#encoding_sjis

#Features

• Shift_JIS to UTF-8 decoding
• Streaming decode API with chunk boundary support
• encoding_rs-compliant behavior
• Character set support: ASCII, half-width katakana, hiragana, katakana, kanji (JIS X 0208)
• Error handling with replacement character (U+FFFD)
• Replacement tracking

#Installation

[dependencies]
encoding_sjis = "{ version = \"0.1.0\" }"

#Usage

#Simple Decode

test "simple decode" {
  let bytes = [0x82, 0xA0, 0x82, 0xA2, 0x82, 0xA4] // "あいう"
  let (result, had_replacements) = decode(src=bytes)
  @test.eq(result, "あいう")
  @test.eq(had_replacements, false)
}

#Streaming Decode

test "streaming decode" {
  let decoder = new_decoder()
  let chunk1 = [0x82, 0xA0] // "あ"
  let chunk2 = [0x82, 0xA2] // "い"

  let result1 = decoder.decode_to_string(src=chunk1, last=false)
  let result2 = decoder.decode_to_string(src=chunk2, last=true)

  @test.eq(result1, "あ")
  @test.eq(result2, "い")
}

#Chunk Boundary Handling

test "chunk boundary" {
  let decoder = new_decoder()
  let chunk1 = [0x82]  // First byte of "あ"
  let chunk2 = [0xA0]  // Second byte of "あ"

  let result1 = decoder.decode_to_string(src=chunk1, last=false)
  let result2 = decoder.decode_to_string(src=chunk2, last=true)

  @test.eq(result1, "")      // Incomplete sequence is buffered
  @test.eq(result2, "あ")    // Completed after second chunk
}

#Error Handling

test "error handling" {
  let bytes = [0x41, 0x80, 0x42] // "A" + invalid + "B"
  let (result, had_replacements) = decode(src=bytes)

  @test.eq(result, "A\ufffdB")  // U+FFFD for invalid byte
  @test.eq(had_replacements, true)
}

#Mixed Content

test "mixed content" {
  let bytes = [
    0x48, 0x65, 0x6C, 0x6C, 0x6F, 0x20,        // "Hello "
    0x82, 0xA0, 0x82, 0xA2,                    // "あい"
    0xB1, 0xB2, 0xB3                           // "ｱｲｳ" (half-width katakana)
  ]
  let (result, _) = decode(src=bytes)
  @test.eq(result, "Hello あいｱｲｳ")
}

#Convenience API

test "shift_jis_to_utf8" {
  let bytes = [0x82, 0xA0, 0x82, 0xA2]
  let result = shift_jis_to_utf8(data=bytes)
  @test.eq(result, "あい")
}

#API

#Functions

#decode(src: Bytes) -> (String, Bool)

• Returns: (decoded string, whether replacement characters were used)

#new_decoder() -> Decoder

#shift_jis_to_utf8(data: Bytes) -> String

#Decoder Methods

#Decoder::decode_to_string(src: Bytes, last: Bool) -> String

• src: Input byte chunk
• last: Set true for the final chunk (flushes any pending incomplete sequence)
• Returns: Decoded string

#Decoder::reset() -> Unit

#Decoder::had_replacements() -> Bool

#Types

#Decoder

• pending_first_byte: Stores the first byte of an incomplete 2-byte sequence
• had_replacements: Tracks if replacement characters were used

#DecodeResult

• result: CoderResult - operation status
• read: Number of bytes read
• written: Number of UTF-8 code units written
• had_replacements: Whether replacement characters were used

#CoderResult

• InputEmpty: Input exhausted (normal completion or waiting for more data)
• OutputFull: Output buffer full (for future buffer-based API)

#Supported Character Sets

• ASCII (0x00-0x7F)
• Half-width Katakana (0xA1-0xDF) → U+FF61-U+FF9F
• Hiragana (0x82A0-0x82F1)
• Full-width Katakana (0x8340-0x8396, 0x83AA-0x83EF, etc.)
• Kanji and symbols (JIS X 0208 compliant)

#License