有很多理由希望将文件拆分成块 - 主要用于网络传输（例如电子邮件附件），但我确信有些场景可能需要我无法想象的事情。 那么如何将文件拆分成可以轻松组装回原始文件的块（包括非Windows系统）？
有什么可能性：
也可以使用GZipStreams，因为它们允许 读取字节（并提供便携式压缩！）。我保证一个upvote如果有人成功（或 提供任何其他方法）： - ）
答案 0 :(得分：4)
前段时间我为此编写了一个名为BinToBat.bat的Batch-JScript混合脚本。这是它的帮助屏幕：
Create an installer Batch program for data files of any type
BINTOBAT [/T:.ext1.ext2...] [/L:lineSize] [/F[:fileSize]] filename ...
/T:.ext1.ext2 Specify the extensions of text type files that will not be
encoded as hexadecimal digits, but preserved as text.
/L:lineSize Specify the size of output lines (default: 78).
/F[:fileSize] /F switch specify to generate a Full installer file.
The optional fileSize specify the maximum output file size.
BinToBat encode the given data files as hexadecimal digits (or preserve they
as compressed text) and insert they into InstallFiles.bat program; when this
program run, it generates the original data files.
You may rename the InstallFiles.bat program as you wish, but preserving the
"Install" prefix is suggested.
You may use wild-cards in the filename list.
If the /F switch is not given, a Partial installer is created:
- You may insert a short description for each file.
- You may insert divisions in the file listing via a dash in the parameters.
- The installer allows to select which files will be downloaded and ask
before overwrite existent files.
If the /F switch is given, a Full installer is created:
- The installer always download all files.
- You may specify commands that will be executed after the files were copied.
- You may specify the maximum size of the output file via /F:fileFize, so in
this case the output file will be divided in parts with a numeric postfix.
If you use /F switch you can NOT rename the InstallFiles??.bat files; the
first one is the installer and the rest just contain data.
您可以从this site下载BinToBat.bat程序。
答案 1 :(得分：3)
所有脚本的语法都相同 - 要拆分的文件和以字节为单位的大小。
1） MAKECAB - 主要限制是在Unix / Mac机器上使用。对于unix最终cabextract或7zip可以使用，但我不确定它是否可以处理拆分CAB文件。即使在Windows上，EXPAND命令也无法执行 应使用EXTRAC32（命令在帮助消息中给出）（或Shell.Application）
;@echo off
;;;;; rem start of the batch part ;;;;;
;;
;;
;; rem the first parameter is the file you want to split the second is the size in bytes.
;; rem size is not guaranteed but will be not overflown
;;
; if "%~2" equ "" (
; call :helpmessage
; exit /b 1
;)
;if not exist "%~dpnx1" (
; call :helpmessage
; exit /b 2
;)
;if exist "%~dpnx1\" (
; call :helpmessage
; exit /b 3
;)
; rem remove leading zeroes
; cmd /c exit /b %~2
; set /a size=%errorlevel%
; if %size% equ 0 (
; echo size must be greater than 0
; exit /b 4
;)
; rem MaxDiskSize must be multiple of 512 and closest possible to desired size.
;if %~2 LSS 512 set diskSize=512 else (
; set /a part=%~2%%512
; set /a diskSize=%~2-part
;)
;makecab /d the_file="%~1" /d diskSize=%diskSize% /d the_size="%~2" /F "%~dpfnxs0"
;exit /b %errorlevel%
;:helpmessage
; echo no existing file has been passed
; echo usage [split a file to cab parts with given size]:
; echo %~nx0 file size
; echo(
; echo size must be greater than 0
; echo (
; echo for extraction use :
; echo extrac32 /a /e file.ext_part1.cab /l .
; exit /b 0
;;
;;;; rem end of the batch part ;;;;;
;;;; directives part ;;;;;
;;
.New Cabinet
.set GenerateInf=OFF
.Set Cabinet=on
.Set Compress=on
.Set MaxDiskSize=%diskSize%;
.Set MaxCabinetSize=%the_size%
.set CabinetFileCountThreshold=1
.set CompressedFileExtensionChar=_
.Set CabinetNameTemplate=%the_file%_part*.cab
.set DestinationDir=.
.Set DiskDirectoryTemplate=;
.set RptFileName=nul
.set UniqueFiles=ON
;.set DoNotCopyFiles=ON
;.set MaxDiskFileCount=1
.set MaxErrors=1
.set GenerateInf=OFF
%the_file% /inf=no
;;
;;;; end of directives part ;;;;;
- 所有其他方法都是直接拆分，文件可以通过正确的顺序将它们相互附加来组装 -
2） JScript - 必须使用.bat扩展名保存的混合文件
@if (@x)==(@y) @end /***** jscript comment ******
@echo off
cscript //E:JScript //nologo "%~f0" "%~nx0" %*
exit /b %errorlevel%
@if (@x)==(@y) @end ****** end comment *********/
//https://github.com/npocmaka/batch.scripts/blob/master/hybrids/jscript/zipjs.bat
var FileSystemObj = new ActiveXObject("Scripting.FileSystemObject");
var AdoDBObj = new ActiveXObject("ADODB.Stream");
var ARGS = WScript.Arguments;
var scriptName=ARGS.Item(0);
if (ARGS.length <3) {
WScript.Echo("Wrong arguments");
WScript.Echo("usage:");
WScript.Echo(scriptName +"file_to_split size_in_bytes");
WScript.Quit(1);
}
var file=ARGS.Item(1);
var max_size=parseInt(ARGS.Item(2));
function getSize(file){
return FileSystemObj.getFile(file).size;
}
function isExist(file){
return FileSystemObj.FileExists(file);
}
function writeFile(fileName,data ){
AdoDBObj.Type = 1;
AdoDBObj.Open();
AdoDBObj.Position=0;
AdoDBObj.Write(data);
AdoDBObj.SaveToFile(fileName,2);
AdoDBObj.Close();
}
function readFile(fileName,size,position){
AdoDBObj.Type = 1;
AdoDBObj.Open();
AdoDBObj.LoadFromFile(fileName);
AdoDBObj.Position=position;
fileBytes=AdoDBObj.Read(size);
AdoDBObj.Close();
return fileBytes;
}
function chunker(file,size){
var part=0;
var position=0;
var buffer=readFile(file,size,0);
file_size=getSize(file);
while (buffer !== null ) {
part++;
WScript.Echo("Creating: "+file+".part."+part);
writeFile(file+".part."+part,buffer);
if (size*part <= file_size) {
position=size*part;
} else {
position=file_size;
}
buffer=readFile(file,size,position);
}
}
if (!isExist(file)){
WScript.Echo(file+" does not exist");
WScript.Quit(2);
}
if(max_size<=0){
WScript.Echo("Size must be bigger than 0.")
WScript.Quit(3);
}
chunker(file,max_size);
3） JScript.net - 必须使用.bat扩展名保存的自编译混合。
@if (@X)==(@Y) @end /* JScript comment
@echo off
setlocal
for /f "tokens=* delims=" %%v in ('dir /b /s /a:-d /o:-n "%SystemRoot%\Microsoft.NET\Framework\*jsc.exe"') do (
set "jsc=%%v"
)
::if not exist "%~n0.exe" (
"%jsc%" /nologo /out:"%~n0.exe" "%~dpsfnx0"
::)
%~n0.exe %*
endlocal & exit /b %errorlevel%
*/
import System;
import System.IO;
var arguments:String[] = Environment.GetCommandLineArgs();
if (arguments.length<3){
Console.WriteLine("Wrong arguments");
Console.WriteLine("usage:");
Console.WriteLine(arguments[0]+"file_to_split size_in_bytes");
Environment.Exit(1);
}
var file=arguments[1];
var max_size=parseInt(arguments[2]);
if (max_size<=0){
Console.WriteLine("size must be bigger than zero");
Environment.Exit(2);
}
if (!File.Exists(file)){
Console.WriteLine("file"+file+" does not exist");
Environment.Exit(3);
}
function writeData(file,data:byte[]){
Console.WriteLine(data.Length);
var writer = new BinaryWriter(File.Open(file, FileMode.Create));
writer.Write(data);
writer.Close();
}
function chunker(inputFile, chunkSize){
var part=0;
var reader= new BinaryReader(File.Open(inputFile, FileMode.Open));
var data:byte[]=reader.ReadBytes(chunkSize);
while(reader.BaseStream.Position !== reader.BaseStream.Length) {
part++;
Console.WriteLine("Processing part " + part);
writeData(inputFile+".part."+part,data);
data=reader.ReadBytes(chunkSize);
}
if (data.Length !== 0) {
part++;
Console.WriteLine("Processing part " + part)
writeData(inputFile+".part."+part,data);
}
reader.Close();
}
chunker(file,max_size);
4） CERTUTIL - 而不是实验性的东西 - 它的速度较慢，因为缓冲区被限制为最大长度，字符串可以包含8,1 **个字符，而且是文本处理：
@echo off
setlocal enableExtensions
rem :-----------------------------------------
rem : check if should prompt the help message
rem :-----------------------------------------
if "%~2" equ "" goto :help
for %%H in (/h -h /help -help) do (
if /I "%~1" equ "%%H" goto :help
)
if not exist "%~1" echo file does not exist & exit /b 1
rem :-----------------------------------------
rem : validate input
rem :-----------------------------------------
set /a size=%~2
if not defined size echo something wrong with size parameter & exit /b 2
if %size%0 LSS 00 echo not a valid number passed as a parameter & exit /b 3
rem : -- two hex symbols and an empty space are 1 byte
rem : -- so the sum of all hex symbols
rem : -- per part should be doubled
set /a len=%size%*2
set "file=%~dfn1"
for %%F in ("%file%") do set fn=%%~nxF
rem : -- clear temp data
del /F /Q "%temp%\file" >nul 2>&1
del /F /Q "%temp%\fn.p.*" >nul 2>&1
certutil -encodehex -f "%file%" "%temp%\file" >nul
set "part=1"
setlocal enableDelayedExpansion
set "hex_str="
set hex_len=0
break>%temp%\fn.p.!part!
rem : -- reads the hex encoded file
rem : -- and make it on a parts that will
rem : -- decoded with certutil
rem :-- the delimitier is <tab> wich separates
rem :-- line number from the rest of the information
rem :-- in the hex file
rem :---------------------------- v <tab>
for /f "usebackq tokens=2 delims= " %%A in ("%temp%\file") do (
set "line=%%A"
rem : -- there's a double space in the middle of the line
rem :-- so here the line is get
set hex_str=!hex_str!!line:~0,48!
rem echo hex_str !hex_str!
rem :-- empty spaces are cleared
set hex_str=!hex_str: =!
rem echo hex_str !hex_str!
rem :-- the length of the hex line 32
set /a hex_len=hex_len+32
rem : -- len/size is reached
rem : -- and the content is printed to a hex file
if !hex_len! GEQ !len! (
echo !hex_len! GEQ !len!
set /a rest=hex_len-len
for %%A in (!rest!) do (
(echo(!hex_str:~0,-%%A!)>>%temp%\fn.p.!part!
rem : -- the rest of the content of the line is saved
set hex_str=!hex_str:~-%%A!
set /a hex_len=rest
echo !hex_len!
)
certutil -decodehex -f %temp%\fn.p.!part! %fn%.part.!part! >nul
echo -- !part!th part created --
rem :-- preprarin next hex file
set /a part=part+1
break>%temp%\fn.p.!part!
rem :-- reinitilization of the len/size of the file part
set /a len=%size%*2
)
rem : -- a buffer that not allows to
rem : -- to enter too long commmands
rem : -- used to reduce disk operations
if !hex_len! GEQ 7800 (
(echo(!hex_str!)>>%temp%\fn.p.!part!
set "hex_str="
set hex_len=0
rem :-- the size that need to be reached is reduces
rem :-- as there's alredy part of the part of the file
rem :-- added to the hex file
set /a len=!len!-7800
if !len! LSS 0 set len=0
)
)
rem : -- adding the rest of the file
echo !hex_str!>>%temp%\fn.p.!part!
certutil -decodehex -f %temp%\fn.p.!part! %fn%.part.!part! >nul
echo -- !part!th part created --
rem : -- clear created temp data
rem del /F /Q %temp%\fn.p.* >nul 2>&1
rem del /F /Q %temp%\file >nul 2>&1
endlocal
endlocal
goto :eof
rem :-----------------------------------------
rem : help message
rem :-----------------------------------------
:help
echo\
echo Splits a file on parts by given size in bytes in 'pure' batch.
echo\
echo\
echo %0 file size
echo\
echo\
这是一个可以汇编用最后三种方法分割的文件的脚本：
@echo off
if "%~1" EQU "" echo parameter not entered & exit /b 1
set "parts=%~1.part"
setlocal enableDelayedExpansion
set numb=0
for /f "delims=." %%P in ('dir /b %parts%*') do (
set /a numb=numb+1
)
rem echo !numb!
setlocal enableDelayedExpansion
set "string=%~1.part.1"
for /l %%n in (2;1;!numb!) do (
set "string=!string!+!parts!.%%n"
)
rem echo !string!
copy /y /b !string! %~1%~x1
endlocal
答案 2 :(得分：2)
下面的示例将拆分文件，生成多个小于所提供的maxChunkSize的输出文件。要重新组合，您可以使用copy /b。
SplitFile.cs
（用c:\Windows\Microsoft.NET\Framework\v2.0.50727\csc /out:splitFile.exe SplitFile.cs编译）
using System;
using System.IO;
namespace SplitFile
{
class Program
{
static void Main(string[] args)
{
long maxChunkSize;
if (args.Length != 3 || !long.TryParse(args[2], out maxChunkSize) || maxChunkSize <= 81920)
{
Console.WriteLine("Usage: splitfile.exe inputFile outputprefix maxchunksize");
Console.WriteLine(" inputfile: File to split");
Console.WriteLine(" outputprefix: Prefix to use for the output name");
Console.WriteLine(" Ex: out -> { out0001.bin, out0002.bin }");
Console.WriteLine(" maxchunksize: Maximum number of bytes in each file");
Console.WriteLine(" Note: this is the maximum size, not an exact size");
Console.WriteLine(" Note: chunk size cannot be smaller than 81920 bytes");
return;
}
string inputFilePath = args[0];
string outputFilePathFormat = string.Format("{0}{{0:0000}}.bin", args[1]);
using (Stream fsInput = File.Open(inputFilePath, FileMode.Open, FileAccess.Read, FileShare.Read))
{
byte[] buffer = new byte[81920 /* default from System.Stream */];
int cOutFileNo = 0;
Stream destination = getOutputFile(ref cOutFileNo, outputFilePathFormat);
try
{
int read;
while ((read = fsInput.Read(buffer, 0, buffer.Length)) != 0)
{
if (destination.Length + read > maxChunkSize)
{
destination.Dispose();
destination = getOutputFile(ref cOutFileNo, outputFilePathFormat);
}
destination.Write(buffer, 0, read);
}
}
finally
{
destination.Dispose();
}
}
}
private static Stream getOutputFile(ref int cOutFileNo, string outFileFormat)
{
string filename = string.Format(outFileFormat, cOutFileNo);
cOutFileNo++;
return File.Open(filename, FileMode.CreateNew, FileAccess.Write, FileShare.None);
}
}
}
使用示例：
C:\drop>splitFile.exe ubuntu-rescue-remix-12-04.iso Ubuntu_Split_ 10485760
C:\drop>dir
01/29/2015 17:21 244,570,112 ubuntu-rescue-remix-12-04.iso
01/30/2015 15:27 10,485,760 Ubuntu_Split_0000.bin
01/30/2015 15:27 10,485,760 Ubuntu_Split_0001.bin
01/30/2015 15:27 10,485,760 Ubuntu_Split_0002.bin
01/30/2015 15:27 10,485,760 Ubuntu_Split_0003.bin
01/30/2015 15:27 10,485,760 Ubuntu_Split_0004.bin
01/30/2015 15:27 10,485,760 Ubuntu_Split_0005.bin
01/30/2015 15:27 10,485,760 Ubuntu_Split_0006.bin
01/30/2015 15:27 10,485,760 Ubuntu_Split_0007.bin
01/30/2015 15:27 10,485,760 Ubuntu_Split_0008.bin
01/30/2015 15:27 10,485,760 Ubuntu_Split_0009.bin
01/30/2015 15:27 10,485,760 Ubuntu_Split_0010.bin
01/30/2015 15:27 10,485,760 Ubuntu_Split_0011.bin
01/30/2015 15:27 10,485,760 Ubuntu_Split_0012.bin
01/30/2015 15:27 10,485,760 Ubuntu_Split_0013.bin
01/30/2015 15:27 10,485,760 Ubuntu_Split_0014.bin
01/30/2015 15:27 10,485,760 Ubuntu_Split_0015.bin
01/30/2015 15:27 10,485,760 Ubuntu_Split_0016.bin
01/30/2015 15:27 10,485,760 Ubuntu_Split_0017.bin
01/30/2015 15:27 10,485,760 Ubuntu_Split_0018.bin
01/30/2015 15:27 10,485,760 Ubuntu_Split_0019.bin
01/30/2015 15:27 10,485,760 Ubuntu_Split_0020.bin
01/30/2015 15:27 10,485,760 Ubuntu_Split_0021.bin
01/30/2015 15:27 10,485,760 Ubuntu_Split_0022.bin
01/30/2015 15:27 3,397,632 Ubuntu_Split_0023.bin
C:\drop>copy /b Ubuntu_Split_* Ubuntu_recombined.iso
Ubuntu_Split_0000.bin
Ubuntu_Split_0001.bin
Ubuntu_Split_0002.bin
Ubuntu_Split_0003.bin
Ubuntu_Split_0004.bin
Ubuntu_Split_0005.bin
Ubuntu_Split_0006.bin
Ubuntu_Split_0007.bin
Ubuntu_Split_0008.bin
Ubuntu_Split_0009.bin
Ubuntu_Split_0010.bin
Ubuntu_Split_0011.bin
Ubuntu_Split_0012.bin
Ubuntu_Split_0013.bin
Ubuntu_Split_0014.bin
Ubuntu_Split_0015.bin
Ubuntu_Split_0016.bin
Ubuntu_Split_0017.bin
Ubuntu_Split_0018.bin
Ubuntu_Split_0019.bin
Ubuntu_Split_0020.bin
Ubuntu_Split_0021.bin
Ubuntu_Split_0022.bin
Ubuntu_Split_0023.bin
1 file(s) copied.
C:\drop>dir Ubuntu*.iso
01/29/2015 17:21 244,570,112 ubuntu-rescue-remix-12-04.iso
01/30/2015 15:27 244,570,112 Ubuntu_recombined.iso
C:\drop>fciv -sha1 ubuntu-rescue-remix-12-04.iso
//
// File Checksum Integrity Verifier version 2.05.
//
02403c37cbdb3e03e00f5176807a793ef63d877c ubuntu-rescue-remix-12-04.iso
C:\drop>fciv -sha1 Ubuntu_recombined.iso
//
// File Checksum Integrity Verifier version 2.05.
//
02403c37cbdb3e03e00f5176807a793ef63d877c ubuntu-rescue-remix-12-04.iso
C:\drop>
答案 3 :(得分：1)
对于原始字节块，powershell 可以拆分文件：
set size=<split size>
set file=<file.in>
for %j in (%file%) do (
set /a chunks=%~zj/%size% >nul
for /l %i in (0,1,!chunks!) do (
set /a tail=%~zj-%i*%size% >nul
powershell gc %file% -Encoding byte -Tail !tail! ^| sc %file%_%i -Encoding byte
if %i lss !chunks! FSUTIL file seteof %file%_%i %size% >nul
)
)
然后可以通过 copy
某些系统可能需要准备好运行 FSUTIL 并启用延迟扩展。
还有另一种方法可以提取文件的较小部分，但 makecab 实用程序以其自己的格式对块进行编码，因此它们不能仅被视为原始字节，类似于平面二进制文件，例如。通过copy。
如某些答案所述，拆分文件的大小不必是 512 的倍数。但是，块可以通过 extrac32 连接。
首先制作一个 ddf（文本）文件或仅通过命令行中的 /d 提供选项：
.Set CabinetNameTemplate=test_*.cab; <-- Enter chunk name format
.Set MaxDiskSize=900000; <-- Enter file split/chunk size (doesn't have to be multiple of 512)
.Set ClusterSize=1000
.Set Cabinet=on;
.Set Compress=off; <-- Optional: set compression on to save disk space
.set CompressionType=LZX;
.set CompressionMemory=21
.Set DiskDirectoryTemplate=;
file.in <-- Enter the file name to be split
那么：
makecab /f ddf.txt
要取回原始文件，请确保所有块都在同一目录中，然后依次调用第一个文件：
extrac32 test_1.cab file.out
MakeCAB 引入了文件夹的概念来引用一组连续的压缩字节。
“MakeCAB 将被压缩的产品或应用程序中的所有文件，将字节放置为一个连续的字节流，压缩整个流，将其切成适当的文件夹，然后用文件夹填充一个或多个文件柜。 "
在 Win 10 上测试。