diff --git a/fauntlet/src/compare_glyphs.rs b/fauntlet/src/compare_glyphs.rs
index 9d8bc1bba..0eaa5248e 100644
--- a/fauntlet/src/compare_glyphs.rs
+++ b/fauntlet/src/compare_glyphs.rs
@@ -1,3 +1,5 @@
+use crate::Hinting;
+
 use super::{FreeTypeInstance, InstanceOptions, RegularizingPen, SkrifaInstance};
 use skrifa::{outline::pen::PathElement, GlyphId};
 use std::{io::Write, path::Path};
@@ -13,6 +15,21 @@ pub fn compare_glyphs(
         // Don't run on bitmap fonts (yet)
         return true;
     }
+    if let Some(Hinting::Auto(_)) = options.hinting {
+        // This font is a pathological case for autohinting due to the
+        // extreme number of generated segments and edges. To be
+        // precise, it takes longer to test this single font than the
+        // entire remainder of the Google Fonts corpus so we skip it
+        // here.
+        // Discussion at <https://github.com/googlefonts/fontations/issues/1147>
+        if ft_instance
+            .family_name()
+            .unwrap_or_default()
+            .contains("Handjet")
+        {
+            return true;
+        }
+    }
     let glyph_count = skrifa_instance.glyph_count();
     let is_scaled = options.ppem != 0;
 
@@ -86,6 +103,7 @@ pub fn compare_glyphs(
             if exit_on_fail {
                 std::process::exit(1);
             }
+            break;
         }
     }
     ok
diff --git a/fauntlet/src/font/freetype.rs b/fauntlet/src/font/freetype.rs
index 2fbd24459..ca6f0be05 100644
--- a/fauntlet/src/font/freetype.rs
+++ b/fauntlet/src/font/freetype.rs
@@ -3,26 +3,12 @@ use freetype::{
     ffi::{FT_Long, FT_Vector},
     Face, Library,
 };
-use skrifa::{
-    outline::{HintingMode, LcdLayout, OutlinePen},
-    GlyphId,
-};
+use skrifa::{outline::OutlinePen, GlyphId};
 
 use std::ffi::{c_int, c_void};
 
 use super::{InstanceOptions, SharedFontData};
 
-fn load_flags_from_hinting(mode: HintingMode) -> LoadFlag {
-    match mode {
-        HintingMode::Strong => LoadFlag::TARGET_MONO,
-        HintingMode::Smooth { lcd_subpixel, .. } => match lcd_subpixel {
-            Some(LcdLayout::Horizontal) => LoadFlag::TARGET_LCD,
-            Some(LcdLayout::Vertical) => LoadFlag::TARGET_LCD_V,
-            None => LoadFlag::TARGET_NORMAL,
-        },
-    }
-}
-
 pub struct FreeTypeInstance {
     face: Face<SharedFontData>,
     load_flags: LoadFlag,
@@ -37,10 +23,10 @@ impl FreeTypeInstance {
         let mut face = library
             .new_memory_face2(data.clone(), options.index as isize)
             .ok()?;
-        let mut load_flags = LoadFlag::NO_AUTOHINT | LoadFlag::NO_BITMAP;
+        let mut load_flags = LoadFlag::NO_BITMAP;
         match options.hinting {
             None => load_flags |= LoadFlag::NO_HINTING,
-            Some(hinting) => load_flags |= load_flags_from_hinting(hinting),
+            Some(hinting) => load_flags |= hinting.freetype_load_flags(),
         };
         if options.ppem != 0 {
             face.set_pixel_sizes(options.ppem, options.ppem).ok()?;
@@ -76,6 +62,10 @@ impl FreeTypeInstance {
         Some(Self { face, load_flags })
     }
 
+    pub fn family_name(&self) -> Option<String> {
+        self.face.family_name()
+    }
+
     pub fn is_scalable(&self) -> bool {
         self.face.is_scalable()
     }
diff --git a/fauntlet/src/font/mod.rs b/fauntlet/src/font/mod.rs
index 6a8975f77..5d8f46a02 100644
--- a/fauntlet/src/font/mod.rs
+++ b/fauntlet/src/font/mod.rs
@@ -4,9 +4,9 @@ use std::{
     sync::Arc,
 };
 
-use ::freetype::Library;
+use ::freetype::{face::LoadFlag, Library};
 use ::skrifa::{
-    outline::HintingMode,
+    outline::{HintingOptions, SmoothMode, Target},
     raw::{types::F2Dot14, FileRef, FontRef, TableProvider},
 };
 
@@ -16,21 +16,76 @@ mod skrifa;
 pub use freetype::FreeTypeInstance;
 pub use skrifa::SkrifaInstance;
 
+#[derive(Copy, Clone, PartialEq, Eq, Default, Debug)]
+pub enum HintingTarget {
+    #[default]
+    Normal,
+    Light,
+    Lcd,
+    VerticalLcd,
+    Mono,
+}
+
+impl HintingTarget {
+    pub fn to_skrifa_target(self) -> Target {
+        match self {
+            Self::Normal => SmoothMode::Normal.into(),
+            Self::Light => SmoothMode::Light.into(),
+            Self::Lcd => SmoothMode::Lcd.into(),
+            Self::VerticalLcd => SmoothMode::VerticalLcd.into(),
+            Self::Mono => Target::Mono,
+        }
+    }
+
+    pub fn to_freetype_load_flags(self) -> LoadFlag {
+        match self {
+            Self::Normal => LoadFlag::TARGET_NORMAL,
+            Self::Light => LoadFlag::TARGET_LIGHT,
+            Self::Lcd => LoadFlag::TARGET_LCD,
+            Self::VerticalLcd => LoadFlag::TARGET_LCD_V,
+            Self::Mono => LoadFlag::TARGET_MONO,
+        }
+    }
+}
+
+#[derive(Copy, Clone, PartialEq, Eq, Debug)]
+pub enum Hinting {
+    Interpreter(HintingTarget),
+    Auto(HintingTarget),
+}
+
+impl Hinting {
+    pub fn skrifa_options(self) -> HintingOptions {
+        match self {
+            Self::Interpreter(target) => HintingOptions {
+                engine: ::skrifa::outline::Engine::Interpreter,
+                target: target.to_skrifa_target(),
+            },
+            Self::Auto(target) => HintingOptions {
+                engine: ::skrifa::outline::Engine::Auto(None),
+                target: target.to_skrifa_target(),
+            },
+        }
+    }
+
+    pub fn freetype_load_flags(self) -> LoadFlag {
+        match self {
+            Self::Interpreter(target) => LoadFlag::NO_AUTOHINT | target.to_freetype_load_flags(),
+            Self::Auto(target) => LoadFlag::FORCE_AUTOHINT | target.to_freetype_load_flags(),
+        }
+    }
+}
+
 #[derive(Copy, Clone, Debug)]
 pub struct InstanceOptions<'a> {
     pub index: usize,
     pub ppem: u32,
     pub coords: &'a [F2Dot14],
-    pub hinting: Option<HintingMode>,
+    pub hinting: Option<Hinting>,
 }
 
 impl<'a> InstanceOptions<'a> {
-    pub fn new(
-        index: usize,
-        ppem: u32,
-        coords: &'a [F2Dot14],
-        hinting: Option<HintingMode>,
-    ) -> Self {
+    pub fn new(index: usize, ppem: u32, coords: &'a [F2Dot14], hinting: Option<Hinting>) -> Self {
         Self {
             index,
             ppem,
diff --git a/fauntlet/src/font/skrifa.rs b/fauntlet/src/font/skrifa.rs
index 44986ebdd..873f1e5d9 100644
--- a/fauntlet/src/font/skrifa.rs
+++ b/fauntlet/src/font/skrifa.rs
@@ -27,8 +27,13 @@ impl<'a> SkrifaInstance<'a> {
         let outlines = font.outline_glyphs();
         let hinter = if options.ppem != 0 && options.hinting.is_some() {
             Some(
-                HintingInstance::new(&outlines, size, options.coords, options.hinting.unwrap())
-                    .ok()?,
+                HintingInstance::new(
+                    &outlines,
+                    size,
+                    options.coords,
+                    options.hinting.unwrap().skrifa_options(),
+                )
+                .ok()?,
             )
         } else {
             None
diff --git a/fauntlet/src/lib.rs b/fauntlet/src/lib.rs
index a2859b009..296a506a0 100644
--- a/fauntlet/src/lib.rs
+++ b/fauntlet/src/lib.rs
@@ -3,5 +3,7 @@ mod font;
 mod pen;
 
 pub use compare_glyphs::compare_glyphs;
-pub use font::{Font, FreeTypeInstance, InstanceOptions, SharedFontData, SkrifaInstance};
+pub use font::{
+    Font, FreeTypeInstance, Hinting, HintingTarget, InstanceOptions, SharedFontData, SkrifaInstance,
+};
 pub use pen::RegularizingPen;
diff --git a/skrifa/generated/generated_autohint_styles.rs b/skrifa/generated/generated_autohint_styles.rs
index 2a1f6733c..1d077f27a 100644
--- a/skrifa/generated/generated_autohint_styles.rs
+++ b/skrifa/generated/generated_autohint_styles.rs
@@ -8,811 +8,751 @@ pub(super) const SCRIPT_CLASSES: &[ScriptClass] = &[
         name: "Adlam",
         group: ScriptGroup::Default,
         tag: Tag::new(b"ADLM"),
-        index: 0,
         hint_top_to_bottom: false,
-        std_chars: &['𞤌', '𞤮', ],
+        std_chars: "𞤌 𞤮",
         blues: &[
-            (&['𞤌', '𞤅', '𞤈', '𞤏', '𞤔', '𞤚', ], TOP),
-            (&['𞤂', '𞤖', ], 0),
-            (&['𞤬', '𞤮', '𞤻', '𞤼', '𞤾', ], TOP | LATIN_X_HEIGHT),
-            (&['𞤤', '𞤨', '𞤩', '𞤭', '𞤴', '𞤸', '𞤺', '𞥀', ], 0),
+            ("𞤌 𞤅 𞤈 𞤏 𞤔 𞤚", TOP),
+            ("𞤂 𞤖", 0),
+            ("𞤬 𞤮 𞤻 𞤼 𞤾", TOP | LATIN_X_HEIGHT),
+            ("𞤤 𞤨 𞤩 𞤭 𞤴 𞤸 𞤺 𞥀", 0),
         ],
     },
     ScriptClass {
         name: "Arabic",
         group: ScriptGroup::Default,
         tag: Tag::new(b"ARAB"),
-        index: 1,
         hint_top_to_bottom: false,
-        std_chars: &['ل', 'ح', 'ـ', ],
+        std_chars: "ل ح ـ",
         blues: &[
-            (&['ا', 'إ', 'ل', 'ك', 'ط', 'ظ', ], TOP),
-            (&['ت', 'ث', 'ط', 'ظ', 'ك', ], 0),
-            (&['ـ', ], LATIN_NEUTRAL),
+            ("ا إ ل ك ط ظ", TOP),
+            ("ت ث ط ظ ك", 0),
+            ("ـ", LATIN_NEUTRAL),
         ],
     },
     ScriptClass {
         name: "Armenian",
         group: ScriptGroup::Default,
         tag: Tag::new(b"ARMN"),
-        index: 2,
         hint_top_to_bottom: false,
-        std_chars: &['ս', 'Ս', ],
+        std_chars: "ս Ս",
         blues: &[
-            (&['Ա', 'Մ', 'Ւ', 'Ս', 'Բ', 'Գ', 'Դ', 'Օ', ], TOP),
-            (&['Ւ', 'Ո', 'Դ', 'Ճ', 'Շ', 'Ս', 'Տ', 'Օ', ], 0),
-            (&['ե', 'է', 'ի', 'մ', 'վ', 'ֆ', 'ճ', ], TOP),
-            (&['ա', 'յ', 'ւ', 'ս', 'գ', 'շ', 'ր', 'օ', ], TOP | LATIN_X_HEIGHT),
-            (&['հ', 'ո', 'ճ', 'ա', 'ե', 'ծ', 'ս', 'օ', ], 0),
-            (&['բ', 'ը', 'ի', 'լ', 'ղ', 'պ', 'փ', 'ց', ], 0),
+            ("Ա Մ Ւ Ս Բ Գ Դ Օ", TOP),
+            ("Ւ Ո Դ Ճ Շ Ս Տ Օ", 0),
+            ("ե է ի մ վ ֆ ճ", TOP),
+            ("ա յ ւ ս գ շ ր օ", TOP | LATIN_X_HEIGHT),
+            ("հ ո ճ ա ե ծ ս օ", 0),
+            ("բ ը ի լ ղ պ փ ց", 0),
         ],
     },
     ScriptClass {
         name: "Avestan",
         group: ScriptGroup::Default,
         tag: Tag::new(b"AVST"),
-        index: 3,
         hint_top_to_bottom: false,
-        std_chars: &['𐬚', ],
+        std_chars: "𐬚",
         blues: &[
-            (&['𐬀', '𐬁', '𐬐', '𐬛', ], TOP),
-            (&['𐬀', '𐬁', ], 0),
+            ("𐬀 𐬁 𐬐 𐬛", TOP),
+            ("𐬀 𐬁", 0),
         ],
     },
     ScriptClass {
         name: "Bamum",
         group: ScriptGroup::Default,
         tag: Tag::new(b"BAMU"),
-        index: 4,
         hint_top_to_bottom: false,
-        std_chars: &['ꛁ', 'ꛯ', ],
+        std_chars: "ꛁ ꛯ",
         blues: &[
-            (&['ꚧ', 'ꚨ', 'ꛛ', 'ꛉ', 'ꛁ', 'ꛈ', 'ꛫ', 'ꛯ', ], TOP),
-            (&['ꚭ', 'ꚳ', 'ꚶ', 'ꛬ', 'ꚢ', 'ꚽ', 'ꛯ', '꛲', ], 0),
+            ("ꚧ ꚨ ꛛ ꛉ ꛁ ꛈ ꛫ ꛯ", TOP),
+            ("ꚭ ꚳ ꚶ ꛬ ꚢ ꚽ ꛯ ꛲", 0),
         ],
     },
     ScriptClass {
         name: "Bengali",
         group: ScriptGroup::Default,
         tag: Tag::new(b"BENG"),
-        index: 5,
         hint_top_to_bottom: true,
-        std_chars: &['০', '৪', ],
+        std_chars: "০ ৪",
         blues: &[
-            (&['ই', 'ট', 'ঠ', 'ি', 'ী', 'ৈ', 'ৗ', ], TOP),
-            (&['ও', 'এ', 'ড', 'ত', 'ন', 'ব', 'ল', 'ক', ], TOP),
-            (&['অ', 'ড', 'ত', 'ন', 'ব', 'ভ', 'ল', 'ক', ], TOP | LATIN_NEUTRAL | LATIN_X_HEIGHT),
-            (&['অ', 'ড', 'ত', 'ন', 'ব', 'ভ', 'ল', 'ক', ], 0),
+            ("ই ট ঠ ি ী ৈ ৗ", TOP),
+            ("ও এ ড ত ন ব ল ক", TOP),
+            ("অ ড ত ন ব ভ ল ক", TOP | LATIN_NEUTRAL | LATIN_X_HEIGHT),
+            ("অ ড ত ন ব ভ ল ক", 0),
         ],
     },
     ScriptClass {
         name: "Buhid",
         group: ScriptGroup::Default,
         tag: Tag::new(b"BUHD"),
-        index: 6,
         hint_top_to_bottom: false,
-        std_chars: &['ᝋ', 'ᝏ', ],
+        std_chars: "ᝋ ᝏ",
         blues: &[
-            (&['ᝐ', 'ᝈ', ], TOP),
-            (&['ᝅ', 'ᝊ', 'ᝎ', ], TOP),
-            (&['ᝂ', 'ᝃ', 'ᝉ', 'ᝌ', ], TOP | LATIN_X_HEIGHT),
-            (&['ᝀ', 'ᝃ', 'ᝆ', 'ᝉ', 'ᝋ', 'ᝏ', 'ᝑ', ], 0),
+            ("ᝐ ᝈ", TOP),
+            ("ᝅ ᝊ ᝎ", TOP),
+            ("ᝂ ᝃ ᝉ ᝌ", TOP | LATIN_X_HEIGHT),
+            ("ᝀ ᝃ ᝆ ᝉ ᝋ ᝏ ᝑ", 0),
         ],
     },
     ScriptClass {
         name: "Chakma",
         group: ScriptGroup::Default,
         tag: Tag::new(b"CAKM"),
-        index: 7,
         hint_top_to_bottom: false,
-        std_chars: &['𑄤', '𑄉', '𑄛', ],
+        std_chars: "𑄤 𑄉 𑄛",
         blues: &[
-            (&['𑄃', '𑄅', '𑄉', '𑄙', '𑄗', ], TOP),
-            (&['𑄅', '𑄛', '𑄝', '𑄗', '𑄓', ], 0),
-            (&['𑄖', '𑄘', '𑄙', '𑄤', '𑄥', ], 0),
+            ("𑄃 𑄅 𑄉 𑄙 𑄗", TOP),
+            ("𑄅 𑄛 𑄝 𑄗 𑄓", 0),
+            ("𑄖𑄳𑄢 𑄘𑄳𑄢 𑄙𑄳𑄢 𑄤𑄳𑄢 𑄥𑄳𑄢", 0),
         ],
     },
     ScriptClass {
         name: "Canadian Syllabics",
         group: ScriptGroup::Default,
         tag: Tag::new(b"CANS"),
-        index: 8,
         hint_top_to_bottom: false,
-        std_chars: &['ᑌ', 'ᓚ', ],
+        std_chars: "ᑌ ᓚ",
         blues: &[
-            (&['ᗜ', 'ᖴ', 'ᐁ', 'ᒣ', 'ᑫ', 'ᑎ', 'ᔑ', 'ᗰ', ], TOP),
-            (&['ᗶ', 'ᖵ', 'ᒧ', 'ᐃ', 'ᑌ', 'ᒍ', 'ᔑ', 'ᗢ', ], 0),
-            (&['ᓓ', 'ᓕ', 'ᓀ', 'ᓂ', 'ᓄ', 'ᕄ', 'ᕆ', 'ᘣ', ], TOP | LATIN_X_HEIGHT),
-            (&['ᕃ', 'ᓂ', 'ᓀ', 'ᕂ', 'ᓗ', 'ᓚ', 'ᕆ', 'ᘣ', ], 0),
-            (&['ᐪ', 'ᙆ', 'ᣘ', 'ᐢ', 'ᒾ', 'ᣗ', 'ᔆ', ], TOP),
-            (&['ᙆ', 'ᗮ', 'ᒻ', 'ᐞ', 'ᔆ', 'ᒡ', 'ᒢ', 'ᓑ', ], 0),
+            ("ᗜ ᖴ ᐁ ᒣ ᑫ ᑎ ᔑ ᗰ", TOP),
+            ("ᗶ ᖵ ᒧ ᐃ ᑌ ᒍ ᔑ ᗢ", 0),
+            ("ᓓ ᓕ ᓀ ᓂ ᓄ ᕄ ᕆ ᘣ", TOP | LATIN_X_HEIGHT),
+            ("ᕃ ᓂ ᓀ ᕂ ᓗ ᓚ ᕆ ᘣ", 0),
+            ("ᐪ ᙆ ᣘ ᐢ ᒾ ᣗ ᔆ", TOP),
+            ("ᙆ ᗮ ᒻ ᐞ ᔆ ᒡ ᒢ ᓑ", 0),
         ],
     },
     ScriptClass {
         name: "Carian",
         group: ScriptGroup::Default,
         tag: Tag::new(b"CARI"),
-        index: 9,
         hint_top_to_bottom: false,
-        std_chars: &['𐊫', '𐋉', ],
+        std_chars: "𐊫 𐋉",
         blues: &[
-            (&['𐊧', '𐊫', '𐊬', '𐊭', '𐊱', '𐊺', '𐊼', '𐊿', ], TOP),
-            (&['𐊣', '𐊧', '𐊷', '𐋀', '𐊫', '𐊸', '𐋉', ], 0),
+            ("𐊧 𐊫 𐊬 𐊭 𐊱 𐊺 𐊼 𐊿", TOP),
+            ("𐊣 𐊧 𐊷 𐋀 𐊫 𐊸 𐋉", 0),
         ],
     },
     ScriptClass {
         name: "Cherokee",
         group: ScriptGroup::Default,
         tag: Tag::new(b"CHER"),
-        index: 10,
         hint_top_to_bottom: false,
-        std_chars: &['Ꭴ', 'Ꮕ', 'ꮕ', ],
+        std_chars: "Ꭴ Ꮕ ꮕ",
         blues: &[
-            (&['Ꮖ', 'Ꮋ', 'Ꭼ', 'Ꮓ', 'Ꭴ', 'Ꮳ', 'Ꭶ', 'Ꮥ', ], TOP),
-            (&['Ꮖ', 'Ꮋ', 'Ꭼ', 'Ꮓ', 'Ꭴ', 'Ꮳ', 'Ꭶ', 'Ꮥ', ], 0),
-            (&['ꮒ', 'ꮤ', 'ꮶ', 'ꭴ', 'ꭾ', 'ꮗ', 'ꮝ', 'ꮿ', ], TOP),
-            (&['ꮖ', 'ꭼ', 'ꮓ', 'ꮠ', 'ꮳ', 'ꭶ', 'ꮥ', 'ꮻ', ], TOP | LATIN_X_HEIGHT),
-            (&['ꮖ', 'ꭼ', 'ꮓ', 'ꮠ', 'ꮳ', 'ꭶ', 'ꮥ', 'ꮻ', ], 0),
-            (&['ᏸ', 'ꮐ', 'ꭹ', 'ꭻ', ], 0),
+            ("Ꮖ Ꮋ Ꭼ Ꮓ Ꭴ Ꮳ Ꭶ Ꮥ", TOP),
+            ("Ꮖ Ꮋ Ꭼ Ꮓ Ꭴ Ꮳ Ꭶ Ꮥ", 0),
+            ("ꮒ ꮤ ꮶ ꭴ ꭾ ꮗ ꮝ ꮿ", TOP),
+            ("ꮖ ꭼ ꮓ ꮠ ꮳ ꭶ ꮥ ꮻ", TOP | LATIN_X_HEIGHT),
+            ("ꮖ ꭼ ꮓ ꮠ ꮳ ꭶ ꮥ ꮻ", 0),
+            ("ᏸ ꮐ ꭹ ꭻ", 0),
         ],
     },
     ScriptClass {
         name: "Coptic",
         group: ScriptGroup::Default,
         tag: Tag::new(b"COPT"),
-        index: 11,
         hint_top_to_bottom: false,
-        std_chars: &['Ⲟ', 'ⲟ', ],
+        std_chars: "Ⲟ ⲟ",
         blues: &[
-            (&['Ⲍ', 'Ⲏ', 'Ⲡ', 'Ⳟ', 'Ⲟ', 'Ⲑ', 'Ⲥ', 'Ⳋ', ], TOP),
-            (&['Ⳑ', 'Ⳙ', 'Ⳟ', 'Ⲏ', 'Ⲟ', 'Ⲑ', 'Ⳝ', 'Ⲱ', ], 0),
-            (&['ⲍ', 'ⲏ', 'ⲡ', 'ⳟ', 'ⲟ', 'ⲑ', 'ⲥ', 'ⳋ', ], TOP | LATIN_X_HEIGHT),
-            (&['ⳑ', 'ⳙ', 'ⳟ', 'ⲏ', 'ⲟ', 'ⲑ', 'ⳝ', 'Ⳓ', ], 0),
+            ("Ⲍ Ⲏ Ⲡ Ⳟ Ⲟ Ⲑ Ⲥ Ⳋ", TOP),
+            ("Ⳑ Ⳙ Ⳟ Ⲏ Ⲟ Ⲑ Ⳝ Ⲱ", 0),
+            ("ⲍ ⲏ ⲡ ⳟ ⲟ ⲑ ⲥ ⳋ", TOP | LATIN_X_HEIGHT),
+            ("ⳑ ⳙ ⳟ ⲏ ⲟ ⲑ ⳝ Ⳓ", 0),
         ],
     },
     ScriptClass {
         name: "Cypriot",
         group: ScriptGroup::Default,
         tag: Tag::new(b"CPRT"),
-        index: 12,
         hint_top_to_bottom: false,
-        std_chars: &['𐠅', '𐠣', ],
+        std_chars: "𐠅 𐠣",
         blues: &[
-            (&['𐠍', '𐠙', '𐠳', '𐠱', '𐠅', '𐠓', '𐠣', '𐠦', ], TOP),
-            (&['𐠃', '𐠊', '𐠛', '𐠣', '𐠳', '𐠵', '𐠐', ], 0),
-            (&['𐠈', '𐠏', '𐠖', ], TOP),
-            (&['𐠈', '𐠏', '𐠖', ], 0),
+            ("𐠍 𐠙 𐠳 𐠱 𐠅 𐠓 𐠣 𐠦", TOP),
+            ("𐠃 𐠊 𐠛 𐠣 𐠳 𐠵 𐠐", 0),
+            ("𐠈 𐠏 𐠖", TOP),
+            ("𐠈 𐠏 𐠖", 0),
         ],
     },
     ScriptClass {
         name: "Cyrillic",
         group: ScriptGroup::Default,
         tag: Tag::new(b"CYRL"),
-        index: 13,
         hint_top_to_bottom: false,
-        std_chars: &['о', 'О', ],
+        std_chars: "о О",
         blues: &[
-            (&['Б', 'В', 'Е', 'П', 'З', 'О', 'С', 'Э', ], TOP),
-            (&['Б', 'В', 'Е', 'Ш', 'З', 'О', 'С', 'Э', ], 0),
-            (&['х', 'п', 'н', 'ш', 'е', 'з', 'о', 'с', ], TOP | LATIN_X_HEIGHT),
-            (&['х', 'п', 'н', 'ш', 'е', 'з', 'о', 'с', ], 0),
-            (&['р', 'у', 'ф', ], 0),
+            ("Б В Е П З О С Э", TOP),
+            ("Б В Е Ш З О С Э", 0),
+            ("х п н ш е з о с", TOP | LATIN_X_HEIGHT),
+            ("х п н ш е з о с", 0),
+            ("р у ф", 0),
         ],
     },
     ScriptClass {
         name: "Devanagari",
         group: ScriptGroup::Default,
         tag: Tag::new(b"DEVA"),
-        index: 14,
         hint_top_to_bottom: true,
-        std_chars: &['ठ', 'व', 'ट', ],
+        std_chars: "ठ व ट",
         blues: &[
-            (&['ई', 'ऐ', 'ओ', 'औ', 'ि', 'ी', 'ो', 'ौ', ], TOP),
-            (&['क', 'म', 'अ', 'आ', 'थ', 'ध', 'भ', 'श', ], TOP),
-            (&['क', 'न', 'म', 'उ', 'छ', 'ट', 'ठ', 'ड', ], TOP | LATIN_NEUTRAL | LATIN_X_HEIGHT),
-            (&['क', 'न', 'म', 'उ', 'छ', 'ट', 'ठ', 'ड', ], 0),
-            (&['ु', 'ृ', ], 0),
+            ("ई ऐ ओ औ ि ी ो ौ", TOP),
+            ("क म अ आ थ ध भ श", TOP),
+            ("क न म उ छ ट ठ ड", TOP | LATIN_NEUTRAL | LATIN_X_HEIGHT),
+            ("क न म उ छ ट ठ ड", 0),
+            ("ु ृ", 0),
         ],
     },
     ScriptClass {
         name: "Deseret",
         group: ScriptGroup::Default,
         tag: Tag::new(b"DSRT"),
-        index: 15,
         hint_top_to_bottom: false,
-        std_chars: &['𐐄', '𐐬', ],
+        std_chars: "𐐄 𐐬",
         blues: &[
-            (&['𐐂', '𐐄', '𐐋', '𐐗', '𐐑', ], TOP),
-            (&['𐐀', '𐐂', '𐐄', '𐐗', '𐐛', ], 0),
-            (&['𐐪', '𐐬', '𐐳', '𐐿', '𐐹', ], TOP | LATIN_X_HEIGHT),
-            (&['𐐨', '𐐪', '𐐬', '𐐿', '𐑃', ], 0),
+            ("𐐂 𐐄 𐐋 𐐗 𐐑", TOP),
+            ("𐐀 𐐂 𐐄 𐐗 𐐛", 0),
+            ("𐐪 𐐬 𐐳 𐐿 𐐹", TOP | LATIN_X_HEIGHT),
+            ("𐐨 𐐪 𐐬 𐐿 𐑃", 0),
         ],
     },
     ScriptClass {
         name: "Ethiopic",
         group: ScriptGroup::Default,
         tag: Tag::new(b"ETHI"),
-        index: 16,
         hint_top_to_bottom: false,
-        std_chars: &['ዐ', ],
+        std_chars: "ዐ",
         blues: &[
-            (&['ሀ', 'ሃ', 'ዘ', 'ፐ', 'ማ', 'በ', 'ዋ', 'ዐ', ], TOP),
-            (&['ለ', 'ሐ', 'በ', 'ዘ', 'ሀ', 'ሪ', 'ዐ', 'ጨ', ], 0),
+            ("ሀ ሃ ዘ ፐ ማ በ ዋ ዐ", TOP),
+            ("ለ ሐ በ ዘ ሀ ሪ ዐ ጨ", 0),
         ],
     },
     ScriptClass {
         name: "Georgian (Mkhedruli)",
         group: ScriptGroup::Default,
         tag: Tag::new(b"GEOR"),
-        index: 17,
         hint_top_to_bottom: false,
-        std_chars: &['ი', 'ე', 'ა', 'Ჿ', ],
+        std_chars: "ი ე ა Ჿ",
         blues: &[
-            (&['გ', 'დ', 'ე', 'ვ', 'თ', 'ი', 'ო', 'ღ', ], TOP | LATIN_X_HEIGHT),
-            (&['ა', 'ზ', 'მ', 'ს', 'შ', 'ძ', 'ხ', 'პ', ], 0),
-            (&['ს', 'ხ', 'ქ', 'ზ', 'მ', 'შ', 'ჩ', 'წ', ], TOP),
-            (&['ე', 'ვ', 'ჟ', 'ტ', 'უ', 'ფ', 'ქ', 'ყ', ], 0),
-            (&['Ნ', 'Ჟ', 'Ჳ', 'Ჸ', 'Გ', 'Ე', 'Ო', 'Ჴ', ], TOP),
-            (&['Ი', 'Ჲ', 'Ო', 'Ჩ', 'Მ', 'Შ', 'Ჯ', 'Ჽ', ], 0),
+            ("გ დ ე ვ თ ი ო ღ", TOP | LATIN_X_HEIGHT),
+            ("ა ზ მ ს შ ძ ხ პ", 0),
+            ("ს ხ ქ ზ მ შ ჩ წ", TOP),
+            ("ე ვ ჟ ტ უ ფ ქ ყ", 0),
+            ("Ნ Ჟ Ჳ Ჸ Გ Ე Ო Ჴ", TOP),
+            ("Ი Ჲ Ო Ჩ Მ Შ Ჯ Ჽ", 0),
         ],
     },
     ScriptClass {
         name: "Georgian (Khutsuri)",
         group: ScriptGroup::Default,
         tag: Tag::new(b"GEOK"),
-        index: 18,
         hint_top_to_bottom: false,
-        std_chars: &['Ⴖ', 'Ⴑ', 'ⴙ', ],
+        std_chars: "Ⴖ Ⴑ ⴙ",
         blues: &[
-            (&['Ⴑ', 'Ⴇ', 'Ⴙ', 'Ⴜ', 'Ⴄ', 'Ⴅ', 'Ⴓ', 'Ⴚ', ], TOP),
-            (&['Ⴄ', 'Ⴅ', 'Ⴇ', 'Ⴈ', 'Ⴆ', 'Ⴑ', 'Ⴊ', 'Ⴋ', ], 0),
-            (&['ⴁ', 'ⴗ', 'ⴂ', 'ⴄ', 'ⴅ', 'ⴇ', 'ⴔ', 'ⴖ', ], TOP | LATIN_X_HEIGHT),
-            (&['ⴈ', 'ⴌ', 'ⴖ', 'ⴎ', 'ⴃ', 'ⴆ', 'ⴋ', 'ⴢ', ], 0),
-            (&['ⴐ', 'ⴑ', 'ⴓ', 'ⴕ', 'ⴙ', 'ⴛ', 'ⴡ', 'ⴣ', ], TOP),
-            (&['ⴄ', 'ⴅ', 'ⴔ', 'ⴕ', 'ⴁ', 'ⴂ', 'ⴘ', 'ⴝ', ], 0),
+            ("Ⴑ Ⴇ Ⴙ Ⴜ Ⴄ Ⴅ Ⴓ Ⴚ", TOP),
+            ("Ⴄ Ⴅ Ⴇ Ⴈ Ⴆ Ⴑ Ⴊ Ⴋ", 0),
+            ("ⴁ ⴗ ⴂ ⴄ ⴅ ⴇ ⴔ ⴖ", TOP | LATIN_X_HEIGHT),
+            ("ⴈ ⴌ ⴖ ⴎ ⴃ ⴆ ⴋ ⴢ", 0),
+            ("ⴐ ⴑ ⴓ ⴕ ⴙ ⴛ ⴡ ⴣ", TOP),
+            ("ⴄ ⴅ ⴔ ⴕ ⴁ ⴂ ⴘ ⴝ", 0),
         ],
     },
     ScriptClass {
         name: "Glagolitic",
         group: ScriptGroup::Default,
         tag: Tag::new(b"GLAG"),
-        index: 19,
         hint_top_to_bottom: false,
-        std_chars: &['Ⱅ', 'ⱅ', ],
+        std_chars: "Ⱅ ⱅ",
         blues: &[
-            (&['Ⰵ', 'Ⱄ', 'Ⱚ', 'Ⰴ', 'Ⰲ', 'Ⰺ', 'Ⱛ', 'Ⰻ', ], TOP),
-            (&['Ⰵ', 'Ⰴ', 'Ⰲ', 'Ⱚ', 'Ⱎ', 'Ⱑ', 'Ⰺ', 'Ⱄ', ], 0),
-            (&['ⰵ', 'ⱄ', 'ⱚ', 'ⰴ', 'ⰲ', 'ⰺ', 'ⱛ', 'ⰻ', ], TOP | LATIN_X_HEIGHT),
-            (&['ⰵ', 'ⰴ', 'ⰲ', 'ⱚ', 'ⱎ', 'ⱑ', 'ⰺ', 'ⱄ', ], 0),
+            ("Ⰵ Ⱄ Ⱚ Ⰴ Ⰲ Ⰺ Ⱛ Ⰻ", TOP),
+            ("Ⰵ Ⰴ Ⰲ Ⱚ Ⱎ Ⱑ Ⰺ Ⱄ", 0),
+            ("ⰵ ⱄ ⱚ ⰴ ⰲ ⰺ ⱛ ⰻ", TOP | LATIN_X_HEIGHT),
+            ("ⰵ ⰴ ⰲ ⱚ ⱎ ⱑ ⰺ ⱄ", 0),
         ],
     },
     ScriptClass {
         name: "Gothic",
         group: ScriptGroup::Default,
         tag: Tag::new(b"GOTH"),
-        index: 20,
         hint_top_to_bottom: true,
-        std_chars: &['𐌴', '𐌾', '𐍃', ],
+        std_chars: "𐌴 𐌾 𐍃",
         blues: &[
-            (&['𐌲', '𐌶', '𐍀', '𐍄', '𐌴', '𐍃', '𐍈', '𐌾', ], TOP),
-            (&['𐌶', '𐌴', '𐍃', '𐍈', ], 0),
+            ("𐌲 𐌶 𐍀 𐍄 𐌴 𐍃 𐍈 𐌾", TOP),
+            ("𐌶 𐌴 𐍃 𐍈", 0),
         ],
     },
     ScriptClass {
         name: "Greek",
         group: ScriptGroup::Default,
         tag: Tag::new(b"GREK"),
-        index: 21,
         hint_top_to_bottom: false,
-        std_chars: &['ο', 'Ο', ],
+        std_chars: "ο Ο",
         blues: &[
-            (&['Γ', 'Β', 'Ε', 'Ζ', 'Θ', 'Ο', 'Ω', ], TOP),
-            (&['Β', 'Δ', 'Ζ', 'Ξ', 'Θ', 'Ο', ], 0),
-            (&['β', 'θ', 'δ', 'ζ', 'λ', 'ξ', ], TOP),
-            (&['α', 'ε', 'ι', 'ο', 'π', 'σ', 'τ', 'ω', ], TOP | LATIN_X_HEIGHT),
-            (&['α', 'ε', 'ι', 'ο', 'π', 'σ', 'τ', 'ω', ], 0),
-            (&['β', 'γ', 'η', 'μ', 'ρ', 'φ', 'χ', 'ψ', ], 0),
+            ("Γ Β Ε Ζ Θ Ο Ω", TOP),
+            ("Β Δ Ζ Ξ Θ Ο", 0),
+            ("β θ δ ζ λ ξ", TOP),
+            ("α ε ι ο π σ τ ω", TOP | LATIN_X_HEIGHT),
+            ("α ε ι ο π σ τ ω", 0),
+            ("β γ η μ ρ φ χ ψ", 0),
         ],
     },
     ScriptClass {
         name: "Gujarati",
         group: ScriptGroup::Default,
         tag: Tag::new(b"GUJR"),
-        index: 22,
         hint_top_to_bottom: false,
-        std_chars: &['ટ', '૦', ],
+        std_chars: "ટ ૦",
         blues: &[
-            (&['ત', 'ન', 'ઋ', 'ઌ', 'છ', 'ટ', 'ર', '૦', ], TOP | LATIN_X_HEIGHT),
-            (&['ખ', 'ગ', 'ઘ', 'ઞ', 'ઇ', 'ઈ', 'ઠ', 'જ', ], 0),
-            (&['ઈ', 'ઊ', 'િ', 'ી', 'લ', 'શ', 'જ', 'સ', ], TOP),
-            (&['ુ', 'ૃ', 'ૄ', 'ખ', 'છ', 'છ', ], 0),
-            (&['૦', '૧', '૨', '૩', '૭', ], TOP),
+            ("ત ન ઋ ઌ છ ટ ર ૦", TOP | LATIN_X_HEIGHT),
+            ("ખ ગ ઘ ઞ ઇ ઈ ઠ જ", 0),
+            ("ઈ ઊ િ ી લી શ્ચિ જિ સી", TOP),
+            ("ુ ૃ ૄ ખુ છૃ છૄ", 0),
+            ("૦ ૧ ૨ ૩ ૭", TOP),
         ],
     },
     ScriptClass {
         name: "Gurmukhi",
         group: ScriptGroup::Default,
         tag: Tag::new(b"GURU"),
-        index: 23,
         hint_top_to_bottom: true,
-        std_chars: &['ਠ', 'ਰ', '੦', ],
+        std_chars: "ਠ ਰ ੦",
         blues: &[
-            (&['ਇ', 'ਈ', 'ਉ', 'ਏ', 'ਓ', 'ੳ', 'ਿ', 'ੀ', ], TOP),
-            (&['ਕ', 'ਗ', 'ਙ', 'ਚ', 'ਜ', 'ਤ', 'ਧ', 'ਸ', ], TOP),
-            (&['ਕ', 'ਗ', 'ਙ', 'ਚ', 'ਜ', 'ਤ', 'ਧ', 'ਸ', ], TOP | LATIN_NEUTRAL | LATIN_X_HEIGHT),
-            (&['ਅ', 'ਏ', 'ਓ', 'ਗ', 'ਜ', 'ਠ', 'ਰ', 'ਸ', ], 0),
-            (&['੦', '੧', '੨', '੩', '੭', ], TOP),
+            ("ਇ ਈ ਉ ਏ ਓ ੳ ਿ ੀ", TOP),
+            ("ਕ ਗ ਙ ਚ ਜ ਤ ਧ ਸ", TOP),
+            ("ਕ ਗ ਙ ਚ ਜ ਤ ਧ ਸ", TOP | LATIN_NEUTRAL | LATIN_X_HEIGHT),
+            ("ਅ ਏ ਓ ਗ ਜ ਠ ਰ ਸ", 0),
+            ("੦ ੧ ੨ ੩ ੭", TOP),
         ],
     },
     ScriptClass {
         name: "Hebrew",
         group: ScriptGroup::Default,
         tag: Tag::new(b"HEBR"),
-        index: 24,
         hint_top_to_bottom: false,
-        std_chars: &['ם', ],
+        std_chars: "ם",
         blues: &[
-            (&['ב', 'ד', 'ה', 'ח', 'ך', 'כ', 'ם', 'ס', ], TOP | LATIN_LONG),
-            (&['ב', 'ט', 'כ', 'ם', 'ס', 'צ', ], 0),
-            (&['ק', 'ך', 'ן', 'ף', 'ץ', ], 0),
+            ("ב ד ה ח ך כ ם ס", TOP | LATIN_LONG),
+            ("ב ט כ ם ס צ", 0),
+            ("ק ך ן ף ץ", 0),
         ],
     },
     ScriptClass {
         name: "Kayah Li",
         group: ScriptGroup::Default,
         tag: Tag::new(b"KALI"),
-        index: 25,
         hint_top_to_bottom: false,
-        std_chars: &['ꤍ', '꤀', ],
+        std_chars: "ꤍ ꤀",
         blues: &[
-            (&['꤅', 'ꤏ', '꤁', 'ꤋ', '꤀', 'ꤍ', ], TOP | LATIN_X_HEIGHT),
-            (&['꤈', 'ꤘ', '꤀', 'ꤍ', 'ꤢ', ], 0),
-            (&['ꤖ', 'ꤡ', ], TOP),
-            (&['ꤑ', 'ꤜ', 'ꤞ', ], 0),
-            (&['ꤑ', 'ꤜ', 'ꤔ', ], 0),
+            ("꤅ ꤏ ꤁ ꤋ ꤀ ꤍ", TOP | LATIN_X_HEIGHT),
+            ("꤈ ꤘ ꤀ ꤍ ꤢ", 0),
+            ("ꤖ ꤡ", TOP),
+            ("ꤑ ꤜ ꤞ", 0),
+            ("ꤑ꤬ ꤜ꤭ ꤔ꤬", 0),
         ],
     },
     ScriptClass {
         name: "Khmer",
         group: ScriptGroup::Default,
         tag: Tag::new(b"KHMR"),
-        index: 26,
         hint_top_to_bottom: false,
-        std_chars: &['០', ],
+        std_chars: "០",
         blues: &[
-            (&['ខ', 'ទ', 'ន', 'ឧ', 'ឩ', 'ា', ], TOP | LATIN_X_HEIGHT),
-            (&['ក', 'ក', 'ក', 'ក', ], LATIN_SUB_TOP),
-            (&['ខ', 'ឃ', 'ច', 'ឋ', 'ប', 'ម', 'យ', 'ឲ', ], 0),
-            (&['ត', 'រ', 'ឲ', 'អ', ], 0),
-            (&['ន', 'ង', 'ក', 'ច', 'ន', 'ល', ], 0),
+            ("ខ ទ ន ឧ ឩ ា", TOP | LATIN_X_HEIGHT),
+            ("ក្ក ក្ខ ក្គ ក្ថ", LATIN_SUB_TOP),
+            ("ខ ឃ ច ឋ ប ម យ ឲ", 0),
+            ("ត្រ រៀ ឲ្យ អឿ", 0),
+            ("ន្ត្រៃ ង្ខ្យ ក្បៀ ច្រៀ ន្តឿ ល្បឿ", 0),
         ],
     },
     ScriptClass {
         name: "Khmer Symbols",
         group: ScriptGroup::Default,
         tag: Tag::new(b"KHMS"),
-        index: 27,
         hint_top_to_bottom: false,
-        std_chars: &['᧡', '᧪', ],
+        std_chars: "᧡ ᧪",
         blues: &[
-            (&['᧠', '᧡', ], TOP | LATIN_X_HEIGHT),
-            (&['᧶', '᧹', ], 0),
+            ("᧠ ᧡", TOP | LATIN_X_HEIGHT),
+            ("᧶ ᧹", 0),
         ],
     },
     ScriptClass {
         name: "Kannada",
         group: ScriptGroup::Default,
         tag: Tag::new(b"KNDA"),
-        index: 28,
         hint_top_to_bottom: false,
-        std_chars: &['೦', 'ಬ', ],
+        std_chars: "೦ ಬ",
         blues: &[
-            (&['ಇ', 'ಊ', 'ಐ', 'ಣ', 'ಸ', 'ನ', 'ದ', 'ರ', ], TOP),
-            (&['ಅ', 'ಉ', 'ಎ', 'ಲ', '೦', '೨', '೬', '೭', ], 0),
+            ("ಇ ಊ ಐ ಣ ಸಾ ನಾ ದಾ ರಾ", TOP),
+            ("ಅ ಉ ಎ ಲ ೦ ೨ ೬ ೭", 0),
         ],
     },
     ScriptClass {
         name: "Lao",
         group: ScriptGroup::Default,
         tag: Tag::new(b"LAOO"),
-        index: 29,
         hint_top_to_bottom: false,
-        std_chars: &['໐', ],
+        std_chars: "໐",
         blues: &[
-            (&['າ', 'ດ', 'ອ', 'ມ', 'ລ', 'ວ', 'ຣ', 'ງ', ], TOP | LATIN_X_HEIGHT),
-            (&['າ', 'ອ', 'ບ', 'ຍ', 'ຣ', 'ຮ', 'ວ', 'ຢ', ], 0),
-            (&['ປ', 'ຢ', 'ຟ', 'ຝ', ], TOP),
-            (&['ໂ', 'ໄ', 'ໃ', ], TOP),
-            (&['ງ', 'ຊ', 'ຖ', 'ຽ', 'ໆ', 'ຯ', ], 0),
+            ("າ ດ ອ ມ ລ ວ ຣ ງ", TOP | LATIN_X_HEIGHT),
+            ("າ ອ ບ ຍ ຣ ຮ ວ ຢ", 0),
+            ("ປ ຢ ຟ ຝ", TOP),
+            ("ໂ ໄ ໃ", TOP),
+            ("ງ ຊ ຖ ຽ ໆ ຯ", 0),
         ],
     },
     ScriptClass {
         name: "Latin",
         group: ScriptGroup::Default,
         tag: Tag::new(b"LATN"),
-        index: 30,
         hint_top_to_bottom: false,
-        std_chars: &['o', 'O', '0', ],
+        std_chars: "o O 0",
         blues: &[
-            (&['T', 'H', 'E', 'Z', 'O', 'C', 'Q', 'S', ], TOP),
-            (&['H', 'E', 'Z', 'L', 'O', 'C', 'U', 'S', ], 0),
-            (&['f', 'i', 'j', 'k', 'd', 'b', 'h', ], TOP),
-            (&['u', 'v', 'x', 'z', 'o', 'e', 's', 'c', ], TOP | LATIN_X_HEIGHT),
-            (&['n', 'r', 'x', 'z', 'o', 'e', 's', 'c', ], 0),
-            (&['p', 'q', 'g', 'j', 'y', ], 0),
+            ("T H E Z O C Q S", TOP),
+            ("H E Z L O C U S", 0),
+            ("f i j k d b h", TOP),
+            ("u v x z o e s c", TOP | LATIN_X_HEIGHT),
+            ("n r x z o e s c", 0),
+            ("p q g j y", 0),
         ],
     },
     ScriptClass {
         name: "Latin Subscript Fallback",
         group: ScriptGroup::Default,
         tag: Tag::new(b"LATB"),
-        index: 31,
         hint_top_to_bottom: false,
-        std_chars: &['ₒ', '₀', ],
+        std_chars: "ₒ ₀",
         blues: &[
-            (&['₀', '₃', '₅', '₇', '₈', ], TOP),
-            (&['₀', '₁', '₂', '₃', '₈', ], 0),
-            (&['ᵢ', 'ⱼ', 'ₕ', 'ₖ', 'ₗ', ], TOP),
-            (&['ₐ', 'ₑ', 'ₒ', 'ₓ', 'ₙ', 'ₛ', 'ᵥ', 'ᵤ', 'ᵣ', ], TOP | LATIN_X_HEIGHT),
-            (&['ₐ', 'ₑ', 'ₒ', 'ₓ', 'ₙ', 'ₛ', 'ᵥ', 'ᵤ', 'ᵣ', ], 0),
-            (&['ᵦ', 'ᵧ', 'ᵨ', 'ᵩ', 'ₚ', ], 0),
+            ("₀ ₃ ₅ ₇ ₈", TOP),
+            ("₀ ₁ ₂ ₃ ₈", 0),
+            ("ᵢ ⱼ ₕ ₖ ₗ", TOP),
+            ("ₐ ₑ ₒ ₓ ₙ ₛ ᵥ ᵤ ᵣ", TOP | LATIN_X_HEIGHT),
+            ("ₐ ₑ ₒ ₓ ₙ ₛ ᵥ ᵤ ᵣ", 0),
+            ("ᵦ ᵧ ᵨ ᵩ ₚ", 0),
         ],
     },
     ScriptClass {
         name: "Latin Superscript Fallback",
         group: ScriptGroup::Default,
         tag: Tag::new(b"LATP"),
-        index: 32,
         hint_top_to_bottom: false,
-        std_chars: &['ᵒ', 'ᴼ', '⁰', ],
+        std_chars: "ᵒ ᴼ ⁰",
         blues: &[
-            (&['⁰', '³', '⁵', '⁷', 'ᵀ', 'ᴴ', 'ᴱ', 'ᴼ', ], TOP),
-            (&['⁰', '¹', '²', '³', 'ᴱ', 'ᴸ', 'ᴼ', 'ᵁ', ], 0),
-            (&['ᵇ', 'ᵈ', 'ᵏ', 'ʰ', 'ʲ', 'ᶠ', 'ⁱ', ], TOP),
-            (&['ᵉ', 'ᵒ', 'ʳ', 'ˢ', 'ˣ', 'ᶜ', 'ᶻ', ], TOP | LATIN_X_HEIGHT),
-            (&['ᵉ', 'ᵒ', 'ʳ', 'ˢ', 'ˣ', 'ᶜ', 'ᶻ', ], 0),
-            (&['ᵖ', 'ʸ', 'ᵍ', ], 0),
+            ("⁰ ³ ⁵ ⁷ ᵀ ᴴ ᴱ ᴼ", TOP),
+            ("⁰ ¹ ² ³ ᴱ ᴸ ᴼ ᵁ", 0),
+            ("ᵇ ᵈ ᵏ ʰ ʲ ᶠ ⁱ", TOP),
+            ("ᵉ ᵒ ʳ ˢ ˣ ᶜ ᶻ", TOP | LATIN_X_HEIGHT),
+            ("ᵉ ᵒ ʳ ˢ ˣ ᶜ ᶻ", 0),
+            ("ᵖ ʸ ᵍ", 0),
         ],
     },
     ScriptClass {
         name: "Lisu",
         group: ScriptGroup::Default,
         tag: Tag::new(b"LISU"),
-        index: 33,
         hint_top_to_bottom: false,
-        std_chars: &['ꓳ', ],
+        std_chars: "ꓳ",
         blues: &[
-            (&['ꓡ', 'ꓧ', 'ꓱ', 'ꓶ', 'ꓩ', 'ꓚ', 'ꓵ', 'ꓳ', ], TOP),
-            (&['ꓕ', 'ꓜ', 'ꓞ', 'ꓡ', 'ꓛ', 'ꓢ', 'ꓳ', 'ꓴ', ], 0),
+            ("ꓡ ꓧ ꓱ ꓶ ꓩ ꓚ ꓵ ꓳ", TOP),
+            ("ꓕ ꓜ ꓞ ꓡ ꓛ ꓢ ꓳ ꓴ", 0),
         ],
     },
     ScriptClass {
         name: "Malayalam",
         group: ScriptGroup::Default,
         tag: Tag::new(b"MLYM"),
-        index: 34,
         hint_top_to_bottom: false,
-        std_chars: &['ഠ', 'റ', ],
+        std_chars: "ഠ റ",
         blues: &[
-            (&['ഒ', 'ട', 'ഠ', 'റ', 'ച', 'പ', 'ച', 'പ', ], TOP),
-            (&['ട', 'ഠ', 'ധ', 'ശ', 'ഘ', 'ച', 'ഥ', 'ല', ], 0),
+            ("ഒ ട ഠ റ ച പ ച്ച പ്പ", TOP),
+            ("ട ഠ ധ ശ ഘ ച ഥ ല", 0),
         ],
     },
     ScriptClass {
         name: "Medefaidrin",
         group: ScriptGroup::Default,
         tag: Tag::new(b"MEDF"),
-        index: 35,
         hint_top_to_bottom: false,
-        std_chars: &['𖹡', '𖹛', '𖹯', ],
+        std_chars: "𖹡 𖹛 𖹯",
         blues: &[
-            (&['𖹀', '𖹁', '𖹂', '𖹃', '𖹏', '𖹚', '𖹟', ], TOP),
-            (&['𖹀', '𖹁', '𖹂', '𖹃', '𖹏', '𖹚', '𖹒', '𖹓', ], 0),
-            (&['𖹤', '𖹬', '𖹧', '𖹴', '𖹶', '𖹾', ], TOP),
-            (&['𖹠', '𖹡', '𖹢', '𖹹', '𖹳', '𖹮', ], TOP | LATIN_X_HEIGHT),
-            (&['𖹠', '𖹡', '𖹢', '𖹳', '𖹭', '𖹽', ], 0),
-            (&['𖹥', '𖹨', '𖹩', ], 0),
-            (&['𖺀', '𖺅', '𖺈', '𖺄', '𖺍', ], TOP),
+            ("𖹀 𖹁 𖹂 𖹃 𖹏 𖹚 𖹟", TOP),
+            ("𖹀 𖹁 𖹂 𖹃 𖹏 𖹚 𖹒 𖹓", 0),
+            ("𖹤 𖹬 𖹧 𖹴 𖹶 𖹾", TOP),
+            ("𖹠 𖹡 𖹢 𖹹 𖹳 𖹮", TOP | LATIN_X_HEIGHT),
+            ("𖹠 𖹡 𖹢 𖹳 𖹭 𖹽", 0),
+            ("𖹥 𖹨 𖹩", 0),
+            ("𖺀 𖺅 𖺈 𖺄 𖺍", TOP),
         ],
     },
     ScriptClass {
         name: "Mongolian",
         group: ScriptGroup::Default,
         tag: Tag::new(b"MONG"),
-        index: 36,
         hint_top_to_bottom: true,
-        std_chars: &['ᡂ', 'ᠪ', ],
+        std_chars: "ᡂ ᠪ",
         blues: &[
-            (&['ᠳ', 'ᠴ', 'ᠶ', 'ᠽ', 'ᡂ', 'ᡊ', '‍', '‍', ], TOP),
-            (&['ᡃ', ], 0),
+            ("ᠳ ᠴ ᠶ ᠽ ᡂ ᡊ ‍ᡡ‍ ‍ᡳ‍", TOP),
+            ("ᡃ", 0),
         ],
     },
     ScriptClass {
         name: "Myanmar",
         group: ScriptGroup::Default,
         tag: Tag::new(b"MYMR"),
-        index: 37,
         hint_top_to_bottom: false,
-        std_chars: &['ဝ', 'င', 'ဂ', ],
+        std_chars: "ဝ င ဂ",
         blues: &[
-            (&['ခ', 'ဂ', 'င', 'ဒ', 'ဝ', 'ၥ', '၊', '။', ], TOP | LATIN_X_HEIGHT),
-            (&['င', 'ဎ', 'ဒ', 'ပ', 'ဗ', 'ဝ', '၊', '။', ], 0),
-            (&['ဩ', 'ြ', '၍', '၏', '၆', 'ါ', 'ိ', ], TOP),
-            (&['ဉ', 'ည', 'ဥ', 'ဩ', 'ဨ', '၂', '၅', '၉', ], 0),
+            ("ခ ဂ င ဒ ဝ ၥ ၊ ။", TOP | LATIN_X_HEIGHT),
+            ("င ဎ ဒ ပ ဗ ဝ ၊ ။", 0),
+            ("ဩ ြ ၍ ၏ ၆ ါ ိ", TOP),
+            ("ဉ ည ဥ ဩ ဨ ၂ ၅ ၉", 0),
         ],
     },
     ScriptClass {
         name: "N'Ko",
         group: ScriptGroup::Default,
         tag: Tag::new(b"NKOO"),
-        index: 38,
         hint_top_to_bottom: false,
-        std_chars: &['ߋ', '߀', ],
+        std_chars: "ߋ ߀",
         blues: &[
-            (&['ߐ', '߉', 'ߒ', 'ߟ', 'ߖ', 'ߜ', 'ߠ', 'ߥ', ], TOP),
-            (&['߀', 'ߘ', 'ߡ', 'ߠ', 'ߥ', ], 0),
-            (&['ߏ', 'ߛ', 'ߋ', ], TOP | LATIN_X_HEIGHT),
-            (&['ߎ', 'ߏ', 'ߛ', 'ߋ', ], 0),
+            ("ߐ ߉ ߒ ߟ ߖ ߜ ߠ ߥ", TOP),
+            ("߀ ߘ ߡ ߠ ߥ", 0),
+            ("ߏ ߛ ߋ", TOP | LATIN_X_HEIGHT),
+            ("ߎ ߏ ߛ ߋ", 0),
         ],
     },
     ScriptClass {
         name: "no script",
         group: ScriptGroup::Default,
         tag: Tag::new(b"NONE"),
-        index: 39,
         hint_top_to_bottom: false,
-        std_chars: &[],
+        std_chars: "",
         blues: &[],
     },
     ScriptClass {
         name: "Ol Chiki",
         group: ScriptGroup::Default,
         tag: Tag::new(b"OLCK"),
-        index: 40,
         hint_top_to_bottom: false,
-        std_chars: &['ᱛ', ],
+        std_chars: "ᱛ",
         blues: &[
-            (&['ᱛ', 'ᱜ', 'ᱝ', 'ᱡ', 'ᱢ', 'ᱥ', ], TOP),
-            (&['ᱛ', 'ᱜ', 'ᱝ', 'ᱡ', 'ᱢ', 'ᱥ', ], 0),
+            ("ᱛ ᱜ ᱝ ᱡ ᱢ ᱥ", TOP),
+            ("ᱛ ᱜ ᱝ ᱡ ᱢ ᱥ", 0),
         ],
     },
     ScriptClass {
         name: "Old Turkic",
         group: ScriptGroup::Default,
         tag: Tag::new(b"ORKH"),
-        index: 41,
         hint_top_to_bottom: false,
-        std_chars: &['𐰗', ],
+        std_chars: "𐰗",
         blues: &[
-            (&['𐰗', '𐰘', '𐰧', ], TOP),
-            (&['𐰉', '𐰗', '𐰦', '𐰧', ], 0),
+            ("𐰗 𐰘 𐰧", TOP),
+            ("𐰉 𐰗 𐰦 𐰧", 0),
         ],
     },
     ScriptClass {
         name: "Osage",
         group: ScriptGroup::Default,
         tag: Tag::new(b"OSGE"),
-        index: 42,
         hint_top_to_bottom: false,
-        std_chars: &['𐓂', '𐓪', ],
+        std_chars: "𐓂 𐓪",
         blues: &[
-            (&['𐒾', '𐓍', '𐓒', '𐓓', '𐒻', '𐓂', '𐒵', '𐓆', ], TOP),
-            (&['𐒰', '𐓍', '𐓂', '𐒿', '𐓎', '𐒹', ], 0),
-            (&['𐒼', '𐒽', '𐒾', ], 0),
-            (&['𐓵', '𐓶', '𐓺', '𐓻', '𐓝', '𐓣', '𐓪', '𐓮', ], TOP | LATIN_X_HEIGHT),
-            (&['𐓘', '𐓚', '𐓣', '𐓵', '𐓡', '𐓧', '𐓪', '𐓶', ], 0),
-            (&['𐓤', '𐓦', '𐓸', '𐓹', '𐓛', ], TOP),
-            (&['𐓤', '𐓥', '𐓦', ], 0),
+            ("𐒾 𐓍 𐓒 𐓓 𐒻 𐓂 𐒵 𐓆", TOP),
+            ("𐒰 𐓍 𐓂 𐒿 𐓎 𐒹", 0),
+            ("𐒼 𐒽 𐒾", 0),
+            ("𐓵 𐓶 𐓺 𐓻 𐓝 𐓣 𐓪 𐓮", TOP | LATIN_X_HEIGHT),
+            ("𐓘 𐓚 𐓣 𐓵 𐓡 𐓧 𐓪 𐓶", 0),
+            ("𐓤 𐓦 𐓸 𐓹 𐓛", TOP),
+            ("𐓤 𐓥 𐓦", 0),
         ],
     },
     ScriptClass {
         name: "Osmanya",
         group: ScriptGroup::Default,
         tag: Tag::new(b"OSMA"),
-        index: 43,
         hint_top_to_bottom: false,
-        std_chars: &['𐒆', '𐒠', ],
+        std_chars: "𐒆 𐒠",
         blues: &[
-            (&['𐒆', '𐒉', '𐒐', '𐒒', '𐒘', '𐒛', '𐒠', '𐒣', ], TOP),
-            (&['𐒀', '𐒂', '𐒆', '𐒈', '𐒊', '𐒒', '𐒠', '𐒩', ], 0),
+            ("𐒆 𐒉 𐒐 𐒒 𐒘 𐒛 𐒠 𐒣", TOP),
+            ("𐒀 𐒂 𐒆 𐒈 𐒊 𐒒 𐒠 𐒩", 0),
         ],
     },
     ScriptClass {
         name: "Hanifi Rohingya",
         group: ScriptGroup::Default,
         tag: Tag::new(b"ROHG"),
-        index: 44,
         hint_top_to_bottom: false,
-        std_chars: &['𐴰', ],
+        std_chars: "𐴰",
         blues: &[
-            (&['𐴃', '𐴀', '𐴆', '𐴖', '𐴕', ], TOP),
-            (&['𐴔', '𐴖', '𐴕', '𐴑', '𐴐', ], 0),
-            (&['ـ', ], LATIN_NEUTRAL),
+            ("𐴃 𐴀 𐴆 𐴖 𐴕", TOP),
+            ("𐴔 𐴖 𐴕 𐴑 𐴐", 0),
+            ("ـ", LATIN_NEUTRAL),
         ],
     },
     ScriptClass {
         name: "Saurashtra",
         group: ScriptGroup::Default,
         tag: Tag::new(b"SAUR"),
-        index: 45,
         hint_top_to_bottom: false,
-        std_chars: &['ꢝ', '꣐', ],
+        std_chars: "ꢝ ꣐",
         blues: &[
-            (&['ꢜ', 'ꢞ', 'ꢳ', 'ꢂ', 'ꢖ', 'ꢒ', 'ꢝ', 'ꢛ', ], TOP),
-            (&['ꢂ', 'ꢨ', 'ꢺ', 'ꢤ', 'ꢎ', ], 0),
+            ("ꢜ ꢞ ꢳ ꢂ ꢖ ꢒ ꢝ ꢛ", TOP),
+            ("ꢂ ꢨ ꢺ ꢤ ꢎ", 0),
         ],
     },
     ScriptClass {
         name: "Shavian",
         group: ScriptGroup::Default,
         tag: Tag::new(b"SHAW"),
-        index: 46,
         hint_top_to_bottom: false,
-        std_chars: &['𐑴', ],
+        std_chars: "𐑴",
         blues: &[
-            (&['𐑕', '𐑙', ], TOP),
-            (&['𐑔', '𐑖', '𐑗', '𐑹', '𐑻', ], 0),
-            (&['𐑟', '𐑣', ], 0),
-            (&['𐑱', '𐑲', '𐑳', '𐑴', '𐑸', '𐑺', '𐑼', ], TOP | LATIN_X_HEIGHT),
-            (&['𐑴', '𐑻', '𐑹', ], 0),
+            ("𐑕 𐑙", TOP),
+            ("𐑔 𐑖 𐑗 𐑹 𐑻", 0),
+            ("𐑟 𐑣", 0),
+            ("𐑱 𐑲 𐑳 𐑴 𐑸 𐑺 𐑼", TOP | LATIN_X_HEIGHT),
+            ("𐑴 𐑻 𐑹", 0),
         ],
     },
     ScriptClass {
         name: "Sinhala",
         group: ScriptGroup::Default,
         tag: Tag::new(b"SINH"),
-        index: 47,
         hint_top_to_bottom: false,
-        std_chars: &['ට', ],
+        std_chars: "ට",
         blues: &[
-            (&['ඉ', 'ක', 'ඝ', 'ඳ', 'ප', 'ය', 'ල', 'ෆ', ], TOP),
-            (&['එ', 'ඔ', 'ඝ', 'ජ', 'ට', 'ථ', 'ධ', 'ර', ], 0),
-            (&['ද', 'ඳ', 'උ', 'ල', 'ත', 'ත', 'බ', 'ද', ], 0),
+            ("ඉ ක ඝ ඳ ප ය ල ෆ", TOP),
+            ("එ ඔ ඝ ජ ට ථ ධ ර", 0),
+            ("ද ඳ උ ල තූ තු බු දු", 0),
         ],
     },
     ScriptClass {
         name: "Sundanese",
         group: ScriptGroup::Default,
         tag: Tag::new(b"SUND"),
-        index: 48,
         hint_top_to_bottom: false,
-        std_chars: &['᮰', ],
+        std_chars: "᮰",
         blues: &[
-            (&['ᮋ', 'ᮞ', 'ᮮ', 'ᮽ', '᮰', 'ᮈ', ], TOP),
-            (&['ᮄ', 'ᮔ', 'ᮕ', 'ᮗ', '᮰', 'ᮆ', 'ᮈ', 'ᮉ', ], 0),
-            (&['ᮼ', '᳄', ], 0),
+            ("ᮋ ᮞ ᮮ ᮽ ᮰ ᮈ", TOP),
+            ("ᮄ ᮔ ᮕ ᮗ ᮰ ᮆ ᮈ ᮉ", 0),
+            ("ᮼ ᳄", 0),
         ],
     },
     ScriptClass {
         name: "Tamil",
         group: ScriptGroup::Default,
         tag: Tag::new(b"TAML"),
-        index: 49,
         hint_top_to_bottom: false,
-        std_chars: &['௦', ],
+        std_chars: "௦",
         blues: &[
-            (&['உ', 'ஒ', 'ஓ', 'ற', 'ஈ', 'க', 'ங', 'ச', ], TOP),
-            (&['க', 'ச', 'ல', 'ஶ', 'உ', 'ங', 'ட', 'ப', ], 0),
+            ("உ ஒ ஓ ற ஈ க ங ச", TOP),
+            ("க ச ல ஶ உ ங ட ப", 0),
         ],
     },
     ScriptClass {
         name: "Tai Viet",
         group: ScriptGroup::Default,
         tag: Tag::new(b"TAVT"),
-        index: 50,
         hint_top_to_bottom: false,
-        std_chars: &['ꪒ', 'ꪫ', ],
+        std_chars: "ꪒ ꪫ",
         blues: &[
-            (&['ꪆ', 'ꪔ', 'ꪒ', 'ꪖ', 'ꪫ', ], TOP),
-            (&['ꪉ', 'ꪫ', 'ꪮ', ], 0),
+            ("ꪆ ꪔ ꪒ ꪖ ꪫ", TOP),
+            ("ꪉ ꪫ ꪮ", 0),
         ],
     },
     ScriptClass {
         name: "Telugu",
         group: ScriptGroup::Default,
         tag: Tag::new(b"TELU"),
-        index: 51,
         hint_top_to_bottom: false,
-        std_chars: &['౦', '౧', ],
+        std_chars: "౦ ౧",
         blues: &[
-            (&['ఇ', 'ఌ', 'ఙ', 'ఞ', 'ణ', 'ఱ', '౯', ], TOP),
-            (&['అ', 'క', 'చ', 'ర', 'ఽ', '౨', '౬', ], 0),
+            ("ఇ ఌ ఙ ఞ ణ ఱ ౯", TOP),
+            ("అ క చ ర ఽ ౨ ౬", 0),
         ],
     },
     ScriptClass {
         name: "Tifinagh",
         group: ScriptGroup::Default,
         tag: Tag::new(b"TFNG"),
-        index: 52,
         hint_top_to_bottom: false,
-        std_chars: &['ⵔ', ],
+        std_chars: "ⵔ",
         blues: &[
-            (&['ⵔ', 'ⵙ', 'ⵛ', 'ⵞ', 'ⴵ', 'ⴼ', 'ⴹ', 'ⵎ', ], TOP),
-            (&['ⵔ', 'ⵙ', 'ⵛ', 'ⵞ', 'ⴵ', 'ⴼ', 'ⴹ', 'ⵎ', ], 0),
+            ("ⵔ ⵙ ⵛ ⵞ ⴵ ⴼ ⴹ ⵎ", TOP),
+            ("ⵔ ⵙ ⵛ ⵞ ⴵ ⴼ ⴹ ⵎ", 0),
         ],
     },
     ScriptClass {
         name: "Thai",
         group: ScriptGroup::Default,
         tag: Tag::new(b"THAI"),
-        index: 53,
         hint_top_to_bottom: false,
-        std_chars: &['า', 'ๅ', '๐', ],
+        std_chars: "า ๅ ๐",
         blues: &[
-            (&['บ', 'เ', 'แ', 'อ', 'ก', 'า', ], TOP | LATIN_X_HEIGHT),
-            (&['บ', 'ป', 'ษ', 'ฯ', 'อ', 'ย', 'ฮ', ], 0),
-            (&['ป', 'ฝ', 'ฟ', ], TOP),
-            (&['โ', 'ใ', 'ไ', ], TOP),
-            (&['ฎ', 'ฏ', 'ฤ', 'ฦ', ], 0),
-            (&['ญ', 'ฐ', ], 0),
-            (&['๐', '๑', '๓', ], 0),
+            ("บ เ แ อ ก า", TOP | LATIN_X_HEIGHT),
+            ("บ ป ษ ฯ อ ย ฮ", 0),
+            ("ป ฝ ฟ", TOP),
+            ("โ ใ ไ", TOP),
+            ("ฎ ฏ ฤ ฦ", 0),
+            ("ญ ฐ", 0),
+            ("๐ ๑ ๓", 0),
         ],
     },
     ScriptClass {
         name: "Vai",
         group: ScriptGroup::Default,
         tag: Tag::new(b"VAII"),
-        index: 54,
         hint_top_to_bottom: false,
-        std_chars: &['ꘓ', 'ꖜ', 'ꖴ', ],
+        std_chars: "ꘓ ꖜ ꖴ",
         blues: &[
-            (&['ꗍ', 'ꘖ', 'ꘙ', 'ꘜ', 'ꖜ', 'ꖝ', 'ꔅ', 'ꕢ', ], TOP),
-            (&['ꗍ', 'ꘖ', 'ꘙ', 'ꗞ', 'ꔅ', 'ꕢ', 'ꖜ', 'ꔆ', ], 0),
+            ("ꗍ ꘖ ꘙ ꘜ ꖜ ꖝ ꔅ ꕢ", TOP),
+            ("ꗍ ꘖ ꘙ ꗞ ꔅ ꕢ ꖜ ꔆ", 0),
         ],
     },
     ScriptClass {
         name: "Limbu",
         group: ScriptGroup::Indic,
         tag: Tag::new(b"LIMB"),
-        index: 55,
         hint_top_to_bottom: false,
-        std_chars: &['o', ],
+        std_chars: "o",
         blues: &[],
     },
     ScriptClass {
         name: "Oriya",
         group: ScriptGroup::Indic,
         tag: Tag::new(b"ORYA"),
-        index: 56,
         hint_top_to_bottom: false,
-        std_chars: &['o', ],
+        std_chars: "o",
         blues: &[],
     },
     ScriptClass {
         name: "Syloti Nagri",
         group: ScriptGroup::Indic,
         tag: Tag::new(b"SYLO"),
-        index: 57,
         hint_top_to_bottom: false,
-        std_chars: &['o', ],
+        std_chars: "o",
         blues: &[],
     },
     ScriptClass {
         name: "Tibetan",
         group: ScriptGroup::Indic,
         tag: Tag::new(b"TIBT"),
-        index: 58,
         hint_top_to_bottom: false,
-        std_chars: &['o', ],
+        std_chars: "o",
         blues: &[],
     },
     ScriptClass {
         name: "CJKV ideographs",
         group: ScriptGroup::Cjk,
         tag: Tag::new(b"HANI"),
-        index: 59,
         hint_top_to_bottom: false,
-        std_chars: &['田', '囗', ],
+        std_chars: "田 囗",
         blues: &[
-            (&['他', '们', '你', '來', '們', '到', '和', '地', '对', '對', '就', '席', '我', '时', '時', '會', '来', '為', '能', '舰', '說', '说', '这', '這', '齊', '|', '军', '同', '已', '愿', '既', '星', '是', '景', '民', '照', '现', '現', '理', '用', '置', '要', '軍', '那', '配', '里', '開', '雷', '露', '面', '顾', ], TOP),
-            (&['个', '为', '人', '他', '以', '们', '你', '來', '個', '們', '到', '和', '大', '对', '對', '就', '我', '时', '時', '有', '来', '為', '要', '說', '说', '|', '主', '些', '因', '它', '想', '意', '理', '生', '當', '看', '着', '置', '者', '自', '著', '裡', '过', '还', '进', '進', '過', '道', '還', '里', '面', ], 0),
-            (&['些', '们', '你', '來', '們', '到', '和', '地', '她', '将', '將', '就', '年', '得', '情', '最', '样', '樣', '理', '能', '說', '说', '这', '這', '通', '|', '即', '吗', '吧', '听', '呢', '品', '响', '嗎', '师', '師', '收', '断', '斷', '明', '眼', '間', '间', '际', '陈', '限', '除', '陳', '随', '際', '隨', ], CJK_HORIZ),
-            (&['事', '前', '學', '将', '將', '情', '想', '或', '政', '斯', '新', '样', '樣', '民', '沒', '没', '然', '特', '现', '現', '球', '第', '經', '谁', '起', '|', '例', '別', '别', '制', '动', '動', '吗', '嗎', '增', '指', '明', '朝', '期', '构', '物', '确', '种', '調', '调', '費', '费', '那', '都', '間', '间', ], CJK_HORIZ | CJK_RIGHT),
+            ("他 们 你 來 們 到 和 地 对 對 就 席 我 时 時 會 来 為 能 舰 說 说 这 這 齊 | 军 同 已 愿 既 星 是 景 民 照 现 現 理 用 置 要 軍 那 配 里 開 雷 露 面 顾", TOP),
+            ("个 为 人 他 以 们 你 來 個 們 到 和 大 对 對 就 我 时 時 有 来 為 要 說 说 | 主 些 因 它 想 意 理 生 當 看 着 置 者 自 著 裡 过 还 进 進 過 道 還 里 面", 0),
+            (" 些 们 你 來 們 到 和 地 她 将 將 就 年 得 情 最 样 樣 理 能 說 说 这 這 通 | 即 吗 吧 听 呢 品 响 嗎 师 師 收 断 斷 明 眼 間 间 际 陈 限 除 陳 随 際 隨", CJK_HORIZ),
+            ("事 前 學 将 將 情 想 或 政 斯 新 样 樣 民 沒 没 然 特 现 現 球 第 經 谁 起 | 例 別 别 制 动 動 吗 嗎 增 指 明 朝 期 构 物 确 种 調 调 費 费 那 都 間 间", CJK_HORIZ | CJK_RIGHT),
         ],
     },
 ];
 
-impl ScriptClass {
+#[allow(unused)]impl ScriptClass {
     pub const ADLM: usize = 0;
     pub const ARAB: usize = 1;
     pub const ARMN: usize = 2;
@@ -969,7 +909,7 @@ pub(super) const STYLE_CLASSES: &[StyleClass] = &[
     StyleClass { name: "CJKV ideographs", index: 89, script: &SCRIPT_CLASSES[59], feature: None },
 ];
 
-impl StyleClass {
+#[allow(unused)]impl StyleClass {
     pub const ADLM: usize = 0;
     pub const ARAB: usize = 1;
     pub const ARMN: usize = 2;
diff --git a/skrifa/scripts/gen_autohint_styles.py b/skrifa/scripts/gen_autohint_styles.py
index 47562b2b0..62f163246 100644
--- a/skrifa/scripts/gen_autohint_styles.py
+++ b/skrifa/scripts/gen_autohint_styles.py
@@ -67,7 +67,7 @@
         "name": "Adlam",
         "tag": "ADLM",
         "hint_top_to_bottom": False,
-        "std_chars": ['𞤌', '𞤮'], # 𞤌 𞤮
+        "std_chars": "𞤌 𞤮", # 𞤌 𞤮
         "base_ranges": [
             (0x1E900, 0x1E95F), # Adlam
         ],
@@ -75,17 +75,17 @@
             (0x1D944, 0x1E94A),
         ],
         "blues": [
-            (['𞤌', '𞤅', '𞤈', '𞤏', '𞤔', '𞤚'], "TOP"),
-            (['𞤂', '𞤖'], "0"),
-            (['𞤬', '𞤮', '𞤻', '𞤼', '𞤾'], "TOP | LATIN_X_HEIGHT"),
-            (['𞤤', '𞤨', '𞤩', '𞤭', '𞤴', '𞤸', '𞤺', '𞥀'], "0"),
+            ("𞤌 𞤅 𞤈 𞤏 𞤔 𞤚", "TOP"),
+            ("𞤂 𞤖", "0"),
+            ("𞤬 𞤮 𞤻 𞤼 𞤾", "TOP | LATIN_X_HEIGHT"),
+            ("𞤤 𞤨 𞤩 𞤭 𞤴 𞤸 𞤺 𞥀", "0"),
         ],
     },
     {
         "name": "Arabic",
         "tag": "ARAB",
         "hint_top_to_bottom": False,
-        "std_chars": ['ل', 'ح', 'ـ'], # ل ح ـ
+        "std_chars": "ل ح ـ", # ل ح ـ
         "base_ranges": [
             (0x0600, 0x06FF), # Arabic
             (0x0750, 0x07FF), # Arabic Supplement
@@ -116,16 +116,16 @@
             (0xFE7E, 0xFE7E),
         ],
         "blues": [
-            (['ا', 'إ', 'ل', 'ك', 'ط', 'ظ'], "TOP"),
-            (['ت', 'ث', 'ط', 'ظ', 'ك'], "0"),
-            (['ـ'], "LATIN_NEUTRAL"),
+            ("ا إ ل ك ط ظ", "TOP"),
+            ("ت ث ط ظ ك", "0"),
+            ("ـ", "LATIN_NEUTRAL"),
         ],
     },
     {
         "name": "Armenian",
         "tag": "ARMN",
         "hint_top_to_bottom": False,
-        "std_chars": ['ս', 'Ս'], # ս Ս
+        "std_chars": "ս Ս", # ս Ս
         "base_ranges": [
             (0x0530, 0x058F), # Armenian
             (0xFB13, 0xFB17), # Alphab. Present. Forms (Armenian)
@@ -134,19 +134,19 @@
             (0x0559, 0x055F),
         ],
         "blues": [
-            (['Ա', 'Մ', 'Ւ', 'Ս', 'Բ', 'Գ', 'Դ', 'Օ'], "TOP"),
-            (['Ւ', 'Ո', 'Դ', 'Ճ', 'Շ', 'Ս', 'Տ', 'Օ'], "0"),
-            (['ե', 'է', 'ի', 'մ', 'վ', 'ֆ', 'ճ'], "TOP"),
-            (['ա', 'յ', 'ւ', 'ս', 'գ', 'շ', 'ր', 'օ'], "TOP | LATIN_X_HEIGHT"),
-            (['հ', 'ո', 'ճ', 'ա', 'ե', 'ծ', 'ս', 'օ'], "0"),
-            (['բ', 'ը', 'ի', 'լ', 'ղ', 'պ', 'փ', 'ց'], "0"),
+            ("Ա Մ Ւ Ս Բ Գ Դ Օ", "TOP"),
+            ("Ւ Ո Դ Ճ Շ Ս Տ Օ", "0"),
+            ("ե է ի մ վ ֆ ճ", "TOP"),
+            ("ա յ ւ ս գ շ ր օ", "TOP | LATIN_X_HEIGHT"),
+            ("հ ո ճ ա ե ծ ս օ", "0"),
+            ("բ ը ի լ ղ պ փ ց", "0"),
         ],
     },
     {
         "name": "Avestan",
         "tag": "AVST",
         "hint_top_to_bottom": False,
-        "std_chars": ['𐬚'], # 𐬚
+        "std_chars": "𐬚", # 𐬚
         "base_ranges": [
             (0x10B00, 0x10B3F), # Avestan
         ],
@@ -154,34 +154,33 @@
             (0x10B39, 0x10B3F),
         ],
         "blues": [
-            (['𐬀', '𐬁', '𐬐', '𐬛'], "TOP"),
-            (['𐬀', '𐬁'], "0"),
+            ("𐬀 𐬁 𐬐 𐬛", "TOP"),
+            ("𐬀 𐬁", "0"),
         ],
     },
     {
         "name": "Bamum",
         "tag": "BAMU",
         "hint_top_to_bottom": False,
-        "std_chars": ['ꛁ', 'ꛯ'], # ꛁ ꛯ
+        "std_chars": "ꛁ ꛯ", # ꛁ ꛯ
         "base_ranges": [
             (0xA6A0, 0xA6FF), # Bamum
-            # https://gitlab.freedesktop.org/freetype/freetype/-/blob/57617782464411201ce7bbc93b086c1b4d7d84a5/src/autofit/afranges.c#L139
-            # "The characters in the Bamum supplement are pictograms, not (directly) related to the syllabic Bamum script"
+            # This is commented out in FreeType
             # (0x16800, 0x16A3F), # Bamum Supplement
         ],
         "non_base_ranges": [
             (0xA6F0, 0xA6F1),
         ],
         "blues": [
-            (['ꚧ', 'ꚨ', 'ꛛ', 'ꛉ', 'ꛁ', 'ꛈ', 'ꛫ', 'ꛯ'], "TOP"),
-            (['ꚭ', 'ꚳ', 'ꚶ', 'ꛬ', 'ꚢ', 'ꚽ', 'ꛯ', '꛲'], "0"),
+            ("ꚧ ꚨ ꛛ ꛉ ꛁ ꛈ ꛫ ꛯ", "TOP"),
+            ("ꚭ ꚳ ꚶ ꛬ ꚢ ꚽ ꛯ ꛲", "0"),
         ],
     },
     {
         "name": "Bengali",
         "tag": "BENG",
         "hint_top_to_bottom": True,
-        "std_chars": ['০', '৪'], # ০ ৪
+        "std_chars": "০ ৪", # ০ ৪
         "base_ranges": [
             (0x0980, 0x09FF), # Bengali
         ],
@@ -194,17 +193,17 @@
             (0x09FE, 0x09FE),
         ],
         "blues": [
-            (['ই', 'ট', 'ঠ', 'ি', 'ী', 'ৈ', 'ৗ'], "TOP"),
-            (['ও', 'এ', 'ড', 'ত', 'ন', 'ব', 'ল', 'ক'], "TOP"),
-            (['অ', 'ড', 'ত', 'ন', 'ব', 'ভ', 'ল', 'ক'], "TOP | LATIN_NEUTRAL | LATIN_X_HEIGHT"),
-            (['অ', 'ড', 'ত', 'ন', 'ব', 'ভ', 'ল', 'ক'], "0"),
+            ("ই ট ঠ ি ী ৈ ৗ", "TOP"),
+            ("ও এ ড ত ন ব ল ক", "TOP"),
+            ("অ ড ত ন ব ভ ল ক", "TOP | LATIN_NEUTRAL | LATIN_X_HEIGHT"),
+            ("অ ড ত ন ব ভ ল ক", "0"),
         ],
     },
     {
         "name": "Buhid",
         "tag": "BUHD",
         "hint_top_to_bottom": False,
-        "std_chars": ['ᝋ', 'ᝏ'], # ᝋ ᝏ
+        "std_chars": "ᝋ ᝏ", # ᝋ ᝏ
         "base_ranges": [
             (0x1740, 0x175F), # Buhid
         ],
@@ -212,17 +211,17 @@
             (0x1752, 0x1753),
         ],
         "blues": [
-            (['ᝐ', 'ᝈ'], "TOP"),
-            (['ᝅ', 'ᝊ', 'ᝎ'], "TOP"),
-            (['ᝂ', 'ᝃ', 'ᝉ', 'ᝌ'], "TOP | LATIN_X_HEIGHT"),
-            (['ᝀ', 'ᝃ', 'ᝆ', 'ᝉ', 'ᝋ', 'ᝏ', 'ᝑ'], "0"),
+            ("ᝐ ᝈ", "TOP"),
+            ("ᝅ ᝊ ᝎ", "TOP"),
+            ("ᝂ ᝃ ᝉ ᝌ", "TOP | LATIN_X_HEIGHT"),
+            ("ᝀ ᝃ ᝆ ᝉ ᝋ ᝏ ᝑ", "0"),
         ],
     },
     {
         "name": "Chakma",
         "tag": "CAKM",
         "hint_top_to_bottom": False,
-        "std_chars": ['𑄤', '𑄉', '𑄛'], # 𑄤 𑄉 𑄛
+        "std_chars": "𑄤 𑄉 𑄛", # 𑄤 𑄉 𑄛
         "base_ranges": [
             (0x11100, 0x1114F), # Chakma
         ],
@@ -232,16 +231,16 @@
             (0x11146, 0x11146),
         ],
         "blues": [
-            (['𑄃', '𑄅', '𑄉', '𑄙', '𑄗'], "TOP"),
-            (['𑄅', '𑄛', '𑄝', '𑄗', '𑄓'], "0"),
-            (['𑄖', '𑄘', '𑄙', '𑄤', '𑄥'], "0"),
+            ("𑄃 𑄅 𑄉 𑄙 𑄗", "TOP"),
+            ("𑄅 𑄛 𑄝 𑄗 𑄓", "0"),
+            ("𑄖𑄳𑄢 𑄘𑄳𑄢 𑄙𑄳𑄢 𑄤𑄳𑄢 𑄥𑄳𑄢", "0"),
         ],
     },
     {
         "name": "Canadian Syllabics",
         "tag": "CANS",
         "hint_top_to_bottom": False,
-        "std_chars": ['ᑌ', 'ᓚ'], # ᑌ ᓚ
+        "std_chars": "ᑌ ᓚ", # ᑌ ᓚ
         "base_ranges": [
             (0x1400, 0x167F), # Unified Canadian Aboriginal Syllabics
             (0x18B0, 0x18FF), # Unified Canadian Aboriginal Syllabics Extended
@@ -249,34 +248,34 @@
         "non_base_ranges": [
         ],
         "blues": [
-            (['ᗜ', 'ᖴ', 'ᐁ', 'ᒣ', 'ᑫ', 'ᑎ', 'ᔑ', 'ᗰ'], "TOP"),
-            (['ᗶ', 'ᖵ', 'ᒧ', 'ᐃ', 'ᑌ', 'ᒍ', 'ᔑ', 'ᗢ'], "0"),
-            (['ᓓ', 'ᓕ', 'ᓀ', 'ᓂ', 'ᓄ', 'ᕄ', 'ᕆ', 'ᘣ'], "TOP | LATIN_X_HEIGHT"),
-            (['ᕃ', 'ᓂ', 'ᓀ', 'ᕂ', 'ᓗ', 'ᓚ', 'ᕆ', 'ᘣ'], "0"),
-            (['ᐪ', 'ᙆ', 'ᣘ', 'ᐢ', 'ᒾ', 'ᣗ', 'ᔆ'], "TOP"),
-            (['ᙆ', 'ᗮ', 'ᒻ', 'ᐞ', 'ᔆ', 'ᒡ', 'ᒢ', 'ᓑ'], "0"),
+            ("ᗜ ᖴ ᐁ ᒣ ᑫ ᑎ ᔑ ᗰ", "TOP"),
+            ("ᗶ ᖵ ᒧ ᐃ ᑌ ᒍ ᔑ ᗢ", "0"),
+            ("ᓓ ᓕ ᓀ ᓂ ᓄ ᕄ ᕆ ᘣ", "TOP | LATIN_X_HEIGHT"),
+            ("ᕃ ᓂ ᓀ ᕂ ᓗ ᓚ ᕆ ᘣ", "0"),
+            ("ᐪ ᙆ ᣘ ᐢ ᒾ ᣗ ᔆ", "TOP"),
+            ("ᙆ ᗮ ᒻ ᐞ ᔆ ᒡ ᒢ ᓑ", "0"),
         ],
     },
     {
         "name": "Carian",
         "tag": "CARI",
         "hint_top_to_bottom": False,
-        "std_chars": ['𐊫', '𐋉'], # 𐊫 𐋉
+        "std_chars": "𐊫 𐋉", # 𐊫 𐋉
         "base_ranges": [
             (0x102A0, 0x102DF), # Carian
         ],
         "non_base_ranges": [
         ],
         "blues": [
-            (['𐊧', '𐊫', '𐊬', '𐊭', '𐊱', '𐊺', '𐊼', '𐊿'], "TOP"),
-            (['𐊣', '𐊧', '𐊷', '𐋀', '𐊫', '𐊸', '𐋉'], "0"),
+            ("𐊧 𐊫 𐊬 𐊭 𐊱 𐊺 𐊼 𐊿", "TOP"),
+            ("𐊣 𐊧 𐊷 𐋀 𐊫 𐊸 𐋉", "0"),
         ],
     },
     {
         "name": "Cherokee",
         "tag": "CHER",
         "hint_top_to_bottom": False,
-        "std_chars": ['Ꭴ', 'Ꮕ', 'ꮕ'], # Ꭴ Ꮕ ꮕ
+        "std_chars": "Ꭴ Ꮕ ꮕ", # Ꭴ Ꮕ ꮕ
         "base_ranges": [
             (0x13A0, 0x13FF), # Cherokee
             (0xAB70, 0xABBF), # Cherokee Supplement
@@ -284,19 +283,19 @@
         "non_base_ranges": [
         ],
         "blues": [
-            (['Ꮖ', 'Ꮋ', 'Ꭼ', 'Ꮓ', 'Ꭴ', 'Ꮳ', 'Ꭶ', 'Ꮥ'], "TOP"),
-            (['Ꮖ', 'Ꮋ', 'Ꭼ', 'Ꮓ', 'Ꭴ', 'Ꮳ', 'Ꭶ', 'Ꮥ'], "0"),
-            (['ꮒ', 'ꮤ', 'ꮶ', 'ꭴ', 'ꭾ', 'ꮗ', 'ꮝ', 'ꮿ'], "TOP"),
-            (['ꮖ', 'ꭼ', 'ꮓ', 'ꮠ', 'ꮳ', 'ꭶ', 'ꮥ', 'ꮻ'], "TOP | LATIN_X_HEIGHT"),
-            (['ꮖ', 'ꭼ', 'ꮓ', 'ꮠ', 'ꮳ', 'ꭶ', 'ꮥ', 'ꮻ'], "0"),
-            (['ᏸ', 'ꮐ', 'ꭹ', 'ꭻ'], "0"),
+            ("Ꮖ Ꮋ Ꭼ Ꮓ Ꭴ Ꮳ Ꭶ Ꮥ", "TOP"),
+            ("Ꮖ Ꮋ Ꭼ Ꮓ Ꭴ Ꮳ Ꭶ Ꮥ", "0"),
+            ("ꮒ ꮤ ꮶ ꭴ ꭾ ꮗ ꮝ ꮿ", "TOP"),
+            ("ꮖ ꭼ ꮓ ꮠ ꮳ ꭶ ꮥ ꮻ", "TOP | LATIN_X_HEIGHT"),
+            ("ꮖ ꭼ ꮓ ꮠ ꮳ ꭶ ꮥ ꮻ", "0"),
+            ("ᏸ ꮐ ꭹ ꭻ", "0"),
         ],
     },
     {
         "name": "Coptic",
         "tag": "COPT",
         "hint_top_to_bottom": False,
-        "std_chars": ['Ⲟ', 'ⲟ'], # Ⲟ ⲟ
+        "std_chars": "Ⲟ ⲟ", # Ⲟ ⲟ
         "base_ranges": [
             (0x2C80, 0x2CFF), # Coptic
         ],
@@ -304,34 +303,34 @@
             (0x2CEF, 0x2CF1),
         ],
         "blues": [
-            (['Ⲍ', 'Ⲏ', 'Ⲡ', 'Ⳟ', 'Ⲟ', 'Ⲑ', 'Ⲥ', 'Ⳋ'], "TOP"),
-            (['Ⳑ', 'Ⳙ', 'Ⳟ', 'Ⲏ', 'Ⲟ', 'Ⲑ', 'Ⳝ', 'Ⲱ'], "0"),
-            (['ⲍ', 'ⲏ', 'ⲡ', 'ⳟ', 'ⲟ', 'ⲑ', 'ⲥ', 'ⳋ'], "TOP | LATIN_X_HEIGHT"),
-            (['ⳑ', 'ⳙ', 'ⳟ', 'ⲏ', 'ⲟ', 'ⲑ', 'ⳝ', 'Ⳓ'], "0"),
+            ("Ⲍ Ⲏ Ⲡ Ⳟ Ⲟ Ⲑ Ⲥ Ⳋ", "TOP"),
+            ("Ⳑ Ⳙ Ⳟ Ⲏ Ⲟ Ⲑ Ⳝ Ⲱ", "0"),
+            ("ⲍ ⲏ ⲡ ⳟ ⲟ ⲑ ⲥ ⳋ", "TOP | LATIN_X_HEIGHT"),
+            ("ⳑ ⳙ ⳟ ⲏ ⲟ ⲑ ⳝ Ⳓ", "0"),
         ],
     },
     {
         "name": "Cypriot",
         "tag": "CPRT",
         "hint_top_to_bottom": False,
-        "std_chars": ['𐠅', '𐠣'], # 𐠅 𐠣
+        "std_chars": "𐠅 𐠣", # 𐠅 𐠣
         "base_ranges": [
             (0x10800, 0x1083F), # Cypriot
         ],
         "non_base_ranges": [
         ],
         "blues": [
-            (['𐠍', '𐠙', '𐠳', '𐠱', '𐠅', '𐠓', '𐠣', '𐠦'], "TOP"),
-            (['𐠃', '𐠊', '𐠛', '𐠣', '𐠳', '𐠵', '𐠐'], "0"),
-            (['𐠈', '𐠏', '𐠖'], "TOP"),
-            (['𐠈', '𐠏', '𐠖'], "0"),
+            ("𐠍 𐠙 𐠳 𐠱 𐠅 𐠓 𐠣 𐠦", "TOP"),
+            ("𐠃 𐠊 𐠛 𐠣 𐠳 𐠵 𐠐", "0"),
+            ("𐠈 𐠏 𐠖", "TOP"),
+            ("𐠈 𐠏 𐠖", "0"),
         ],
     },
     {
         "name": "Cyrillic",
         "tag": "CYRL",
         "hint_top_to_bottom": False,
-        "std_chars": ['о', 'О'], # о О
+        "std_chars": "о О", # о О
         "base_ranges": [
             (0x0400, 0x04FF), # Cyrillic
             (0x0500, 0x052F), # Cyrillic Supplement
@@ -346,18 +345,18 @@
             (0xA69E, 0xA69F),
         ],
         "blues": [
-            (['Б', 'В', 'Е', 'П', 'З', 'О', 'С', 'Э'], "TOP"),
-            (['Б', 'В', 'Е', 'Ш', 'З', 'О', 'С', 'Э'], "0"),
-            (['х', 'п', 'н', 'ш', 'е', 'з', 'о', 'с'], "TOP | LATIN_X_HEIGHT"),
-            (['х', 'п', 'н', 'ш', 'е', 'з', 'о', 'с'], "0"),
-            (['р', 'у', 'ф'], "0"),
+            ("Б В Е П З О С Э", "TOP"),
+            ("Б В Е Ш З О С Э", "0"),
+            ("х п н ш е з о с", "TOP | LATIN_X_HEIGHT"),
+            ("х п н ш е з о с", "0"),
+            ("р у ф", "0"),
         ],
     },
     {
         "name": "Devanagari",
         "tag": "DEVA",
         "hint_top_to_bottom": True,
-        "std_chars": ['ठ', 'व', 'ट'], # ठ व ट
+        "std_chars": "ठ व ट", # ठ व ट
         "base_ranges": [
             (0x0900, 0x093B), # Devanagari
             (0x093D, 0x0950), # ... continued
@@ -377,35 +376,35 @@
             (0xA8FF, 0xA8FF),
         ],
         "blues": [
-            (['ई', 'ऐ', 'ओ', 'औ', 'ि', 'ी', 'ो', 'ौ'], "TOP"),
-            (['क', 'म', 'अ', 'आ', 'थ', 'ध', 'भ', 'श'], "TOP"),
-            (['क', 'न', 'म', 'उ', 'छ', 'ट', 'ठ', 'ड'], "TOP | LATIN_NEUTRAL | LATIN_X_HEIGHT"),
-            (['क', 'न', 'म', 'उ', 'छ', 'ट', 'ठ', 'ड'], "0"),
-            (['ु', 'ृ'], "0"),
+            ("ई ऐ ओ औ ि ी ो ौ", "TOP"),
+            ("क म अ आ थ ध भ श", "TOP"),
+            ("क न म उ छ ट ठ ड", "TOP | LATIN_NEUTRAL | LATIN_X_HEIGHT"),
+            ("क न म उ छ ट ठ ड", "0"),
+            ("ु ृ", "0"),
         ],
     },
     {
         "name": "Deseret",
         "tag": "DSRT",
         "hint_top_to_bottom": False,
-        "std_chars": ['𐐄', '𐐬'], # 𐐄 𐐬
+        "std_chars": "𐐄 𐐬", # 𐐄 𐐬
         "base_ranges": [
             (0x10400, 0x1044F), # Deseret
         ],
         "non_base_ranges": [
         ],
         "blues": [
-            (['𐐂', '𐐄', '𐐋', '𐐗', '𐐑'], "TOP"),
-            (['𐐀', '𐐂', '𐐄', '𐐗', '𐐛'], "0"),
-            (['𐐪', '𐐬', '𐐳', '𐐿', '𐐹'], "TOP | LATIN_X_HEIGHT"),
-            (['𐐨', '𐐪', '𐐬', '𐐿', '𐑃'], "0"),
+            ("𐐂 𐐄 𐐋 𐐗 𐐑", "TOP"),
+            ("𐐀 𐐂 𐐄 𐐗 𐐛", "0"),
+            ("𐐪 𐐬 𐐳 𐐿 𐐹", "TOP | LATIN_X_HEIGHT"),
+            ("𐐨 𐐪 𐐬 𐐿 𐑃", "0"),
         ],
     },
     {
         "name": "Ethiopic",
         "tag": "ETHI",
         "hint_top_to_bottom": False,
-        "std_chars": ['ዐ'], # ዐ
+        "std_chars": "ዐ", # ዐ
         "base_ranges": [
             (0x1200, 0x137F), # Ethiopic
             (0x1380, 0x139F), # Ethiopic Supplement
@@ -416,15 +415,15 @@
             (0x135D, 0x135F),
         ],
         "blues": [
-            (['ሀ', 'ሃ', 'ዘ', 'ፐ', 'ማ', 'በ', 'ዋ', 'ዐ'], "TOP"),
-            (['ለ', 'ሐ', 'በ', 'ዘ', 'ሀ', 'ሪ', 'ዐ', 'ጨ'], "0"),
+            ("ሀ ሃ ዘ ፐ ማ በ ዋ ዐ", "TOP"),
+            ("ለ ሐ በ ዘ ሀ ሪ ዐ ጨ", "0"),
         ],
     },
     {
         "name": "Georgian (Mkhedruli)",
         "tag": "GEOR",
         "hint_top_to_bottom": False,
-        "std_chars": ['ი', 'ე', 'ა', 'Ჿ'], # ი ე ა Ი
+        "std_chars": "ი ე ა Ჿ", # ი ე ა Ი
         "base_ranges": [
             (0x10D0, 0x10FF), # Georgian (Mkhedruli)
             (0x1C90, 0x1CBF), # Georgian Extended (Mtavruli)
@@ -432,19 +431,19 @@
         "non_base_ranges": [
         ],
         "blues": [
-            (['გ', 'დ', 'ე', 'ვ', 'თ', 'ი', 'ო', 'ღ'], "TOP | LATIN_X_HEIGHT"),
-            (['ა', 'ზ', 'მ', 'ს', 'შ', 'ძ', 'ხ', 'პ'], "0"),
-            (['ს', 'ხ', 'ქ', 'ზ', 'მ', 'შ', 'ჩ', 'წ'], "TOP"),
-            (['ე', 'ვ', 'ჟ', 'ტ', 'უ', 'ფ', 'ქ', 'ყ'], "0"),
-            (['Ნ', 'Ჟ', 'Ჳ', 'Ჸ', 'Გ', 'Ე', 'Ო', 'Ჴ'], "TOP"),
-            (['Ი', 'Ჲ', 'Ო', 'Ჩ', 'Მ', 'Შ', 'Ჯ', 'Ჽ'], "0"),
+            ("გ დ ე ვ თ ი ო ღ", "TOP | LATIN_X_HEIGHT"),
+            ("ა ზ მ ს შ ძ ხ პ", "0"),
+            ("ს ხ ქ ზ მ შ ჩ წ", "TOP"),
+            ("ე ვ ჟ ტ უ ფ ქ ყ", "0"),
+            ("Ნ Ჟ Ჳ Ჸ Გ Ე Ო Ჴ", "TOP"),
+            ("Ი Ჲ Ო Ჩ Მ Შ Ჯ Ჽ", "0"),
         ],
     },
     {
         "name": "Georgian (Khutsuri)",
         "tag": "GEOK",
         "hint_top_to_bottom": False,
-        "std_chars": ['Ⴖ', 'Ⴑ', 'ⴙ'], # Ⴖ Ⴑ ⴙ
+        "std_chars": "Ⴖ Ⴑ ⴙ", # Ⴖ Ⴑ ⴙ
         "base_ranges": [
             (0x10A0, 0x10CD), # Georgian (Asomtavruli)
             (0x2D00, 0x2D2D), # Georgian Supplement (Nuskhuri)
@@ -452,19 +451,19 @@
         "non_base_ranges": [
         ],
         "blues": [
-            (['Ⴑ', 'Ⴇ', 'Ⴙ', 'Ⴜ', 'Ⴄ', 'Ⴅ', 'Ⴓ', 'Ⴚ'], "TOP"),
-            (['Ⴄ', 'Ⴅ', 'Ⴇ', 'Ⴈ', 'Ⴆ', 'Ⴑ', 'Ⴊ', 'Ⴋ'], "0"),
-            (['ⴁ', 'ⴗ', 'ⴂ', 'ⴄ', 'ⴅ', 'ⴇ', 'ⴔ', 'ⴖ'], "TOP | LATIN_X_HEIGHT"),
-            (['ⴈ', 'ⴌ', 'ⴖ', 'ⴎ', 'ⴃ', 'ⴆ', 'ⴋ', 'ⴢ'], "0"),
-            (['ⴐ', 'ⴑ', 'ⴓ', 'ⴕ', 'ⴙ', 'ⴛ', 'ⴡ', 'ⴣ'], "TOP"),
-            (['ⴄ', 'ⴅ', 'ⴔ', 'ⴕ', 'ⴁ', 'ⴂ', 'ⴘ', 'ⴝ'], "0"),
+            ("Ⴑ Ⴇ Ⴙ Ⴜ Ⴄ Ⴅ Ⴓ Ⴚ", "TOP"),
+            ("Ⴄ Ⴅ Ⴇ Ⴈ Ⴆ Ⴑ Ⴊ Ⴋ", "0"),
+            ("ⴁ ⴗ ⴂ ⴄ ⴅ ⴇ ⴔ ⴖ", "TOP | LATIN_X_HEIGHT"),
+            ("ⴈ ⴌ ⴖ ⴎ ⴃ ⴆ ⴋ ⴢ", "0"),
+            ("ⴐ ⴑ ⴓ ⴕ ⴙ ⴛ ⴡ ⴣ", "TOP"),
+            ("ⴄ ⴅ ⴔ ⴕ ⴁ ⴂ ⴘ ⴝ", "0"),
         ],
     },
     {
         "name": "Glagolitic",
         "tag": "GLAG",
         "hint_top_to_bottom": False,
-        "std_chars": ['Ⱅ', 'ⱅ'], # Ⱅ ⱅ
+        "std_chars": "Ⱅ ⱅ", # Ⱅ ⱅ
         "base_ranges": [
             (0x2C00, 0x2C5F), # Glagolitic
             (0x1E000, 0x1E02F), # Glagolitic Supplement
@@ -473,32 +472,32 @@
             (0x1E000, 0x1E02F),
         ],
         "blues": [
-            (['Ⰵ', 'Ⱄ', 'Ⱚ', 'Ⰴ', 'Ⰲ', 'Ⰺ', 'Ⱛ', 'Ⰻ'], "TOP"),
-            (['Ⰵ', 'Ⰴ', 'Ⰲ', 'Ⱚ', 'Ⱎ', 'Ⱑ', 'Ⰺ', 'Ⱄ'], "0"),
-            (['ⰵ', 'ⱄ', 'ⱚ', 'ⰴ', 'ⰲ', 'ⰺ', 'ⱛ', 'ⰻ'], "TOP | LATIN_X_HEIGHT"),
-            (['ⰵ', 'ⰴ', 'ⰲ', 'ⱚ', 'ⱎ', 'ⱑ', 'ⰺ', 'ⱄ'], "0"),
+            ("Ⰵ Ⱄ Ⱚ Ⰴ Ⰲ Ⰺ Ⱛ Ⰻ", "TOP"),
+            ("Ⰵ Ⰴ Ⰲ Ⱚ Ⱎ Ⱑ Ⰺ Ⱄ", "0"),
+            ("ⰵ ⱄ ⱚ ⰴ ⰲ ⰺ ⱛ ⰻ", "TOP | LATIN_X_HEIGHT"),
+            ("ⰵ ⰴ ⰲ ⱚ ⱎ ⱑ ⰺ ⱄ", "0"),
         ],
     },
     {
         "name": "Gothic",
         "tag": "GOTH",
         "hint_top_to_bottom": True,
-        "std_chars": ['𐌴', '𐌾', '𐍃'], # 𐌴 𐌾 𐍃
+        "std_chars": "𐌴 𐌾 𐍃", # 𐌴 𐌾 𐍃
         "base_ranges": [
             (0x10330, 0x1034F), # Gothic
         ],
         "non_base_ranges": [
         ],
         "blues": [
-            (['𐌲', '𐌶', '𐍀', '𐍄', '𐌴', '𐍃', '𐍈', '𐌾'], "TOP"),
-            (['𐌶', '𐌴', '𐍃', '𐍈'], "0"),
+            ("𐌲 𐌶 𐍀 𐍄 𐌴 𐍃 𐍈 𐌾", "TOP"),
+            ("𐌶 𐌴 𐍃 𐍈", "0"),
         ],
     },
     {
         "name": "Greek",
         "tag": "GREK",
         "hint_top_to_bottom": False,
-        "std_chars": ['ο', 'Ο'], # ο Ο
+        "std_chars": "ο Ο", # ο Ο
         "base_ranges": [
             (0x0370, 0x03FF), # Greek and Coptic
             (0x1F00, 0x1FFF), # Greek Extended
@@ -513,19 +512,19 @@
             (0x1FFD, 0x1FFE),
         ],
         "blues": [
-            (['Γ', 'Β', 'Ε', 'Ζ', 'Θ', 'Ο', 'Ω'], "TOP"),
-            (['Β', 'Δ', 'Ζ', 'Ξ', 'Θ', 'Ο'], "0"),
-            (['β', 'θ', 'δ', 'ζ', 'λ', 'ξ'], "TOP"),
-            (['α', 'ε', 'ι', 'ο', 'π', 'σ', 'τ', 'ω'], "TOP | LATIN_X_HEIGHT"),
-            (['α', 'ε', 'ι', 'ο', 'π', 'σ', 'τ', 'ω'], "0"),
-            (['β', 'γ', 'η', 'μ', 'ρ', 'φ', 'χ', 'ψ'], "0"),
+            ("Γ Β Ε Ζ Θ Ο Ω", "TOP"),
+            ("Β Δ Ζ Ξ Θ Ο", "0"),
+            ("β θ δ ζ λ ξ", "TOP"),
+            ("α ε ι ο π σ τ ω", "TOP | LATIN_X_HEIGHT"),
+            ("α ε ι ο π σ τ ω", "0"),
+            ("β γ η μ ρ φ χ ψ", "0"),
         ],
     },
     {
         "name": "Gujarati",
         "tag": "GUJR",
         "hint_top_to_bottom": False,
-        "std_chars": ['ટ', '૦'], # ટ ૦
+        "std_chars": "ટ ૦", # ટ ૦
         "base_ranges": [
             (0x0A80, 0x0AFF), # Gujarati
         ],
@@ -538,18 +537,18 @@
             (0x0AFA, 0x0AFF),
         ],
         "blues": [
-            (['ત', 'ન', 'ઋ', 'ઌ', 'છ', 'ટ', 'ર', '૦'], "TOP | LATIN_X_HEIGHT"),
-            (['ખ', 'ગ', 'ઘ', 'ઞ', 'ઇ', 'ઈ', 'ઠ', 'જ'], "0"),
-            (['ઈ', 'ઊ', 'િ', 'ી', 'લ', 'શ', 'જ', 'સ'], "TOP"),
-            (['ુ', 'ૃ', 'ૄ', 'ખ', 'છ', 'છ'], "0"),
-            (['૦', '૧', '૨', '૩', '૭'], "TOP"),
+            ("ત ન ઋ ઌ છ ટ ર ૦", "TOP | LATIN_X_HEIGHT"),
+            ("ખ ગ ઘ ઞ ઇ ઈ ઠ જ", "0"),
+            ("ઈ ઊ િ ી લી શ્ચિ જિ સી", "TOP"),
+            ("ુ ૃ ૄ ખુ છૃ છૄ", "0"),
+            ("૦ ૧ ૨ ૩ ૭", "TOP"),
         ],
     },
     {
         "name": "Gurmukhi",
         "tag": "GURU",
         "hint_top_to_bottom": True,
-        "std_chars": ['ਠ', 'ਰ', '੦'], # ਠ ਰ ੦
+        "std_chars": "ਠ ਰ ੦", # ਠ ਰ ੦
         "base_ranges": [
             (0x0A00, 0x0A7F), # Gurmukhi
         ],
@@ -561,18 +560,18 @@
             (0x0A75, 0x0A75),
         ],
         "blues": [
-            (['ਇ', 'ਈ', 'ਉ', 'ਏ', 'ਓ', 'ੳ', 'ਿ', 'ੀ'], "TOP"),
-            (['ਕ', 'ਗ', 'ਙ', 'ਚ', 'ਜ', 'ਤ', 'ਧ', 'ਸ'], "TOP"),
-            (['ਕ', 'ਗ', 'ਙ', 'ਚ', 'ਜ', 'ਤ', 'ਧ', 'ਸ'], "TOP | LATIN_NEUTRAL | LATIN_X_HEIGHT"),
-            (['ਅ', 'ਏ', 'ਓ', 'ਗ', 'ਜ', 'ਠ', 'ਰ', 'ਸ'], "0"),
-            (['੦', '੧', '੨', '੩', '੭'], "TOP"),
+            ("ਇ ਈ ਉ ਏ ਓ ੳ ਿ ੀ", "TOP"),
+            ("ਕ ਗ ਙ ਚ ਜ ਤ ਧ ਸ", "TOP"),
+            ("ਕ ਗ ਙ ਚ ਜ ਤ ਧ ਸ", "TOP | LATIN_NEUTRAL | LATIN_X_HEIGHT"),
+            ("ਅ ਏ ਓ ਗ ਜ ਠ ਰ ਸ", "0"),
+            ("੦ ੧ ੨ ੩ ੭", "TOP"),
         ],
     },
     {
         "name": "Hebrew",
         "tag": "HEBR",
         "hint_top_to_bottom": False,
-        "std_chars": ['ם'], # ם
+        "std_chars": "ם", # ם
         "base_ranges": [
             (0x0590, 0x05FF), # Hebrew
             (0xFB1D, 0xFB4F), # Alphab. Present. Forms (Hebrew)
@@ -585,16 +584,16 @@
             (0xFB1E, 0xFB1E),
         ],
         "blues": [
-            (['ב', 'ד', 'ה', 'ח', 'ך', 'כ', 'ם', 'ס'], "TOP | LATIN_LONG"),
-            (['ב', 'ט', 'כ', 'ם', 'ס', 'צ'], "0"),
-            (['ק', 'ך', 'ן', 'ף', 'ץ'], "0"),
+            ("ב ד ה ח ך כ ם ס", "TOP | LATIN_LONG"),
+            ("ב ט כ ם ס צ", "0"),
+            ("ק ך ן ף ץ", "0"),
         ],
     },
     {
         "name": "Kayah Li",
         "tag": "KALI",
         "hint_top_to_bottom": False,
-        "std_chars": ['ꤍ', '꤀'], # ꤍ ꤀
+        "std_chars": "ꤍ ꤀", # ꤍ ꤀
         "base_ranges": [
             (0xA900, 0xA92F), # Kayah Li
         ],
@@ -602,18 +601,18 @@
             (0xA926, 0xA92D),
         ],
         "blues": [
-            (['꤅', 'ꤏ', '꤁', 'ꤋ', '꤀', 'ꤍ'], "TOP | LATIN_X_HEIGHT"),
-            (['꤈', 'ꤘ', '꤀', 'ꤍ', 'ꤢ'], "0"),
-            (['ꤖ', 'ꤡ'], "TOP"),
-            (['ꤑ', 'ꤜ', 'ꤞ'], "0"),
-            (['ꤑ', 'ꤜ', 'ꤔ'], "0"),
+            ("꤅ ꤏ ꤁ ꤋ ꤀ ꤍ", "TOP | LATIN_X_HEIGHT"),
+            ("꤈ ꤘ ꤀ ꤍ ꤢ", "0"),
+            ("ꤖ ꤡ", "TOP"),
+            ("ꤑ ꤜ ꤞ", "0"),
+            ("ꤑ꤬ ꤜ꤭ ꤔ꤬", "0"),
         ],
     },
     {
         "name": "Khmer",
         "tag": "KHMR",
         "hint_top_to_bottom": False,
-        "std_chars": ['០'], # ០
+        "std_chars": "០", # ០
         "base_ranges": [
             (0x1780, 0x17FF), # Khmer
         ],
@@ -624,33 +623,33 @@
             (0x17DD, 0x17DD),
         ],
         "blues": [
-            (['ខ', 'ទ', 'ន', 'ឧ', 'ឩ', 'ា'], "TOP | LATIN_X_HEIGHT"),
-            (['ក', 'ក', 'ក', 'ក'], "LATIN_SUB_TOP"),
-            (['ខ', 'ឃ', 'ច', 'ឋ', 'ប', 'ម', 'យ', 'ឲ'], "0"),
-            (['ត', 'រ', 'ឲ', 'អ'], "0"),
-            (['ន', 'ង', 'ក', 'ច', 'ន', 'ល'], "0"),
+            ("ខ ទ ន ឧ ឩ ា", "TOP | LATIN_X_HEIGHT"),
+            ("ក្ក ក្ខ ក្គ ក្ថ", "LATIN_SUB_TOP"),
+            ("ខ ឃ ច ឋ ប ម យ ឲ", "0"),
+            ("ត្រ រៀ ឲ្យ អឿ", "0"),
+            ("ន្ត្រៃ ង្ខ្យ ក្បៀ ច្រៀ ន្តឿ ល្បឿ", "0"),
         ],
     },
     {
         "name": "Khmer Symbols",
         "tag": "KHMS",
         "hint_top_to_bottom": False,
-        "std_chars": ['᧡', '᧪'], # ᧡ ᧪
+        "std_chars": "᧡ ᧪", # ᧡ ᧪
         "base_ranges": [
             (0x19E0, 0x19FF), # Khmer Symbols
         ],
         "non_base_ranges": [
         ],
         "blues": [
-            (['᧠', '᧡'], "TOP | LATIN_X_HEIGHT"),
-            (['᧶', '᧹'], "0"),
+            ("᧠ ᧡", "TOP | LATIN_X_HEIGHT"),
+            ("᧶ ᧹", "0"),
         ],
     },
     {
         "name": "Kannada",
         "tag": "KNDA",
         "hint_top_to_bottom": False,
-        "std_chars": ['೦', 'ಬ'], # ೦ ಬ
+        "std_chars": "೦ ಬ", # ೦ ಬ
         "base_ranges": [
             (0x0C80, 0x0CFF), # Kannada
         ],
@@ -663,15 +662,15 @@
             (0x0CE2, 0x0CE3),
         ],
         "blues": [
-            (['ಇ', 'ಊ', 'ಐ', 'ಣ', 'ಸ', 'ನ', 'ದ', 'ರ'], "TOP"),
-            (['ಅ', 'ಉ', 'ಎ', 'ಲ', '೦', '೨', '೬', '೭'], "0"),
+            ("ಇ ಊ ಐ ಣ ಸಾ ನಾ ದಾ ರಾ", "TOP"),
+            ("ಅ ಉ ಎ ಲ ೦ ೨ ೬ ೭", "0"),
         ],
     },
     {
         "name": "Lao",
         "tag": "LAOO",
         "hint_top_to_bottom": False,
-        "std_chars": ['໐'], # ໐
+        "std_chars": "໐", # ໐
         "base_ranges": [
             (0x0E80, 0x0EFF), # Lao
         ],
@@ -681,18 +680,18 @@
             (0x0EC8, 0x0ECD),
         ],
         "blues": [
-            (['າ', 'ດ', 'ອ', 'ມ', 'ລ', 'ວ', 'ຣ', 'ງ'], "TOP | LATIN_X_HEIGHT"),
-            (['າ', 'ອ', 'ບ', 'ຍ', 'ຣ', 'ຮ', 'ວ', 'ຢ'], "0"),
-            (['ປ', 'ຢ', 'ຟ', 'ຝ'], "TOP"),
-            (['ໂ', 'ໄ', 'ໃ'], "TOP"),
-            (['ງ', 'ຊ', 'ຖ', 'ຽ', 'ໆ', 'ຯ'], "0"),
+            ("າ ດ ອ ມ ລ ວ ຣ ງ", "TOP | LATIN_X_HEIGHT"),
+            ("າ ອ ບ ຍ ຣ ຮ ວ ຢ", "0"),
+            ("ປ ຢ ຟ ຝ", "TOP"),
+            ("ໂ ໄ ໃ", "TOP"),
+            ("ງ ຊ ຖ ຽ ໆ ຯ", "0"),
         ],
     },
     {
         "name": "Latin",
         "tag": "LATN",
         "hint_top_to_bottom": False,
-        "std_chars": ['o', 'O', '0'],
+        "std_chars": "o O 0",
         "base_ranges": [
             (0x0020, 0x007F), # Basic Latin (no control chars)
             (0x00A0, 0x00A9), # Latin-1 Supplement (no control chars)
@@ -746,19 +745,19 @@
             (0xA7F8, 0xA7FA),
         ],
         "blues": [
-            (['T', 'H', 'E', 'Z', 'O', 'C', 'Q', 'S'], "TOP"),
-            (['H', 'E', 'Z', 'L', 'O', 'C', 'U', 'S'], "0"),
-            (['f', 'i', 'j', 'k', 'd', 'b', 'h'], "TOP"),
-            (['u', 'v', 'x', 'z', 'o', 'e', 's', 'c'], "TOP | LATIN_X_HEIGHT"),
-            (['n', 'r', 'x', 'z', 'o', 'e', 's', 'c'], "0"),
-            (['p', 'q', 'g', 'j', 'y'], "0"),
+            ("T H E Z O C Q S", "TOP"),
+            ("H E Z L O C U S", "0"),
+            ("f i j k d b h", "TOP"),
+            ("u v x z o e s c", "TOP | LATIN_X_HEIGHT"),
+            ("n r x z o e s c", "0"),
+            ("p q g j y", "0"),
         ],
     },
     {
         "name": "Latin Subscript Fallback",
         "tag": "LATB",
         "hint_top_to_bottom": False,
-        "std_chars": ['ₒ', '₀'], # ₒ ₀
+        "std_chars": "ₒ ₀", # ₒ ₀
         "base_ranges": [
             (0x1D62, 0x1D6A), # some small subscript letters
             (0x2080, 0x209C), # subscript digits and letters
@@ -767,19 +766,19 @@
         "non_base_ranges": [
         ],
         "blues": [
-            (['₀', '₃', '₅', '₇', '₈'], "TOP"),
-            (['₀', '₁', '₂', '₃', '₈'], "0"),
-            (['ᵢ', 'ⱼ', 'ₕ', 'ₖ', 'ₗ'], "TOP"),
-            (['ₐ', 'ₑ', 'ₒ', 'ₓ', 'ₙ', 'ₛ', 'ᵥ', 'ᵤ', 'ᵣ'], "TOP | LATIN_X_HEIGHT"),
-            (['ₐ', 'ₑ', 'ₒ', 'ₓ', 'ₙ', 'ₛ', 'ᵥ', 'ᵤ', 'ᵣ'], "0"),
-            (['ᵦ', 'ᵧ', 'ᵨ', 'ᵩ', 'ₚ'], "0"),
+            ("₀ ₃ ₅ ₇ ₈", "TOP"),
+            ("₀ ₁ ₂ ₃ ₈", "0"),
+            ("ᵢ ⱼ ₕ ₖ ₗ", "TOP"),
+            ("ₐ ₑ ₒ ₓ ₙ ₛ ᵥ ᵤ ᵣ", "TOP | LATIN_X_HEIGHT"),
+            ("ₐ ₑ ₒ ₓ ₙ ₛ ᵥ ᵤ ᵣ", "0"),
+            ("ᵦ ᵧ ᵨ ᵩ ₚ", "0"),
         ],
     },
     {
         "name": "Latin Superscript Fallback",
         "tag": "LATP",
         "hint_top_to_bottom": False,
-        "std_chars": ['ᵒ', 'ᴼ', '⁰'], # ᵒ ᴼ ⁰
+        "std_chars": "ᵒ ᴼ ⁰", # ᵒ ᴼ ⁰
         "base_ranges": [
             (0x00AA, 0x00AA), # feminine ordinal indicator
             (0x00B2, 0x00B3), # superscript two and three
@@ -798,34 +797,34 @@
         "non_base_ranges": [
         ],
         "blues": [
-            (['⁰', '³', '⁵', '⁷', 'ᵀ', 'ᴴ', 'ᴱ', 'ᴼ'], "TOP"),
-            (['⁰', '¹', '²', '³', 'ᴱ', 'ᴸ', 'ᴼ', 'ᵁ'], "0"),
-            (['ᵇ', 'ᵈ', 'ᵏ', 'ʰ', 'ʲ', 'ᶠ', 'ⁱ'], "TOP"),
-            (['ᵉ', 'ᵒ', 'ʳ', 'ˢ', 'ˣ', 'ᶜ', 'ᶻ'], "TOP | LATIN_X_HEIGHT"),
-            (['ᵉ', 'ᵒ', 'ʳ', 'ˢ', 'ˣ', 'ᶜ', 'ᶻ'], "0"),
-            (['ᵖ', 'ʸ', 'ᵍ'], "0"),
+            ("⁰ ³ ⁵ ⁷ ᵀ ᴴ ᴱ ᴼ", "TOP"),
+            ("⁰ ¹ ² ³ ᴱ ᴸ ᴼ ᵁ", "0"),
+            ("ᵇ ᵈ ᵏ ʰ ʲ ᶠ ⁱ", "TOP"),
+            ("ᵉ ᵒ ʳ ˢ ˣ ᶜ ᶻ", "TOP | LATIN_X_HEIGHT"),
+            ("ᵉ ᵒ ʳ ˢ ˣ ᶜ ᶻ", "0"),
+            ("ᵖ ʸ ᵍ", "0"),
         ],
     },
     {
         "name": "Lisu",
         "tag": "LISU",
         "hint_top_to_bottom": False,
-        "std_chars": ['ꓳ'], # ꓳ
+        "std_chars": "ꓳ", # ꓳ
         "base_ranges": [
             (0xA4D0, 0xA4FF), # Lisu
         ],
         "non_base_ranges": [
         ],
         "blues": [
-            (['ꓡ', 'ꓧ', 'ꓱ', 'ꓶ', 'ꓩ', 'ꓚ', 'ꓵ', 'ꓳ'], "TOP"),
-            (['ꓕ', 'ꓜ', 'ꓞ', 'ꓡ', 'ꓛ', 'ꓢ', 'ꓳ', 'ꓴ'], "0"),
+            ("ꓡ ꓧ ꓱ ꓶ ꓩ ꓚ ꓵ ꓳ", "TOP"),
+            ("ꓕ ꓜ ꓞ ꓡ ꓛ ꓢ ꓳ ꓴ", "0"),
         ],
     },
     {
         "name": "Malayalam",
         "tag": "MLYM",
         "hint_top_to_bottom": False,
-        "std_chars": ['ഠ', 'റ'], # ഠ റ
+        "std_chars": "ഠ റ", # ഠ റ
         "base_ranges": [
             (0x0D00, 0x0D7F), # Malayalam
         ],
@@ -836,35 +835,35 @@
             (0x0D62, 0x0D63),
         ],
         "blues": [
-            (['ഒ', 'ട', 'ഠ', 'റ', 'ച', 'പ', 'ച', 'പ'], "TOP"),
-            (['ട', 'ഠ', 'ധ', 'ശ', 'ഘ', 'ച', 'ഥ', 'ല'], "0"),
+            ("ഒ ട ഠ റ ച പ ച്ച പ്പ", "TOP"),
+            ("ട ഠ ധ ശ ഘ ച ഥ ല", "0"),
         ],
     },
     {
         "name": "Medefaidrin",
         "tag": "MEDF",
         "hint_top_to_bottom": False,
-        "std_chars": ['𖹡', '𖹛', '𖹯'], # 𖹡 𖹛 𖹯
+        "std_chars": "𖹡 𖹛 𖹯", # 𖹡 𖹛 𖹯
         "base_ranges": [
             (0x16E40, 0x16E9F), # Medefaidrin
         ],
         "non_base_ranges": [
         ],
         "blues": [
-            (['𖹀', '𖹁', '𖹂', '𖹃', '𖹏', '𖹚', '𖹟'], "TOP"),
-            (['𖹀', '𖹁', '𖹂', '𖹃', '𖹏', '𖹚', '𖹒', '𖹓'], "0"),
-            (['𖹤', '𖹬', '𖹧', '𖹴', '𖹶', '𖹾'], "TOP"),
-            (['𖹠', '𖹡', '𖹢', '𖹹', '𖹳', '𖹮'], "TOP | LATIN_X_HEIGHT"),
-            (['𖹠', '𖹡', '𖹢', '𖹳', '𖹭', '𖹽'], "0"),
-            (['𖹥', '𖹨', '𖹩'], "0"),
-            (['𖺀', '𖺅', '𖺈', '𖺄', '𖺍'], "TOP"),
+            ("𖹀 𖹁 𖹂 𖹃 𖹏 𖹚 𖹟", "TOP"),
+            ("𖹀 𖹁 𖹂 𖹃 𖹏 𖹚 𖹒 𖹓", "0"),
+            ("𖹤 𖹬 𖹧 𖹴 𖹶 𖹾", "TOP"),
+            ("𖹠 𖹡 𖹢 𖹹 𖹳 𖹮", "TOP | LATIN_X_HEIGHT"),
+            ("𖹠 𖹡 𖹢 𖹳 𖹭 𖹽", "0"),
+            ("𖹥 𖹨 𖹩", "0"),
+            ("𖺀 𖺅 𖺈 𖺄 𖺍", "TOP"),
         ],
     },
     {
         "name": "Mongolian",
         "tag": "MONG",
         "hint_top_to_bottom": True,
-        "std_chars": ['ᡂ', 'ᠪ'], # ᡂ ᠪ
+        "std_chars": "ᡂ ᠪ", # ᡂ ᠪ
         "base_ranges": [
             (0x1800, 0x18AF), # Mongolian
             (0x11660, 0x1167F), # Mongolian Supplement
@@ -874,15 +873,15 @@
             (0x18A9, 0x18A9),
         ],
         "blues": [
-            (['ᠳ', 'ᠴ', 'ᠶ', 'ᠽ', 'ᡂ', 'ᡊ', '‍', '‍'], "TOP"),
-            (['ᡃ'], "0"),
+            ("ᠳ ᠴ ᠶ ᠽ ᡂ ᡊ ‍ᡡ‍ ‍ᡳ‍", "TOP"),
+            ("ᡃ", "0"),
         ],
     },
     {
         "name": "Myanmar",
         "tag": "MYMR",
         "hint_top_to_bottom": False,
-        "std_chars": ['ဝ', 'င', 'ဂ'], # ဝ င ဂ
+        "std_chars": "ဝ င ဂ", # ဝ င ဂ
         "base_ranges": [
             (0x1000, 0x109F), # Myanmar
             (0xA9E0, 0xA9FF), # Myanmar Extended-B
@@ -903,17 +902,17 @@
             (0xAA7C, 0xAA7C),
         ],
         "blues": [
-            (['ခ', 'ဂ', 'င', 'ဒ', 'ဝ', 'ၥ', '၊', '။'], "TOP | LATIN_X_HEIGHT"),
-            (['င', 'ဎ', 'ဒ', 'ပ', 'ဗ', 'ဝ', '၊', '။'], "0"),
-            (['ဩ', 'ြ', '၍', '၏', '၆', 'ါ', 'ိ'], "TOP"),
-            (['ဉ', 'ည', 'ဥ', 'ဩ', 'ဨ', '၂', '၅', '၉'], "0"),
+            ("ခ ဂ င ဒ ဝ ၥ ၊ ။", "TOP | LATIN_X_HEIGHT"),
+            ("င ဎ ဒ ပ ဗ ဝ ၊ ။", "0"),
+            ("ဩ ြ ၍ ၏ ၆ ါ ိ", "TOP"),
+            ("ဉ ည ဥ ဩ ဨ ၂ ၅ ၉", "0"),
         ],
     },
     {
         "name": "N'Ko",
         "tag": "NKOO",
         "hint_top_to_bottom": False,
-        "std_chars": ['ߋ', '߀'], # ߋ ߀
+        "std_chars": "ߋ ߀", # ߋ ߀
         "base_ranges": [
             (0x07C0, 0x07FF), # N'Ko
         ],
@@ -922,17 +921,17 @@
             (0x07FD, 0x07FD),
         ],
         "blues": [
-            (['ߐ', '߉', 'ߒ', 'ߟ', 'ߖ', 'ߜ', 'ߠ', 'ߥ'], "TOP"),
-            (['߀', 'ߘ', 'ߡ', 'ߠ', 'ߥ'], "0"),
-            (['ߏ', 'ߛ', 'ߋ'], "TOP | LATIN_X_HEIGHT"),
-            (['ߎ', 'ߏ', 'ߛ', 'ߋ'], "0"),
+            ("ߐ ߉ ߒ ߟ ߖ ߜ ߠ ߥ", "TOP"),
+            ("߀ ߘ ߡ ߠ ߥ", "0"),
+            ("ߏ ߛ ߋ", "TOP | LATIN_X_HEIGHT"),
+            ("ߎ ߏ ߛ ߋ", "0"),
         ],
     },
     {
         "name": "no script",
         "tag": "NONE",
         "hint_top_to_bottom": False,
-        "std_chars": [],
+        "std_chars": "",
         "base_ranges": [
         ],
         "non_base_ranges": [
@@ -944,88 +943,88 @@
         "name": "Ol Chiki",
         "tag": "OLCK",
         "hint_top_to_bottom": False,
-        "std_chars": ['ᱛ'], # ᱛ
+        "std_chars": "ᱛ", # ᱛ
         "base_ranges": [
             (0x1C50, 0x1C7F), # Ol Chiki
         ],
         "non_base_ranges": [
         ],
         "blues": [
-            (['ᱛ', 'ᱜ', 'ᱝ', 'ᱡ', 'ᱢ', 'ᱥ'], "TOP"),
-            (['ᱛ', 'ᱜ', 'ᱝ', 'ᱡ', 'ᱢ', 'ᱥ'], "0"),
+            ("ᱛ ᱜ ᱝ ᱡ ᱢ ᱥ", "TOP"),
+            ("ᱛ ᱜ ᱝ ᱡ ᱢ ᱥ", "0"),
         ],
     },
     {
         "name": "Old Turkic",
         "tag": "ORKH",
         "hint_top_to_bottom": False,
-        "std_chars": ['𐰗'], # 𐰗
+        "std_chars": "𐰗", # 𐰗
         "base_ranges": [
             (0x10C00, 0x10C4F), # Old Turkic
         ],
         "non_base_ranges": [
         ],
         "blues": [
-            (['𐰗', '𐰘', '𐰧'], "TOP"),
-            (['𐰉', '𐰗', '𐰦', '𐰧'], "0"),
+            ("𐰗 𐰘 𐰧", "TOP"),
+            ("𐰉 𐰗 𐰦 𐰧", "0"),
         ],
     },
     {
         "name": "Osage",
         "tag": "OSGE",
         "hint_top_to_bottom": False,
-        "std_chars": ['𐓂', '𐓪'], # 𐓂 𐓪
+        "std_chars": "𐓂 𐓪", # 𐓂 𐓪
         "base_ranges": [
             (0x104B0, 0x104FF), # Osage
         ],
         "non_base_ranges": [
         ],
         "blues": [
-            (['𐒾', '𐓍', '𐓒', '𐓓', '𐒻', '𐓂', '𐒵', '𐓆'], "TOP"),
-            (['𐒰', '𐓍', '𐓂', '𐒿', '𐓎', '𐒹'], "0"),
-            (['𐒼', '𐒽', '𐒾'], "0"),
-            (['𐓵', '𐓶', '𐓺', '𐓻', '𐓝', '𐓣', '𐓪', '𐓮'], "TOP | LATIN_X_HEIGHT"),
-            (['𐓘', '𐓚', '𐓣', '𐓵', '𐓡', '𐓧', '𐓪', '𐓶'], "0"),
-            (['𐓤', '𐓦', '𐓸', '𐓹', '𐓛'], "TOP"),
-            (['𐓤', '𐓥', '𐓦'], "0"),
+            ("𐒾 𐓍 𐓒 𐓓 𐒻 𐓂 𐒵 𐓆", "TOP"),
+            ("𐒰 𐓍 𐓂 𐒿 𐓎 𐒹", "0"),
+            ("𐒼 𐒽 𐒾", "0"),
+            ("𐓵 𐓶 𐓺 𐓻 𐓝 𐓣 𐓪 𐓮", "TOP | LATIN_X_HEIGHT"),
+            ("𐓘 𐓚 𐓣 𐓵 𐓡 𐓧 𐓪 𐓶", "0"),
+            ("𐓤 𐓦 𐓸 𐓹 𐓛", "TOP"),
+            ("𐓤 𐓥 𐓦", "0"),
         ],
     },
     {
         "name": "Osmanya",
         "tag": "OSMA",
         "hint_top_to_bottom": False,
-        "std_chars": ['𐒆', '𐒠'], # 𐒆 𐒠
+        "std_chars": "𐒆 𐒠", # 𐒆 𐒠
         "base_ranges": [
             (0x10480, 0x104AF), # Osmanya
         ],
         "non_base_ranges": [
         ],
         "blues": [
-            (['𐒆', '𐒉', '𐒐', '𐒒', '𐒘', '𐒛', '𐒠', '𐒣'], "TOP"),
-            (['𐒀', '𐒂', '𐒆', '𐒈', '𐒊', '𐒒', '𐒠', '𐒩'], "0"),
+            ("𐒆 𐒉 𐒐 𐒒 𐒘 𐒛 𐒠 𐒣", "TOP"),
+            ("𐒀 𐒂 𐒆 𐒈 𐒊 𐒒 𐒠 𐒩", "0"),
         ],
     },
     {
         "name": "Hanifi Rohingya",
         "tag": "ROHG",
         "hint_top_to_bottom": False,
-        "std_chars": ['𐴰'], # 𐴰
+        "std_chars": "𐴰", # 𐴰
         "base_ranges": [
             (0x10D00, 0x10D3F), # Hanifi Rohingya
         ],
         "non_base_ranges": [
         ],
         "blues": [
-            (['𐴃', '𐴀', '𐴆', '𐴖', '𐴕'], "TOP"),
-            (['𐴔', '𐴖', '𐴕', '𐴑', '𐴐'], "0"),
-            (['ـ'], "LATIN_NEUTRAL"),
+            ("𐴃 𐴀 𐴆 𐴖 𐴕", "TOP"),
+            ("𐴔 𐴖 𐴕 𐴑 𐴐", "0"),
+            ("ـ", "LATIN_NEUTRAL"),
         ],
     },
     {
         "name": "Saurashtra",
         "tag": "SAUR",
         "hint_top_to_bottom": False,
-        "std_chars": ['ꢝ', '꣐'], # ꢝ ꣐
+        "std_chars": "ꢝ ꣐", # ꢝ ꣐
         "base_ranges": [
             (0xA880, 0xA8DF), # Saurashtra
         ],
@@ -1034,33 +1033,33 @@
             (0xA8B4, 0xA8C5),
         ],
         "blues": [
-            (['ꢜ', 'ꢞ', 'ꢳ', 'ꢂ', 'ꢖ', 'ꢒ', 'ꢝ', 'ꢛ'], "TOP"),
-            (['ꢂ', 'ꢨ', 'ꢺ', 'ꢤ', 'ꢎ'], "0"),
+            ("ꢜ ꢞ ꢳ ꢂ ꢖ ꢒ ꢝ ꢛ", "TOP"),
+            ("ꢂ ꢨ ꢺ ꢤ ꢎ", "0"),
         ],
     },
     {
         "name": "Shavian",
         "tag": "SHAW",
         "hint_top_to_bottom": False,
-        "std_chars": ['𐑴'], # 𐑴
+        "std_chars": "𐑴", # 𐑴
         "base_ranges": [
             (0x10450, 0x1047F), # Shavian
         ],
         "non_base_ranges": [
         ],
         "blues": [
-            (['𐑕', '𐑙'], "TOP"),
-            (['𐑔', '𐑖', '𐑗', '𐑹', '𐑻'], "0"),
-            (['𐑟', '𐑣'], "0"),
-            (['𐑱', '𐑲', '𐑳', '𐑴', '𐑸', '𐑺', '𐑼'], "TOP | LATIN_X_HEIGHT"),
-            (['𐑴', '𐑻', '𐑹'], "0"),
+            ("𐑕 𐑙", "TOP"),
+            ("𐑔 𐑖 𐑗 𐑹 𐑻", "0"),
+            ("𐑟 𐑣", "0"),
+            ("𐑱 𐑲 𐑳 𐑴 𐑸 𐑺 𐑼", "TOP | LATIN_X_HEIGHT"),
+            ("𐑴 𐑻 𐑹", "0"),
         ],
     },
     {
         "name": "Sinhala",
         "tag": "SINH",
         "hint_top_to_bottom": False,
-        "std_chars": ['ට'], # ට
+        "std_chars": "ට", # ට
         "base_ranges": [
             (0x0D80, 0x0DFF), # Sinhala
         ],
@@ -1069,16 +1068,16 @@
             (0x0DD2, 0x0DD6),
         ],
         "blues": [
-            (['ඉ', 'ක', 'ඝ', 'ඳ', 'ප', 'ය', 'ල', 'ෆ'], "TOP"),
-            (['එ', 'ඔ', 'ඝ', 'ජ', 'ට', 'ථ', 'ධ', 'ර'], "0"),
-            (['ද', 'ඳ', 'උ', 'ල', 'ත', 'ත', 'බ', 'ද'], "0"),
+            ("ඉ ක ඝ ඳ ප ය ල ෆ", "TOP"),
+            ("එ ඔ ඝ ජ ට ථ ධ ර", "0"),
+            ("ද ඳ උ ල තූ තු බු දු", "0"),
         ],
     },
     {
         "name": "Sundanese",
         "tag": "SUND",
         "hint_top_to_bottom": False,
-        "std_chars": ['᮰'], # ᮰
+        "std_chars": "᮰", # ᮰
         "base_ranges": [
             (0x1B80, 0x1BBF), # Sundanese
             (0x1CC0, 0x1CCF), # Sundanese Supplement
@@ -1088,16 +1087,16 @@
             (0x1BA1, 0x1BAD),
         ],
         "blues": [
-            (['ᮋ', 'ᮞ', 'ᮮ', 'ᮽ', '᮰', 'ᮈ'], "TOP"),
-            (['ᮄ', 'ᮔ', 'ᮕ', 'ᮗ', '᮰', 'ᮆ', 'ᮈ', 'ᮉ'], "0"),
-            (['ᮼ', '᳄'], "0"),
+            ("ᮋ ᮞ ᮮ ᮽ ᮰ ᮈ", "TOP"),
+            ("ᮄ ᮔ ᮕ ᮗ ᮰ ᮆ ᮈ ᮉ", "0"),
+            ("ᮼ ᳄", "0"),
         ],
     },
     {
         "name": "Tamil",
         "tag": "TAML",
         "hint_top_to_bottom": False,
-        "std_chars": ['௦'], # ௦
+        "std_chars": "௦", # ௦
         "base_ranges": [
             (0x0B80, 0x0BFF), # Tamil
         ],
@@ -1107,15 +1106,15 @@
             (0x0BCD, 0x0BCD),
         ],
         "blues": [
-            (['உ', 'ஒ', 'ஓ', 'ற', 'ஈ', 'க', 'ங', 'ச'], "TOP"),
-            (['க', 'ச', 'ல', 'ஶ', 'உ', 'ங', 'ட', 'ப'], "0"),
+            ("உ ஒ ஓ ற ஈ க ங ச", "TOP"),
+            ("க ச ல ஶ உ ங ட ப", "0"),
         ],
     },
     {
         "name": "Tai Viet",
         "tag": "TAVT",
         "hint_top_to_bottom": False,
-        "std_chars": ['ꪒ', 'ꪫ'], # ꪒ ꪫ
+        "std_chars": "ꪒ ꪫ", # ꪒ ꪫ
         "base_ranges": [
             (0xAA80, 0xAADF), # Tai Viet
         ],
@@ -1127,15 +1126,15 @@
             (0xAAC1, 0xAAC1),
         ],
         "blues": [
-            (['ꪆ', 'ꪔ', 'ꪒ', 'ꪖ', 'ꪫ'], "TOP"),
-            (['ꪉ', 'ꪫ', 'ꪮ'], "0"),
+            ("ꪆ ꪔ ꪒ ꪖ ꪫ", "TOP"),
+            ("ꪉ ꪫ ꪮ", "0"),
         ],
     },
     {
         "name": "Telugu",
         "tag": "TELU",
         "hint_top_to_bottom": False,
-        "std_chars": ['౦', '౧'], # ౦ ౧
+        "std_chars": "౦ ౧", # ౦ ౧
         "base_ranges": [
             (0x0C00, 0x0C7F), # Telugu
         ],
@@ -1147,30 +1146,30 @@
             (0x0C62, 0x0C63),
         ],
         "blues": [
-            (['ఇ', 'ఌ', 'ఙ', 'ఞ', 'ణ', 'ఱ', '౯'], "TOP"),
-            (['అ', 'క', 'చ', 'ర', 'ఽ', '౨', '౬'], "0"),
+            ("ఇ ఌ ఙ ఞ ణ ఱ ౯", "TOP"),
+            ("అ క చ ర ఽ ౨ ౬", "0"),
         ],
     },
     {
         "name": "Tifinagh",
         "tag": "TFNG",
         "hint_top_to_bottom": False,
-        "std_chars": ['ⵔ'], # ⵔ
+        "std_chars": "ⵔ", # ⵔ
         "base_ranges": [
             (0x2D30, 0x2D7F), # Tifinagh
         ],
         "non_base_ranges": [
         ],
         "blues": [
-            (['ⵔ', 'ⵙ', 'ⵛ', 'ⵞ', 'ⴵ', 'ⴼ', 'ⴹ', 'ⵎ'], "TOP"),
-            (['ⵔ', 'ⵙ', 'ⵛ', 'ⵞ', 'ⴵ', 'ⴼ', 'ⴹ', 'ⵎ'], "0"),
+            ("ⵔ ⵙ ⵛ ⵞ ⴵ ⴼ ⴹ ⵎ", "TOP"),
+            ("ⵔ ⵙ ⵛ ⵞ ⴵ ⴼ ⴹ ⵎ", "0"),
         ],
     },
     {
         "name": "Thai",
         "tag": "THAI",
         "hint_top_to_bottom": False,
-        "std_chars": ['า', 'ๅ', '๐'], # า ๅ ๐
+        "std_chars": "า ๅ ๐", # า ๅ ๐
         "base_ranges": [
             (0x0E00, 0x0E7F), # Thai
         ],
@@ -1180,35 +1179,35 @@
             (0x0E47, 0x0E4E),
         ],
         "blues": [
-            (['บ', 'เ', 'แ', 'อ', 'ก', 'า'], "TOP | LATIN_X_HEIGHT"),
-            (['บ', 'ป', 'ษ', 'ฯ', 'อ', 'ย', 'ฮ'], "0"),
-            (['ป', 'ฝ', 'ฟ'], "TOP"),
-            (['โ', 'ใ', 'ไ'], "TOP"),
-            (['ฎ', 'ฏ', 'ฤ', 'ฦ'], "0"),
-            (['ญ', 'ฐ'], "0"),
-            (['๐', '๑', '๓'], "0"),
+            ("บ เ แ อ ก า", "TOP | LATIN_X_HEIGHT"),
+            ("บ ป ษ ฯ อ ย ฮ", "0"),
+            ("ป ฝ ฟ", "TOP"),
+            ("โ ใ ไ", "TOP"),
+            ("ฎ ฏ ฤ ฦ", "0"),
+            ("ญ ฐ", "0"),
+            ("๐ ๑ ๓", "0"),
         ],
     },
     {
         "name": "Vai",
         "tag": "VAII",
         "hint_top_to_bottom": False,
-        "std_chars": ['ꘓ', 'ꖜ', 'ꖴ'], # ꘓ ꖜ ꖴ
+        "std_chars": "ꘓ ꖜ ꖴ", # ꘓ ꖜ ꖴ
         "base_ranges": [
             (0xA500, 0xA63F), # Vai
         ],
         "non_base_ranges": [
         ],
         "blues": [
-            (['ꗍ', 'ꘖ', 'ꘙ', 'ꘜ', 'ꖜ', 'ꖝ', 'ꔅ', 'ꕢ'], "TOP"),
-            (['ꗍ', 'ꘖ', 'ꘙ', 'ꗞ', 'ꔅ', 'ꕢ', 'ꖜ', 'ꔆ'], "0"),
+            ("ꗍ ꘖ ꘙ ꘜ ꖜ ꖝ ꔅ ꕢ", "TOP"),
+            ("ꗍ ꘖ ꘙ ꗞ ꔅ ꕢ ꖜ ꔆ", "0"),
         ],
     },
     {
         "name": "Limbu",
         "tag": "LIMB",
         "hint_top_to_bottom": False,
-        "std_chars": ['o'], # XXX
+        "std_chars": "o", # XXX
         "base_ranges": [
             (0x1900, 0x194F), # Limbu
         ],
@@ -1223,7 +1222,7 @@
         "name": "Oriya",
         "tag": "ORYA",
         "hint_top_to_bottom": False,
-        "std_chars": ['o'], # XXX
+        "std_chars": "o", # XXX
         "base_ranges": [
             (0x0B00, 0x0B7F), # Oriya
         ],
@@ -1241,7 +1240,7 @@
         "name": "Syloti Nagri",
         "tag": "SYLO",
         "hint_top_to_bottom": False,
-        "std_chars": ['o'], # XXX
+        "std_chars": "o", # XXX
         "base_ranges": [
             (0xA800, 0xA82F), # Syloti Nagri
         ],
@@ -1257,7 +1256,7 @@
         "name": "Tibetan",
         "tag": "TIBT",
         "hint_top_to_bottom": False,
-        "std_chars": ['o'], # XXX
+        "std_chars": "o", # XXX
         "base_ranges": [
             (0x0F00, 0x0FFF), # Tibetan
         ],
@@ -1278,7 +1277,7 @@
         "name": "CJKV ideographs",
         "tag": "HANI",
         "hint_top_to_bottom": False,
-        "std_chars": ['田', '囗'], # 田 囗
+        "std_chars": "田 囗", # 田 囗
         "base_ranges": [
             (0x1100, 0x11FF), # Hangul Jamo
             (0x2E80, 0x2EFF), # CJK Radicals Supplement
@@ -1319,10 +1318,10 @@
             (0x3190, 0x319F),
         ],
         "blues": [
-            (['他', '们', '你', '來', '們', '到', '和', '地', '对', '對', '就', '席', '我', '时', '時', '會', '来', '為', '能', '舰', '說', '说', '这', '這', '齊', '|', '军', '同', '已', '愿', '既', '星', '是', '景', '民', '照', '现', '現', '理', '用', '置', '要', '軍', '那', '配', '里', '開', '雷', '露', '面', '顾'], "TOP"),
-            (['个', '为', '人', '他', '以', '们', '你', '來', '個', '們', '到', '和', '大', '对', '對', '就', '我', '时', '時', '有', '来', '為', '要', '說', '说', '|', '主', '些', '因', '它', '想', '意', '理', '生', '當', '看', '着', '置', '者', '自', '著', '裡', '过', '还', '进', '進', '過', '道', '還', '里', '面'], "0"),
-            (['些', '们', '你', '來', '們', '到', '和', '地', '她', '将', '將', '就', '年', '得', '情', '最', '样', '樣', '理', '能', '說', '说', '这', '這', '通', '|', '即', '吗', '吧', '听', '呢', '品', '响', '嗎', '师', '師', '收', '断', '斷', '明', '眼', '間', '间', '际', '陈', '限', '除', '陳', '随', '際', '隨'], "CJK_HORIZ"),
-            (['事', '前', '學', '将', '將', '情', '想', '或', '政', '斯', '新', '样', '樣', '民', '沒', '没', '然', '特', '现', '現', '球', '第', '經', '谁', '起', '|', '例', '別', '别', '制', '动', '動', '吗', '嗎', '增', '指', '明', '朝', '期', '构', '物', '确', '种', '調', '调', '費', '费', '那', '都', '間', '间'], "CJK_HORIZ | CJK_RIGHT"),
+            ("他 们 你 來 們 到 和 地 对 對 就 席 我 时 時 會 来 為 能 舰 說 说 这 這 齊 | 军 同 已 愿 既 星 是 景 民 照 现 現 理 用 置 要 軍 那 配 里 開 雷 露 面 顾", "TOP"),
+            ("个 为 人 他 以 们 你 來 個 們 到 和 大 对 對 就 我 时 時 有 来 為 要 說 说 | 主 些 因 它 想 意 理 生 當 看 着 置 者 自 著 裡 过 还 进 進 過 道 還 里 面", "0"),
+            (" 些 们 你 來 們 到 和 地 她 将 將 就 年 得 情 最 样 樣 理 能 說 说 这 這 通 | 即 吗 吧 听 呢 品 响 嗎 师 師 收 断 斷 明 眼 間 间 际 陈 限 除 陳 随 際 隨", "CJK_HORIZ"),
+            ("事 前 學 将 將 情 想 或 政 斯 新 样 樣 民 沒 没 然 特 现 現 球 第 經 谁 起 | 例 別 别 制 动 動 吗 嗎 增 指 明 朝 期 构 物 确 种 調 调 費 费 那 都 間 间", "CJK_HORIZ | CJK_RIGHT"),
         ],
     },
 ]
@@ -1356,23 +1355,16 @@ def generate() -> str:
         buf += "        name: \"{}\",\n".format(script["name"])
         buf += "        group: ScriptGroup::{},\n".format(group)
         buf += "        tag: Tag::new(b\"{}\"),\n".format(tag)
-        buf += "        index: {},\n".format(i)
         buf += "        hint_top_to_bottom: {},\n".format(str(script["hint_top_to_bottom"]).lower())
         # standard characters
-        buf += "        std_chars: &["
-        if len(std_chars) != 0:
-            for std_char in std_chars:
-                buf += "'{}', ".format(std_char)
-        buf += "],\n"
+        buf += "        std_chars: \"{}\",\n".format(script["std_chars"])
         # blue characters
         buf += "        blues: &["
         if len(blues) != 0:
             buf += "\n";
             for blue in blues:
-                buf += "            (&["
-                for ch in blue[0]:
-                    buf += "'{}', ".format(ch)
-                buf += "], {}),\n".format(blue[1])
+                buf += "            (\"" + blue[0] + "\""
+                buf += ", {}),\n".format(blue[1])
             buf += "        ],\n"
         else:
             buf += "],\n"
@@ -1403,6 +1395,7 @@ def generate() -> str:
 
     # Add some symbolic indices for each script so they can be
     # referenced by ScriptClass::LATN for example
+    buf += "#[allow(unused)]"
     buf += "impl ScriptClass {\n"
     for i, script in enumerate(SCRIPT_CLASSES):
         buf += "    pub const {}: usize = {};\n".format(script["tag"], i)
@@ -1430,6 +1423,7 @@ def generate() -> str:
     buf += "];\n\n";
 
     # Symbolic indices for style classes
+    buf += "#[allow(unused)]"
     buf += "impl StyleClass {\n"
     for (i, tag) in enumerate(style_class_tags):
         buf += "    pub const {}: usize = {};\n".format(tag, i)
diff --git a/skrifa/src/outline/autohint/axis.rs b/skrifa/src/outline/autohint/axis.rs
index 5d53e9217..6f5922539 100644
--- a/skrifa/src/outline/autohint/axis.rs
+++ b/skrifa/src/outline/autohint/axis.rs
@@ -40,6 +40,7 @@ impl Axis {
 }
 
 impl Axis {
+    #[cfg(test)]
     pub fn new(dim: Dimension, orientation: Option<Orientation>) -> Self {
         let mut axis = Self::default();
         axis.reset(dim, orientation);
@@ -164,10 +165,6 @@ impl Segment {
         &points[self.first()]
     }
 
-    pub fn first_point_mut<'a>(&self, points: &'a mut [Point]) -> &'a mut Point {
-        &mut points[self.first()]
-    }
-
     pub fn last(&self) -> usize {
         self.last_ix as usize
     }
@@ -176,10 +173,6 @@ impl Segment {
         &points[self.last()]
     }
 
-    pub fn last_point_mut<'a>(&self, points: &'a mut [Point]) -> &'a mut Point {
-        &mut points[self.last()]
-    }
-
     pub fn edge<'a>(&self, edges: &'a [Edge]) -> Option<&'a Edge> {
         edges.get(self.edge_ix.map(|ix| ix as usize)?)
     }
@@ -237,10 +230,6 @@ impl Edge {
 }
 
 impl Edge {
-    pub fn first_segment<'a>(&self, segments: &'a [Segment]) -> Option<&'a Segment> {
-        segments.get(self.first_ix as usize)
-    }
-
     pub fn link<'a>(&self, edges: &'a [Edge]) -> Option<&'a Edge> {
         edges.get(self.link_ix.map(|ix| ix as usize)?)
     }
diff --git a/skrifa/src/outline/autohint/hint.rs b/skrifa/src/outline/autohint/hint.rs
index f61c379e0..b151b3d5a 100644
--- a/skrifa/src/outline/autohint/hint.rs
+++ b/skrifa/src/outline/autohint/hint.rs
@@ -117,23 +117,45 @@ pub(crate) fn align_strong_points(outline: &mut Outline, axis: &mut Axis) -> Opt
             store_point(point, dim, edge.pos + (ou - edge.opos));
             continue;
         }
-        // Find enclosing edges
-        let mut min_ix = 0;
-        let mut max_ix = edges.len();
-        while min_ix < max_ix {
-            let mid_ix = (min_ix + max_ix) >> 1;
-            let edge = &edges[mid_ix];
-            let fpos = edge.fpos as i32;
-            match u.cmp(&fpos) {
-                Ordering::Less => max_ix = mid_ix,
-                Ordering::Greater => min_ix = mid_ix + 1,
-                Ordering::Equal => {
-                    // We are on an edge
+        // Find enclosing edges; for a small number of edges, use a linear
+        // search.
+        // Note: this is actually critical for matching FreeType in cases where
+        // we have more than one edge with the same fpos. When this happens,
+        // linear and binary searches can produce different results.
+        // See <https://gitlab.freedesktop.org/freetype/freetype/-/blob/57617782464411201ce7bbc93b086c1b4d7d84a5/src/autofit/afhints.c#L1489>
+        let min_ix = if edges.len() <= 8 {
+            if let Some((min_ix, edge)) = edges
+                .iter()
+                .enumerate()
+                .find(|(_ix, edge)| edge.fpos as i32 >= u)
+            {
+                if edge.fpos as i32 == u {
                     store_point(point, dim, edge.pos);
                     continue 'points;
                 }
+                min_ix
+            } else {
+                0
             }
-        }
+        } else {
+            let mut min_ix = 0;
+            let mut max_ix = edges.len();
+            while min_ix < max_ix {
+                let mid_ix = (min_ix + max_ix) >> 1;
+                let edge = &edges[mid_ix];
+                let fpos = edge.fpos as i32;
+                match u.cmp(&fpos) {
+                    Ordering::Less => max_ix = mid_ix,
+                    Ordering::Greater => min_ix = mid_ix + 1,
+                    Ordering::Equal => {
+                        // We are on an edge
+                        store_point(point, dim, edge.pos);
+                        continue 'points;
+                    }
+                }
+            }
+            min_ix
+        };
         // Point is not on an edge
         if let Some(before_ix) = min_ix.checked_sub(1) {
             let edge_before = edges.get(before_ix)?;
@@ -628,7 +650,7 @@ mod tests {
             Default::default(),
             metrics.style_class().script.group,
         );
-        let hinted_metrics = latin::hint_outline(&mut outline, &metrics, &scale);
+        let hinted_metrics = latin::hint_outline(&mut outline, &metrics, &scale, None);
         (outline, hinted_metrics)
     }
 }
diff --git a/skrifa/src/outline/autohint/instance.rs b/skrifa/src/outline/autohint/instance.rs
index 997476e01..e1231585a 100644
--- a/skrifa/src/outline/autohint/instance.rs
+++ b/skrifa/src/outline/autohint/instance.rs
@@ -5,7 +5,7 @@ use crate::{attribute::Style, prelude::Size, MetadataProvider};
 use super::{
     super::{
         pen::PathStyle, AdjustedMetrics, DrawError, OutlineGlyph, OutlineGlyphCollection,
-        OutlinePen, SmoothMode, Target,
+        OutlinePen, Target,
     },
     metrics::{fixed_mul, pix_round, Scale, UnscaledStyleMetricsSet},
     outline::Outline,
@@ -113,7 +113,8 @@ impl Instance {
         );
         let mut outline = Outline::default();
         outline.fill(glyph, coords)?;
-        let hinted_metrics = super::latin::hint_outline(&mut outline, &metrics, &scale);
+        let hinted_metrics =
+            super::latin::hint_outline(&mut outline, &metrics, &scale, Some(style));
         let h_advance = common.advance_width(glyph_id, coords);
         let mut pp1x = 0;
         let mut pp2x = fixed_mul(h_advance, hinted_metrics.x_scale);
diff --git a/skrifa/src/outline/autohint/latin/blues.rs b/skrifa/src/outline/autohint/latin/blues.rs
index f94e8a6eb..7ffad1733 100644
--- a/skrifa/src/outline/autohint/latin/blues.rs
+++ b/skrifa/src/outline/autohint/latin/blues.rs
@@ -4,7 +4,7 @@ use super::super::{
     super::{unscaled::UnscaledOutlineBuf, OutlineGlyphCollection},
     cycling::{cycle_backward, cycle_forward},
     metrics::{UnscaledBlue, UnscaledBlues, MAX_BLUES},
-    style::{blue_flags, ScriptClass, ScriptGroup, StyleClass},
+    style::{blue_flags, ScriptGroup, StyleClass},
 };
 use crate::{charmap::Charmap, FontRef, MetadataProvider};
 use raw::types::F2Dot14;
@@ -50,7 +50,7 @@ fn compute_default_blues(font: &FontRef, coords: &[F2Dot14], style: &StyleClass)
     let (glyphs, charmap, units_per_em) = things_all_blues_need(font);
     let flat_threshold = units_per_em / 14;
     // Walk over each of the blue character sets for our script.
-    for (blue_chars, blue_flags) in style.script.blues {
+    for (blue_str, blue_flags) in style.script.blues {
         let is_top_like = (blue_flags & (blue_flags::TOP | blue_flags::LATIN_SUB_TOP)) != 0;
         let is_top = blue_flags & blue_flags::TOP != 0;
         let is_x_height = blue_flags & blue_flags::LATIN_X_HEIGHT != 0;
@@ -60,10 +60,19 @@ fn compute_default_blues(font: &FontRef, coords: &[F2Dot14], style: &StyleClass)
         let mut descender = i16::MAX;
         let mut n_flats = 0;
         let mut n_rounds = 0;
-        for ch in *blue_chars {
+        for cluster in blue_str.split(' ') {
             // TODO shaping: https://github.com/googlefonts/fontations/issues/1128
             let y_offset = 0;
-            let Some(gid) = charmap.map(*ch) else {
+            let mut cluster_chars = cluster.chars();
+            let Some(ch) = cluster_chars.next() else {
+                continue;
+            };
+            // Without shaping, we need to skip multi-character clusters
+            // See <https://gitlab.freedesktop.org/freetype/freetype/-/blob/57617782464411201ce7bbc93b086c1b4d7d84a5/src/autofit/afshaper.c#L639>
+            if cluster_chars.next().is_some() {
+                continue;
+            }
+            let Some(gid) = charmap.map(ch) else {
                 continue;
             };
             if gid.to_u32() == 0 {
@@ -182,8 +191,9 @@ fn compute_default_blues(font: &FontRef, coords: &[F2Dot14], style: &StyleClass)
                 // See <https://gitlab.freedesktop.org/freetype/freetype/-/blob/57617782464411201ce7bbc93b086c1b4d7d84a5/src/autofit/aflatin.c#L641>
                 // heuristic threshold value
                 let length_threshold = units_per_em / 25;
-                let dist =
-                    best_contour[segment_last].x as i32 - best_contour[segment_first].x as i32;
+                let dist = (best_contour[segment_last].x as i32
+                    - best_contour[segment_first].x as i32)
+                    .abs();
                 if dist < length_threshold && segment_last - segment_first + 2 <= best_contour.len()
                 {
                     // heuristic threshold value
@@ -237,6 +247,9 @@ fn compute_default_blues(font: &FontRef, coords: &[F2Dot14], style: &StyleClass)
                                 <= 20 * dist
                         {
                             hit = false;
+                            if last == segment_first {
+                                break;
+                            }
                             continue;
                         }
                         if best_contour[last].is_on_curve() {
@@ -442,7 +455,7 @@ fn compute_cjk_blues(font: &FontRef, coords: &[F2Dot14], style: &StyleClass) ->
     let (mut outline_buf, mut flats, mut fills) = buffers();
     let (glyphs, charmap, _) = things_all_blues_need(font);
     // Walk over each of the blue character sets for our script.
-    for (blue_chars, blue_flags) in style.script.blues {
+    for (blue_str, blue_flags) in style.script.blues {
         let is_horizontal = blue_flags & blue_flags::CJK_HORIZ != 0;
         // Note: horizontal blue zones are disabled by default and have been
         // for many years in FreeType:
@@ -460,16 +473,25 @@ fn compute_cjk_blues(font: &FontRef, coords: &[F2Dot14], style: &StyleClass) ->
         let mut n_flats = 0;
         let mut n_fills = 0;
         let mut is_fill = true;
-        for ch in *blue_chars {
+        for cluster in blue_str.split(' ') {
             // The '|' character is used as a sentinel in the blue string that
             // signifies a switch to characters that define "flat" values
             // <https://gitlab.freedesktop.org/freetype/freetype/-/blob/57617782464411201ce7bbc93b086c1b4d7d84a5/src/autofit/afcjk.c#L380>
-            if *ch == '|' {
+            if cluster == "|" {
                 is_fill = false;
                 continue;
             }
             // TODO shaping: https://github.com/googlefonts/fontations/issues/1128
-            let Some(gid) = charmap.map(*ch) else {
+            let mut cluster_chars = cluster.chars();
+            let Some(ch) = cluster_chars.next() else {
+                continue;
+            };
+            // Without shaping, we need to skip multi-character clusters
+            // See <https://gitlab.freedesktop.org/freetype/freetype/-/blob/57617782464411201ce7bbc93b086c1b4d7d84a5/src/autofit/afshaper.c#L639>
+            if cluster_chars.next().is_some() {
+                continue;
+            }
+            let Some(gid) = charmap.map(ch) else {
                 continue;
             };
             if gid.to_u32() == 0 {
diff --git a/skrifa/src/outline/autohint/latin/edges.rs b/skrifa/src/outline/autohint/latin/edges.rs
index c79e6611a..15d12f765 100644
--- a/skrifa/src/outline/autohint/latin/edges.rs
+++ b/skrifa/src/outline/autohint/latin/edges.rs
@@ -116,7 +116,7 @@ pub(crate) fn compute_edges(
                             break;
                         }
                     }
-                    if dist2 > edge_distance_threshold {
+                    if dist2 >= edge_distance_threshold {
                         continue;
                     }
                 }
@@ -207,6 +207,9 @@ fn compute_edge_properties(axis: &mut Axis) {
         let mut segment_ix = edge.first_ix as usize;
         let last_segment_ix = edge.last_ix as usize;
         loop {
+            // This loop can modify the current edge, so make sure we
+            // reload it here
+            let edge = edges[edge_ix];
             let segment = &segments[segment_ix];
             let next_segment_ix = segment.edge_next_ix;
             // Check roundness
diff --git a/skrifa/src/outline/autohint/latin/hint.rs b/skrifa/src/outline/autohint/latin/hint.rs
index 30e92e3ad..96893c777 100644
--- a/skrifa/src/outline/autohint/latin/hint.rs
+++ b/skrifa/src/outline/autohint/latin/hint.rs
@@ -5,7 +5,7 @@
 //! the original outline points.
 
 use super::super::{
-    axis::{Axis, Dimension, Edge},
+    axis::{Axis, Edge},
     metrics::{fixed_mul_div, pix_floor, pix_round, Scale, ScaledAxisMetrics, ScaledWidth},
     style::ScriptGroup,
 };
@@ -39,7 +39,7 @@ pub(crate) fn hint_edges(
     let edges = axis.edges.as_mut_slice();
     // Special case for lowercase m
     if axis.dim == Axis::HORIZONTAL && (edges.len() == 6 || edges.len() == 12) {
-        hint_lowercase_m(edges);
+        hint_lowercase_m(edges, group);
     }
     // Handle serifs and single segment edges
     if serif_count > 0 || anchor_ix.is_none() {
@@ -277,7 +277,7 @@ fn align_stem_edges(
 /// Make sure that lowercase m's maintain symmetry.
 ///
 /// See <https://gitlab.freedesktop.org/freetype/freetype/-/blob/57617782464411201ce7bbc93b086c1b4d7d84a5/src/autofit/aflatin.c#L3365>
-fn hint_lowercase_m(edges: &mut [Edge]) {
+fn hint_lowercase_m(edges: &mut [Edge], group: ScriptGroup) {
     let (edge1_ix, edge2_ix, edge3_ix) = if edges.len() == 6 {
         (0, 2, 4)
     } else {
@@ -289,6 +289,15 @@ fn hint_lowercase_m(edges: &mut [Edge]) {
     let dist1 = edge2.opos - edge1.opos;
     let dist2 = edge3.opos - edge2.opos;
     let span = (dist1 - dist2).abs();
+    if group != ScriptGroup::Default {
+        // CJK has additional conditions on the following...
+        // See <https://gitlab.freedesktop.org/freetype/freetype/-/blob/57617782464411201ce7bbc93b086c1b4d7d84a5/src/autofit/afcjk.c#L2090>
+        for (edge, ix) in [(edge1, edge1_ix), (edge2, edge2_ix), (edge3, edge3_ix)] {
+            if edge.link_ix != Some((ix + 1) as u16) {
+                return;
+            }
+        }
+    }
     if span < 8 {
         let delta = edge3.pos - (2 * edge2.pos - edge1.pos);
         let link_ix = edge3.link_ix.map(|ix| ix as usize);
@@ -317,7 +326,7 @@ fn align_remaining_edges(
     mut anchor_ix: Option<usize>,
 ) {
     if group == ScriptGroup::Default {
-        /// See <https://gitlab.freedesktop.org/freetype/freetype/-/blob/57617782464411201ce7bbc93b086c1b4d7d84a5/src/autofit/aflatin.c#L3418>
+        // See <https://gitlab.freedesktop.org/freetype/freetype/-/blob/57617782464411201ce7bbc93b086c1b4d7d84a5/src/autofit/aflatin.c#L3418>
         for edge_ix in 0..axis.edges.len() {
             let edges = &mut axis.edges;
             let edge = &edges[edge_ix];
@@ -576,25 +585,26 @@ fn stem_width(
                     }
                     dist = (dist - new_base_delta.abs() + 32) & !63;
                 }
-            } else {
-                // Divergent CJK behavior
-                // See <https://gitlab.freedesktop.org/freetype/freetype/-/blob/57617782464411201ce7bbc93b086c1b4d7d84a5/src/autofit/afcjk.c#L1544>
-                if dist < 54 {
-                    dist += (54 - dist) / 2;
-                } else if dist < 3 * 64 {
-                    let delta = dist & 63;
-                    dist &= -64;
-                    if delta < 10 {
-                        dist += delta;
-                    } else if delta < 22 {
-                        dist += 10;
-                    } else if delta < 42 {
-                        dist += delta;
-                    } else if delta < 54 {
-                        dist += 54;
-                    } else {
-                        dist += delta;
-                    }
+            }
+        }
+        if group != ScriptGroup::Default {
+            // Divergent CJK behavior
+            // See <https://gitlab.freedesktop.org/freetype/freetype/-/blob/57617782464411201ce7bbc93b086c1b4d7d84a5/src/autofit/afcjk.c#L1544>
+            if dist < 54 {
+                dist += (54 - dist) / 2;
+            } else if dist < 3 * 64 {
+                let delta = dist & 63;
+                dist &= -64;
+                if delta < 10 {
+                    dist += delta;
+                } else if delta < 22 {
+                    dist += 10;
+                } else if delta < 42 {
+                    dist += delta;
+                } else if delta < 54 {
+                    dist += 54;
+                } else {
+                    dist += delta;
                 }
             }
         }
diff --git a/skrifa/src/outline/autohint/latin/metrics.rs b/skrifa/src/outline/autohint/latin/metrics.rs
index 43aa8379f..c53beffe5 100644
--- a/skrifa/src/outline/autohint/latin/metrics.rs
+++ b/skrifa/src/outline/autohint/latin/metrics.rs
@@ -10,10 +10,10 @@ use super::super::{
     axis::{Axis, Dimension},
     metrics::{
         fixed_div, fixed_mul, fixed_mul_div, pix_round, Scale, ScaledAxisMetrics, ScaledBlue,
-        ScaledStyleMetrics, ScaledWidth, UnscaledAxisMetrics, UnscaledBlue, UnscaledBlues,
-        UnscaledStyleMetrics, WidthMetrics,
+        ScaledStyleMetrics, ScaledWidth, UnscaledAxisMetrics, UnscaledBlue, UnscaledStyleMetrics,
+        WidthMetrics,
     },
-    style::{blue_flags, ScriptClass, ScriptGroup, StyleClass},
+    style::{blue_flags, ScriptGroup, StyleClass},
 };
 use crate::{prelude::Size, MetadataProvider};
 use raw::{types::F2Dot14, FontRef};
@@ -26,9 +26,28 @@ pub(crate) fn compute_unscaled_style_metrics(
     coords: &[F2Dot14],
     style: &StyleClass,
 ) -> UnscaledStyleMetrics {
+    let charmap = font.charmap();
+    // We don't attempt to produce any metrics if we don't have a Unicode
+    // cmap
+    // See <https://gitlab.freedesktop.org/freetype/freetype/-/blob/57617782464411201ce7bbc93b086c1b4d7d84a5/src/autofit/aflatin.c#L1146>
+    if charmap.is_symbol() {
+        return UnscaledStyleMetrics {
+            class_ix: style.index as u16,
+            axes: [
+                UnscaledAxisMetrics {
+                    dim: Axis::HORIZONTAL,
+                    ..Default::default()
+                },
+                UnscaledAxisMetrics {
+                    dim: Axis::VERTICAL,
+                    ..Default::default()
+                },
+            ],
+            ..Default::default()
+        };
+    }
     let [hwidths, vwidths] = super::widths::compute_widths(font, coords, style.script);
     let [hblues, vblues] = super::blues::compute_unscaled_blues(font, coords, style);
-    let charmap = font.charmap();
     let glyph_metrics = font.glyph_metrics(Size::unscaled(), coords);
     let mut digit_advance = None;
     let mut digits_have_same_width = true;
@@ -89,11 +108,7 @@ pub(crate) fn scale_style_metrics(
         scale_axis(&unscaled_metrics.axes[0]),
         scale_axis(&unscaled_metrics.axes[1]),
     ];
-    ScaledStyleMetrics {
-        scale,
-        group: unscaled_metrics.style_class().script.group,
-        axes,
-    }
+    ScaledStyleMetrics { scale, axes }
 }
 
 /// Computes scaled metrics for a single axis.
@@ -132,7 +147,7 @@ fn scale_default_axis_metrics(
             for blue in blues {
                 max_height = max_height.max(blue.ascender).max(-blue.descender);
             }
-            let mut dist = fixed_mul(max_height, new_scale - axis.scale);
+            let mut dist = fixed_mul(max_height, new_scale - axis.scale).abs();
             dist &= !127;
             if dist == 0 {
                 axis.scale = new_scale;
@@ -257,7 +272,7 @@ fn scale_cjk_axis_metrics(
         };
         // A blue zone is only active if it is less than 3/4 pixels tall
         let dist = fixed_mul(unscaled_blue.position - unscaled_blue.overshoot, scale);
-        if dist <= 48 || dist >= -48 {
+        if (-48..=48).contains(&dist) {
             blue.position.fitted = pix_round(blue.position.scaled);
             // For CJK, "overshoot" is actually undershoot
             let delta1 = fixed_div(blue.position.fitted, scale) - unscaled_blue.overshoot;
diff --git a/skrifa/src/outline/autohint/latin/mod.rs b/skrifa/src/outline/autohint/latin/mod.rs
index 9bc2d6a13..1ec8ad1f6 100644
--- a/skrifa/src/outline/autohint/latin/mod.rs
+++ b/skrifa/src/outline/autohint/latin/mod.rs
@@ -11,6 +11,7 @@ use super::{
     axis::Axis,
     metrics::{Scale, UnscaledStyleMetrics},
     outline::Outline,
+    style::{GlyphStyle, ScriptGroup},
 };
 
 pub(crate) use metrics::compute_unscaled_style_metrics;
@@ -49,14 +50,24 @@ pub(crate) fn hint_outline(
     outline: &mut Outline,
     metrics: &UnscaledStyleMetrics,
     scale: &Scale,
+    glyph_style: Option<GlyphStyle>,
 ) -> HintedMetrics {
     let scaled_metrics = metrics::scale_style_metrics(metrics, *scale);
-    let y_scale = scaled_metrics.axes[1].scale;
+    let scale = &scaled_metrics.scale;
     let mut axis = Axis::default();
     let hint_top_to_bottom = metrics.style_class().script.hint_top_to_bottom;
     outline.scale(&scaled_metrics.scale);
-    let mut hinted_metrics = HintedMetrics::default();
+    let mut hinted_metrics = HintedMetrics {
+        x_scale: scale.x_scale,
+        ..Default::default()
+    };
     let group = metrics.style_class().script.group;
+    // For default script group, we don't proceed with hinting if we're
+    // missing alignment zones. FreeType swaps in a "dummy" hinter here
+    // See <https://gitlab.freedesktop.org/freetype/freetype/-/blob/57617782464411201ce7bbc93b086c1b4d7d84a5/src/autofit/afglobal.c#L475>
+    if group == ScriptGroup::Default && scaled_metrics.axes[1].blues.is_empty() {
+        return hinted_metrics;
+    }
     for dim in 0..2 {
         if (dim == Axis::HORIZONTAL && scale.flags & Scale::NO_HORIZONTAL != 0)
             || (dim == Axis::VERTICAL && scale.flags & Scale::NO_VERTICAL != 0)
@@ -80,13 +91,19 @@ pub(crate) fn hint_outline(
             group,
         );
         if dim == Axis::VERTICAL {
-            edges::compute_blue_edges(
-                &mut axis,
-                scale,
-                &metrics.axes[dim].blues,
-                &scaled_metrics.axes[dim].blues,
-                group,
-            );
+            if group != ScriptGroup::Default
+                || glyph_style
+                    .map(|style| !style.is_non_base())
+                    .unwrap_or(true)
+            {
+                edges::compute_blue_edges(
+                    &mut axis,
+                    scale,
+                    &metrics.axes[dim].blues,
+                    &scaled_metrics.axes[dim].blues,
+                    group,
+                );
+            }
         } else {
             hinted_metrics.x_scale = scaled_metrics.axes[0].scale;
         }
diff --git a/skrifa/src/outline/autohint/latin/segments.rs b/skrifa/src/outline/autohint/latin/segments.rs
index 27d5a6970..860261996 100644
--- a/skrifa/src/outline/autohint/latin/segments.rs
+++ b/skrifa/src/outline/autohint/latin/segments.rs
@@ -6,14 +6,13 @@
 //! The linking stage associates pairs of segments to form stems and
 //! identifies serifs with a post-process pass.
 
-use raw::tables::glyf::PointFlags;
-
 use super::super::{
     axis::{Axis, Dimension, Segment},
     metrics::fixed_div,
-    outline::{Outline, Point},
+    outline::Outline,
     style::ScriptGroup,
 };
+use raw::tables::glyf::PointFlags;
 
 // Bounds for score, position and coordinate values.
 // See <https://gitlab.freedesktop.org/freetype/freetype/-/blob/57617782464411201ce7bbc93b086c1b4d7d84a5/src/autofit/aflatin.c#L1598>
@@ -23,7 +22,11 @@ const MIN_SCORE: i32 = -32000;
 /// Computes segments for the Latin writing system.
 ///
 /// See <https://gitlab.freedesktop.org/freetype/freetype/-/blob/57617782464411201ce7bbc93b086c1b4d7d84a5/src/autofit/aflatin.c#L1537>
-pub(crate) fn compute_segments(outline: &mut Outline, axis: &mut Axis, group: ScriptGroup) -> bool {
+pub(crate) fn compute_segments(
+    outline: &mut Outline,
+    axis: &mut Axis,
+    _group: ScriptGroup,
+) -> bool {
     assign_point_uvs(outline, axis.dim);
     if !build_segments(outline, axis) {
         return false;
@@ -229,7 +232,7 @@ fn link_segments_cjk(outline: &Outline, axis: &mut Axis, scale: i32) {
             let Some(link2) = seg2.link(segments).copied() else {
                 continue;
             };
-            if link2.link_ix != Some(ix2 as u16) || link2.pos <= link1.pos {
+            if link2.link_ix != Some(ix2 as u16) || link2.pos < link1.pos {
                 continue;
             }
             if seg1.pos == seg2.pos && link1.pos == link2.pos {
@@ -398,10 +401,10 @@ fn build_segments(outline: &mut Outline, axis: &mut Axis) -> bool {
                                 // Discard previous segment
                                 state.min_pos = state.min_pos.min(prev_state.min_pos);
                                 state.max_pos = state.max_pos.max(prev_state.max_pos);
-                                let segment = &mut axis.segments[segment_ix];
+                                let mut segment = axis.segments[segment_ix];
                                 segment.last_ix = point_ix as u16;
-                                state.apply_to_segment(segment, flat_threshold);
-                                prev_segment_ix = Some(segment_ix);
+                                state.apply_to_segment(&mut segment, flat_threshold);
+                                axis.segments[prev_segment_ix.unwrap()] = segment;
                                 prev_state = state;
                             }
                         }
diff --git a/skrifa/src/outline/autohint/latin/widths.rs b/skrifa/src/outline/autohint/latin/widths.rs
index b00f1563f..3e65a30b4 100644
--- a/skrifa/src/outline/autohint/latin/widths.rs
+++ b/skrifa/src/outline/autohint/latin/widths.rs
@@ -32,11 +32,14 @@ pub(super) fn compute_widths(
     // We take the first available glyph from the standard character set.
     if let Some(glyph) = script
         .std_chars
-        .iter()
-        .filter_map(|&ch| glyphs.get(charmap.map(ch)?))
+        .split(' ')
+        .map(|cluster| cluster.chars())
+        .filter_map(|mut chars| Some((chars.next()?, chars.count())))
+        .filter(|(_ch, remaining_count)| *remaining_count == 0)
+        .filter_map(|(ch, _)| glyphs.get(charmap.map(ch)?))
         .next()
     {
-        if outline.fill(&glyph, coords).is_ok() {
+        if outline.fill(&glyph, coords).is_ok() && !outline.points.is_empty() {
             // Now process each dimension
             for (dim, (_metrics, widths)) in result.iter_mut().enumerate() {
                 axis.reset(dim, outline.orientation);
@@ -60,6 +63,13 @@ pub(super) fn compute_widths(
                         }
                     }
                 }
+                // FreeTypes `af_sort_and_quantize_widths()`` has the side effect
+                // of always updating the width count to 1 when we don't find
+                // any...
+                // See <https://gitlab.freedesktop.org/freetype/freetype/-/blob/57617782464411201ce7bbc93b086c1b4d7d84a5/src/autofit/afhints.c#L121>
+                if widths.is_empty() {
+                    widths.push(0);
+                }
                 // The value 100 is heuristic
                 metrics::sort_and_quantize_widths(widths, units_per_em / 100);
             }
diff --git a/skrifa/src/outline/autohint/metrics.rs b/skrifa/src/outline/autohint/metrics.rs
index cc0cd01b5..275ef1ab6 100644
--- a/skrifa/src/outline/autohint/metrics.rs
+++ b/skrifa/src/outline/autohint/metrics.rs
@@ -153,8 +153,6 @@ impl UnscaledStyleMetricsSet {
 pub(crate) struct ScaledStyleMetrics {
     /// Multidimensional scaling factors and deltas.
     pub scale: Scale,
-    /// Script set for the associated style.
-    pub group: ScriptGroup,
     /// Per-dimension scaled metrics.
     pub axes: [ScaledAxisMetrics; 2],
 }
@@ -256,10 +254,17 @@ impl Scale {
         if is_mono {
             flags |= Self::MONO;
         }
-        // Disable horizontal hinting completely for LCD, light hinting
-        // and italic fonts.
-        if target.is_lcd() || is_light || is_italic {
-            flags |= Self::NO_HORIZONTAL;
+        if group == ScriptGroup::Default {
+            // Disable horizontal hinting completely for LCD, light hinting
+            // and italic fonts
+            // See <https://gitlab.freedesktop.org/freetype/freetype/-/blob/57617782464411201ce7bbc93b086c1b4d7d84a5/src/autofit/aflatin.c#L2674>
+            if target.is_lcd() || is_light || is_italic {
+                flags |= Self::NO_HORIZONTAL;
+            }
+        } else {
+            // CJK doesn't hint advances
+            // See <https://gitlab.freedesktop.org/freetype/freetype/-/blob/57617782464411201ce7bbc93b086c1b4d7d84a5/src/autofit/afcjk.c#L1432>
+            flags |= Self::NO_ADVANCE;
         }
         // CJK doesn't hint advances
         // See <https://gitlab.freedesktop.org/freetype/freetype/-/blob/57617782464411201ce7bbc93b086c1b4d7d84a5/src/autofit/afcjk.c#L1432>
@@ -375,7 +380,6 @@ mod tests {
     use super::{super::style::STYLE_CLASSES, *};
     use crate::MetadataProvider;
     use raw::TableProvider;
-    use std::sync::Arc;
 
     #[test]
     fn sort_widths() {
diff --git a/skrifa/src/outline/autohint/mod.rs b/skrifa/src/outline/autohint/mod.rs
index 7a561aa1b..8b760f63b 100644
--- a/skrifa/src/outline/autohint/mod.rs
+++ b/skrifa/src/outline/autohint/mod.rs
@@ -1,8 +1,5 @@
 //! Runtime autohinting support.
 
-// Remove when the work is complete.
-#![allow(dead_code, unused)]
-
 mod axis;
 mod cycling;
 mod hint;
diff --git a/skrifa/src/outline/autohint/outline.rs b/skrifa/src/outline/autohint/outline.rs
index 3b2182739..d6ecb453c 100644
--- a/skrifa/src/outline/autohint/outline.rs
+++ b/skrifa/src/outline/autohint/outline.rs
@@ -266,7 +266,7 @@ impl Outline {
             let mut first_ix = contour.first();
             let mut ix = first_ix;
             let mut prev_ix = contour.prev(first_ix);
-            let mut point = points[0];
+            let mut point = points[first_ix];
             while prev_ix != first_ix {
                 let prev = points[prev_ix];
                 let out_x = point.fx - prev.fx;
@@ -341,6 +341,9 @@ impl Outline {
         let points = self.points.as_mut_slice();
         for i in 0..points.len() {
             let point = points[i];
+            if point.flags.has_marker(PointMarker::WEAK_INTERPOLATION) {
+                continue;
+            }
             if point.in_dir == Direction::None && point.out_dir == Direction::None {
                 let u_index = point.u as usize;
                 let v_index = point.v as usize;
@@ -761,7 +764,7 @@ mod tests {
             let mut base_svg = SvgPen::default();
             let settings = DrawSettings::unhinted(Size::unscaled(), LocationRef::default())
                 .with_path_style(path_style);
-            glyph.draw(settings, &mut base_svg);
+            glyph.draw(settings, &mut base_svg).unwrap();
             let base_svg = base_svg.to_string();
             // Autohinter outline code path
             let mut outline = Outline::default();
@@ -774,7 +777,7 @@ mod tests {
                 point.y = point.fy << 6;
             }
             let mut autohint_svg = SvgPen::default();
-            outline.to_path(path_style, &mut autohint_svg);
+            outline.to_path(path_style, &mut autohint_svg).unwrap();
             let autohint_svg = autohint_svg.to_string();
             if base_svg != autohint_svg {
                 results.push((gid, base_svg, autohint_svg));
diff --git a/skrifa/src/outline/autohint/style.rs b/skrifa/src/outline/autohint/style.rs
index 7ff8ffbad..30515fdc9 100644
--- a/skrifa/src/outline/autohint/style.rs
+++ b/skrifa/src/outline/autohint/style.rs
@@ -43,6 +43,21 @@ impl GlyphStyle {
             None
         }
     }
+
+    fn maybe_assign(&mut self, other: Self) {
+        // FreeType walks the style array in order so earlier styles
+        // have precedence. Since we walk the cmap and binary search
+        // on the full range mapping, our styles are mapped in a
+        // different order. This check allows us to replace a currently
+        // mapped style if the new style index is lower which matches
+        // FreeType's behavior.
+        //
+        // Note that we keep the extra bits because FreeType allows
+        // setting the NON_BASE bit to an already mapped style.
+        if other.0 & Self::STYLE_INDEX_MASK <= self.0 & Self::STYLE_INDEX_MASK {
+            self.0 = (self.0 & !Self::STYLE_INDEX_MASK) | other.0;
+        }
+    }
 }
 
 impl Default for GlyphStyle {
@@ -89,14 +104,11 @@ impl GlyphStyleMap {
             let Some(style) = map.styles.get_mut(gid.to_u32() as usize) else {
                 continue;
             };
-            if !style.is_unassigned() {
-                continue;
-            }
             // Charmaps enumerate in order so we're likely to encounter at least
             // a few codepoints in the same range.
             if let Some(last) = last_range {
                 if last.1.contains(ch) {
-                    *style = last.1.style;
+                    style.maybe_assign(last.1.style);
                     continue;
                 }
             }
@@ -108,7 +120,7 @@ impl GlyphStyleMap {
                 continue;
             };
             if range.contains(ch) {
-                *style = range.style;
+                style.maybe_assign(range.style);
                 if let Some(style_ix) = range.style.style_index() {
                     let style_ix = style_ix as usize;
                     if map.metrics_map[style_ix] == 0xFF {
@@ -215,25 +227,19 @@ pub(crate) enum ScriptGroup {
 /// autohinter.
 #[derive(Clone, Debug)]
 pub(crate) struct ScriptClass {
+    #[allow(unused)]
     pub name: &'static str,
     /// Group that defines how glyphs belonging to this script are hinted.
     pub group: ScriptGroup,
     /// Unicode tag for the script.
+    #[allow(unused)]
     pub tag: Tag,
-    /// Index of self in the SCRIPT_CLASSES array.
-    pub index: usize,
     /// True if outline edges are processed top to bottom.
     pub hint_top_to_bottom: bool,
     /// Characters used to define standard width and height of stems.
-    pub std_chars: &'static [char],
+    pub std_chars: &'static str,
     /// "Blue" characters used to define alignment zones.
-    pub blues: &'static [(&'static [char], u32)],
-}
-
-impl ScriptClass {
-    pub fn from_index(index: u16) -> Option<&'static ScriptClass> {
-        SCRIPT_CLASSES.get(index as usize)
-    }
+    pub blues: &'static [(&'static str, u32)],
 }
 
 /// Defines the basic properties for each style supported by the
@@ -244,6 +250,7 @@ impl ScriptClass {
 /// coverage.
 #[derive(Clone, Debug)]
 pub(crate) struct StyleClass {
+    #[allow(unused)]
     pub name: &'static str,
     /// Index of self in the STYLE_CLASSES array.
     pub index: usize,
@@ -251,6 +258,7 @@ pub(crate) struct StyleClass {
     pub script: &'static ScriptClass,
     /// OpenType feature tag for styles that derive coverage from layout
     /// tables.
+    #[allow(unused)]
     pub feature: Option<Tag>,
 }
 
diff --git a/skrifa/src/outline/mod.rs b/skrifa/src/outline/mod.rs
index 19cbeb1f9..b6d73af63 100644
--- a/skrifa/src/outline/mod.rs
+++ b/skrifa/src/outline/mod.rs
@@ -446,7 +446,7 @@ impl<'a> OutlineGlyph<'a> {
                     sink.try_reserve(outline.points.len())?;
                     let mut contour_start = 0;
                     for contour_end in outline.contours.iter().map(|contour| *contour as usize) {
-                        if contour_end > contour_start {
+                        if contour_end >= contour_start {
                             if let Some(points) = outline.points.get(contour_start..=contour_end) {
                                 let flags = &outline.flags[contour_start..=contour_end];
                                 sink.extend(points.iter().zip(flags).enumerate().map(
diff --git a/skrifa/src/outline/path.rs b/skrifa/src/outline/path.rs
index 5adb2138b..052511bdb 100644
--- a/skrifa/src/outline/path.rs
+++ b/skrifa/src/outline/path.rs
@@ -373,7 +373,7 @@ mod tests {
         // For this test case (in 26.6 fixed point): [(640, 128) + (128, 128)] / 2 = (384, 128)
         // which becomes (6.0, 2.0) when converted to floating point.
         let points = [pt(640, 128), pt(256, 64), pt(640, 64), pt(128, 128)];
-        let mut pen = SvgPen::default();
+        let mut pen = SvgPen::with_precision(1);
         to_path(&points, &flags, &contours, path_style, &mut pen).unwrap();
         assert_eq!(pen.as_ref(), expected);
     }
diff --git a/skrifa/src/outline/pen.rs b/skrifa/src/outline/pen.rs
index 5a06cf146..4b12b2b98 100644
--- a/skrifa/src/outline/pen.rs
+++ b/skrifa/src/outline/pen.rs
@@ -111,9 +111,21 @@ impl OutlinePen for NullPen {
 
 /// Pen that generates SVG style path data.
 #[derive(Clone, Default, Debug)]
-pub struct SvgPen(String);
+pub struct SvgPen(String, Option<usize>);
 
 impl SvgPen {
+    /// Creates a new SVG pen that formats floating point values with the
+    /// standard behavior.
+    pub fn new() -> Self {
+        Self::default()
+    }
+
+    /// Creates a new SVG pen with the given precision (the number of digits
+    /// that will be printed after the decimal).
+    pub fn with_precision(precision: usize) -> Self {
+        Self(String::default(), Some(precision))
+    }
+
     /// Clears the content of the internal string.
     pub fn clear(&mut self) {
         self.0.clear();
@@ -137,22 +149,42 @@ impl core::ops::Deref for SvgPen {
 impl OutlinePen for SvgPen {
     fn move_to(&mut self, x: f32, y: f32) {
         self.maybe_push_space();
-        let _ = write!(self.0, "M{x:.1},{y:.1}");
+        let _ = if let Some(prec) = self.1 {
+            write!(self.0, "M{x:.0$},{y:.0$}", prec)
+        } else {
+            write!(self.0, "M{x},{y}")
+        };
     }
 
     fn line_to(&mut self, x: f32, y: f32) {
         self.maybe_push_space();
-        let _ = write!(self.0, "L{x:.1},{y:.1}");
+        let _ = if let Some(prec) = self.1 {
+            write!(self.0, "L{x:.0$},{y:.0$}", prec)
+        } else {
+            write!(self.0, "L{x},{y}")
+        };
     }
 
     fn quad_to(&mut self, cx0: f32, cy0: f32, x: f32, y: f32) {
         self.maybe_push_space();
-        let _ = write!(self.0, "Q{cx0:.1},{cy0:.1} {x:.1},{y:.1}");
+        let _ = if let Some(prec) = self.1 {
+            write!(self.0, "Q{cx0:.0$},{cy0:.0$} {x:.0$},{y:.0$}", prec)
+        } else {
+            write!(self.0, "Q{cx0},{cy0} {x},{y}")
+        };
     }
 
     fn curve_to(&mut self, cx0: f32, cy0: f32, cx1: f32, cy1: f32, x: f32, y: f32) {
         self.maybe_push_space();
-        let _ = write!(self.0, "C{cx0:.1},{cy0:.1} {cx1:.1},{cy1:.1} {x:.1},{y:.1}");
+        let _ = if let Some(prec) = self.1 {
+            write!(
+                self.0,
+                "C{cx0:.0$},{cy0:.0$} {cx1:.0$},{cy1:.0$} {x:.0$},{y:.0$}",
+                prec
+            )
+        } else {
+            write!(self.0, "C{cx0},{cy0} {cx1},{cy1} {x},{y}")
+        };
     }
 
     fn close(&mut self) {
@@ -169,7 +201,7 @@ impl AsRef<str> for SvgPen {
 
 impl From<String> for SvgPen {
     fn from(value: String) -> Self {
-        Self(value)
+        Self(value, None)
     }
 }
 
@@ -184,3 +216,32 @@ impl fmt::Display for SvgPen {
         write!(f, "{}", self.0)
     }
 }
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn svg_pen_precision() {
+        let svg_data = [None, Some(1), Some(4)].map(|prec| {
+            let mut pen = match prec {
+                None => SvgPen::new(),
+                Some(prec) => SvgPen::with_precision(prec),
+            };
+            pen.move_to(1.0, 2.45556);
+            pen.line_to(1.2, 4.0);
+            pen.quad_to(2.0345, 3.56789, -0.157, -425.07);
+            pen.curve_to(-37.0010, 4.5, 2.0, 1.0, -0.5, -0.25);
+            pen.close();
+            pen.to_string()
+        });
+        let expected = [
+            "M1,2.45556 L1.2,4 Q2.0345,3.56789 -0.157,-425.07 C-37.001,4.5 2,1 -0.5,-0.25 Z", 
+            "M1.0,2.5 L1.2,4.0 Q2.0,3.6 -0.2,-425.1 C-37.0,4.5 2.0,1.0 -0.5,-0.2 Z", 
+            "M1.0000,2.4556 L1.2000,4.0000 Q2.0345,3.5679 -0.1570,-425.0700 C-37.0010,4.5000 2.0000,1.0000 -0.5000,-0.2500 Z"
+        ];
+        for (result, expected) in svg_data.iter().zip(&expected) {
+            assert_eq!(result, expected);
+        }
+    }
+}
diff --git a/skrifa/src/outline/unscaled.rs b/skrifa/src/outline/unscaled.rs
index 6281cef28..d5eca5332 100644
--- a/skrifa/src/outline/unscaled.rs
+++ b/skrifa/src/outline/unscaled.rs
@@ -115,6 +115,12 @@ impl<'a> UnscaledOutlineRef<'a> {
                 }
                 cur_contour = point_ix..point_ix;
                 found_best_in_cur_contour = false;
+                // Ignore single point contours
+                match self.points.get(point_ix + 1) {
+                    Some(next_point) if next_point.is_contour_start => continue,
+                    None => continue,
+                    _ => {}
+                }
             }
             cur_contour.end += 1;
             if f(point) {