Tcl

Convert utf-8 Text File to Windows-1252

Demonstrates how to convert a text file using the utf-8 byte representation to windows-1252. Tcl

load ./chilkat.dll

set success 0

# Converts a file containing the following to windows-1252:

# <greetings>
#     <message>Hello, world!</message>
#     <message>¡Hola, mundo!</message>
#     <message>Bonjour, le monde!</message>
#     <message>Hallo, Welt!</message>
#     <message>Olá, mundo!</message>
#     <message>Привет, мир!</message>
#     <message>你好，世界！</message>
#     <message>こんにちは、世界！</message>
#     <message>안녕하세요, 세계!</message>
#     <message>😊🌍</message>
# </greetings>

# --------------------------------------------------------------------------------------------------------------------------
# Note:
# Windows-1252 is an 8-bit single-byte encoding. It can only encode:
# 
#     The basic ASCII set (0x00–0x7F).
#     Latin-1 Supplement (0xA0–0xFF), plus some extra printable characters (like curly quotes, €, etc.).
#     In total: 256 possible code points, covering most Western European languages but nothing outside of Latin script.

# --------------------------------------------------------------------------------------------------------------------------
# Characters in your XML that are representable
# 
#     Hello, world! ✅ (ASCII only)
#     ¡Hola, mundo! ✅ (inverted exclamation mark U+00A1 is in Windows-1252)
#     Bonjour, le monde! ✅
#     Hallo, Welt! ✅
#     Olá, mundo! ✅ (U+00E1 á and U+00F3 ó are in Windows-1252)

# --------------------------------------------------------------------------------------------------------------------------
# Characters that break conversion
# 
#     Russian / Cyrillic: Привет, мир!
#     → These are Cyrillic characters (U+041F … U+0440). Not representable in Windows-1252. Conversion would require replacement (e.g. with ? or XML character references).
#     Chinese: 你好，世界！
#     → CJK ideographs (U+4F60, U+597D, etc.). Not in Windows-1252.
#     Japanese: こんにちは、世界！
#     → Hiragana + CJK. Not in Windows-1252.
#     Korean: 안녕하세요, 세계!
#     → Hangul syllables. Not in Windows-1252.
#     Emoji: 😊🌍
#     → Unicode Supplementary Multilingual Plane (U+1F60A, U+1F30D). Windows-1252 cannot encode any emoji.

set bd [new_CkBinData]

# Load the utf-8 bytes.
set success [CkBinData_LoadFile $bd "qa_data/xml/utf8test.xml"]
if {$success == 0} then {
    puts [CkBinData_lastErrorText $bd]
    delete_CkBinData $bd
    exit
}

# If allOrNone = 1, then the conversion fails and the contents of the BinData
# are left unchanged if any char is unconvertable.

# If allOrNone = 0, then non-convertable chars are discarded.
set allOrNone 0
set fromCharset "utf-8"
set toCharset "windows-1252"
set success [CkBinData_CharsetConvert $bd $fromCharset $toCharset $allOrNone]

# The return value will be 0 if any utf-8 chars were discarded because of non-convertability.
if {$success == 0} then {
    puts "Some utf-8 chars could not be converted to windows-1252"
} else {
    puts "All utf-8 chars were converted to windows-1252"
}

set success [CkBinData_WriteFile $bd "c:/temp/qa_output/out.xml"]

# The output file contains the following, where all non-convertable chars were discarded

# <greetings>
#     <message>Hello, world!</message>
#     <message>¡Hola, mundo!</message>
#     <message>Bonjour, le monde!</message>
#     <message>Hallo, Welt!</message>
#     <message>Olá, mundo!</message>
#     <message>, !</message>
#     <message></message>
#     <message></message>
#     <message>, !</message>
#     <message></message>
# </greetings>

delete_CkBinData $bd