脚本在不同目录中的行为不同

Question

我有一个简单的脚本，是从一个较大的脚本中提取出来的，该脚本专门用于重命名名称中包含奇数字符的文件。我已经在同一台计算机上的不同目录中成功运行了它，正如我们所说，它在一个目录中成功运行，但由于某种原因，当针对更高级别的目录运行它时，它会错误地重命名文件。例如，我已经针对名为 UnsortedMusic 的文件夹运行了此命令。它应该做的是：

/UnsortedMusic/Dir 001/SubDir 001/Song To Sing.flac > /UnsortedMusic/Dir 001/SubDir 001/SongToSing.flac

它从名称中取出空格，然后我有本质上相同的代码，但专用于目录以删除目录名称中的空格：

/UnsortedMusic/Dir 001/SubDir 001/SongToSing.flac > /UnsortedMusic/Dir001/SubDir001/SongToSing.flac

这在其他目录上成功运行，但由于某种原因在一个特定目录上它正在执行如下操作：

 /UnsortedMusic/Dir 001/SubDir 001/Song To Sing.flac > /UnsortedMusic/Dir001/SubDir001/UnsortedMusicDir001SubDir001SongToSingflacSongToSingflac

正如您所看到的，它不仅使用所有目录名称重命名文件，还删除了“.”。在 .flac 中，这意味着我必须手动更改数千个文件。

下面列出了这段代码：

#!/bin/bash
#set -e

FolderName="Unsorted" #Dir location to be sorted

# Log Location on Server.
LOG_LOCATION=$(pwd) #/home/user/scripts/logs
exec > >(tee -i $LOG_LOCATION/RenameFile.log)
exec 2>&1
Start by seeing if the new directory allready exists, if not, make it

echo "Looking in $FolderName for files with abnormal characters"
#Rename any directories that might have a weird name structure that could conflict
find ${FolderName} -type f \( -name "* *" -o -name "*(*" -o -name "*[*" -o -name "*&*" \) -print0 | sort -rz | while read -d $'\0' file;
    do
        mv -v "$file" "$(dirname "$file")/$(basename "${file//[^a-zA-z0-9_-.]/}")";
done

#echo "Log Location should be: [ $LOG_LOCATION ]"
echo "Finished Process"

这是一个更大的流程脚本的一部分，如下所示：

#!/bin/bash
#set -e

NewDirName="TestDirV2" #New Dir name
FolderName="TestDir" #Dir location to be sorted

# Log Location on Server.
LOG_LOCATION=$(pwd) #/home/user/scripts/logs
exec > >(tee -i $LOG_LOCATION/Organizer.log)
exec 2>&1
#Start by seeing if the new directory allready exists, if not, make it
if [ ! -d ${NewDirName} ];
then 
    mkdir -p ${NewDirName}
    chmod -R a+rwx ${NewDirName}
fi

echo "Looking in $FolderName for files with abnormal characters"
#Rename any files that might have a weird name structure that could conflict 
find ${FolderName} -type f \( -name "* *" -o -name "*(*" -o -name "*[*" -o -name "*&*" \) -print0 | sort -rz | while read -d $'\0' file;
    do
        mv -v "$file" "$(dirname "$file")/$(basename "${file//[^a-zA-z0-9_-.]/}")";
done

echo "Looking in $FolderName for directories with abnormal characters"
#Rename any directories that might have a weird name structure that could conflict 
find ${FolderName} -type d \( -name "* *" -o -name "*(*" -o -name "*[*" -o -name "*&*" \) -print0 | sort -rz | while read -d $'\0' file;
    do
        mv -v "$file" "$(dirname "$file")/$(basename "${file//[^a-zA-z0-9_-.]/}")";
done


for FileType in $(find $FolderName -type f | perl -ne 'print $1 if m/\.([^.\/]+)$/' | sort -u)
do


    echo "Finished removing odd characters that may result in problems"

    NewDir=$(echo "${NewDirName}/${FileType}");      

    if [ ! -d ${NewDirName}/${FileType} ];
        then
            echo "$NewDir does not exist"      
            echo "Making $NewDir"
            mkdir -p ${NewDir}
            chmod -R a+rwx ${NewDir}
    fi

    echo "Now looking for renamed $FileType files"
    for FileToMove in $(find ${FolderName} -name "*.${FileType}");
    do  
        echo "Dir to move ${FileToMove}"        
        echo "File to move ${FileToMove##*/}"

        SpecificFile=$(echo "${FileToMove##*/}");       

        echo "Checking to see if $SpecificFile already exist in $NewDir"

        if [ -f ${NewDir}/${SpecificFile} ];
        then    
            x=0
            while [ -f ${NewDir}/${SpecificFile} ]        
                do
                echo "File ${SpecificFile} already exists, renaming"
                SpecificFile=${x}${SpecificFile}
                x=$(( x++ ));
                echo "Renamed to $SpecificFile"
            done
        fi    

    #rsync -rav --progress ${FileToMove} ${NewDir}/${SpecificFile}
    mv -v ${FileToMove} ${NewDir}/${SpecificFile}
    done

done
chmod -R a+rwx ${NewDir}

echo "Log Location should be: [ $LOG_LOCATION ]"
echo "Finished Process"