treeVw.set_model(None) batch_insertdata(dataList) treeVw.set_model(dataList)
creating project
$ django-admin startproject DOC $ ls DOC/ __init__.py manage.py settings.py urls.pytest the server
DOC$ ./manage.py runserverget this page http://localhost:8000/
 |
| untouch project page |
$ cd DOC/ DOC$ ./manage.py startapp doc_ui new folder doc_ui created DOC$ ls doc_ui __init__.py __init__.pyc manage.py settings.py settings.pyc urls.py DOC$ ls doc_ui/ __init__.py models.py tests.py views.pyshowing a "hello, world":
(r'^doc_ui/', 'DOC.doc_ui.views.index'),to url_patterns in file urls.py
'DOC.doc_ui',to INSTALLED_APPS in file settings.py
from django.http import HttpResponse
def index(request):
return HttpResponse("Hello, world.")
run/restart the server
DOC$ ./manage.py runserver
Before shorten menu
Two rows of toolbar while working with xournal on small screen might be difficult. They consume too much space. Xournal manual point out how to hide these toolbar.
check your ~user/.xournal/config. It can be generated by saving preference.
Then edit the value of this line:
shorten_menu_items=toolbarMain toolbarPen hbox1
toolbarMain for top toolbar
toolbarPen for below top toolbar
hbox1 for status bar
After shorten menu
I got these values from xournal.glade of the xournal source code as pointed out by xournal user manual. The lines on xournal.glade are:
widget class="GtkToolbar" id="toolbarMain"
widget class="GtkToolbar" id="toolbarPen"
widget class="GtkHBox" id="hbox1"
last updated: 1,2 May 2011
Update: I was wrong. It was caused by additional input device that shift the device numbering N on /dev/input/eventN. However the BIOS rumour is not wrong.
Multiprocessors are now dominant, but real multiprocessors do not provide the sequentially consistent memory that is assumed by most work on semantics and verification. Instead, they have subtle relaxed (or weak) memory models, usually described only in ambiguous (and sometimes flawed) prose, leading to widespread confusion.
It has been around 1 year I've been using my USB-disk linux installation as my main computing platform. Generally it is slow, particularly for writing. I even feel it runs much slower during shutdown. Just like general flash-disk devices, USB-disks have limited read-write lifetime, therefore I try to minimize any access to the disk. The idea is maximizing the usage of memory for read-write operations that occur very often.
Here are my configuration:
Note: I usually use those two browser, haven't found out where are the cache of opera nor firefox/iceweasel.
...
### END INIT INFO
. /lib/init/vars.sh
. /lib/lsb/init-functions
make_tmp() {
#preparing /tmp
mkdir /dev/shm/tmp
chmod 777 /dev/shm/tmp
#we also prepare the browsers cache
mkdir /dev/shm/tmp/google-chrome
chmod 777 /dev/shm/tmp/google-chrome
mkdir /dev/shm/tmp/epiphany
chmod 777 /dev/shm/tmp/epiphany
}
case "$1" in
start)
# Using 'no !=' instead of 'yes =' to make sure async nfs
# mounting is the default even without a value in
# /etc/default/rcS
make_tmp
;;
...
modprobe slram map=VRAM,0xc4000000,+0x0c000000
modprobe mtdblock
mkswap /dev/mtdblock0
swapon /dev/mtdblock0
swapon -s
lp -d dest_print -o media=a4 -o sides=two-sided-long-edge -P 1-3 file.pdf
lp -d dest_print -o media=a4 -o sides=two-sided-short-edge -o number-up=2 -P 1-4 file.pdf
Last update: 22 Sept 2010, 1 Oct 2010, 2,4,17 Jan 2011, 1 Feb 2011, 25 March 2011
cp libflashplayer.so /usr/lib/mozilla/plugins/
alias snd-card-0 snd-hda-intelinto
options snd-hda-intel index=0
alias snd-card-0 snd-hda-intel
options snd-hda-intel model=laptop
make clean && make mrproper
make menuconfig
kompie# ls /usr/lib|grep gsl
libgsl.a
libgslcblas.a
libgslcblas.so
libgslcblas.so.0
libgslcblas.so.0.0.0
libgsl.so
libgsl.so.0
libgsl.so.0.12.0
kompie# rm /usr/lib/libgsl*
kompie# ls /usr/lib|grep gsl
Warning!!! All this hacking trick DID NOT work! Finally I do this following to compile my GSL library into 32 bit
kompie# cd gsl-1.0/
kompie/gsl-1.0# cp .libs/libgsl.a /usr/lib
kompie/gsl-1.0# cp .libs/libgsl.so.0.0.0 /usr/lib
kompie/gsl-1.0# cp cblas/.libs/libgslcblas.
libgslcblas.a libgslcblas.lai libgslcblas.so.0
libgslcblas.la libgslcblas.so libgslcblas.so.0.0.0
kompie/gsl-1.0# cp cblas/.libs/libgslcblas.a /usr/lib
kompie/gsl-1.0# cp cblas/.libs/libgslcblas.so.0.0.0 /usr/lib/
kompie/gsl-1.0# ls /usr/lib|grep gsl
libgsl.a
libgslcblas.a
libgslcblas.so.0.0.0
libgsl.so.0.0.0
kompie/gsl-1.0# ldconfig
kompie/gsl-1.0# ls /usr/lib|grep gsl
libgsl.a
libgslcblas.a
libgslcblas.so.0
libgslcblas.so.0.0.0
libgsl.so.0
libgsl.so.0.0.0
kompie/gsl-1.0#
$ CFLAGS="-m32 -g -O2" LDFLAGS="-m32" ./configure i386
Section "Device"
Identifier "Configured Video Device"
Driver "intel"
EndSection
Note that the log from /var/log/Xorg.0.log has an entry "Setting screen physical size to 270 x 203"
Option "MinX" "50"
Option "MaxX" "3900"
Option "MinY" "150"
Option "MaxY" "4050"
xrandr -o left
kernel /boot/vmlinuz-2.6.35.3-intel-amd64 root=UUID-...-xxxx rothen change into [the "=" sign is not a cosmetic]
kernel /boot/vmlinuz-2.6.35.3-intel-amd64 root=UUID-...-xxxx ro acpi_osi=
root# echo BRIGHTNESS > /sys/class/backlight/panasonic/brightnessNote root access is mandatory, therefore we need to fix the acpi script ... where BRIGHTNESS value between 0 to 21
First of all, in debian, 64 bit for ordinary intel processor is denoted as AMD64
instead of IA64, which is totally different architecture that is die casted
on unpopular intel Itanium processor only.
The procedure is quite simple.
Last update: 13 Feb 2011
I wrote several simple scripts to manipulate the content of files. Most of them are written in bash, but due to my lack of expertise sometimes I mixed bash with make, and perl. It will be very practical.. almost no explanation, but I'll try to gradually improve this article. Here are they:
find . -type f -name "*.bash" -print0|xargs -0 grep -ne "keyword"
#!/bin/bash
TMPDIR=/tmp
fname=${1}
old=${2}
new=${3}
tmpfile=${TMPDIR}/$$.tmp
#echo "ls ${fname}"
for mydataset in `ls ${fname}`; do
echo "${mydataset}"
cat ${mydataset} | sed -e "s/${old}/${new}/g" > ${tmpfile}
mv ${tmpfile} ${mydataset}
done
clean:
find . -name "*.log" -print -exec rm {} \;
find . -name "*.[chCH]" -print | etags -
etags file1 file2 ...
:%s/[0-9]*\/[0-9]*\t//g
cat fileinput |sed -n 'H;${g;s/\n/crlf/g;p}'| sed -n 'H;${g;s/crlfcrlf/\n\n/g;p}'| sed -n 'H;${g;s/crlf/ /g;p}' > file output
mogrify -resize 1024 *.jpg
montage *.jpg -tile 1x2 -geometry 1024 combine.jpg
mogrify -format pdf *.jpg
#usage:
# want to download albaqarah ayat 1 to 10
# to download other ayat you must change this script accordingly
# just to list the download commands
#$ bash dlayat.sh 1 10
# actually download
#$ bash dlayat.sh 1 10 download
i="${1}"
j="${2}"
download=0
if [ "${3}" == "download" ]
then
download=1
echo download
fi
while [ $i -le $j ]
do
padd=""
if [ $i -lt 100 ]
then
padd="0"
fi
if [ $i -lt 10 ]
then
padd="00"
fi
if [ $download -eq 1 ]
then
wget http://www.everyayah.com/data/Ghamadi_40kbps/002${padd}${i}.mp3
else
echo wget http://www.everyayah.com/data/Ghamadi_40kbps/002${padd}${i}.mp3
fi
i=$[$i+1]
done
#change the raw file extension accordingly
thepath=${1}
for file in `ls ${thepath}*.CR2`; do
ufraw-batch --out-type=jpg ${file}
done
