sebgod/mercury-unicode

Unicode character database parser and library for the Mercury language

mercury-unicode

Introduction

Unicode character database parser and library for the Mercury language. The implementation uses the latest Unicode release, version 8.0. As Unicode updates are rather frequent, this code uses a generative approach, to facilitate the frequent addition of characters, blocks and scripts.

The project can be build using any current Mercury compiler in any legal grade.

A consequence of this approach is that this library can also be used as a common ground for portable libraries which must ensure a consistent behaviour across different execution environments. This means that no native code is involved, and all facts and data are purely derived from the UCD file.

Components

1. ucd_type_compiler.m

• Input: PropertyValueAliases.txt
• Output: ucd_types.m

2. ucd_compiler.m

• The UCD compiler itself is modular, and links together a list of processors:
  • process_scripts.m
  • process_blocks.m
  • process_unicode_data.m

3. Makefile

• Variables:
  • MMC: the Mercury compiler executable
  • MCFLAGS: the flags for the given Mercury compiler, default: --use-grade-subdirs
• Targets:
  • clean: removes all generated code files and binary output
  • update: forces an update of all downloaded files
  • realclean: invokes clean, but also removes downloaded files
  • libucd: A library containing all expored functions and predicates dealing with Unicode, sourcing ucd.m