2) First Solution – Conky and devilspie
2）第一种解决方法 – Conky和devilspie

By default, conky will draw to the “root” window (your desktop). This is what causes the conflict with nautilus. That being said, conky can operate fairly well with just it’s -o (own window) option.
conky默认会描绘“根”窗口（你的桌面）。这就是与nautilus冲突的原因。也就是说conky加上-o（own window）参数将能够很好的清楚的运行。

NOTE: In order to get conky to draw it’s own window, you must either run conky as `conky -o` or set “own_window yes” in your .conkyrc
注解：为了让conky描绘自己的窗口（own window），你必须使用‘conky -o’运行conky或者在~/.conkyrc中设置成“own_window yes”

The only problem with telling conky to draw it’s own window in gnome, is that it will appear on only a single virtual desktop, and it will show on the pager and takbar. In order to fix this, you must use a Window Management program (some examples include wmctrl and devilspie). For this example, I will show you how to configure devilspie (versions <=0.12 and >=0.13) to spread conky across all desktops and hide it from the pager and taskbar. First, Install devilspie. (This may vary by distribution)
告诉conky在gnome中描绘自己的own window的唯一问题是它只能显示在单一的虚拟桌面上，并且它将显示在页面调度程序和任务栏里。为了修复这些，你必须使用窗口管理程序（例如 wmctrl和devilspie）。在这个例子里，我会向你展示如何配置devilspie（版本<=0.12 和 >=0.13）让conky穿越所有桌面并且在页面调度程序和任务栏里隐藏。首先，安装devilspie。（不同发行版本的安装方法是多样化的）

Homepage: http://www.burtonini.com/blog/computers/devilspie
主页：http://www.burtonini.com/blog/computers/devilspie
Gentoo Install: emerge -av devilspie
Gentoo安装：emerge -av devilspie
Debian Install: apt-get install devilspie
Debian安装：apt-get install devilspie
All others: Find a binary package for your distribuiton or build from source, that is beyond the scope of this document.
其它：找到你所使用发行版本的二进制包或者从源码编译，这些超出了本文的范围。

Next, you will have to configure devilspie. Please make sure you configure it for the CORRECT VERSION. The configuration CHANGED as of devilspie version 0.13, and it is NOT compatible with previous configurations.
接着，你需要配置devilspie。请确定你所配置的是正确版本。0.13的版本配置有所改变，它将不在兼容与以前的配置文件。

Version 0.12 and below:
版本0.12和以下：

Create a file called .devilspie.xml in your user’s home (~) directory. A basic configuration of devilspie is as follows:
在你的家目录下创建一个名为.devilspie.xml的文件。基本的配置如下：

Code: <?xml version=”1.0″?>
<!DOCTYPE devilspie SYSTEM “devilspie.dtd”>
<devilspie>
<!– This flurb configures devilspie to watch for a window called “conky” –>
<flurb name=”conky”>
<matchers>
<matcher name=”DevilsPieMatcherWindowName”>
<property name=”application_name” value=”conky”/>
</matcher>
</matchers>
<!– The following controls what actions are preformed –>
<actions>
<!– Hide conky from the taskbar and the pager –>
<action name=”DevilsPieActionHide”>
<property name=”skip-tasklist” value=”TRUE”/>
<property name=”skip-pager” value=”TRUE”/>
</action>
<!– Show conky on all workspaces –>
<action name=”DevilsPieActionSetWorkspace”>
<property name=”pinned” value=”TRUE”/>
</action>
</actions>
</flurb>
</devilspie>

Version 0.13 and later:
版本0.13和以后：

Create a file called conky.ds in a directory called .devilspie your user’s home (~) directory (to create the directory, use `mkdir .devilspie` and to create the file, `touch .devilspie/conky.ds`). A basic configuration of devilspie is as follows:
在你的家目录下创建一个名为.devilspie的目录，并在其下创建conky.ds的配置文件（使用‘mkdir .devilspie’创建目录，使用‘touch .devilspie/conky.ds’创建文件）。基本的devilspie配置如下：

Code:

(if (matches (window_name) “.*conky”) (begin pin (skip_pager) (skip_tasklist)))

Configuring Gnome
配置Gnome

Once conky and devilspie are properly configured, you can make them both automatically start when Gnome starts.
一旦conky和evilspie配置好了，你就能让它们在Gnome启动的时候同时自动运行。
Click Desktop -> Preferences -> Sessions, select the “Startup Programs” tab, then click the “Add” button – use “devilspie” for the startup command, then click “Ok”. Click the “Add” button again, and use “conky” for the startup command (or use “conky -o” if you didn’t set “own_window yes” in .conkyrc) click close, and the next time you login, devilspie and conky will automatically start and work right!
点击 桌面 ->首选项 ->会话，选择”启动程序“栏，点击”添加“按钮 – 在启动命令中输入”devilspie“，然后点击确定。再以同样的方式添加”conky“（如果在.conkyrc中没有设置”own_window yes“，输入”conky -o“）点击关闭，这样当你下次登陆时，devilspie和conky将会自动启动并正确工作！

3) Second Solution – Disabling Nautilus
3） 第二种解决方法 – 禁用Nautilus

Rather than installing devilspie (or another window management application), some users may find it helpful to instead keep nautilus from drawing to the root window. To do this, run gconf-editor and uncheck show_desktop in Apps/Nautilus/Preferences/.
与其安装devilspie（或者其它窗口管理程序），一些使用者发现不如禁止nautilus绘制根窗口来得有用。为了做到这些，运行配置编辑器（gconf-editor），在App/Nautilus/Perferences里的show_desktop里设置。

The downside of disabling nautilus from drawing to the root window is that you will no longer have icons on the desktop, or be able to right-click on the desktop.
禁止nautilus绘制根窗口的结果，你将不在拥有桌面上的图标或者是右键点击。

This project is also generously hosted by sourceforge. SourceForge.net Logo

Valid XHTML 1.0 Strict Valid CSS!

#保存mp3的目录
SAVE=”${HOME}/baidump3″

#下载重试次数
TRYCOUNT=2

#临时目录
TMP=”/tmp/${USER}”

#创建下载目录
if [ ! -d “${SAVE}” ];then
mkdir -p “${SAVE}”
fi

#创建临时下载目录
if [ -d “${TMP}” ];then
rm -rf “${TMP}”
fi
mkdir -p “${TMP}”

echo “开始下载百度最新100首歌曲列表”
wget -O ${TMP}/mp3.html ${SOURCE}
echo “下载百度最新100首歌曲列表完成。”

#转换网页编码
iconv -f gbk -t utf8 ${TMP}/mp3.html |\

grep ” href=\”http://mp3.baidu.com/m” |\

#将mp3list.txt所有开头的空格去掉
sed -e ‘s/ *//’ |\

#将mp3list.txt所有开头的tab去掉
sed -e ‘s/\t*//’ |\

#将mp3list.txt所有全角空格去掉
sed -e ‘s/　//g’ |\

#将所有的回车符去掉
sed ‘:a;N;$!ba;s/\n/,/g’ |\

#在td>,后面加上回车符，一行表示一个mp3文件。
sed -e ‘s/,<td/\n<td/g’ |\
sed -e ‘s/td>,/td>\n/g’ |\

#删除<td width=”30%”> <td> </td> <td…FFFFFF”> <p> </p>
sed -e ‘s/<td width=”30%”>//g’ |\
sed -e ‘s/<td>//g’ |\
sed -e ‘s/<\/td>//g’ |\
sed -e ‘s/<p>//g’ |\
sed -e ‘s/<\/p>//g’ |\
sed -e ‘s/<td.*”border”>//g’ |\

#删除</a>…”_blank”>
sed -e ‘s/<\/A>\/<A.*_blank>/、/g’ |\
sed -e ‘s/<\/A>/<\/a>/g’ |\
sed -e ‘s/<\/a>.*_blank>/-/g’ |\
#sed -e ‘s/<\/a>.*_blank”>/-/g’ |\
#删除)
sed -e ‘s/<\/a>)/<\/a>/g’ |\

#删除&
sed -e ‘s/\&\;/\//g’ >${TMP}/mp3list.txt

#得到：<a href=”http://mp3.baidu.com/m?tn=baidump3&ct=134217728&lm=-1&li=2&word=Baby%20Baby%20tell%20me%20%CD%F5%D0%C4%C1%E8″ target=”_blank”>Baby ,Baby tell me-王心凌</a>

#取得行号，循环
line=$(awk ‘END{print NR}’ ${TMP}/mp3list.txt)
i=1;
while((i<=line));do
downed=0;
mpline=`awk ‘NR=='”$i”” ${TMP}/mp3list.txt`
url=`echo $mpline | sed -e ‘s/<a href=”//g’ | sed ‘s/\ target.*//g’ | sed ‘s/”//g’ | cat`
name=`echo $mpline | sed -e ‘s/.*_blank”>//g’ | sed -e ‘s/.*_blank>//g’ |\
sed -e ‘s/<\/a>//g’ | sed -e ‘s/\//-/g’ | sed -e ‘s/:/-/g’ | sed -e ‘s/”/’\”/g’ | cat`

#检查是否已经下载过这首歌，如果下载过，放弃
if [ -e “${SAVE}/${name}.mp3” ] || [ -e “${SAVE}/${name}.wma” ]; then
echo -e “\e[1;6m\e[1;31m发现 ${name} 下载过，忽略，继续下一首。\e[1;6m\e[00m”
((i++))
continue;
fi

echo “开始通过 $url 下载 $name”;
wget -O ${TMP}/down.html $url
echo “获取 $name 下载列表完成。”;

#down.txt为有效的下载地址
iconv -f gbk -t utf8 -c ${TMP}/down.html | grep “onclick=\”return ow(this)\”” |\
sed -e ‘s/.*<a href=”//g’ | sed -e ‘s/” target=”_blank”.*//g’ > ${TMP}/down.txt

#size.txt为有效的下载文件大小
iconv -f gbk -t utf8 -c ${TMP}/down.html | grep “M<\/td>” |\
sed -e ‘s/<td>//g’ | sed -e ‘s/ M<\/td>//g’ > ${TMP}/size.txt

#down.txt与size.txt合并而在的down_size.txt文件中字段之间以”`”作为分隔符
paste -d ‘`’ ${TMP}/size.txt ${TMP}/down.txt > ${TMP}/down_size.txt

#排序
sort -n -r ${TMP}/down_size.txt > ${TMP}/down_size_sort.txt

#去掉后面的尺寸
sed ‘s/.*`//’ ${TMP}/down_size_sort.txt > ${TMP}/temp.txt

##### 析取出mp3 的下载地址或 wma的下载地址 ##############
grep -i “word=mp3” ${TMP}/temp.txt > ${TMP}/down_mp3.txt
grep -i “word=wma” ${TMP}/temp.txt > ${TMP}/down_wma.txt

downline_mp3=$(awk ‘END{print NR}’ ${TMP}/down_mp3.txt);
downline_wma=$(awk ‘END{print NR}’ ${TMP}/down_wma.txt);
echo -e “\e[1;6m\e[1;31m发现 ${downline_mp3} 个名为 ${name}.mp3 下载地址。\e[1;6m\e[00m”
echo -e “\e[1;6m\e[1;31m发现 ${downline_wma} 个名为 ${name}.wma 下载地址。\e[1;6m\e[00m”
# 初始化计数器
j=1;
# 优先下载mp3格式的歌曲
while((j<=downline_mp3)); do
mp3=`awk ‘NR=='”$j”” ${TMP}/down_mp3.txt | sed -e ‘s/ /\\ /g’`
echo -e “\e[1;6m\e[1;31m正在下载${name}.mp3\e[1;6m\e[00m”
#echo -e “\e[1;6m\e[1;31m中转页面地址为${mp3}\e[1;6m\e[00m”
wget -O “${TMP}/transit.html” “$mp3”
realURL=`iconv -f gbk -t utf8 -c ${TMP}/transit.html | grep “<li class=\”li\” style=\”margin-right:10px;\”>” | sed ‘s/.*href=”//’ | sed ‘s/” target=”_blank”>.*//’`
#echo -e “\e[1;6m\e[1;31m真实下载地址为${realURL}\e[1;6m\e[00m”
if [ x`which axel` != x”” ];then
axel -n 7 -a -o “${TMP}/${name}.mp3” “${realURL}”
else
#wget太慢了。但是如果没有安装axel,可以把上面一行注释掉，用下面一行代替
wget -c –tries=$TRYCOUNT $realURL -O “${TMP}/${name}.mp3”
fi
if [ “$?” = 0 ]; then
if [ `file -ib “${TMP}/${name}.mp3” | sed -e ‘s/\/.*//g’` = “audio” ]; then
if [ x`which mid3v2` != x”” ];then
title=`echo $name | sed -e ‘s/-.*//g’`
artist=`echo $name | sed -e ‘s/.*-//g’ | sed -e ‘s/.mp3//g’ | sed -e ‘s/.wma//g’`
mid3v2 -D “${TMP}/${name}.mp3”
mid3v2 -t “${title}” -a “${artist}” “${TMP}/${name}.mp3”
fi
mv “${TMP}/${name}.mp3” “${SAVE}/${name}.mp3”
downed=1;
break;
else
echo -e “\e[1;6m\e[1;31m下载 ${name}.mp3 文件无效，正在删除重新下载\e[1;6m\e[00m”
rm “${TMP}/${name}.mp3”;
((j++))
fi
else
echo -e “\e[1;6m\e[1;31m下载 ${name}.mp3 文件无效，正在删除重新下载\e[1;6m\e[00m”
rm “${TMP}/${name}.mp3”;
((j++))
fi
done

#如果下载成功继续下其余的歌
#continue用于跳过循环体中的后续命令
if [ “$downed” = 1 ] ; then
((i++))
echo -e “\e[1;7m\e[1;41m下载 $name 成功\e[1;7m\e[00m”
continue;
fi

# 如果没有mp3格式的则下载wma格式的歌
j=1;
while((j<=downline_wma)); do
wma=`awk ‘NR=='”$j”” ${TMP}/down_wma.txt`
echo -e “\e[1;6m\e[1;31m正在下载${name}.wma\e[1;6m\e[00m”

#echo -e “\e[1;6m\e[1;31m中转页面地址为${mp3}\e[1;6m\e[00m”
wget -O “${TMP}/transit.html” “$wma”
realURL=`iconv -f gbk -t utf8 -c ${TMP}/transit.html | grep “<li class=\”li\” style=\”margin-right:10px;\”>” | sed ‘s/.*href=”//’ | sed ‘s/” target=”_blank”>.*//’`
#echo -e “\e[1;6m\e[1;31m真实下载地址为${realURL}\e[1;6m\e[00m”
if [ x`which axel` != x”” ];then
axel -n 7 -a -o “${TMP}/${name}.wma” “${realURL}”
else
#wget太慢了。但是如果没有安装axel,可以把上面一行注释掉，用下面一行代替
wget -c –tries=$TRYCOUNT $realURL -O “${TMP}/${name}.wma”
fi
if [ “$?” = 0 ]; then
if [ `file -ib “${TMP}/${name}.wma” | sed -e ‘s/\/.*//g’` = “application” ]; then
#title=`echo $name | sed -e ‘s/-.*//g’`
#artist=`echo $name | sed -e ‘s/.*-//g’ | sed -e ‘s/.mp3//g’ | sed -e ‘s/.wma//g’`
#mid3v2 -D “${TMP}/${name}.wma”
#mid3v2 -t “${title}” -a “${artist}” “${TMP}/${name}.wma”
mv “${TMP}/${name}.wma” “${SAVE}/${name}.wma”
downed=1;
break;
else
echo -e “\e[1;6m\e[1;31m下载 ${name}.wma 文件无效，正在删除重新下载\e[1;6m\e[00m”
rm “${TMP}/${name}.wma”;
((j++))
fi
else
echo -e “\e[1;6m\e[1;31m下载 ${name}.wma 文件无效，正在删除重新下载\e[1;6m\e[00m”
rm “${TMP}/${name}.wma”;
((j++))
fi
done

((i++))
if [ “$downed” = 1 ] ; then
echo -e “\e[1;7m\e[1;41m下载 $name 成功\e[1;7m\e[00m”
else
echo -e “\e[1;7m\e[1;41m下载 $name 失败\e[1;7m\e[00m”
fi
done
rm -fr ${TMP}
exit 0