Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
33 changes: 33 additions & 0 deletions babelfish/converters/name.py
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,7 @@

from typing import ClassVar

from babelfish.compat import resource_stream
from babelfish.language import LANGUAGE_MATRIX

from . import LanguageEquivalenceConverter
Expand All @@ -18,3 +19,35 @@ class NameConverter(LanguageEquivalenceConverter):
for iso_language in LANGUAGE_MATRIX:
if iso_language.name:
SYMBOLS[iso_language.alpha3] = iso_language.name

#: Common/colloquial names that no ISO 639-3 data source provides but that
#: users routinely expect to resolve. Keyed by alpha3 -> list of names.
COMMON_NAMES: ClassVar[dict[str, list[str]]] = {
'ell': ['Greek'], # SIL only lists 'Modern Greek (1453-)'
}

def __init__(self) -> None:
# The base class builds the forward map (alpha3 -> ISO 639-3 reference
# name) and seeds the reverse map with those reference names.
super().__init__()

# Register the additional ISO 639-3 print names as reverse-only aliases,
# so e.g. both 'Pashto' and 'Pushto' resolve to 'pus' while the forward
# name (Language('pus').name) stays the canonical reference name.
with resource_stream('babelfish', 'data/iso-639-3_Name_Index.tab') as f:
f.readline()
for line in f:
alpha3, print_name, _inverted_name = line.decode('utf-8').rstrip('\r\n').split('\t')
self._add_name(alpha3, print_name)

for alpha3, names in self.COMMON_NAMES.items():
for name in names:
self._add_name(alpha3, name)

def _add_name(self, alpha3: str, name: str) -> None:
# Skip names whose language is not in the loaded matrix (e.g. codes the
# Name Index knows but the bundled iso-639-3.tab does not), and never
# let an alias override a reference name already claimed by a language.
if name and alpha3 in self.to_symbol and name not in self.from_symbol:
self.from_symbol[name] = (alpha3, None, None)
self.codes.add(name)
Loading