读了几遍这个话题,我得出的结论是,同一文档的所有版本都可以识别,因为它们都具有相同的文档编号(您应该说明这一点,而不是向我们展示多个示例并让我们猜测)。
下面的批处理文件解决了您的问题。
@echo off
setlocal EnableDelayedExpansion
set prevName=
for %%a in (*.pdf) do (
if not defined prevName (
rem Initialize process with first name and DocumentNumber
set prevName=%%a
for /F "delims=_" %%b in ("%%a") do set prevNumber=%%b
) else (
rem Check if next name have the same DocumentNumber
set nextName=%%a
for /F "delims=_" %%b in ("%%a") do set nextNumber=%%b
if "!prevNumber!" equ "!nextNumber!" (
rem Yes: Two versions of same file, move the older one
ECHO move "!prevName!" "C:\dest\dir"
) else (
rem No: Different files, update DocumentNumber
set prevNumber=!nextNumber!
)
rem Anyway, previous name was processed
set prevname=!nextName!
)
)
磁盘上的文件:
00000056_NEWER WIDGET_Gray_W2.pdf
D00003456_BTW-FW001_OPTB_A12.pdf
D00003456_BTW-FW001_OPTB_B9.pdf
D00003456_BTW-FW001_OPTB_C6.pdf
D00003456_BTW-FW001_OPTB_D2.pdf
DPP-456_BTW-FW001_OPTB_A1.pdf
DPP-456_BTW-FW001_OPTB_C45.pdf
DPP-456_NEW WIDGET_F6.pdf
DPP-456_NEWER WIDGET_Blue_W2.pdf
SD0001_ClassyWidget_C45.pdf
SD0001_I001_A1.pdf
SD0034_WIDGET_F6.pdf
程序结果:
move "D00003456_BTW-FW001_OPTB_A12.pdf" "C:\dest\dir"
move "D00003456_BTW-FW001_OPTB_B9.pdf" "C:\dest\dir"
move "D00003456_BTW-FW001_OPTB_C6.pdf" "C:\dest\dir"
move "DPP-456_BTW-FW001_OPTB_A1.pdf" "C:\dest\dir"
move "DPP-456_BTW-FW001_OPTB_C45.pdf" "C:\dest\dir"
move "DPP-456_NEW WIDGET_F6.pdf" "C:\dest\dir"
move "SD0001_ClassyWidget_C45.pdf" "C:\dest\dir"
编辑:回答新评论
正如我之前所说,您的解释不清楚,因此我的解决方案基于您的示例。我之前的代码可以正确地与您的原始示例一起使用,但有一种情况除外:Ex3 中的文档编号 SD0001。但是,在这种情况下,您以相反的顺序列出了文件 I001 和 ClassyWidget !. DIR 和 FOR 命令都首先列出 ClassyWidget,然后是 I001。你为什么把它们颠倒过来?当我阅读问题描述时,我专注于如何解决它,而不是检查OP给出的数据是对还是错!这样,我的解决方案将最后列出的文件作为旧版本提供(如所有示例中所示)。
这个新的“细节”导致了一个完全不同的解决方案,因为它现在需要两次遍历所有文件来首先识别每个 DocNumber 的最后一个版本。尽管在问题描述中您说“DocumentName 可以包含下划线”,但显示的所有示例都只包含一个下划线。当 DocumentName 只有一个或没有下划线时,下面的新解决方案可以正确获取旧版本;如果它可能有多个下划线,则需要进行“小”修改。
@echo off
setlocal EnableDelayedExpansion
rem First pass: get oldest version of each DocNumber
for %%a in (*.pdf) do (
rem docNumber=First token (%%b), version=fourth (%%e) or third (%%d) token
for /F "tokens=1-4 delims=_" %%b in ("%%~Na") do (
set version=%%e
if not defined version set version=%%d
if "!version!" gtr "!oldestVersion[%%b]!" set oldestVersion[%%b]=!version!
)
)
rem Second pass: move all versions of each DocNumber, except the oldest one
for %%a in (*.pdf) do (
for /F "tokens=1-4 delims=_" %%b in ("%%~Na") do (
set version=%%e
if not defined version set version=%%d
if "!version!" neq "!oldestVersion[%%b]!" ECHO move "%%a" "C:\dest\dir"
)
)
磁盘上的文件:
00000056_NEWER WIDGET_Gray_W2.pdf
BI0000018307_MW531-ABZ2-Bond-Resc_xls_D2.pdf
BI0000018307_MW531-Triton-Bond-Res_xls_B9.pdf
BI0000018307_MW531-Triton-Bond-Res_xls_C5.pdf
D00003456_BTW-FW001_OPTB_A12.pdf
D00003456_BTW-FW001_OPTB_B9.pdf
D00003456_BTW-FW001_OPTB_C6.pdf
D00003456_BTW-FW001_OPTB_D2.pdf
DPP-456_BTW-FW001_OPTB_A1.pdf
DPP-456_BTW-FW001_OPTB_C45.pdf
DPP-456_NEW WIDGET_F6.pdf
DPP-456_NEWER WIDGET_Blue_W2.pdf
SD0001_ClassyWidget_C45.pdf
SD0001_I001_A1.pdf
SD0034_WIDGET_F6.pdf
程序结果:
move "BI0000018307_MW531-Triton-Bond-Res_xls_B9.pdf" "C:\dest\dir"
move "BI0000018307_MW531-Triton-Bond-Res_xls_C5.pdf" "C:\dest\dir"
move "D00003456_BTW-FW001_OPTB_A12.pdf" "C:\dest\dir"
move "D00003456_BTW-FW001_OPTB_B9.pdf" "C:\dest\dir"
move "D00003456_BTW-FW001_OPTB_C6.pdf" "C:\dest\dir"
move "DPP-456_BTW-FW001_OPTB_A1.pdf" "C:\dest\dir"
move "DPP-456_BTW-FW001_OPTB_C45.pdf" "C:\dest\dir"
move "DPP-456_NEW WIDGET_F6.pdf" "C:\dest\dir"
move "SD0001_I001_A1.pdf" "C:\dest\dir"
编辑:相同问题的第三个解决方案...
@echo off
setlocal EnableDelayedExpansion
rem Filename syntax: DocumentNumber_DocumentName_Version.pdf
rem May be spaces and underscores in DocumentName
rem Goal: Find last Version of same DocumentNumber and keep it (DocumentName don't cares)
rem Move the rest to other folder
rem First pass: Create *ordered* FILE array with this format: FILE[docNumber,Version.pdf]=docName
for %%a in (*.pdf) do (
set "fileName=%%a"
rem Change spaces by slashes in DocumentName
set fileName=!fileName: =/!
rem DocumentNumber=first token, Version=last token; separated by underscores
set docNumber=
for %%b in (!fileName:_^= !) do (
if not defined docNumber (
set docNumber=%%b
) else (
set Version=%%b
)
)
rem Eliminate DocumentNumber and Version from filename (get DocumentName)
set fileName=!fileName:*_=!
for %%v in (!Version!) do set docName=!filename:_%%v=!
rem Create the array element, restoring spaces in DocumentName
set FILE[!docNumber!,!Version!]=!docName:/= !
)
rem Delete REM part in two next commands if you want to review the created FILE array
REM SET FILE[
REM ECHO/
rem Second pass: Move all array elements of same DocumentNumber, except the last one
set docNumber=/
for /F "tokens=2-4 delims=[,]=" %%a in ('set FILE[') do (
rem %%a=docNumber, %%b=Version.pdf, %%c=docName
if "!docNumber!" equ "%%a" (
ECHO move "!docNumber!_!docName!_!Version!" "C:\dest\dir"
)
set docNumber=%%a
set docName=%%c
set Version=%%b
)