File size: 5,915 Bytes
f9bc0c1 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 | """Canonical writing-system profiles for Newari / Nepal Bhasa OCR.
The profiles distinguish source-image appearance from OCR transcription encoding.
That distinction matters for Bhujimol and Ranjana: neither may be treated as a
Prachalit visual model merely because its transcription can use an encoded script.
"""
from __future__ import annotations
import re
from dataclasses import dataclass
ScriptCodepointRange = tuple[int, int]
DEVANAGARI_RANGES: tuple[ScriptCodepointRange, ...] = (
(0x0900, 0x097F),
(0xA8E0, 0xA8FF),
)
NEWA_RANGES: tuple[ScriptCodepointRange, ...] = ((0x11400, 0x1147F),)
@dataclass(frozen=True, slots=True)
class NewariWritingSystemProfile:
"""Static encoding and routing semantics for one Newari writing system."""
profile_id: str
display_name: str
aliases: tuple[str, ...]
source_unicode_status: str
source_unicode_script: str | None
source_unicode_ranges: tuple[ScriptCodepointRange, ...]
transcription_unicode_script: str
transcription_unicode_ranges: tuple[ScriptCodepointRange, ...]
route_script: str
visual_profile_required: bool
input_policy: str
output_policy: str
NEWARI_WRITING_SYSTEM_PROFILES: tuple[NewariWritingSystemProfile, ...] = (
NewariWritingSystemProfile(
profile_id="newa_prachalit",
display_name="Newar / Prachalit",
aliases=(
"newa",
"newar",
"newar script",
"newari script",
"nepal lipi",
"nepaalalipi",
"prachalit",
"prachalit newa",
),
source_unicode_status="encoded",
source_unicode_script="Newa",
source_unicode_ranges=NEWA_RANGES,
transcription_unicode_script="Newa",
transcription_unicode_ranges=NEWA_RANGES,
route_script="newa",
visual_profile_required=True,
input_policy="unicode_text_or_source_image",
output_policy="native_newa_unicode",
),
NewariWritingSystemProfile(
profile_id="bhujimol",
display_name="Bhujimol / Bhujinmol",
aliases=("bhujimol", "bhujinmol", "bhujimol script", "bhujinmol script"),
source_unicode_status="no_separate_script_encoding",
source_unicode_script="Newa",
source_unicode_ranges=NEWA_RANGES,
transcription_unicode_script="Newa",
transcription_unicode_ranges=NEWA_RANGES,
route_script="bhujimol",
visual_profile_required=True,
input_policy="source_image_or_newa_text_with_explicit_bhujimol_style",
output_policy="newa_unicode_transcription_with_source_style_metadata",
),
NewariWritingSystemProfile(
profile_id="ranjana",
display_name="Ranjana",
aliases=("ranjana", "ranjana script", "ranjana lipi", "rañjana", "rañjanā"),
source_unicode_status="unencoded",
source_unicode_script=None,
source_unicode_ranges=(),
transcription_unicode_script="Newa",
transcription_unicode_ranges=NEWA_RANGES,
route_script="ranjana",
visual_profile_required=True,
input_policy="source_image_or_audited_legacy_font_mapping_only",
output_policy="newa_unicode_transcription_with_source_image_and_ranjana_style_metadata",
),
NewariWritingSystemProfile(
profile_id="devanagari",
display_name="Devanagari",
aliases=("devanagari", "deva", "devanagari script", "देवनागरी"),
source_unicode_status="encoded",
source_unicode_script="Devanagari",
source_unicode_ranges=DEVANAGARI_RANGES,
transcription_unicode_script="Devanagari",
transcription_unicode_ranges=DEVANAGARI_RANGES,
route_script="devanagari",
visual_profile_required=True,
input_policy="unicode_text_or_source_image",
output_policy="native_devanagari_unicode",
),
)
def _normalize_alias(value: str) -> str:
return re.sub(r"[\s_/-]+", " ", value.strip().casefold())
def _build_alias_index() -> dict[str, NewariWritingSystemProfile]:
aliases: dict[str, NewariWritingSystemProfile] = {}
for profile in NEWARI_WRITING_SYSTEM_PROFILES:
values = (profile.profile_id, profile.display_name, *profile.aliases)
for value in values:
normalized = _normalize_alias(value)
previous = aliases.get(normalized)
if previous is not None and previous.profile_id != profile.profile_id:
raise RuntimeError(
f"ambiguous Newari writing-system alias {value!r}: "
f"{previous.profile_id!r} and {profile.profile_id!r}"
)
aliases[normalized] = profile
return aliases
NEWARI_WRITING_SYSTEM_ALIASES = _build_alias_index()
NEWARI_WRITING_SYSTEM_BY_ID = {profile.profile_id: profile for profile in NEWARI_WRITING_SYSTEM_PROFILES}
# These keys score a recognizer's Unicode transcription, not the visual identity
# of the source glyphs. In particular, Ranjana remains unencoded.
NEWARI_TRANSCRIPTION_ROUTING_RANGES: dict[str, tuple[ScriptCodepointRange, ...]] = {
"newar": NEWA_RANGES,
"newar-script": NEWA_RANGES,
"newa": NEWA_RANGES,
"prachalit": NEWA_RANGES,
"newa/prachalit": NEWA_RANGES,
"newa-prachalit": NEWA_RANGES,
"bhujimol": NEWA_RANGES,
"bhujinmol": NEWA_RANGES,
"ranjana": NEWA_RANGES,
"ranjana-newa-transcription": NEWA_RANGES,
}
def resolve_newari_writing_system(value: str) -> NewariWritingSystemProfile:
"""Resolve a user/config alias to one of the four supported profiles."""
profile = NEWARI_WRITING_SYSTEM_ALIASES.get(_normalize_alias(value))
if profile is None:
supported = ", ".join(profile.profile_id for profile in NEWARI_WRITING_SYSTEM_PROFILES)
raise ValueError(f"unsupported Newari writing system {value!r}; supported profiles: {supported}")
return profile
|