chore: note tasks and predefine function signatures

mklookup.py

@@ -6,6 +6,7 @@
 
 # FIX: moum to jungseong can be 1:1 mapped; it does not need a lookup table
 
+
 def jamo_to_compat_jamo(jamo: str, /) -> str | None:
     """Maps a Jamo character to a Compatibility Jamo character."""
     name = ud.name(jamo).split(" ")[-1]  # HANGUL CHOSEONG "KIYEOK"

ricecake/hangul/compose.py

@@ -1 +1,21 @@
 """(De)composition of Hangul Jamo and Syllable."""
+
+
+def compose(cho: str, jung: str, jong: str | None) -> str:
+    """Composes Choseong, Jungseong, and Jongseong into a Syllable."""
+    ...
+
+
+def decompose(syl: str) -> tuple[str, str, str | None]:
+    """Decomposes a Syllable into Choseong, Jungseong, and Jongseong."""
+    ...
+
+
+# FEAT: decompose composite Jaum and Moum into tuple of str
+# |
+# | - [ ] There are 5 cases, cho/jung/jong and compat jaum/moum.
+# |   Which should be included and which should be not?
+# |
+# | - [ ] Should it be keyboard-based or shape-based?
+# |   "ㅐ" is "ㅏ" + "ㅣ" but it can be typed at once with a keyboard.
+# |   If so, should "ㅙ" be decomposed to ("ㅗ", "ㅐ") or ("ㅗ", "ㅏ", "ㅣ")?

ricecake/hangul/convert.py

@@ -1 +1,21 @@
 """Conversion between Hangul Jamo and Compatibility Jamo."""
+
+
+def jamo_to_compat_jamo(c: str, /) -> str | None:
+    """Maps a Jamo character to a Compatibility Jamo character."""
+    ...
+
+
+def compat_jaum_to_choseong(c: str, /) -> str | None:
+    """Maps a Compatibility Jaum character to a Jamo Choseong character."""
+    ...
+
+
+def compat_moum_to_jungseong(c: str, /) -> str:
+    """Maps a Compatibility Moum character to a Jamo Jungseong character."""
+    ...
+
+
+def compat_jaum_to_jongseong(c: str, /) -> str:
+    """Maps a Compatibility Jaum character to a Jamo Jongseong character."""
+    ...

ricecake/hangul/detect.py

@@ -9,6 +9,7 @@
 JONGSEONG_COEF = 1
 
 
+# FIX: MAYBE: MODERN_{kind} vs {kind}_MODERN
 # https://en.wikipedia.org/wiki/Hangul_Jamo_(Unicode_block)
 JAMO_BASE = 0x1100  # 'ᄀ'
 JAMO_END = 0x11FF  # 'ᇿ'