Sorry, this entry is only available in Français.
This directory contains several files where that’s possible to select desired fonts by prefered order.
Cyrillic mongolian input method for Ibus
I made a packet ibus-table-mongolian or Maral (Марал), cyrillique mongolian input method for ibus for ArchLinux (tested on ARM with ArchLinuxARM and x86_64 (with Antergos) in fact in “any”), but I obtain something strange. I will probably have to make a bug report.
Keyboard in this method should be disposition independant, it should be only based on US keyboard and don’t adapt to local keyboard, because the phonetic disposition is totally unrelated to latin caracter phonetic disposition. Mongolian keyboard strangly works perfectly with on ARM architecture, but adapt on locale keyboard on x86_64. The same problem appear on Ubuntu 14.04 for x86_64 that uses a really older version.
The defaut configuration file by author, Ochko, use the former ibus-table 1.3 version :
KEYBOARD_LAYOUT = US_Default
On more recent version, this should be :
LAYOUT = us
I trie both definitions, and each time, I have the same difference between ARM et x86_64. I suppose this is a subtile problem like related to Endianess or something like that ça, but I don’t understand why for? You can still use default linux mongolian keyboard that works perfectly fine, but change between method is not smooth like with ibus.
Update of some archlinuxARM packages
Download all this stuff
Thanks to the hardware and software development work of AllWinner, to the Linux mainlining effort of the Linux-sunxi community to improve and push driver into main line Linux kernel sources and to the integration work of ArchLinux and ArchLinuxARM communities, the Cubieboard2 (using Allwinner A20 SoC) since this night officially use 4.0 kernel as default with ArchlinuxARM. Before this date an old Linux kernel 3.x branch was used.
* Linux XXX 4.0.5-2-ARCH #1 SMP Fri Jun 12 20:03:44 MDT 2015 armv7l GNU/Linux
This means that :
* Lots of NAND flash can be used for system
* CPU frequency is well managed now
* Most of driver are integrated and probably really stable.
25 june update, 11 days after the 4.0 : A new update occured today with linux kernel 4.1
* Linux XXX 4.1.0-1-ARCH #1 SMP Tue Jun 23 23:24:14 MDT 2015 armv7l GNU/Linux).
After a fist test of Inkscape 0.91pre, due to the number of changes, and improvement of reactivity, to resolved bugs, better multithread (with OpenMP), etc… I wanted to try it on my Cubieboard2, where it was usable for basic function, but a little slow for some advanced fucntions (like calligraphic tool).
I then compiled it on ArchlinuxARM to test it on my Cubieboard2, I put the paquet in mon micro-reprository ArchlinuxARM, because I still didn’t understand the process to submit it in yaourthouse of AUR (Arch User Repository) or inArchlinuxARM repository.
So then… that’s really fast. This uses both 2 Cortex-A7 CPU (need to be tested on quadcore far faster Cortex-A17 architectures like Rockchip RK3288 like Rikomagic MK802II LE (for Linux Edition), see demo) :). That’s now fully usable, blurs are still slow to draw, but display is relatively asyncrone.
For information, my video is recorded on a x86_64 with application remote display, using X11/tcp (via 100 Mb switch ), this explain some slowness and time shift in display.