jbhannah/agate

Format ruby characters (currently only furigana) in text for display

agate's Issues

Testing and coverage

Test with RSpec and Travis-CI against all available Ruby versions, and use SimpleCov/Coveralls.io for coverage.

Extract HTML formatter to its own class

Create a class Agate::Formatter::HTML, have :formatter => :html be a default option with Agate::Parser, and load and run the specified formatter automatically in Agate::Parser#parse.

Strip ruby text

Formatter to strip ruby text from a passed string.

Plain-text formatter

"Formatter" that echoes the text passed to it, mostly for testing purposes.

Unicode interlinear annotation characters

http://eev.ee/blog/2015/09/12/dark-corners-of-unicode/#interesting-characters

Command-line parser

Take input via command line (piped through echo or as an argument) and return formatted text.

Usage:

agate (-h|[-f|--format <format>] [-d|--delimiters <delimiters>] <string>)

Flags:

• -d|--delimiters <delimiters> Specify delimiters to use for matching. Can be a single character or pair of characters wrapped around ruby characters (see #2).
• -f|--format <format> Specify a format to return. Possible values for <format>:
  • html (see #3)
• -h|--help Show usage information.

Pre-release cleanup

After all other issues in the v1.0.0 milestone have been resolved:

• All builds passing on Travis-CI
• 100% test coverage
• 4.0 Code Climate score

Refactor tests

• Test parser using plain text formatter (#12)
• Test individual formatters
• Test CLI
• Minimize redundant options, use defaults unless it's necessary to explicitly specify

Rdoc documentation

Write Rdoc documentation for inclusion on rdoc.info.

Only load desired formatters

Possibilities:

• Lazily load requested formatter when instantiating a new Agate::Parser
• Refactor so that a formatter is instantiated by user, and Agate::Parser is used internally

Custom delimiters

Allow user specification of custom delimiters to be used in matching Regexp (see #1) and as ruby parentheses (contents of rp element). Default to【 】.

Unicode regexp character properties don't work out of the box with Rubinius

From Travis-CI:

invalid character property name {Han}: (\p{Han}+)((?-mix:\())([\p{Hiragana}\p{Katakana}]+)((?-mix:\)))

String#chars returns an enumerator in Ruby 1.9

String#chars behaves differently in Ruby 1.9 from 2.0, returning an enumerator for use with a block instead of an array of characters. Breaks Agate::Parser at line 27:

undefined method `last' for #<Enumerator: "【】":chars>

Automatically generate furigana for text

Inspired by the furigana gem, but use a pure-Ruby implementation of or alternative to MeCab if possible, instead of needing to install and run an external binary. Add as an option to the parser and command-line (see #5).

Use regular expressions to match kanji followed by furigana

Unicode block property regex: \p{Hiragana}
Unicode block property regex: \p{Katakana}
Unicode block property regex: \p{Han}

(via) e.g.

/\p{Han}+【[\p{Hiragana}\p{Katakana}]+】/.match("勉強【べんきょう】します") # 勉強【べんきょう】

Language detection/abstraction

Detect the base language of the string being parsed to determine how to detect ruby characters. e.g. which regex to use: single character followed by delimited ruby text (all languages), or multiple characters followed by combined ruby text (e.g. Japanese kanji compounds).