Shell script to fix weird characters in folder/filenames and remove duplicate entries

fix-duplicate-names-in-zip.sh

#!/bin/bash
# This script was created for fixing zip files downloaded from using (https://github.com/dulldusk/phpfm/)
# This is a known issue but not fixed yet (https://github.com/dulldusk/phpfm/issues/27)
# The problem I have faced so far is that the php program replaces all _ characters with some unicode character ž and that
# all folders are accompanied by extra files with the same name (as mentioned in the issue page also)

# Requires: bash, 7z, zip

# Usage: make it executable ofcourse, and
# ./fix-duplicate-names-in-zip.sh test.zip
# and chillax. It will perform required actions on folders/files inside the zip itself.
# For safer side, backup your zip before running it. 
# You can modify the script according to your need if you observe some other problem
# with the zip.

renamelist=renamelist.txt

>| $renamelist

rencount=0

replaceChar(){
# replaces all 'ž' (/u009e) characters from folder/file names with '_' character
   
    filelist=`7z -slt l $1 | grep -oP "(?<=Path = ).+" | tail -n +2 | grep $'\u009e' | sort`

    IFS=$'\n'
    for f in $filelist
    do
       
        fbasename=`basename $f`
        fdirpath=`dirname $f`
        newpath=''
        newname=''
        
        if echo "$fbasename" | grep $'\u009e'; then
            rencount=$((rencount+1))
            newname=`echo $fbasename | sed 's/\xc2\x9e/_/g'` 

            if [ ! -z "$fdirpath" ] && [ "$fdirpath" != "." ]; then
                newpath=`echo $fdirpath | sed 's/\xc2\x9e/_/g'` #works with 7z
                newpath="$newpath/"
            fi

            echo $newpath$fbasename >> $renamelist
            echo $newpath$newname >> $renamelist
        fi

    done

    
    
    renameContent=`cat $renamelist`

    if [ ! -z "$renameContent" ]; then
        echo -e "\e[7;38;5;34mRenaming files and folders to fix ž characters..\e[0m"
        7z rn $1 @$renamelist
        sleep 1
    else
        echo -e "\e[7;38;5;34mNothing to rename..\e[0m"
        return
    fi

    if 7z -slt l $1 | grep -oP "(?<=Path = ).+" | tail -n +2 | grep $'\u009e'; then
        replaceChar $1
    fi
}



replaceChar $1
echo -e "\e[7;38;5;34mTotal $rencount files/folders renamed..\e[0m"

# exit




removeRedund(){
   # removes extra files with same names accompanying folders
    
    #works fine with either of below
    filelist=`zipinfo -1 $1 | sort` #preserves order
    # filelist=`7z -slt l $1 | grep -oP "(?<=Path = ).+" | tail -n +2 | sort`

    echo -e "\e[7;38;5;34mRemoving duplicate entries of files named same as folder names recursively..\e[0m"

    IFS=$'\n'
    for file in $filelist
    do
        fdir=`dirname $file`
        flist=`echo "$filelist" | grep $fdir | sort`
        if [ ! -z "$fdir" ]; then 
            echo -en "\r\033[KChecking $file... "
        fi
        
        for fname in $flist
        do
                if [[ "$fname" =~ "${file}/" ]]; then
                        zip -d $1 "$file" -UN=UTF8
                        break
                fi
        done

    done
}



removeRedund $1

echo ""