USE data overrides

[behdad]

From an opentype-layout thread, from Andrew Glass:

Here is the override:

  # ================================================
  # Indic_Matra_Category=Bottom
  0F72        ; Bottom # Mn      TIBETAN VOWEL SIGN I
  0F7A..0F7D  ; Bottom # Mn  [4] TIBETAN VOWEL SIGN E..TIBETAN VOWEL SIGN OO
  0F80        ; Bottom # Mn      TIBETAN VOWEL SIGN REVERSED I
  11127..11129; Bottom # Mn  [3] CHAKMA VOWEL SIGN A..CHAKMA VOWEL SIGN II
  1112D       ; Bottom # Mn      CHAKMA VOWEL SIGN AI
  11130       ; Bottom # Mn      CHAKMA VOWEL SIGN OI

[behdad]

From another email from Andrew Glass, also:

  # Indic_Syllabic_Category=Gemination_Mark
  11134      ; Gemination_Mark # Mc      CHAKMA MAAYYAA

[behdad]

Actually, those are now listed in USE spec itself:
https://docs.microsoft.com/en-gb/typography/script-development/use#overrides

Overrides to Indic_Syllabic_Category

  # Indic_Syllabic_Category=Bindu
  AA29       ; Bindu          # Mn       CHAM VOWEL SIGN AA
  # ================================================
  # Indic_Syllabic_Category=Nukta
  0F71       ; Nukta          # Mn       TIBETAN VOWEL SIGN AA
  # ================================================
  # Indic_Syllabic_Category=Tone_Mark
  A982       ; Tone_Mark      # Mn       JAVANESE SIGN LAYAR
  # ================================================
  # Indic_Syllabic_Category=Consonant_Dead
  0F7F       ; Consonant_Dead # Mc       TIBETAN SIGN RNAM BCAD
  # ================================================
  # Indic_Syllabic_Category=Gemination_Mark 
  11134      ; Gemination_Mark # Mc      CHAKMA MAAYYAA

Overrides to Indic_Positional_Category

  # Indic_Matra_Category=Top
  0F74        ; Top     # Mn      TIBETAN VOWEL SIGN U
  AA35        ; Top     # Mn      CHAM CONSONANT SIGN
  1A18        ; Top     # Mn      BUGINESE VOWEL SIGN U

  # ================================================
  # Indic_Matra_Category=Bottom
  0F72        ; Bottom  # Mn      TIBETAN VOWEL SIGN I
  0F7A..0F7D  ; Bottom  # Mn  [4] TIBETAN VOWEL SIGN E..TIBETAN VOWEL SIGN OO
  0F80        ; Bottom  # Mn      TIBETAN VOWEL SIGN REVERSED I
  11127..11129; Bottom  # Mn  [3] CHAKMA VOWEL SIGN A..CHAKMA VOWEL SIGN II
  1112D       ; Bottom  # Mn      CHAKMA VOWEL SIGN AI
  11130       ; Bottom  # Mn      CHAKMA VOWEL SIGN OI

[behdad]

@jfkthame Any comments here? For Tibetan we decided to apply the override based on category (all top matras) instead of hardcoding. Maybe I should just implement override files and ship these verbatim?

[behdad]

From Ned email from Jan 3, 2018:

# Indic_Syllabic_Category=Bindu
AA29       ; Bindu          # Mn       CHAM VOWEL SIGN AA
# ================================================
# Indic_Syllabic_Category=Nukta
0F71       ; Nukta          # Mn       TIBETAN VOWEL SIGN AA
1BF2..1BF3 ; Nukta          # Mc [2]   BATAK PANGOLAT..BATAK PANONGONAN
# ================================================
# Indic_Syllabic_Category=Tone_Mark
1CED       ; Tone_Mark      # Mn       VEDIC SIGN TIRYAK
A982       ; Tone_Mark      # Mn       JAVANESE SIGN LAYAR
# ================================================
# Indic_Syllabic_Category=Consonant_Dead
002D       ; Consonant_Dead # Pd       HYPHEN MINUS
0F7F       ; Consonant_Dead # Mc       TIBETAN SIGN RNAM BCAD
# ================================================
# Indic_Syllabic_Category=Gemination_Mark
11134      ; Gemination_Mark # Mc      CHAKMA MAAYYAA
# ================================================
# Indic_Syllabic_Category=Symbol_Combining
1B6B..1B73 ; Symbol_Combining # Mn [9] BALINESE MUSICAL SYMBOL COMBINING TEGEH..BALINESE MUSICAL SYMBOL COMBINING GONG
# ================================================
# Indic_Syllabic_Category=Cantillation_Mark
1CE2..1CE8 ; Cantillation_Mark # Mn [7] VEDIC SIGN VISARGA SVARITA..VEDIC SIGN VISARGA ANUDATTA WITH TAIL
# ================================================
# Indic_Syllabic_Category=Consonant
A8F2..A8F7 ; Consonant      # Lo [6]   DEVANAGARI SIGN SPACING CANDRABINDU..DEVANAGARI SIGN CANDRABINDU AVAGRAHA
A8FB       ; Consonant      # Lo       DEVANAGARI HEADSTROKE
A8FD       ; Consonant      # Lo       DEVANAGARI JAIN OM
# ================================================
# Indic_Syllabic_Category=Consonant_Medial
0A51       ; Consonant_Medial # Mn     GURMUKHI SIGN UDAAT
A8B4       ; Consonant_Medial # Mc     SAURASHTRA CONSONANT SIGN HAARU
# ================================================
# Indic_Syllabic_Category=Consonant_Placeholder
0C80       ; Consonant_Placeholder # Lo     KANNADA SIGN SPACING CANDRABINDU
2015       ; Consonant_Placeholder # Pd     HORIZONTAL BAR
2022       ; Consonant_Placeholder # Po     BULLET
25FB..25FE ; Consonant_Placeholder # Sm [4] WHITE MEDIUM SQUARE..BLACK MEDIUM SMALL SQUARE

# Indic_Positional_Category=Left
1C29          ; Left # Mc       LEPCHA VOWEL SIGN OO
# ================================================
# Indic_Positional_Category=Top
0F74       ; Top     # Mn      TIBETAN VOWEL SIGN U
1A18       ; Top     # Mn      BUGINESE VOWEL SIGN U
1B6B       ; Top     # Mn      BALINESE MUSICAL SYMBOL COMBINING TEGEH
1B6D..1B73 ; Top     # Mn  [7] BALINESE MUSICAL SYMBOL COMBINING KEMPUL..BALINESE MUSICAL SYMBOL COMBINING GONG
AA35        ; Top     # Mn      CHAM CONSONANT SIGN
# ================================================
# Indic_Positional_Category=Bottom
0A51       ; Bottom  # Mn      GURMUKHI SIGN UDAAT
0F72       ; Bottom  # Mn      TIBETAN VOWEL SIGN I
0F7A..0F7D ; Bottom  # Mn  [4] TIBETAN VOWEL SIGN E..TIBETAN VOWEL SIGN OO
0F80       ; Bottom  # Mn      TIBETAN VOWEL SIGN REVERSED I
1B6C       ; Bottom  # Mn      BALINESE MUSICAL SYMBOL COMBINING ENDEP
1BF2..1BF3 ; Bottom  # Mc  [2] BATAK PANGOLAT..BATAK PANONGONAN
11127..11129; Bottom # Mn  [3] CHAKMA VOWEL SIGN A..CHAKMA VOWEL SIGN II
1112D       ; Bottom # Mn      CHAKMA VOWEL SIGN AI
11130       ; Bottom # Mn      CHAKMA VOWEL SIGN OI

[brawer]

Is there somewhere a complete list of overrides? (For getting the data fixed upstream, at Unicode).

[behdad]

I've been maintaining what I've received from other implementations here, to be cross-checked with what we do in HarfBuzz at some point.

[KrasnayaPloshchad]

CC @domschl

[KrasnayaPloshchad]

CC @eroux

[Richard57]

Is there somewhere a complete list of overrides? (For getting the data fixed upstream, at Unicode).

Last UTC meeting I was at (by phone) that discussed the issue declared that the needs of the USE did not determine the Unicode categories. For example, one couldn't change Tai Tham Consonant_Medial to Consonant_Subjoined just because that gave better results from the USE.