请教shell读写XML问题

现有 123.xml文件，内容是：
<?xml version="1.0" encoding="GBK"?>
<videoinfo>
<videoname>呵呵</videoname>
<videopath>20080110</videopath>
<videosize>0.0</videosize>
<widthOfImage>320</widthOfImage>
<heightOfImage>240</heightOfImage>
<kbps>260</kbps>
<logo>260</logo>
<videodesc>呵呵恶化和</videodesc>
<state>1</state>
<uploadtime>Thu Jan 10 16:00:53 CST 2008</uploadtime>
</videoinfo>

我想用shell 脚本编写一个程序，修改里面的内容。比如修改<state>1</state>为<state>2</state> ，
修改<videosize>0.0</videosize>为<videosize>333</videosize> 该怎么操作呢。谢谢各位高手。

davistar 发表于 2008-01-10 17:47

eg
[code]
awk '{if(/state/){sub(/>[^<]*</,">2<")}print}' urfile
sed 's//(<state>/)[^<]*/(<//state>/)//12/2/' urfile
[/code]

hyhappy 发表于 2008-01-10 17:53

楼上的好厉害，只是看的不太明白。能大概给解释一下吗？
awk '{if(/state/){sub(/>[^<]*</,">2<")}print}' urfile
sed 's//(<state>/)[^<]*/(<//state>/)//12/2/' urfile 这两行是都可以实现上述操作吗？我第一个测试可以，第2个测试没成功。

linuxnextyear 发表于 2008-01-10 18:16

[quote]awk '{if(/state/){sub(/>[^<]*</,">2<")}print}' urfile
[/quote]

我也刚弄完xml的，帮你解释一下这个：
if(/state/)：匹配state行才执行后面的操作
sub(/>[^<]*</,">2<")
=sub(/正则/,"b")，用b替换正则匹配的内容
其中的 />[^<]*</ 为：
匹配 >...< 的意思
[^<]* 及不是<的多个字符

所以这句话就是：匹配以<开头，直到第一个> !

hyhappy 发表于 2008-01-10 21:31

首先感谢光明使者，精灵你们二位，哈哈。
<state>1</state>
awk '{if(/state/){sub(/>[^<]*</,">2<")}print}' urfile

这条命令难理解的就是/>[^<]*</这里了。我理解的是，匹配〉〈这个之间的内容，但是为什么[^<]，看精灵解释说不是<的多个字符，但是〉〈里面已经没有〈这个符号了，为什么还要排除呢。　搞不懂。

7717060 发表于 2008-01-10 22:07

<?xml version="1.0" encoding="UTF-8"?>
<address xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:noNamespaceSchemaLocation="add.xsd">
<group groupname="default">
<entry>
<id>0001</id>
<name>wds</name>
<persioninfo>
<age>24</age>
<work>student</work>
<birtyday>1984</birtyday>
<qq>125153052</qq>
</persioninfo>
<connect>
<address>neusoft institute of informaction </address>
<tel>
<office>84835207</office>
</tel>
<email href="mailto:7717060@163.com"/>
</connect>
</entry>
</group>
</address>

#!/bin/bash
#-----------------------------------------------------------------------------------
#    filename: main.sh                                                          |
#    author: wds                                                 |
#       begin: 2008.1.1                                                          |
#       end: 2008.1.4                                                          |
#    version: v.1                                                             |
#-----------------------------------------------------------------------------------
xmladd=add.xml
xmltemp=.temp2
tim=

(d a t e) c o u n t =

$(date) count=$ (more

$xmladd | wc -l) function exitt() { rm -rf $xmltemp exit } function memo() { clear echo echo "----------------------------------------------------------------------------------------------" echo "| **************************** |" echo "| * This is XML Homework * |" echo "| * Author: wds * |" echo "| * Address List * |" echo "| **************************** |" echo "| |" echo "| 1.Show Informaction |" echo "| 2.Add Informaction |" echo "| 3.Del Informaction |" echo "| 4.Change Infromaction |" echo "| 5.Select Infromcation |" echo "| 6.Help |" echo " 7.Exit |" echo "| |" echo "______________________________________________________________________________________________" } function add() { clear echo "please input your informaction!" echo -n "id: -->" ; read id echo -n "name:-->"; read nam echo -n "age: -->"; read age echo -n "work:-->"; read work echo -n "birthday[year-month-day]: -->" ; read birthday echo -n "qq: -->"; read qq echo -n "address: -->>" ; read address echo -n "office: -->>" ; read office echo -n "email: -->>" ; read mail echo " " >> $xmladd echo "<entry>" >> $xmladd echo "<id>$id</id>" >> $xmladd echo "<name>$name</name>" >> $xmladd echo "<persion info>" >>$xmladd echo "<age>$age</age>" >> $xmladd echo "<work>$work</work>" echo "<birthday>$birthday</birthday>" >> $xmladd echo "<qq>$qq</qq>" >>$xmladd echo "</persion info>" >> $xmladd echo "<connect>" >> $xmladd echo "<address>$address</address>" >> $xmladd echo "<tel>" >>$xmladd echo "<office>$office</office>" >> $xmladd echo "</tel>" >> $xmladd echo "<email href="mailto:$mail">" >> $xmladd echo "</connect>" >>$xmladd echo "</entry>" >>$xmladd } function error() { clear echo "sorry please input [1-6]:" sleep 2 memo } #------------------------------------------------------------------------------------------------- function show() { echo "--------------------------------------------------------------------------------------------" cat $xmladd echo "--------------------------------------------------------------------------------------------" echo "Now is $tim page have $count line:" echo "--------------------------------------------------------------------------------------------" echo "Do you want save your informcation(y/Y)" echo -n "-->"; read answer if [ "$answer" = "y" -a "$answer" = "Y" ] then echo "</group>" >> $xmladd echo "</address>" >> $xmladd else continue fi } function del() { echo "please input id:" echo -n "id: -->"; read id delnum=$(more $xmladd | grep -n "$id" | cut -d: -f1) echo "$delnum" >> ww delnum=$(expr $delnum - 1 ) pageup=$delnum echo "$pageup" >>ww delnum=$(expr $delnum + 16 ) pagedown=$delnum echo "$pagedown" >>ww cat $xmladd | sed "$pageup,"$pagedown"d" > .temp more .temp > $xmladd rm -rf .temp echo "delete OK!!!" sleep 1 } function change() echo "Please Input Id:" echo -n "id: -->"; read id delnum=$ (more

x m l a d d | g r e p - n "

$xmladd | grep -n "$ id" | cut -d: -f1)
delnum=

(e x p r

$( expr$ delnum - 1 )
pageup=

d e l n u m d e l n u m =

$delnum delnum=$ ( expr

d e l n u m + 15) p a g e d o w n =

$delnum + 15 ) pagedown=$ delnum
cat

x m l a d d | s e d - n "

$xmladd | sed -n "$ pageup,"

p a g e d o w n " p " e c h o - n " N e w i d : - - >"; r e a d i d e c h o - n " N e w n a m e : - - >"; r e a d n a m e e c h o - n " N e w a g e : - - >"; r e a d a g e e c h o - n " N e w w o r k : - - >"; r e a d w o r k e c h o - n " N e w b i r t h d a y [y e a r - m o n t h - d a y] : - - >"; r e a d b i r t h d a y e c h o - n " N e w q q : - - >"; r e a d q q e c h o - n " N e w a d d r e s s : - - >>"; r e a d a d d r e s s e c h o - n " N e w o f f i c e : - - >>"; r e a d o f f i c e e c h o - n " N e w e m a i l : - - >>"; r e a d m a i l e c h o "< e n t r y >">>

$pagedown"p" echo -n "New id: -->" ; read id echo -n "New name:-->"; read name echo -n "New age: -->"; read age echo -n "New work:-->"; read work echo -n "New birthday[year-month-day]: -->" ; read birthday echo -n "New qq: -->"; read qq echo -n "New address: -->>" ; read address echo -n "New office: -->>" ; read office echo -n "New email: -->>" ; read mail echo "<entry>" >>$ xmltemp
echo "<id>

i d < / i d >">>

$id</id>" >>$ xmltemp
echo "<name>

n a m e < / n a m e >">>

$name</name>" >>$ xmltemp
echo "<persion info>" >>

x m l t e m p e c h o "< a g e >

$xmltemp echo "<age>$ age</age>" >>

x m l t e m p e c h o "< w o r k >

$xmltemp echo "<work>$ work</work>" >>

x m l t e m p e c h o "< b i r t h d a y >

$xmltemp echo "<birthday>$ birthday</birthday>" >>

x m l t e m p e c h o "< q q >

$xmltemp echo "<qq>$ qq</qq>" >>

x m l t e m p e c h o "< / p e r s i o n i n f o >">>

$xmltemp echo "</persion info>" >>$ xmltemp
echo "<connect>" >>

x m l t e m p e c h o "< a d d r e s s >

$xmltemp echo "<address>$ address</address>" >>

x m l t e m p e c h o "< t e l >">>

$xmltemp echo "<tel>" >>$ xmltemp
echo "<office>

o f f i c e < / o f f i c e >">>

$office</office>" >>$ xmltemp
echo "</tel>" >>

x m l t e m p e c h o "< e m a i l h r e f =" m a i l t o :

$xmltemp echo "<email href="mailto:$ mail">" >>

x m l t e m p e c h o "< / c o n n e c t >">>

$xmltemp echo "</connect>" >>$ xmltemp
echo "</entry>" >>

x m l t e m p c a t

$xmltemp cat$ xmladd | sed "

p a g e u p, "

$pageup,"$ pagedown"d" > .temp
more .temp >

x m l a d d m o r e

$xmladd more$ xmltemp >>

$xmladd } function selectt() { while : do echo "Please Input Id:" echo -n "id: -->"; read id delnum=$( expr $delnum - 1 ) pageup=$delnum delnum=$( expr $delnum + 15 ) pagedown=$delnum cat $xmladd | sed -n "$pageup,"$pagedown"p" echo -n "Back enter (e/E) continue select (*)"; read answer if [ "$answer" = "E" -o "$answer" = "e" ] then return memo else continue fi done } function help() { while : do clear echo " 1.show: funciton show it's can list all the informaction of xml. 2.add: function add it's can insert into your need informcation to the xml. 3.del: function del it's can delete informaction from you xml. 4.change: function change if xml have some informcation you dont need you can change use this function. 5.select: function select if your xml have a lots of informcation you can use this function to select your need informcaction. 6.help: fucntion help is help your how to use this software 7.exit: quit " echo -n "Go back to then memo!(Y/y)"; read answer if [ "$answer" = "y" -o "$answer" = "Y" ] then return memo else continue fi done } while : do memo echo "please input your choice:" read choice case "$ choice"
in
1 ) show ;;
2 ) add ;;
3 ) del ;;
4 ) change ;;
5 ) selectt;;
6 ) help ;;
7 ) exitt ;;
* ) error;;
esac
done

[[i] 本帖最后由 7717060 于 2008-1-10 22:12 编辑 [/i]]

jinl 发表于 2008-01-10 23:25

[quote]原帖由 [i]hyhappy[/i] 于 2008-1-10 21:31 发表 [url=http://bbs.chinaunix.net/redirect.php?goto=findpost&pid=7835443&ptid=1040744][img]http://bbs.chinaunix.net/images/common/back.gif[/img][/url]
首先感谢光明使者，精灵你们二位，哈哈。
1
awk '{if(/state/){sub(/>[^ [/quote]
/>[^<]*/</ 匹配> 开始任意个非<字符，并以<结束
为什么要用[^<]来匹配>...<这样形式的内容，又涉及到正则表达式的最长匹配原则
如果用/>.*</来匹配，那么如果遇到 <>1<>2<>这样的字符串，将匹配最长的>..<
即1<>2为匹配内容，而不是你需要的1作为匹配内容，所以/>[^<]*</将确保匹配第一个><中间的内容