需求场景：

某个站点删除大量稿件，但是这些稿件已经被百度收录，这样用户访问将会出现404，用户体验不太好，所以需要将删除的稿件生成为xml格式文件，并且每个文件为5000条数据，然后提交至百度进行收录删除。

Shell脚本将普通文件转成xml格式文件

普通文件：

https://www.abc.com//ys/13003183/20191115/123456.html

https://www.abc.com/html/ys/13003183/20191115/123765.html

https://www.abc.com/html/ys/13003183/20191115/567567.html

https://www.abc.com/html/ys/13003183/20191115/456456.html

https://www.abc.com/html/ys/13003183/20191115/374456.html

https://www.abc.com/html/ys/13003183/20191115/37456645.html

xml格式文件：

<url> <loc> https://www.abc.com/html/ys/13003183/20191115/37404973.html </loc> </url>

<url> <loc> https://www.abc.com/html/jb/13003184/20191115/37404988.html </loc> </url>

<url> <loc> https://www.abc.com/html/jb/13003184/20191115/37404968.html </loc> </url>

<url> <loc> https://www.abc.com/ylaq/13003182/20191115/37404860.html </loc> </url>

<url> <loc> https://www.abc.com/ylaq/13003182/20191115/37404861.html </loc> </url>

</urlset>

脚本信息：

cat xml.sh
#!/bin/bash
#
sed -i 's/^/<url> <loc> /g' $1
sed -i 's/$/ </loc> </url>/g' $1
name=`echo $1 | awk -F"." '{print $1}'`
echo $name
split -l 5000 $1 ${name}_xml
for filename in `find ./ -name "${name}_xml*"`
do
 sed -i '1 i\<urlset>' $filename
 echo "</urlset>" >> $filename
 mv $filename ${filename}.xml
done