如何在 Bash 中比较和递归修改字符串

Question

I need to write a bash code performing some tasks I am going to explain. The input: two uppercase strings of same length, no matter their length is. Es:

 CYVFGDDAS -->  string1   ,   unchangeable reference string
 CRFDGVEAT -->  string2   ,   modifiable string

I am trying to write Bash code that is able to compare the characters with same index recursively starting from the first position:

-- beginnig of the cycle --

if the characters are the same skip any action and go to the
  the next position, 
  
while
  
if the characters are not the same the character of string1
  replaces the character of string2 at that position
  
the new string2 is saved in a file

a substituion code is also written in the same file (I will 
explain this below)

the old string2 is replaced by the new string2 in such a way 
its changes are retained

start anothe cycle from the beginning

------

Repeat the cycle until the last character is processed.

So, for the example above, the code should start checking from the first position where two C characters are placed. They match so no action is taken and both strings are left unchanged.

Going to he second position Y should replace R in the second string, the modified string should be saved and written in a text file togheter with the substitution code YA2V ( Y is the replacing character of string1, A is a costant character that must be present in all substitutions codes, 2 is the positional index where the substitution occurred, and V is the replaced character of string2).

I am proficient in Python which has a large number of modules for string manipulation but because the code should be added to a pre-existing Bash program I need to get this done in Bash environment (builtin commands, awk, sed etc, does not matter). Looks to me that Bash does not have an extended arsenal of tools like Python, so I am first of all wondering if this project is feasible or not.

However, what I tried so far is to convert the strings in blank separated fields by inserting spaces between the characters in such a way awk can deal better with them as fields but I did not go very far with this. Sorry for the lengthy explanation. Any help is greatly appreciated.

Answer 1

No recursion is needed, just iterate over the strings. You can use parameter expansion with a for loop:

#!/bin/bash
s1=CYVFGDDAS
s2=CRFDGVEAT

for ((i=0; i<${#s1} ; ++i)) ; do
    if [[ ${s1:i:1} != ${s2:i:1} ]] ; then
        printf '%s\n' "${s1:0:i+1}${s2:i+1}"
        printf '%s\n' "${s1:i:1}A$((i+1))${s2:i:1}"
    fi
done

${s1:i:1} means extract the substring of $s1 from position $i of length 1 . If the length is omitted, it extracts as much as it can.

It just outputs the strings, redirect them to files as you need.

CYFDGVEAT
YA2R
CYVDGVEAT
VA3F
CYVFGVEAT
FA4D
CYVFGDEAT
DA6V
CYVFGDDAT
DA7E
CYVFGDDAS
SA9T