Iconvex ICU

iconvex_icu provides the current ICU extension for Iconvex. It contains 212 native Elixir codecs and 191 package-owned mapping tables. Runtime conversion uses no ICU library, NIF, port, operating-system iconv, or external process.

This package owns current ICU 78.3 converter profiles and native implementations. The separate iconvex_icu_archive package owns 1,050 historical revision snapshots.

Package contents

The checked-in surface manifest defines all 212 canonical registrations. It also defines aliases, implementation modules, table IDs, and source artifacts.

Installation

def deps do
  [
    {:iconvex, "~> 0.1.1"},
    {:iconvex_icu, "~> 0.1.0"}
  ]
end

The OTP application registers the entire package in one transaction. A failed name or table-provider claim rolls back the operation. Stopping the application removes its routes and providers.

Examples

All codecs use Iconvex.convert/4:

{:ok, encoded} =
  Iconvex.convert("AあいB", "UTF-8", "ibm-930_P120-1999")

<<0xC1, 0x0E, 0x44, 0x81, 0x44, 0x82, 0x0F, 0xC2>> = encoded

{:ok, "AあいB"} =
  Iconvex.convert(encoded, "ibm-930_P120-1999", "UTF-8")

ICU's S/390 newline surface keeps the LF/NL distinction explicit:

{:ok, <<0x15, 0x25>>} =
  Iconvex.convert(<<0x0A::utf8, 0x85::utf8>>, "UTF-8",
    "ibm-1047_P100-1995,swaplfnl"
  )

LMBCS profiles are separate encodings because their optimization groups choose different canonical byte sequences:

{:ok, lmbcs} = Iconvex.convert("Lotus Ελληνικά", "UTF-8", "LMBCS-6")
{:ok, "Lotus Ελληνικά"} = Iconvex.convert(lmbcs, "LMBCS-6", "UTF-8")

The package preserves versioned and host-endian names exactly.

{:ok, bytes} = Iconvex.convert("A😀", "UTF-8", "UTF-16,version=2")
{:ok, "A😀"} = Iconvex.convert(bytes, "UTF-16,version=2", "UTF-8")

Provenance and verification

The mapping corpus uses ICU release 78.3 at Git revision 21d1eb0f306e1141c10931e914dfc038c06121da. Tests verify every source digest and reconstruct directional mappings from UCM files. They execute these cases:

• Every slot in all single-byte and swaplfnl tables.
• All 689,977 generated decoder-domain cases.
• All 691,381 canonical generated encoder mappings.
• Every stateful SI/SO source mapping.
• Original ICU-derived JIS, LMBCS, Unicode-variant, and Compound Text suites.
• Complete registration and provider lifecycle rollback.

The full Unicode scalar fixture covers LMBCS and Unicode variants. See BENCHMARKS.md for the executable ICU-oracle performance gates. All 24 LMBCS encode and decode paths remain below the 30x limit.

The package loads generated tables only when needed. It caches them in persistent_term. It shares Core's table and stateful engines instead of duplicating the conversion runtime.

Licensing

The original Iconvex source uses LGPL-2.1-or-later. The Unicode License v3 covers ICU/Unicode data and derived mapping tables. See LICENSE, LICENSE.UNICODE, and NOTICE.

Complete codec catalog

The generator builds this table from the release inventories. Do not edit this section by hand. Codec names ignore ASCII letter case. You can use each alias to look up its codec. Aliases do not increase the canonical codec count.

212 canonical codecs · 755 aliases · 967 accepted names.

Registered codec names