you? > > > > Rob > > > > I'm not sure I understood, but it seems that you need an easy way > to reset all the RadioButtons of a Form. > Easy workaround: put an invisible RadioButton control in the form. > When you want to reset all the other visible RadioButton controls, set the > invisible = TRUE. > Sorry, if I didn't understand. > > Stefano Correct to add one but _never_ invisible !!! This is the one you need to cancel the choise. Save file picture.bmp as: (visible) o) save as *.jpg o) save as *.gif (your invisible set by opening the form to true to prevent the jpg/gif) o) save as default asis In real life it is a choise too. Ron From rospolosco at ...29... Fri Nov 9 08:51:31 2007 From: rospolosco at ...29... (Stefano Palmeri) Date: Fri, 9 Nov 2007 08:51:31 +0100 Subject: [Gambas-devel] Bug with radiobutton and gtk In-Reply-To: <200711090431.06786.ronstk@...124...> References: <8eb28a500711070151r4d4b53efme83b4385cafc0584@...178...> <200711081901.02962.rospolosco@...29...> <200711090431.06786.ronstk@...124...> Message-ID: <200711090851.31820.rospolosco@...29...> Alle 04:31, venerd? 9 novembre 2007, ron ha scritto: > On Thursday 08 November 2007 19:01, Stefano Palmeri wrote: > > Alle 13:35, gioved? 8 novembre 2007, Rob ha scritto: > > > On Thursday 08 November 2007 07:29, Benoit Minisini wrote: > > > > I see no reason for manually setting a RadioButton value to FALSE. > > > > If you need that, you should use a CheckBox, don't you? > > > > > > Well, if you present the user with a set of choices via RadioButton > > > controls, and don't want there to be a default choice, then when you > > > start entering a new record or something you would want all the > > > radio buttons to be cleared, wouldn't you? > > > > > > Rob > > > > I'm not sure I understood, but it seems that you need an easy way > > to reset all the RadioButtons of a Form. > > Easy workaround: put an invisible RadioButton control in the form. > > When you want to reset all the other visible RadioButton controls, set > > the invisible = TRUE. > > Sorry, if I didn't understand. > > > > Stefano > > Correct to add one but _never_ invisible !!! > This is the one you need to cancel the choise. > > Save file picture.bmp as: > (visible) > o) save as *.jpg > o) save as *.gif > > (your invisible set by opening the form to true to prevent the jpg/gif) > o) save as default asis > > In real life it is a choise too. > > Ron De gustibus... Why _never_ invisible? You can put a Button labelled "Reset" that set the invisible = TRUE. It's just something relying on what you like to see in your form :-) and how many Parent controls with RadioButtons children you need to reset once for all, with just one press. Stefano > > Ron > > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Splunk Inc. > Still grepping through log files to find problems? Stop. > Now Search log events and configuration files using AJAX and a browser. > Download your FREE copy of Splunk now >> http://get.splunk.com/ > _______________________________________________ > Gambas-devel mailing list > Gambas-devel at lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/gambas-devel From robert at ...554... Mon Nov 12 10:44:39 2007 From: robert at ...554... (=?ISO-8859-15?Q?Robert_Ch=E9ramy?=) Date: Mon, 12 Nov 2007 10:44:39 +0100 Subject: [Gambas-devel] Reload in Editor Message-ID: <47382087.4000902@...554...> Hi, the "reload" button in the Editor (1.9.90, svn trunk) does nothing if the source has not been modified through the editor. If the file has just been modified on the file system (for example a generated Database Module from the Database Manager), it won't be reloaded. I think the "reload" button should reload the file from disk every time (after confirmation has been asked to the user). I propose attached patch to change this behaviour. Cheers, tibob -------------- next part -------------- A non-text attachment was scrubbed... Name: editor_reload.diff Type: text/x-patch Size: 614 bytes Desc: not available URL: From gambas at ...1... Mon Nov 12 10:59:52 2007 From: gambas at ...1... (Benoit Minisini) Date: Mon, 12 Nov 2007 10:59:52 +0100 Subject: [Gambas-devel] Reload in Editor In-Reply-To: <47382087.4000902@...554...> References: <47382087.4000902@...554...> Message-ID: <200711121059.52241.gambas@...1...> On lundi 12 novembre 2007, Robert Ch?ramy wrote: > Hi, > > the "reload" button in the Editor (1.9.90, svn trunk) does nothing if > the source has not been modified through the editor. > If the file has just been modified on the file system (for example a > generated Database Module from the Database Manager), it won't be reloaded. > > I think the "reload" button should reload the file from disk every time > (after confirmation has been asked to the user). > > I propose attached patch to change this behaviour. > > Cheers, > > tibob OK, thanks. I will apply the same patch for other editors (form editor, text editor) for coherency. Regards, -- Benoit Minisini From gambas at ...1... Tue Nov 13 12:03:18 2007 From: gambas at ...1... (Benoit Minisini) Date: Tue, 13 Nov 2007 12:03:18 +0100 Subject: [Gambas-devel] Problem with subversion logs In-Reply-To: <200711131042.52685.gambas@...1...> References: <200711131008.43479.gambas@...1...> <200711131038.34627.lordheavym@...176...> <200711131042.52685.gambas@...1...> Message-ID: <200711131203.18659.gambas@...1...> Since 1.9.90, the ChangeLog is automatically created from the Subversion commits. Alas, this works reliably only if committers follow the revision log format described in a previous post, and there: http://gambasdoc.org/help/howto/svn. So please follow it, because otherwise some changes may not appear in the ChangeLog, especially the lines that do not begin with "NEW:", "BUG:" or "OPT:" in a component change. It takes me a lot of time to fix everything manually. Regards, -- Benoit Minisini From david_villalobos_c at ...7... Tue Nov 13 18:12:08 2007 From: david_villalobos_c at ...7... (David Villalobos Cambronero) Date: Tue, 13 Nov 2007 09:12:08 -0800 (PST) Subject: [Gambas-devel] Develop in C/C++ Message-ID: <444167.5794.qm@...555...> Hi all, Some days ago I suscribed to the developer list because I want to start developing somethig for Gambas, or at least, report a bug and try to fix it. Maybe this post is not Gambas related, (sorry for that) but I want to know if you can tell me a good Manual to learn C/C++ for Linux, some years ago I started to develop in C/C++ but always Windows oriented, now I want to start for Linux. Can anyone helpme please??? Best regards David ____________________________________________________________________________________ Be a better sports nut! Let your teams follow you with Yahoo Mobile. Try it now. http://mobile.yahoo.com/sports;_ylt=At9_qDKvtAbMuh1G1SQtBI7ntAcJ -------------- next part -------------- An HTML attachment was scrubbed... URL: From lordheavym at ...176... Tue Nov 13 18:19:58 2007 From: lordheavym at ...176... (Laurent Carlier) Date: Tue, 13 Nov 2007 18:19:58 +0100 Subject: [Gambas-devel] Develop in C/C++ In-Reply-To: <444167.5794.qm@...555...> References: <444167.5794.qm@...555...> Message-ID: <200711131820.06706.lordheavym@...176...> Le Tuesday 13 November 2007 18:12:08 David Villalobos Cambronero, vous avez ?crit?: > Hi all, > > Some days ago I suscribed to the developer list because I want to start > developing somethig for Gambas, or at least, report a bug and try to fix > it. > > Maybe this post is not Gambas related, (sorry for that) but I want to know > if you can tell me a good Manual to learn C/C++ for Linux, some years ago I > started to develop in C/C++ but always Windows oriented, now I want to > start for Linux. > > Can anyone helpme please??? > > Best regards > > David > > thinking in c++ ? http://www.mindviewinc.com/ -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part. URL: From dcamposf at ...176... Tue Nov 13 18:22:58 2007 From: dcamposf at ...176... (Daniel Campos) Date: Tue, 13 Nov 2007 18:22:58 +0100 Subject: [Gambas-devel] Develop in C/C++ In-Reply-To: <200711131820.06706.lordheavym@...176...> References: <444167.5794.qm@...555...> <200711131820.06706.lordheavym@...176...> Message-ID: <7259b5ae0711130922q88c748amf5a8f8f0f6ab237c@...178...> A good starting point: http://www.ibiblio.org/pub/Linux/docs/linux-doc-project/programmers-guide/lpg-0.4.pdf All the C basis: the different capabilities of the stanrd C library for Unix are here. Start with this book, then try socket programming, GTK+ or QT programming and so on... Daniel 2007/11/13, Laurent Carlier : > Le Tuesday 13 November 2007 18:12:08 David Villalobos Cambronero, vous avez > ?crit: > > Hi all, > > > > Some days ago I suscribed to the developer list because I want to start > > developing somethig for Gambas, or at least, report a bug and try to fix > > it. > > > > Maybe this post is not Gambas related, (sorry for that) but I want to know > > if you can tell me a good Manual to learn C/C++ for Linux, some years ago I > > started to develop in C/C++ but always Windows oriented, now I want to > > start for Linux. > > > > Can anyone helpme please??? > > > > Best regards > > > > David > > > > > > thinking in c++ ? http://www.mindviewinc.com/ > > > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Splunk Inc. > Still grepping through log files to find problems? Stop. > Now Search log events and configuration files using AJAX and a browser. > Download your FREE copy of Splunk now >> http://get.splunk.com/ > _______________________________________________ > Gambas-devel mailing list > Gambas-devel at lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/gambas-devel > > > From david_villalobos_c at ...7... Wed Nov 14 13:28:21 2007 From: david_villalobos_c at ...7... (David Villalobos Cambronero) Date: Wed, 14 Nov 2007 04:28:21 -0800 (PST) Subject: [Gambas-devel] New translation for Spanish Message-ID: <154252.76814.qm@...556...> Hi all, I compliled Gambas yesterday and see some new strings, attached is an updated language file. Can anyone upload it? Best regards. David ____________________________________________________________________________________ Get easy, one-click access to your favorites. Make Yahoo! your homepage. http://www.yahoo.com/r/hs -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: gambas2-es-1.9.91.po Type: application/octet-stream Size: 68830 bytes Desc: not available URL: From gambas at ...1... Thu Nov 15 15:26:08 2007 From: gambas at ...1... (Benoit Minisini) Date: Thu, 15 Nov 2007 15:26:08 +0100 Subject: [Gambas-devel] New translation for Spanish In-Reply-To: <154252.76814.qm@...556...> References: <154252.76814.qm@...556...> Message-ID: <200711151526.08211.gambas@...1...> On mercredi 14 novembre 2007, David Villalobos Cambronero wrote: > Hi all, I compliled Gambas yesterday and see some new strings, attached is > an updated language file. > > Can anyone upload it? > > Best regards. > > David > OK, done. -- Benoit Minisini From jredrejo at ...176... Mon Nov 19 09:39:29 2007 From: jredrejo at ...176... (=?ISO-8859-1?Q?Jos=E9_Luis_Redrejo?=) Date: Mon, 19 Nov 2007 09:39:29 +0100 Subject: [Gambas-devel] Perfomance differences between gb.gtk and gb.qt components Message-ID: <8eb28a500711190039o2b1dd06bod1b995fac21d6b42@...178...> Hi, I'm doing some tests with a pretty complex application made in gambas. It has quite a bunch of classes and about one thousand lines of code. I was surprised when I checked the just starting the application and let it show the graphical interface, without interacting with it, took a lot of cpu use in a quite powerful computer (checked with the gnome-system monitor and later with top), but I was much more surprised when changing from using gb.gtk to gb.qt the cpu use was as little as it can not be appreciate, so it was around 0. This can be easily appreciated just creating a clean new project with a form, a couple of buttons and a textbox, without adding any code. Running it with gb.qt, top only shows gbx2 comsumption at the starting and very fast it disappears from the view, running it using gb.gtk the application stays on top of the processes all the time, with about a 4% of CPU use. So, it seems there is some problem with the event loop in gb.gtk component. I think this is a very important bug in order to use gambas under Gnome, as many users do. Regards. -------------- next part -------------- An HTML attachment was scrubbed... URL: From gambas at ...1... Mon Nov 19 09:46:58 2007 From: gambas at ...1... (Benoit Minisini) Date: Mon, 19 Nov 2007 09:46:58 +0100 Subject: [Gambas-devel] Perfomance differences between gb.gtk and gb.qt components In-Reply-To: <8eb28a500711190039o2b1dd06bod1b995fac21d6b42@...178...> References: <8eb28a500711190039o2b1dd06bod1b995fac21d6b42@...178...> Message-ID: <200711190946.58272.gambas@...1...> On lundi 19 novembre 2007, Jos? Luis Redrejo wrote: > Hi, I'm doing some tests with a pretty complex application made in gambas. > It has quite a bunch of classes and about one thousand lines of code. > I was surprised when I checked the just starting the application and let it > show the graphical interface, without interacting with it, took a lot of > cpu use in a quite powerful computer (checked with the gnome-system monitor > and later with top), but I was much more surprised when changing from using > gb.gtk to gb.qt the cpu use was as little as it can not be appreciate, so > it was around 0. > This can be easily appreciated just creating a clean new project with a > form, a couple of buttons and a textbox, without adding any code. Running > it with gb.qt, top only shows gbx2 comsumption at the starting and very > fast it disappears from the view, running it using gb.gtk the application > stays on top of the processes all the time, with about a 4% of CPU use. > > So, it seems there is some problem with the event loop in gb.gtk component. > I think this is a very important bug in order to use gambas under Gnome, as > many users do. > > Regards. Maybe Daniel could give us some details about the gb.gtk event loop. I noticed that my GTK+ programs were always far slower than my QT programs, using Gambas or not, so I didn't notice that specific event loop problem. Regards, -- Benoit Minisini From jredrejo at ...176... Mon Nov 19 10:00:53 2007 From: jredrejo at ...176... (=?ISO-8859-1?Q?Jos=E9_Luis_Redrejo?=) Date: Mon, 19 Nov 2007 10:00:53 +0100 Subject: [Gambas-devel] Perfomance differences between gb.gtk and gb.qt components In-Reply-To: <200711190946.58272.gambas@...1...> References: <8eb28a500711190039o2b1dd06bod1b995fac21d6b42@...178...> <200711190946.58272.gambas@...1...> Message-ID: <8eb28a500711190100x5eb642d0id1b80c1feb70b9f8@...178...> 2007/11/19, Benoit Minisini : > > On lundi 19 novembre 2007, Jos? Luis Redrejo wrote: > > Hi, I'm doing some tests with a pretty complex application made in > gambas. > > It has quite a bunch of classes and about one thousand lines of code. > > I was surprised when I checked the just starting the application and let > it > > show the graphical interface, without interacting with it, took a lot of > > cpu use in a quite powerful computer (checked with the gnome-system > monitor > > and later with top), but I was much more surprised when changing from > using > > gb.gtk to gb.qt the cpu use was as little as it can not be appreciate, > so > > it was around 0. > > This can be easily appreciated just creating a clean new project with a > > form, a couple of buttons and a textbox, without adding any code. > Running > > it with gb.qt, top only shows gbx2 comsumption at the starting and very > > fast it disappears from the view, running it using gb.gtk the > application > > stays on top of the processes all the time, with about a 4% of CPU use. > > > > So, it seems there is some problem with the event loop in gb.gtkcomponent. > > I think this is a very important bug in order to use gambas under Gnome, > as > > many users do. > > > > Regards. > > Maybe Daniel could give us some details about the gb.gtk event loop. > > I noticed that my GTK+ programs were always far slower than my QT > programs, > using Gambas or not, so I didn't notice that specific event loop problem. > I run mostly gtk programs and can assure you that no one (except firefox when using the flash plugin) takes 6%-15% of the CPU only by showing its interface. I don't think the problem is in gtk, but in the gb.gtk component (well, you can say that gtk makes much harder implementing an event loop but there are much more bindings to gtk that have not that problem). As you say, maybe Daniel can enlighten us. -------------- next part -------------- An HTML attachment was scrubbed... URL: From gambas at ...1... Mon Nov 19 10:05:57 2007 From: gambas at ...1... (Benoit Minisini) Date: Mon, 19 Nov 2007 10:05:57 +0100 Subject: [Gambas-devel] Perfomance differences between gb.gtk and gb.qt components In-Reply-To: <8eb28a500711190100x5eb642d0id1b80c1feb70b9f8@...178...> References: <8eb28a500711190039o2b1dd06bod1b995fac21d6b42@...178...> <200711190946.58272.gambas@...1...> <8eb28a500711190100x5eb642d0id1b80c1feb70b9f8@...178...> Message-ID: <200711191005.57503.gambas@...1...> On lundi 19 novembre 2007, Jos? Luis Redrejo wrote: > 2007/11/19, Benoit Minisini : > > On lundi 19 novembre 2007, Jos? Luis Redrejo wrote: > > > Hi, I'm doing some tests with a pretty complex application made in > > > > gambas. > > > > > It has quite a bunch of classes and about one thousand lines of code. > > > I was surprised when I checked the just starting the application and > > > let > > > > it > > > > > show the graphical interface, without interacting with it, took a lot > > > of cpu use in a quite powerful computer (checked with the gnome-system > > > > monitor > > > > > and later with top), but I was much more surprised when changing from > > > > using > > > > > gb.gtk to gb.qt the cpu use was as little as it can not be appreciate, > > > > so > > > > > it was around 0. > > > This can be easily appreciated just creating a clean new project with a > > > form, a couple of buttons and a textbox, without adding any code. > > > > Running > > > > > it with gb.qt, top only shows gbx2 comsumption at the starting and very > > > fast it disappears from the view, running it using gb.gtk the > > > > application > > > > > stays on top of the processes all the time, with about a 4% of CPU use. > > > > > > So, it seems there is some problem with the event loop in > > > gb.gtkcomponent. I think this is a very important bug in order to use > > > gambas under Gnome, > > > > as > > > > > many users do. > > > > > > Regards. > > > > Maybe Daniel could give us some details about the gb.gtk event loop. > > > > I noticed that my GTK+ programs were always far slower than my QT > > programs, > > using Gambas or not, so I didn't notice that specific event loop problem. > > I run mostly gtk programs and can assure you that no one (except firefox > when using the flash plugin) takes 6%-15% of the CPU only by showing its > interface. I don't think the problem is in gtk, but in the gb.gtk component > (well, you can say that gtk makes much harder implementing an event loop > but there are much more bindings to gtk that have not that problem). As you > say, maybe Daniel can enlighten us. I didn't say that there is no problem in gb.gtk, I just said that I didn't notice because of the general slowness of GTK+ applications on my Athlon 2000. By reading the source code, I know that there is a problem, but I don't know how to fix it yet. -- Benoit Minisini From jredrejo at ...176... Mon Nov 19 10:48:38 2007 From: jredrejo at ...176... (=?ISO-8859-1?Q?Jos=E9_Luis_Redrejo?=) Date: Mon, 19 Nov 2007 10:48:38 +0100 Subject: [Gambas-devel] Perfomance differences between gb.gtk and gb.qt components In-Reply-To: <200711191005.57503.gambas@...1...> References: <8eb28a500711190039o2b1dd06bod1b995fac21d6b42@...178...> <200711190946.58272.gambas@...1...> <8eb28a500711190100x5eb642d0id1b80c1feb70b9f8@...178...> <200711191005.57503.gambas@...1...> Message-ID: <8eb28a500711190148l33820997p7aa7e8d31786463e@...178...> 2007/11/19, Benoit Minisini : > > On lundi 19 novembre 2007, Jos? Luis Redrejo wrote: > > 2007/11/19, Benoit Minisini : > > > On lundi 19 novembre 2007, Jos? Luis Redrejo wrote: > > > > Hi, I'm doing some tests with a pretty complex application made in > > > > > > gambas. > > > > > > > It has quite a bunch of classes and about one thousand lines of > code. > > > > I was surprised when I checked the just starting the application and > > > > let > > > > > > it > > > > > > > show the graphical interface, without interacting with it, took a > lot > > > > of cpu use in a quite powerful computer (checked with the > gnome-system > > > > > > monitor > > > > > > > and later with top), but I was much more surprised when changing > from > > > > > > using > > > > > > > gb.gtk to gb.qt the cpu use was as little as it can not be > appreciate, > > > > > > so > > > > > > > it was around 0. > > > > This can be easily appreciated just creating a clean new project > with a > > > > form, a couple of buttons and a textbox, without adding any code. > > > > > > Running > > > > > > > it with gb.qt, top only shows gbx2 comsumption at the starting and > very > > > > fast it disappears from the view, running it using gb.gtk the > > > > > > application > > > > > > > stays on top of the processes all the time, with about a 4% of CPU > use. > > > > > > > > So, it seems there is some problem with the event loop in > > > > gb.gtkcomponent. I think this is a very important bug in order to > use > > > > gambas under Gnome, > > > > > > as > > > > > > > many users do. > > > > > > > > Regards. > > > > > > Maybe Daniel could give us some details about the gb.gtk event loop. > > > > > > I noticed that my GTK+ programs were always far slower than my QT > > > programs, > > > using Gambas or not, so I didn't notice that specific event loop > problem. > > > > I run mostly gtk programs and can assure you that no one (except firefox > > when using the flash plugin) takes 6%-15% of the CPU only by showing its > > interface. I don't think the problem is in gtk, but in the gb.gtkcomponent > > (well, you can say that gtk makes much harder implementing an event loop > > but there are much more bindings to gtk that have not that problem). As > you > > say, maybe Daniel can enlighten us. > > I didn't say that there is no problem in gb.gtk, I just said that I didn't > notice because of the general slowness of GTK+ applications on my Athlon > 2000. I see By reading the source code, I know that there is a problem, but I don't know > how to fix it yet. > > great, that's already a progress, so Daniel, can you help? -------------- next part -------------- An HTML attachment was scrubbed... URL: From dcamposf at ...176... Mon Nov 19 12:18:18 2007 From: dcamposf at ...176... (Daniel Campos) Date: Mon, 19 Nov 2007 12:18:18 +0100 Subject: [Gambas-devel] Perfomance differences between gb.gtk and gb.qt components In-Reply-To: <8eb28a500711190148l33820997p7aa7e8d31786463e@...178...> References: <8eb28a500711190039o2b1dd06bod1b995fac21d6b42@...178...> <200711190946.58272.gambas@...1...> <8eb28a500711190100x5eb642d0id1b80c1feb70b9f8@...178...> <200711191005.57503.gambas@...1...> <8eb28a500711190148l33820997p7aa7e8d31786463e@...178...> Message-ID: <7259b5ae0711190318o6419838fh6cfba83d7144f09@...178...> I'm working on it... 2007/11/19, Jos? Luis Redrejo : > > > > 2007/11/19, Benoit Minisini : > > On lundi 19 novembre 2007, Jos? Luis Redrejo wrote: > > > 2007/11/19, Benoit Minisini : > > > > On lundi 19 novembre 2007, Jos? Luis Redrejo wrote: > > > > > Hi, I'm doing some tests with a pretty complex application made in > > > > > > > > gambas. > > > > > > > > > It has quite a bunch of classes and about one thousand lines of > code. > > > > > I was surprised when I checked the just starting the application and > > > > > let > > > > > > > > it > > > > > > > > > show the graphical interface, without interacting with it, took a > lot > > > > > of cpu use in a quite powerful computer (checked with the > gnome-system > > > > > > > > monitor > > > > > > > > > and later with top), but I was much more surprised when changing > from > > > > > > > > using > > > > > > > > > gb.gtk to gb.qt the cpu use was as little as it can not be > appreciate, > > > > > > > > so > > > > > > > > > it was around 0. > > > > > This can be easily appreciated just creating a clean new project > with a > > > > > form, a couple of buttons and a textbox, without adding any code. > > > > > > > > Running > > > > > > > > > it with gb.qt, top only shows gbx2 comsumption at the starting and > very > > > > > fast it disappears from the view, running it using gb.gtk the > > > > > > > > application > > > > > > > > > stays on top of the processes all the time, with about a 4% of CPU > use. > > > > > > > > > > So, it seems there is some problem with the event loop in > > > > > gb.gtkcomponent. I think this is a very important bug in order to > use > > > > > gambas under Gnome, > > > > > > > > as > > > > > > > > > many users do. > > > > > > > > > > Regards. > > > > > > > > Maybe Daniel could give us some details about the gb.gtk event loop. > > > > > > > > I noticed that my GTK+ programs were always far slower than my QT > > > > programs, > > > > using Gambas or not, so I didn't notice that specific event loop > problem. > > > > > > I run mostly gtk programs and can assure you that no one (except firefox > > > when using the flash plugin) takes 6%-15% of the CPU only by showing its > > > interface. I don't think the problem is in gtk, but in the gb.gtk > component > > > (well, you can say that gtk makes much harder implementing an event loop > > > but there are much more bindings to gtk that have not that problem). As > you > > > say, maybe Daniel can enlighten us. > > > > I didn't say that there is no problem in gb.gtk, I just said that I didn't > > notice because of the general slowness of GTK+ applications on my Athlon > > 2000. > > > I see > > > > By reading the source code, I know that there is a problem, but I don't > know > > how to fix it yet. > > > > > > great, that's already a progress, so Daniel, can you help? > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2005. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > _______________________________________________ > Gambas-devel mailing list > Gambas-devel at lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/gambas-devel > > From gambas at ...1... Mon Nov 19 12:24:18 2007 From: gambas at ...1... (Benoit Minisini) Date: Mon, 19 Nov 2007 12:24:18 +0100 Subject: [Gambas-devel] Perfomance differences between gb.gtk and gb.qt components In-Reply-To: <7259b5ae0711190318o6419838fh6cfba83d7144f09@...178...> References: <8eb28a500711190039o2b1dd06bod1b995fac21d6b42@...178...> <8eb28a500711190148l33820997p7aa7e8d31786463e@...178...> <7259b5ae0711190318o6419838fh6cfba83d7144f09@...178...> Message-ID: <200711191224.18497.gambas@...1...> On lundi 19 novembre 2007, Daniel Campos wrote: > I'm working on it... > I think the problem comes from the event loop being too active. I think gApplication::iteration() should block while waiting for events. The gbx2 event loop works this way: it blocks on select() while waiting for watched file descriptors, with a timeout if a timer must be triggered. I think the QT event loop works the same way, by watching the X11 file descriptor to know when X events arrive. Regards, -- Benoit Minisini From dcamposf at ...176... Mon Nov 19 13:26:51 2007 From: dcamposf at ...176... (Daniel Campos) Date: Mon, 19 Nov 2007 13:26:51 +0100 Subject: [Gambas-devel] Perfomance differences between gb.gtk and gb.qt components In-Reply-To: <200711191224.18497.gambas@...1...> References: <8eb28a500711190039o2b1dd06bod1b995fac21d6b42@...178...> <8eb28a500711190148l33820997p7aa7e8d31786463e@...178...> <7259b5ae0711190318o6419838fh6cfba83d7144f09@...178...> <200711191224.18497.gambas@...1...> Message-ID: <7259b5ae0711190426y618b8e16qbb117e4c40eae4c5@...178...> > I think the problem comes from the event loop being too active. I think > gApplication::iteration() should block while waiting for events. > > The gbx2 event loop works this way: it blocks on select() while waiting for > watched file descriptors, with a timeout if a timer must be triggered. > > I think the QT event loop works the same way, by watching the X11 file > descriptor to know when X events arrive. > Yes,yes, of course that's the problem, but I have to think in the way to manage watched descriptors if I block the main event loop... Daniel From dcamposf at ...176... Mon Nov 19 14:17:04 2007 From: dcamposf at ...176... (Daniel Campos) Date: Mon, 19 Nov 2007 14:17:04 +0100 Subject: [Gambas-devel] Perfomance differences between gb.gtk and gb.qt components In-Reply-To: <7259b5ae0711190426y618b8e16qbb117e4c40eae4c5@...178...> References: <8eb28a500711190039o2b1dd06bod1b995fac21d6b42@...178...> <8eb28a500711190148l33820997p7aa7e8d31786463e@...178...> <7259b5ae0711190318o6419838fh6cfba83d7144f09@...178...> <200711191224.18497.gambas@...1...> <7259b5ae0711190426y618b8e16qbb117e4c40eae4c5@...178...> Message-ID: <7259b5ae0711190517o168edc5fm987f244000599255@...178...> Well, there are functions to watch file descriptor inide the main event loop of GTK+,I will rewrite that code, but It will take a bit, stay tuned... 2007/11/19, Daniel Campos : > > I think the problem comes from the event loop being too active. I think > > gApplication::iteration() should block while waiting for events. > > > > The gbx2 event loop works this way: it blocks on select() while waiting for > > watched file descriptors, with a timeout if a timer must be triggered. > > > > I think the QT event loop works the same way, by watching the X11 file > > descriptor to know when X events arrive. > > > > Yes,yes, of course that's the problem, but I have to think in the way > to manage watched descriptors if I block the main event loop... > > Daniel > From gambas at ...1... Tue Nov 20 01:05:07 2007 From: gambas at ...1... (Benoit Minisini) Date: Tue, 20 Nov 2007 01:05:07 +0100 Subject: [Gambas-devel] Perfomance differences between gb.gtk and gb.qt components In-Reply-To: <7259b5ae0711190517o168edc5fm987f244000599255@...178...> References: <8eb28a500711190039o2b1dd06bod1b995fac21d6b42@...178...> <7259b5ae0711190426y618b8e16qbb117e4c40eae4c5@...178...> <7259b5ae0711190517o168edc5fm987f244000599255@...178...> Message-ID: <200711200105.08002.gambas@...1...> On lundi 19 novembre 2007, Daniel Campos wrote: > Well, there are functions to watch file descriptor inide the main > event loop of GTK+,I will rewrite that code, but It will take a bit, > stay tuned... > Do you know that there are functions to manage dynamic arrays in the Gambas API? (Be careful, they have nothing to do with the functions that manage Gambas arrays). See http://gambasdoc.org/help/api/cat/marray With that, you don't have to always write the same piece of code with GB.Alloc(), GB.Realloc()... Regards, -- Benoit Minisini From gambas at ...1... Tue Nov 20 01:09:38 2007 From: gambas at ...1... (Benoit Minisini) Date: Tue, 20 Nov 2007 01:09:38 +0100 Subject: [Gambas-devel] Perfomance differences between gb.gtk and gb.qt components In-Reply-To: <7259b5ae0711190517o168edc5fm987f244000599255@...178...> References: <8eb28a500711190039o2b1dd06bod1b995fac21d6b42@...178...> <7259b5ae0711190426y618b8e16qbb117e4c40eae4c5@...178...> <7259b5ae0711190517o168edc5fm987f244000599255@...178...> Message-ID: <200711200109.38901.gambas@...1...> On lundi 19 novembre 2007, Daniel Campos wrote: > Well, there are functions to watch file descriptor inide the main > event loop of GTK+,I will rewrite that code, but It will take a bit, > stay tuned... > With the last revision, I get 100% CPU usage... -- Benoit Minisini From gambas at ...1... Tue Nov 20 02:21:17 2007 From: gambas at ...1... (Benoit Minisini) Date: Tue, 20 Nov 2007 02:21:17 +0100 Subject: [Gambas-devel] Perfomance differences between gb.gtk and gb.qt components In-Reply-To: <200711200109.38901.gambas@...1...> References: <8eb28a500711190039o2b1dd06bod1b995fac21d6b42@...178...> <7259b5ae0711190517o168edc5fm987f244000599255@...178...> <200711200109.38901.gambas@...1...> Message-ID: <200711200221.17454.gambas@...1...> On mardi 20 novembre 2007, Benoit Minisini wrote: > On lundi 19 novembre 2007, Daniel Campos wrote: > > Well, there are functions to watch file descriptor inide the main > > event loop of GTK+,I will rewrite that code, but It will take a bit, > > stay tuned... > > With the last revision, I get 100% CPU usage... If I put a "return;" just to the beginning of the CWatcher::Add() function, then I get normal CPU usage. This function is called because my GTK+ project uses gb.form, that calls external processes during the initialization of the Stock class, and so the file descriptor of the processes input/output are watched. So there is something in g_io_* functions that makes gtk_main_iteration() not blocking after having been used, even if the GIOChannel are correctly unreferenced, and so destroyed. Regards, -- Benoit Minisini From dcamposf at ...176... Tue Nov 20 10:32:25 2007 From: dcamposf at ...176... (Daniel Campos) Date: Tue, 20 Nov 2007 10:32:25 +0100 Subject: [Gambas-devel] Perfomance differences between gb.gtk and gb.qt components In-Reply-To: <200711200221.17454.gambas@...1...> References: <8eb28a500711190039o2b1dd06bod1b995fac21d6b42@...178...> <7259b5ae0711190517o168edc5fm987f244000599255@...178...> <200711200109.38901.gambas@...1...> <200711200221.17454.gambas@...1...> Message-ID: <7259b5ae0711200132m3b2bcbfco8b32fee14a6c8678@...178...> Yes, I see... well, checking again what's happening, the documentation is not clear, as usual Stay tuned :-) 2007/11/20, Benoit Minisini : > On mardi 20 novembre 2007, Benoit Minisini wrote: > > On lundi 19 novembre 2007, Daniel Campos wrote: > > > Well, there are functions to watch file descriptor inide the main > > > event loop of GTK+,I will rewrite that code, but It will take a bit, > > > stay tuned... > > > > With the last revision, I get 100% CPU usage... > > If I put a "return;" just to the beginning of the CWatcher::Add() function, > then I get normal CPU usage. > > This function is called because my GTK+ project uses gb.form, that calls > external processes during the initialization of the Stock class, and so the > file descriptor of the processes input/output are watched. > > So there is something in g_io_* functions that makes gtk_main_iteration() not > blocking after having been used, even if the GIOChannel are correctly > unreferenced, and so destroyed. > > Regards, > > -- > Benoit Minisini > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2005. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > _______________________________________________ > Gambas-devel mailing list > Gambas-devel at lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/gambas-devel > From dcamposf at ...176... Tue Nov 20 11:06:44 2007 From: dcamposf at ...176... (Daniel Campos) Date: Tue, 20 Nov 2007 11:06:44 +0100 Subject: [Gambas-devel] Perfomance differences between gb.gtk and gb.qt components In-Reply-To: <7259b5ae0711200132m3b2bcbfco8b32fee14a6c8678@...178...> References: <8eb28a500711190039o2b1dd06bod1b995fac21d6b42@...178...> <7259b5ae0711190517o168edc5fm987f244000599255@...178...> <200711200109.38901.gambas@...1...> <200711200221.17454.gambas@...1...> <7259b5ae0711200132m3b2bcbfco8b32fee14a6c8678@...178...> Message-ID: <7259b5ae0711200206m9378badgf520331d3d6743aa@...178...> Well, yes, your fix seems to be the only way to free the source... extrange 2007/11/20, Daniel Campos : > Yes, I see... well, checking again what's happening, the documentation > is not clear, as usual > > Stay tuned :-) > > 2007/11/20, Benoit Minisini : > > On mardi 20 novembre 2007, Benoit Minisini wrote: > > > On lundi 19 novembre 2007, Daniel Campos wrote: > > > > Well, there are functions to watch file descriptor inide the main > > > > event loop of GTK+,I will rewrite that code, but It will take a bit, > > > > stay tuned... > > > > > > With the last revision, I get 100% CPU usage... > > > > If I put a "return;" just to the beginning of the CWatcher::Add() function, > > then I get normal CPU usage. > > > > This function is called because my GTK+ project uses gb.form, that calls > > external processes during the initialization of the Stock class, and so the > > file descriptor of the processes input/output are watched. > > > > So there is something in g_io_* functions that makes gtk_main_iteration() not > > blocking after having been used, even if the GIOChannel are correctly > > unreferenced, and so destroyed. > > > > Regards, > > > > -- > > Benoit Minisini > > > > ------------------------------------------------------------------------- > > This SF.net email is sponsored by: Microsoft > > Defy all challenges. Microsoft(R) Visual Studio 2005. > > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > > _______________________________________________ > > Gambas-devel mailing list > > Gambas-devel at lists.sourceforge.net > > https://lists.sourceforge.net/lists/listinfo/gambas-devel > > > From dcamposf at ...176... Wed Nov 21 17:18:27 2007 From: dcamposf at ...176... (Daniel Campos) Date: Wed, 21 Nov 2007 17:18:27 +0100 Subject: [Gambas-devel] Improved gbs2 Message-ID: <7259b5ae0711210818x49923952q43c309093a033187@...178...> Hi: I've made a complete reimplementation of gbs2, the main features: -> It can detect and warn about syntax errors prior to invoke the compiler. -> It is able to redirect the errors from gbc2 to the right source script file in which the error was done, including the right line number. -> As the prior implementation it can use the "INCLUDE" keyword to include more script files, but here you can use "-i=file-to-include" too, so you can reimplement classes and launch the main script using them -> The implementation of a class with a given name can be divided in more than a file, the gbs2 program will join them, so you can reimplement methods including different files, or expand the capabilities of a class with new methods. I send the source code as well as an example with an error, now It is easy to find as you can detect the file and line number in which the problem exists. 1) Compile the gbs2 implementation, then put it into /usr/bin 2) Run the example: gbs2 program.g 3) QUIZ! Fix the error and then run it again Daniel -------------- next part -------------- A non-text attachment was scrubbed... Name: gbs2-1.9.91.tar.gz Type: application/x-gzip Size: 22863 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: example.tar.gz Type: application/x-gzip Size: 673 bytes Desc: not available URL: From gambas.fr at ...176... Wed Nov 21 17:50:16 2007 From: gambas.fr at ...176... (Fabien Bodard) Date: Wed, 21 Nov 2007 17:50:16 +0100 Subject: [Gambas-devel] Improved gbs2 In-Reply-To: <7259b5ae0711210818x49923952q43c309093a033187@...178...> References: <7259b5ae0711210818x49923952q43c309093a033187@...178...> Message-ID: <200711211750.17109.gambas.fr@...176...> Le Wednesday 21 November 2007 17:18:27 Daniel Campos, vous avez ?crit?: > Hi: > > I've made a complete reimplementation of gbs2, the main features: > > -> It can detect and warn about syntax errors prior to invoke the compiler. > > -> It is able to redirect the errors from gbc2 to the right source > script file in which the error was done, including the right line > number. > > -> As the prior implementation it can use the "INCLUDE" keyword to > include more script files, but here you can use "-i=file-to-include" > too, so you can reimplement classes and launch the main script using > them > > -> The implementation of a class with a given name can be divided in > more than a file, the gbs2 program will join them, so you can > reimplement methods including different files, or expand the > capabilities of a class with new methods. > > I send the source code as well as an example with an error, now It is > easy to find as you can detect the file and line number in which the > problem exists. > > 1) Compile the gbs2 implementation, then put it into /usr/bin > 2) Run the example: gbs2 program.g > 3) QUIZ! Fix the error and then run it again > > Daniel Marvellous ! It is now quicker to find error :) Another great job from you Daniel So for that this is a present for you : The kate highlighting template file... Sorry , i've not the gtksourceview one as it is so hard to do... Put it in /usr/share/apps/katepart/syntax To avoid problem on kate color... delete the highlight*.rc file in your home/../.kde/share/app or .kde/conf... i don't remember The .lang file is for gtksourceview... put it in /usr/share/gtksourceview-2.0/language-specs/ But it is not finished... if you have time... -------------- next part -------------- A non-text attachment was scrubbed... Name: gambas.xml Type: text/xml Size: 17752 bytes Desc: not available URL: -------------- next part -------------- application/x-gambasscript *.g '