Wāpuro Rōmaji

https://gist.github.com/grantovich/4aa38233fb3862301cb1285e18ce57ee

This exercise is an interesting real-world text conversion problem involving English and Japanese characters. No advance knowledge of Japanese is required!

• Background
• Base Requirements
• Extra: Geminates
• Extra: Digraphs
• Extra: Obsolete
• Appendix: Conversion Table

Background

(this is only context; you can skip this section entirely if you want)

The English alphabet has separate letters for consonant sounds (like k) and vowel sounds (like a). The Japanese "alphabet" instead has characters called kana that represent whole syllables (like か, which is pronounced "ka"). Kana come in two flavors, hiragana and katakana — in this exercise we'll only be using hiragana.

To help English-speakers pronounce Japanese words without knowing kana, systems of "romanization" were developed to express them using the English alphabet. The result is called rōmaji (fun fact: the ji is the same one as in "emoji" — it means "characters" or "writing"). For example, the greeting "こんにちは" can be rendered in rōmaji as konnichiwa.

English-speakers interested in writing kana have a related problem: how do you compose Japanese text using an English QWERTY keyboard? You can't "write rōmaji" — it's only a pronunciation guide, so it can't be cleanly converted back into kana. For example, the は above is actually the kana for ha, but is sometimes pronounced wa, despite wa having its own kana. If you type w followed by a, which character did you mean?

Wāpuro rōmaji solves this problem by providing a one-to-one mapping of English letters to kana. For example, with a wāpuro rōmaji input system, if you want は you must type an h followed by an a on a QWERTY keyboard. This lack of ambiguity makes it easier for machines, and humans, to process.

Base Requirements

At the bottom of this document you'll find a table of hiragana and their corresponding wāpuro rōmaji. This can be easily copied into your project by viewing the source of this file (the Raw button in the upper-right).

• Your program should accept valid strings of wāpuro rōmaji as input. A string is valid when it consists only of spaces and letter sequences that appear in the rōmaji column of the conversion table.

• If the input is valid, your program should output its hiragana conversion. Output should consist only of spaces and characters in the hiragana column of the conversion table.

• If the input is not valid, your program should output an error saying so (it doesn't need to explain why).

• Spaces act as word breaks in the input. Your output doesn't have to preserve the spaces, but they may change the conversion — see the examples below.

Examples

Input	Output	Notes
konnichiha	こんいちは	ko+n+ni+chi+ha
oyasuminasai	おやすみなさい	o+ya+su+mi+na+sa+i
yoroshiku ne	よろしくね	preserving spaces is optional
kana	かな	ka+na, however...
kan a	かんあ	...adding a space makes it ka+n+a
exodia	error	x is not a valid letter
cheese	error	che is not a valid syllable

Extra: Geminates

(extras can be done in any order)

Some consonants in Japanese can be "held" for an extra syllable-length (like the "k" sound in "bookkeeper"). These are called geminate consonants, and they are spelled differently from their "short" counterparts.

Geminate consonants are represented in rōmaji by doubling the first letter of a syllable, like the pp in the word ippai. To convert these to hiragana, look up the kana for the "short" version of the syllable (pa in this example), then add the special character "small tsu" (っ) before it.

Geminate consonants have some restrictions:

1. They cannot appear in the first syllable of a word
2. They can only appear in syllables starting with k, s, t, p, or ch

Examples

Input	Output	Notes
matte	まって	ma + (small tsu) + te
hippu	ひっぷ	hi + (small tsu) + pu
kocchi	こっち	only the c is doubled for ch
summo	error	m is not one of the allowed consonants
ma tte	error	cannot be in the first syllable of a word

Extra: Digraphs

(extras can be done in any order)

Our conversion table only covers the monographs of hiragana — syllables represented by a single character. There is another category called digraphs, which require two characters. kya, ju, and cho are examples of digraphs.

Digraphs are represented in rōmaji as:

1. Consonant (must have an i kana, cannot be w)
2. y (except when the consonant is sh, ch, or j)
3. Vowel (must be a, u, or o)

To convert these to hiragana, look up the i kana for the consonant (1), then append one of the special characters "small ya" (ゃ), "small yu" (ゅ), or "small yo" (ょ), according to the vowel sound (3).

Examples

Input	Output	Notes
kya	きゃ	ki + (small ya)
ju	じゅ	ji + (small yu)
cho	ちょ	chi + (small yo)
wya	error	w is not allowed in digraphs
nye	error	e is not one of the allowed vowels
tya	error	although t exists as a consonant, it has no i kana
shyo	error	correct spelling is sho, since the y is omitted with sh

Extra: Obsolete

(extras can be done in any order)

There are two extra syllables not present in the conversion table that are considered obsolete, and rarely seen outside of old place names or family names: wi (ゐ) and we (ゑ).

Add an option to your program that allows and converts wi and we. When the option is not enabled, obsolete syllables in the input should result in an error, consistent with the base requirements.

With this option enabled, you can convert the ancient poem Iroha: a "perfect pangram" that uses every monographic syllable, including wi and we, exactly once.

Input

iro ha nihoheto
chirinuru wo
wa ka yo tare so
tsune naramu
uwi no okuyama
kefu koete
asaki yume mishi
wehi mo sesu

Output

いろはにほへと
ちりぬるを
わかよたれそ
つねならむ
うゐのおくやま
けふこえて
あさきゆめみし
ゑひもせす

Appendix: Conversion Table

hiragana	rōmaji
ん	n
あ	a
い	i
う	u
え	e
お	o
か	ka
き	ki
く	ku
け	ke
こ	ko
さ	sa
し	shi
す	su
せ	se
そ	so
た	ta
ち	chi
つ	tsu
て	te
と	to
な	na
に	ni
ぬ	nu
ね	ne
の	no
は	ha
ひ	hi
ふ	fu
へ	he
ほ	ho
ま	ma
み	mi
む	mu
め	me
も	mo
や	ya
ゆ	yu
よ	yo
ら	ra
り	ri
る	ru
れ	re
ろ	ro
わ	wa
ゐ	wi
ゑ	we
を	wo
が	ga
ぎ	gi
ぐ	gu
げ	ge
ご	go
ざ	za
じ	ji
ず	zu
ぜ	ze
ぞ	zo
だ	da
ぢ	di
づ	du
で	de
ど	do
ば	ba
び	bi
ぶ	bu
べ	be
ぼ	bo
ぱ	pa
ぴ	pi
ぷ	pu
ぺ	pe
ぽ	po