请教P处理文本的效率问题 - 中国DOS联盟论坛

要处理的文本内容剪辑如下：
<hgsdp:msisdn=8615082687528,loc;
HLR SUBSCRIBER DATA

SUBSCRIBER IDENTITY
MSISDN IMSI STATE AUTHD
8615082687528 460023826698378 CONNECTED AVAILABLE

NAM
0

LOCATION DATA
VLR ADDRESS MSRN MSC NUMBER LMSID
4-8613744818 8613744818

SGSN NUMBER
4-8613741803

END
<hgsdp:msisdn=8613980325588,loc;
HLR SUBSCRIBER DATA

SUBSCRIBER IDENTITY
MSISDN IMSI STATE AUTHD
8613980325588 460000324808588 CONNECTED AVAILABLE

NAM
0

LOCATION DATA
VLR ADDRESS MSRN MSC NUMBER LMSID
4-99292370000 99292370000

SGSN NUMBER
4-8613746991
MS PURGED IN SGSN

END
<hgsdp:msisdn=8615884933063,loc;
HLR SUBSCRIBER DATA

SUBSCRIBER IDENTITY
MSISDN IMSI STATE AUTHD
8615884933063 460028825794364 CONNECTED AVAILABLE

NAM
0

LOCATION DATA
VLR ADDRESS MSRN MSC NUMBER LMSID
4-8613741249 8613741249
MS PURGED IN VLR

SGSN NUMBER
4-8613742297

END
<hgsdp:msisdn=8613608272886,loc;
HLR SUBSCRIBER DATA

SUBSCRIBER IDENTITY
MSISDN IMSI STATE AUTHD
8613608272886 460008480450049 CONNECTED AVAILABLE

NAM
0

LOCATION DATA
VLR ADDRESS MSRN MSC NUMBER LMSID
4-8613744818 8613744818

SGSN NUMBER
UNKNOWN

END
<hgsdp:msisdn=8613882685885,loc;
HLR SUBSCRIBER DATA

SUBSCRIBER IDENTITY
MSISDN IMSI STATE AUTHD
8613882685885 460028825960924 CONNECTED AVAILABLE

NAM
1

LOCATION DATA
VLR ADDRESS MSRN MSC NUMBER LMSID
4-8613740848 8613740848

END

我要选出含有MS PURGED IN SGSN、MS PURGED IN VLR、UNKNOWN或无SGSN NUMBER字符串对应的<hgsdp开头的这行。

我的代码如下：


::findmsisdn.bat - 查找固定模式文本内容中含有MS PURGED IN SGSN、MS PURGED IN VLR、UNKNOWN或无SGSN NUMBER字符串中的第一行
::genteman - 2009-09-22 -CMD@WinXP Pro
::contact amdaround@163.com
::错误处理部分借用了Will Sort的代码，在此表示感谢！

@echo off & setlocal enabledelayedexpansion

if == goto :error

if == %0 :error 0 "Incomplete argument - Usage:%~n0 <filename>"
if not exist "%~1" %0 :error 1 "%~n1 does not exist"

for /f "delims=" %%i in (%~s1) do (
echo "%%i" | find "hgsdp" >nul && set tmpstr=%%i
set str=!str!,%%i
if %%i==END (
echo "!str!" | find "SGSN NUMBER" >nul || echo !tmpstr! >>"%~dp1NO SGSN NUMBER.txt"
echo "!str!" | find "MS PURGED IN SGSN" >nul && echo !tmpstr! >>"%~dp1MS PURGED IN SGSN.txt"
echo "!str!" | find "MS PURGED IN VLR" >nul && echo !tmpstr! >>"%~dp1MS PURGED IN VLR.txt"
echo "!str!" | find "SGSN NUMBER,UNKNOWN" >nul && echo !tmpstr! >>"%~dp1"UNKNOWN.txt
set str=
)
)
goto :EOF

:error - 错误处理
echo.
echo Error %2: %3
echo.
exit /b %2

现在的问题是运行效率特别低，对一个大小超过10M的文件处理需要至少24小时的执行时间，如果是在UNIX下用sed来处理的话只需要几秒钟就能搞定。

上面的代码问题是因为每次循环都要进行多次判断，所以造成执行效率低下，但我实在想不出有更好的算法。

请教大家有没有优化的代码？

第1楼　genteman	发表于 2008-09-24 09:39
初级用户　发帖 43　积分 90
[讨论]请教P处理文本的效率问题要处理的文本内容剪辑如下： <hgsdp:msisdn=8615082687528,loc; HLR SUBSCRIBER DATA SUBSCRIBER IDENTITY MSISDN IMSI STATE AUTHD 8615082687528 460023826698378 CONNECTED AVAILABLE NAM 0 LOCATION DATA VLR ADDRESS MSRN MSC NUMBER LMSID 4-8613744818 8613744818 SGSN NUMBER 4-8613741803 END <hgsdp:msisdn=8613980325588,loc; HLR SUBSCRIBER DATA SUBSCRIBER IDENTITY MSISDN IMSI STATE AUTHD 8613980325588 460000324808588 CONNECTED AVAILABLE NAM 0 LOCATION DATA VLR ADDRESS MSRN MSC NUMBER LMSID 4-99292370000 99292370000 SGSN NUMBER 4-8613746991 MS PURGED IN SGSN END <hgsdp:msisdn=8615884933063,loc; HLR SUBSCRIBER DATA SUBSCRIBER IDENTITY MSISDN IMSI STATE AUTHD 8615884933063 460028825794364 CONNECTED AVAILABLE NAM 0 LOCATION DATA VLR ADDRESS MSRN MSC NUMBER LMSID 4-8613741249 8613741249 MS PURGED IN VLR SGSN NUMBER 4-8613742297 END <hgsdp:msisdn=8613608272886,loc; HLR SUBSCRIBER DATA SUBSCRIBER IDENTITY MSISDN IMSI STATE AUTHD 8613608272886 460008480450049 CONNECTED AVAILABLE NAM 0 LOCATION DATA VLR ADDRESS MSRN MSC NUMBER LMSID 4-8613744818 8613744818 SGSN NUMBER UNKNOWN END <hgsdp:msisdn=8613882685885,loc; HLR SUBSCRIBER DATA SUBSCRIBER IDENTITY MSISDN IMSI STATE AUTHD 8613882685885 460028825960924 CONNECTED AVAILABLE NAM 1 LOCATION DATA VLR ADDRESS MSRN MSC NUMBER LMSID 4-8613740848 8613740848 END 我要选出含有MS PURGED IN SGSN、MS PURGED IN VLR、UNKNOWN或无SGSN NUMBER字符串对应的<hgsdp开头的这行。我的代码如下： ::findmsisdn.bat - 查找固定模式文本内容中含有MS PURGED IN SGSN、MS PURGED IN VLR、UNKNOWN或无SGSN NUMBER字符串中的第一行 ::genteman - 2009-09-22 -CMD@WinXP Pro ::contact amdaround@163.com ::错误处理部分借用了Will Sort的代码，在此表示感谢！ @echo off & setlocal enabledelayedexpansion if == goto :error if == %0 :error 0 "Incomplete argument - Usage:%~n0 <filename>" if not exist "%~1" %0 :error 1 "%~n1 does not exist" for /f "delims=" %%i in (%~s1) do ( echo "%%i" \| find "hgsdp" >nul && set tmpstr=%%i set str=!str!,%%i if %%i==END ( echo "!str!" \| find "SGSN NUMBER" >nul \|\| echo !tmpstr! >>"%~dp1NO SGSN NUMBER.txt" echo "!str!" \| find "MS PURGED IN SGSN" >nul && echo !tmpstr! >>"%~dp1MS PURGED IN SGSN.txt" echo "!str!" \| find "MS PURGED IN VLR" >nul && echo !tmpstr! >>"%~dp1MS PURGED IN VLR.txt" echo "!str!" \| find "SGSN NUMBER,UNKNOWN" >nul && echo !tmpstr! >>"%~dp1"UNKNOWN.txt set str= ) ) goto :EOF :error - 错误处理 echo. echo Error %2: %3 echo. exit /b %2 现在的问题是运行效率特别低，对一个大小超过10M的文件处理需要至少24小时的执行时间，如果是在UNIX下用sed来处理的话只需要几秒钟就能搞定。上面的代码问题是因为每次循环都要进行多次判断，所以造成执行效率低下，但我实在想不出有更好的算法。请教大家有没有优化的代码？