mongfontbuilder 0.4.0__tar.gz

mongfontbuilder-0.4.0/PKG-INFO

@@ -0,0 +1,40 @@
+Metadata-Version: 2.3
+Name: mongfontbuilder
+Version: 0.4.0
+Summary: Mongolian Font Builder.
+License: MIT
+Author: Kushim Jiang
+Author-email: kushim_jiang@outlook.com
+Requires-Python: >= 3.10.0, < 4.0.0
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Requires-Dist: PyYAML (>=6.0.1,<7.0.0)
+Requires-Dist: cattrs (>=24.1.2,<25.0.0)
+Requires-Dist: fontmake (>=3.7.1,<4.0.0)
+Requires-Dist: tptq.feacomposer (>=1.5.0,<2.0.0)
+Requires-Dist: ufonormalizer (>=0.6.2,<0.7.0)
+Requires-Dist: unicodedata2 (>=16.0.0)
+Project-URL: Repository, https://github.com/Kushim-Jiang/mongfontbuilder
+Project-URL: releasenotes, https://github.com/Kushim-Jiang/mongfontbuilder/releases
+Description-Content-Type: text/markdown
+
+# Mongolian Font Builder
+
+# Tooling
+
+The Python package `mongfontbuilder` is maintained in the [lib/](lib/) directory.
+
+# Documentation
+
+The documentation of this project is maintained in the [docs](docs/) directory, and will be published as a future version of the Mongolian UTN ([UTN \#57: Encoding an Shaping of the Mongolian Script](https://www.unicode.org/notes/tn57/)).
+
+See [CONTRIBUTING.md](CONTRIBUTING.md) for guide.
+
+## Data
+
+The source of truth of the UTN’s data is maintained in the [data](data/) directory, which is then exported to [lib/mongfontbuilder/data.json](/lib/mongfontbuilder/data.json) for consuming from Python.
+

mongfontbuilder-0.4.0/README.md

@@ -0,0 +1,15 @@
+# Mongolian Font Builder
+
+# Tooling
+
+The Python package `mongfontbuilder` is maintained in the [lib/](lib/) directory.
+
+# Documentation
+
+The documentation of this project is maintained in the [docs](docs/) directory, and will be published as a future version of the Mongolian UTN ([UTN \#57: Encoding an Shaping of the Mongolian Script](https://www.unicode.org/notes/tn57/)).
+
+See [CONTRIBUTING.md](CONTRIBUTING.md) for guide.
+
+## Data
+
+The source of truth of the UTN’s data is maintained in the [data](data/) directory, which is then exported to [lib/mongfontbuilder/data.json](/lib/mongfontbuilder/data.json) for consuming from Python.

mongfontbuilder-0.4.0/lib/mongfontbuilder/__init__.py

@@ -0,0 +1,275 @@
+from __future__ import annotations
+
+import re
+from collections.abc import Iterator
+from dataclasses import dataclass, field, replace
+from typing import Iterable
+
+from fontTools import unicodedata
+from ufoLib2.objects import Component, Font, Glyph
+
+from .data import (
+    CharacterName,
+    JoiningPosition,
+    LocaleID,
+    VariantData,
+    VariantReference,
+    WrittenUnitID,
+    codePointToCmapVariant,
+)
+from .data.misc import fina, init, isol, medi
+
+
+def constructFont(font: Font, locales: list[LocaleID]) -> None:
+    from .otl import MongFeaComposer
+
+    constructPredefinedGlyphs(font, locales)
+
+    composer = MongFeaComposer(font, locales)
+    assert not font.features.text, font.features.text
+    font.features.text = composer.asFeatureFile().asFea().replace(":", "-")  # HACK
+
+
+def splitWrittens(writtens: str | Iterable[str]) -> list[WrittenUnitID]:
+    """
+    >>> splitWrittens("ABbCcc")
+    ['A', 'Bb', 'Ccc']
+    """
+
+    if isinstance(writtens, str):
+        return re.sub(r"[A-Z]", lambda x: " " + x[0], writtens).removeprefix(" ").split(" ")
+    return list(writtens)
+
+
+def getPosition(index: int, length: int):
+    return isol if length == 1 else (init if index == 0 else fina if index == length - 1 else medi)
+
+
+def writtenCombinations(writtens: str, position: JoiningPosition) -> Iterator[list[str]]:
+    """
+    >>> [*writtenCombinations("ABCD", "isol")]
+    [['A.init', 'B.medi', 'C.medi', 'D.fina'], ['A.init', 'B.medi', 'CD.fina'], ['A.init', 'BC.medi', 'D.fina'], ['A.init', 'BCD.fina'], ['AB.init', 'C.medi', 'D.fina'], ['AB.init', 'CD.fina'], ['ABC.init', 'D.fina'], ['ABCD.isol']]
+    """
+
+    parts = splitWrittens(writtens)
+    if "Lv" in parts:
+        index = parts.index("Lv")
+        if index > 0:
+            parts[index - 1] += parts.pop(index)
+
+    leftJoin = 1 if position in (medi, fina) else 0
+    rightJoin = 1 if position in (init, medi) else 0
+    placeholder = "X"
+    if leftJoin:
+        parts = [placeholder] + parts
+    if rightJoin:
+        parts += [placeholder]
+
+    combinations = [[]]
+    for part in parts:
+        newCombinations = []
+        for comb in combinations:
+            newCombinations.append(comb + [part])
+            if comb:
+                newCombinations.append(comb[:-1] + [comb[-1] + part])
+        combinations = newCombinations
+
+    for comb in combinations:
+        result = [
+            f"{written}.{getPosition(index, len(comb))}" for index, written in enumerate(comb)
+        ][leftJoin : (len(comb) - rightJoin)]
+        if (
+            len([l for w in result for l in w if l.isupper()])
+            == len(list(l for l in str(writtens) if l.isupper()))
+            and result
+        ):
+            yield result
+
+
+@dataclass
+class GlyphDescriptor:
+    codePoints: list[int]
+    units: list[WrittenUnitID]
+    position: JoiningPosition
+    suffixes: list[str] = field(default_factory=list)
+
+    @classmethod
+    def parse(cls, name: str) -> GlyphDescriptor:
+        """
+        >>> GlyphDescriptor.parse('u1820.A.init')
+        GlyphDescriptor(codePoints=[6176], units=['A'], position='init', suffixes=[])
+        >>> GlyphDescriptor.parse('_A.init')
+        GlyphDescriptor(codePoints=[], units=['A'], position='init', suffixes=[])
+        """
+
+        x, y, position, *suffixes = (
+            "." + name.removeprefix("_")  # _A.init
+            if name.startswith("_")
+            else name  # u1820.A.init
+        ).split(".")
+        units = splitWrittens(y)
+        assert units and all(i in data.writtenUnits for i in units), name
+        assert position in data.types.joiningPositions, name
+        instance = cls(
+            codePoints=[int(i.removeprefix("u"), 16) for i in x.split("_")] if x else [],
+            units=units,
+            position=position,
+            suffixes=suffixes,
+        )
+        assert str(instance) == name
+        return instance
+
+    @classmethod
+    def fromData(
+        cls,
+        charName: CharacterName,
+        position: JoiningPosition,
+        variantData: VariantData | None = None,
+        suffixes: list[str] = [],
+        locale: LocaleID | None = None,
+    ) -> GlyphDescriptor:
+        from .data import variantFromReference
+
+        if not variantData:
+            variantData = next(i for i in data.variants[charName][position].values() if i.default)
+
+        written = None
+        if locale and locale in variantData.locales:
+            written = variantData.locales[locale].written
+        if not written:
+            written = variantData.written
+        assert written, variantData
+
+        if isinstance(written, VariantReference):
+            units = variantFromReference(written, data.variants[charName])
+            suffixes = ["_" + position, *suffixes]
+            position = written.position
+        else:
+            units = written
+        return cls([ord(unicodedata.lookup(charName))], units, position, suffixes)
+
+    def __str__(self) -> str:
+        assert self.units, self
+        if self.codePoints:
+            name = "_".join(uNameFromCodePoint(i) for i in self.codePoints) + "."
+        else:
+            name = "_"
+        return name + ".".join(["".join(self.units), self.position, *self.suffixes])
+
+    def __add__(self, other: GlyphDescriptor) -> GlyphDescriptor:
+        """
+        >>> GlyphDescriptor.parse('u1820.A.init') + GlyphDescriptor.parse('u1820.A.medi')
+        GlyphDescriptor(codePoints=[6176, 6176], units=['A', 'A'], position='init', suffixes=[])
+        """
+        joiningType: dict[tuple[JoiningPosition, JoiningPosition], JoiningPosition] = {
+            ("init", "fina"): "isol",
+            ("init", "medi"): "init",
+            ("medi", "medi"): "medi",
+            ("medi", "fina"): "fina",
+        }
+        assert (self.position, other.position) in joiningType
+        return GlyphDescriptor(
+            self.codePoints + other.codePoints,
+            self.units + other.units,
+            joiningType[(self.position, other.position)],
+        )
+
+    def pseudoPosition(self) -> JoiningPosition | None:
+        if self.suffixes:
+            suffix = self.suffixes[0]
+            if suffix in pseudoPositionSuffixes:
+                position = suffix.removeprefix("_")
+                assert position in data.misc.joiningPositions
+                return position
+
+    def __hash__(self) -> int:
+        return hash(self.__str__())
+
+
+def uNameFromCodePoint(codePoint: int) -> str:
+    return f"u{codePoint:04X}"
+
+
+pseudoPositionSuffixes = ["_" + i for i in data.misc.joiningPositions]
+
+
+def constructPredefinedGlyphs(
+    font: Font,
+    locales: list[LocaleID],
+    *,
+    initPadding: float = 40,
+    finaPadding: float = 100,
+) -> None:
+    sources = list[GlyphDescriptor]()
+    for name in font.keys():
+        try:
+            target = GlyphDescriptor.parse(name)
+        except:
+            continue
+        sources.append(target)
+
+    targetedLocales = {*locales}
+    for charName, positionToFVSToVariant in data.variants.items():
+        variantNames = list[str]()
+        for position, fvsToVariant in positionToFVSToVariant.items():
+            for variant in fvsToVariant.values():
+                if not targetedLocales.intersection(variant.locales):
+                    continue
+
+                target = GlyphDescriptor.fromData(charName, position, variant)
+                targetName = str(target)
+                variantNames.append(targetName)
+
+                glyph = font.get(targetName)
+                if glyph is not None:
+                    glyph.unicode = None
+                    continue
+
+                memberNames: list[str]
+                writtenTarget = replace(target, codePoints=[], suffixes=[])
+                for source in sources:
+                    if source == writtenTarget:
+                        memberNames = [str(source)]
+                        break
+                else:
+                    for source in sources:
+                        if replace(source, codePoints=[]) == writtenTarget:
+                            memberNames = [str(source)]
+                            break
+                    else:
+                        for writtenVariants in writtenCombinations(
+                            "".join(writtenTarget.units), writtenTarget.position
+                        ):
+                            if len(writtenVariants) == len(writtenTarget.units):
+                                memberNames = ["_" + i for i in writtenVariants]
+                                break
+                        else:
+                            raise NotImplementedError(target)
+
+                members: list[Glyph | float] = [font[i] for i in memberNames]
+                if pseudoPosition := target.pseudoPosition():
+                    if pseudoPosition in ["isol", "init"]:
+                        members = [initPadding, *members]
+                    if pseudoPosition in ["isol", "fina"]:
+                        members = [*members, finaPadding]
+                composeGlyph(font, targetName, members)
+
+        if variantNames:
+            codePoint = ord(unicodedata.lookup(charName))
+            variant = GlyphDescriptor([codePoint], *codePointToCmapVariant[codePoint])
+            glyph = composeGlyph(font, uNameFromCodePoint(codePoint), [font[str(variant)]])
+            glyph.unicode = codePoint
+
+
+def composeGlyph(font: Font, name: str, members: list[Glyph | float]) -> Glyph:
+    font.lib.get("public.glyphOrder", []).append(name)
+    glyph = font.newGlyph(name)
+    for member in members:
+        if isinstance(member, Glyph):
+            component = Component(str(member.name))
+            component.move((glyph.width, 0))
+            glyph.components.append(component)
+            glyph.width += member.width
+        else:
+            glyph.width += member
+    return glyph

mongfontbuilder-0.4.0/lib/mongfontbuilder/__main__.py

@@ -0,0 +1,36 @@
+"""
+poetry run python -m mongfontbuilder
+"""
+
+from argparse import ArgumentParser
+from pathlib import Path
+
+from ufoLib2 import Font
+
+from . import constructFont
+from .data import locales
+
+parser = ArgumentParser()
+parser.add_argument(
+    "input",
+    type=Path,
+    help="path to read source UFO font from",
+)
+parser.add_argument(
+    "output",
+    type=Path,
+    help="path to write constructed UFO font to",
+)
+parser.add_argument(
+    "--locales",
+    metavar="LOCALE",
+    choices=locales,
+    nargs="+",
+    required=True,
+    help="targeted locales, one or more from: " + ", ".join(locales),
+)
+
+args = parser.parse_args()
+font = Font.open(args.input)
+constructFont(font, args.locales)
+font.save(args.output, overwrite=True)

mongfontbuilder-0.4.0/lib/mongfontbuilder/data/README.md

@@ -0,0 +1,3 @@
+# Data files
+
+JSON files in this directory are not manually maintained. They are exported from [/data/](/data/) with [/data/export.ts](/data/export.ts).

mongfontbuilder-0.4.0/lib/mongfontbuilder/data/__init__.py

@@ -0,0 +1,100 @@
+import json
+from importlib.resources import files
+from typing import Literal
+
+from cattrs import structure
+from fontTools import unicodedata
+
+from .misc import JoiningPosition, joiningPositions
+from .types import (
+    FVS,
+    AliasData,
+    CharacterName,
+    LocaleData,
+    LocaleID,
+    VariantData,
+    VariantReference,
+    WrittenUnitID,
+)
+
+assert __package__
+dir = files(__package__)
+
+with (dir / "writtenUnits.json").open(encoding="utf-8") as f:
+    writtenUnits: list[WrittenUnitID] = json.load(f)
+
+with (dir / "ligatures.json").open(encoding="utf-8") as f:
+    ligatures: dict[
+        Literal["required", "optional"],
+        dict[str, list[JoiningPosition]],
+    ] = json.load(f)
+
+with (dir / "locales.json").open(encoding="utf-8") as f:
+    locales = structure(
+        json.load(f),
+        dict[LocaleID, LocaleData],
+    )
+
+with (dir / "aliases.json").open(encoding="utf-8") as f:
+    aliases = structure(
+        json.load(f),
+        dict[CharacterName, AliasData],
+    )
+
+with (dir / "variants.json").open(encoding="utf-8") as f:
+    variants = structure(
+        json.load(f),
+        dict[CharacterName, dict[JoiningPosition, dict[FVS, VariantData]]],
+    )
+
+with (dir / "particles.json").open(encoding="utf-8") as f:
+    particles = structure(json.load(f), dict[LocaleID, dict[str, list[FVS]]])
+
+
+def variantFromReference(
+    reference: VariantReference,
+    positionToFVSToVariantData: dict[JoiningPosition, dict[FVS, VariantData]],
+) -> list[WrittenUnitID]:
+    position, fvs, locale = reference
+    if not locale:
+        written = positionToFVSToVariantData[position][fvs].written
+    else:
+        written = positionToFVSToVariantData[position][fvs].locales[locale].written
+    assert isinstance(written, list)
+    return written
+
+
+def _resolveCmapVariants() -> dict[int, tuple[list[WrittenUnitID], JoiningPosition]]:
+    codePointToPositionToVariant = dict[
+        int, dict[JoiningPosition, tuple[list[WrittenUnitID], JoiningPosition]]
+    ]()
+    for charName, positionToFVSToVariantData in variants.items():
+        codePoint = ord(unicodedata.lookup(charName))
+        for position, fvsToVariantData in positionToFVSToVariantData.items():
+            for data in fvsToVariantData.values():
+                if data.default:
+                    written = data.written
+                    if isinstance(written, VariantReference):
+                        variant = (
+                            variantFromReference(written, positionToFVSToVariantData),
+                            written.position,
+                        )
+                    else:
+                        variant = written, position
+                    codePointToPositionToVariant.setdefault(codePoint, {})[position] = variant
+                    break
+
+    codePointToVariant = dict[int, tuple[list[WrittenUnitID], JoiningPosition]]()
+    for codePoint, positionToVariant in sorted(codePointToPositionToVariant.items()):
+        for position in joiningPositions:
+            if variant := positionToVariant.get(position):
+                if variant not in codePointToVariant.values():
+                    codePointToVariant[codePoint] = variant
+                    break
+        else:
+            raise NotImplementedError
+
+    return codePointToVariant
+
+
+codePointToCmapVariant = _resolveCmapVariants()