Re: [MSEide-MSEgui-talk] Rounded slider button
I'm polishing slider button. Made new shape bitmaps with gimp. This time gimp created different pixels on each side, some of them have transparency (base color was black for circle shape, it is a result of smooth / antyalias option). TImagelist size is 7x7 because this is size of top-left, bottom-left, bottom-right and top-right corners. For rest edges it is enough to have 1px width/height because I can use "stretch" option while adding png files into timagelist. Unfortunately these 1 pixel bitmaps have transparency. When adding them as stretched, they becomes some kind of gradient. Don't know if this is just normal stretch policy (algorithm) or I need to check something when exporting to PNG in gimp. Currently my options are as on second screenshot. Translation from Polish is "save background color" and "save color's values for transparent pixels", compression is set to 0. In first screenshot, indexes 1 and 3 are added with stretch mode, they sizes are 7x1 and 1x7, just simply one black pixel with transparency arround 60%. I can solve this by making these 1x7px bitmaps as 7x7 by simply copy and paste in gimp and add without stretch check. That is not a problem because size of my slider buttons is fixed. But just wondering if it's possible to add such bitmaps as "repeated", not "stretched". If not then it could be cool feature, for example make small pattern in gimp which is repeated and looks uniformly, like grip in splitters. BTW: How to make transparent colorty at runtime? For example, I would like to make shape in gimp with base color stricte black (it may have black transparency pixels due to smoothing / antyaliasing). Add it to image list as base shape and then manipulate at runtime, for example set color to green (or still black) with alpha 50%: timagelist.bitmap.init(rgbtocolor(0,127,0,50)). Can't find any rgbtocolor related function with alpha param -- Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot___ mseide-msegui-talk mailing list mseide-msegui-talk@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mseide-msegui-talk
Re: [MSEide-MSEgui-talk] Bugs I mentioned in fgGUI Android thread
On 04/04/2017 11:39 AM, Martin Schreiber wrote: > On Tuesday 04 April 2017 17:12:37 Jon Foster wrote: >> Martin, >> >> Per your request in the discussion we were having in the fpGUI news groups >> about the "need" for form files and streaming form layouts I will detail >> what I remember of my trial runs of recent MSEgui versions. I used 4.4 >> (4.4.1 I think) and 3.8.8. 4.4.x worked better in the visuals department. >> The problems I ran into fall roughly in three categories: > [...] > > Which operating system/window manager? > Can you try again with current git master version: > https://gitlab.com/mseide-msegui/mseide-msegui/tree/master > so we can sort out the problems? > Please compile MSEide in clone-directory with > fpc -Fulib/common/* -Fulib/common/kernel/linux apps/ide/mseide.pas OK, I cloned the git repo and compiled MSEide. Almost all of the problems persist. Still too numerous for me to do much with it. My current desktop setup and the one I've been using for years: Debian 7 i386. Enlightenment 0.16.x (aka E16-1.0.0). Intel Core 2 x 4 @3GHz. Using Xorg 1.12.4 with ATI video card & 2 monitors. Looks like it is using the Xorg supplied ATI Radeon driver, with RandR and such. Also have an Intel based notebook in basically the same configuration... except its using the "fbdev" video driver instead of Intel's since it was too buggy. The hardware is too new for the driver. >> 1. Z-Ordering problems: Forms falling / rising unexpectedly. The worst >> [...] > Maybe a window manager with buggy NET_RESTACK_WINDOW protocol handling. > Starting MSEgui applications with --NORESTACKWINDOW activates a workaround, > see README.TXT. > >> 2. Form content disappearing. Particularly with the 3.x version using the >> [...] > That looks similar like an effect of a buggy RADEON driver > https://bugs.freedesktop.org/show_bug.cgi?id=69543 > https://bugs.freedesktop.org/show_bug.cgi?id=84253 > Switching off EXAPIXMAPS helps, see README.TXT. Fascinating this appears to be the case. The fbdev driver on the notebook does not have this issue. Things sure seem to have gone downhill since Xorg took over for Xfree86. Assuming your also correct about WM issues with #1 above I still have to say that all of the other software on my system is running without these issues. I'm not going to invest any more time into this ... unless I can find the time to get the current Enlightenment going. Everything is too fine tuned to disrupt. > >> 3. Won't connect to MySQL: In 3.x I'd get an "Access Violation" after >> setting the properties and then setting the "connected" property to true. >> [...] > That should already been fixed. > http://www.mail-archive.com/mseide-msegui-talk@lists.sourceforge.net/msg10791.html > A heavy regression, sorry. This is fixed in "master". > [...] THX! -- Jon Foster JF Possibilities, Inc. -- Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot ___ mseide-msegui-talk mailing list mseide-msegui-talk@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mseide-msegui-talk
