How to Convert EUC-KR Source Code to UTF-8

I found old STM32 firmware code. The files were in EUC-KR encoding. This is an old Korean text format. Modern tools use UTF-8. So I needed to change all files to UTF-8.

The Problem

• My firmware project (STM32F10x chip)
• 358 source files (.c, .cpp, .h files)
• Most files were EUC-KR or ISO-8859-1
• I wanted to change everything to UTF-8

Solution: A Simple Script

I made a bash script. It uses iconv command. This command can change text encoding safely.

The Conversion Script

#!/bin/bash

# Change source files from EUC-KR to UTF-8

BASE_DIR="/path/to/your/project"
SUCCESS_COUNT=0
FAIL_COUNT=0
SKIP_COUNT=0

echo "=== Starting UTF-8 conversion ==="
echo "Working directory: $BASE_DIR"

# Find all source files
find "$BASE_DIR" -type f \( \
    -name "*.c" -o -name "*.cpp" -o -name "*.cc" -o \
    -name "*.h" -o -name "*.hpp" -o \
    -name "*.txt" -o -name "*.md" \
\) | while read -r file; do

    # Check current encoding
    ENCODING=$(file -b --mime-encoding "$file")

    # Skip if already UTF-8
    if [[ "$ENCODING" == "utf-8" || "$ENCODING" == "us-ascii" ]]; then
        echo "[SKIP] $file (already $ENCODING)"
        ((SKIP_COUNT++))
        continue
    fi

    # Convert EUC-KR to UTF-8
    if iconv -f EUC-KR -t UTF-8 "$file" > "$file.utf8.tmp" 2>/dev/null; then
        # Success: replace original file
        mv "$file.utf8.tmp" "$file"
        echo "[OK] $file ($ENCODING -> UTF-8)"
        ((SUCCESS_COUNT++))
    else
        # Failed: remove temp file
        rm -f "$file.utf8.tmp"
        echo "[FAIL] $file (conversion failed)"
        ((FAIL_COUNT++))
    fi
done

echo ""
echo "=== Conversion done ==="
echo "Success: $SUCCESS_COUNT"
echo "Failed: $FAIL_COUNT"
echo "Skipped: $SKIP_COUNT"

How to Use

1. Save the script

vi /tmp/convert_to_utf8.sh
# Paste the script above
# Change BASE_DIR to your project path

2. Make it executable

chmod +x /tmp/convert_to_utf8.sh

3. Run it (save log)

/tmp/convert_to_utf8.sh 2>&1 | tee /tmp/convert_utf8.log

4. Check results

# Count statistics
grep -c "^\[OK\]" /tmp/convert_utf8.log      # success
grep -c "^\[SKIP\]" /tmp/convert_utf8.log    # skipped
grep -c "^\[FAIL\]" /tmp/convert_utf8.log    # failed

# See failed files
grep "^\[FAIL\]" /tmp/convert_utf8.log

5. Verify encoding

# Check encoding after conversion
file -b --mime-encoding your_source_file.c

# Output: utf-8

Why This Script is Safe

1. Safe Conversion

• First converts to temporary file (.utf8.tmp)
• Only replaces original if successful
• Original file is kept if conversion fails

2. No Duplicate Work

• Uses file command to check current encoding
• Skips files already in UTF-8 or ASCII

3. Clear Logs

• Shows status for each file
• Gives final statistics

4. Many File Types

• C/C++ source: .c, .cpp, .cc, .h, .hpp
• Documents: .txt, .md

Important Tips

1. Make a backup first: Save your project before converting

   tar -czf project_backup_$(date +%Y%m%d).tar.gz /path/to/project

2. If using Git: Convert before committing

   git status  # check status before
   ./convert_to_utf8.sh
   git diff    # see what changed
   git add .
   git commit -m "Convert source files to UTF-8"

3. Text files only: The script only works on text files, not binary files

Other Uses

You can change the file types to convert:

# Python projects
-name "*.py" -o -name "*.pyx" -o -name "*.pyi"

# Java projects
-name "*.java" -o -name "*.xml" -o -name "*.properties"

# Web projects
-name "*.html" -o -name "*.css" -o -name "*.js" -o -name "*.jsx"

Conclusion

Using iconv and find together makes it easy to update old projects. This is very useful for embedded projects with old code. The script is safe and works well.

---

Tags: linux, encoding, utf8, euckr, bash, script, embedded