Windows批处理文件有哪些不太为人所知,但很重要和有用的特性?

指南:

每个答案一个特征 给出特性的简短描述和示例,而不仅仅是文档链接 将答案限制在本地功能,即不需要额外的软件,如Windows资源包

澄清:这里我们指的是由cmd.exe处理的脚本,这是WinNT变体的默认值。

(请参见:Windows批处理文件:.bat vs .cmd?)


当前回答

脚本所在的路径(带驱动器):~dp0

set BAT_HOME=%~dp0
echo %BAT_HOME%
cd %BAT_HOME%

其他回答

对于具有数字计数器的循环(输出1到10):

for /l %i in (1,1,10) do echo %i

基于行的执行

虽然在大多数情况下没有明显的好处,但当试图在运行时更新内容时,它会有所帮助。例如:

UpdateSource.bat

copy UpdateSource.bat Current.bat
echo "Hi!"

Current.bat

copy UpdateSource.bat Current.bat

现在,执行Current.bat会产生这样的输出。

HI!

注意,批处理执行是按行号进行的。如果基本行没有完全相同的行号,这样的更新可能会导致跳过或向后移动一行。

子例程(输出42):

    @echo off
    call :answer 42
    goto :eof
:do_something
    echo %1
    goto :eof

子例程返回一个值(输出0、1、2,等等):

    @echo off
    setlocal enableextensions enabledelayedexpansion
    call :seq_init seq1
:loop1
    if not %seq1%== 10 (
        call :seq_next seq1
        echo !seq1!
        goto :loop1
    )
    endlocal
    goto :eof

:seq_init
    set /a "%1 = -1"
    goto :eof
:seq_next
    set /a "seq_next_tmp1 = %1"
    set /a "%1 = %seq_next_tmp1% + 1"
    set seq_next_tmp1=
    goto :eof

抽取随机的文本行

@echo off

:: Get time (alas, it's only HH:MM xM

for /f %%a in ('time /t') do set zD1=%%a



:: Get last digit of MM

set zD2=%zD1:~4,1%



:: Seed the randomizer, if needed

if not defined zNUM1 set /a zNUM1=%zD2%


:: Get a kinda random number

set /a zNUM1=zNUM1 * 214013 + 2531011

set /a zNUM2=zNUM1 ^>^> 16 ^& 0x7fff


:: Pull off the first digit

:: (Last digit would be better, but it's late, and I'm tired)

set zIDX=%zNUM2:~0,1%


:: Map it down to 0-3

set /a zIDX=zIDX/3


:: Finally, we can set do some proper initialization

set /a zIIDX=0

set zLO=

set zLL=""


:: Step through each line in the file, looking for line zIDX

for /f "delims=@" %%a in (c:\lines.txt) do call :zoo  %zIDX%  %%a


:: If line zIDX wasn't found, we'll settle for zee LastLine

if "%zLO%"=="" set zLO=%zLL%

goto awdun


:: See if the current line is line zIDX

:zoo


:: Save string of all parms

set zALL=%*


:: Strip off the first parm (sure hope lines aren't longer than 254 chars)

set zWORDS=%zALL:~2,255%


:: Make this line zee LastLine

set zLL=%zWORDS%


:: If this is the line we're looking for, make it zee LineOut

if {%1}=={%zIIDX%} set zLO=%zWORDS%


:: Keep track of line numbers

set /a zIIDX=%zIIDX% + 1

goto :eof




:awdun

echo ==%zLO%==


:: Be socially responsible

set zALL=

set zD1=

set zD2=

set zIDX=

set zIIDX=

set zLL=

set zLO=

:: But don't mess with seed

::set zNUM1=

set zNUM2=

set zWORDS=

批处理脚本最常见的需求之一是记录生成的输出,以供以后检查。是的,您可以将stdout和stderr重定向到一个文件,但是您无法看到发生了什么,除非您跟踪日志文件。

因此,考虑使用stdout/stderr日志记录工具运行您的批处理脚本,例如logger,它将使用时间戳记录输出,并且您仍然能够看到脚本进程。

另一个stdout/stderr日志记录实用程序

Yet another stdout/stderr logging utility [2010-08-05]
Copyright (C) 2010 LoRd_MuldeR <MuldeR2@GMX.de>
Released under the terms of the GNU General Public License (see License.txt)

Usage:
  logger.exe [logger options] : program.exe [program arguments]
  program.exe [program arguments] | logger.exe [logger options] : -

Options:
  -log <file name>  Name of the log file to create (default: "<program> <time>.log")
  -append           Append to the log file instead of replacing the existing file
  -mode <mode>      Write 'stdout' or 'stderr' or 'both' to log file (default: 'both')
  -format <format>  Format of log file, 'raw' or 'time' or 'full' (default: 'time')
  -filter <filter>  Don't write lines to log file that contain this string
  -invert           Invert filter, i.e. write only lines to log file that match filter
  -ignorecase       Apply filter in a case-insensitive way (default: case-sensitive)
  -nojobctrl        Don't add child process to job object (applies to Win2k and later)
  -noescape         Don't escape double quotes when forwarding command-line arguments
  -silent           Don't print additional information to the console
  -priority <flag>  Change process priority (idle/belownormal/normal/abovenormal/high)
  -inputcp <cpid>   Use the specified codepage for input processing (default: 'utf8')
  -outputcp <cpid>  Use the specified codepage for log file output (default: 'utf8')