mirror/OCRmyPDF
1
0
Fork 0
mirror of https://github.com/ocrmypdf/OCRmyPDF.git synced 2026-05-18 19:47:48 -04:00
Code Issues Packages Projects Releases Wiki Activity

validation: add proper list of languages supported by hocr

Browse Source 
Based on Latin-1 support in default PDF fonts.
This commit is contained in:
James R. Barlow
2021-05-23 01:14:15 -07:00
parent c409fa5825
commit c87221a4e6
1 changed files with 54 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

55
src/ocrmypdf/_validation.py
View File

@@ -37,7 +37,60 @@ from ocrmypdf.subprocess import check_external_program
# -------------
# External dependencies
HOCR_OK_LANGS = frozenset(['eng', 'deu', 'spa', 'ita', 'por'])
# According to Wikipedia these languages are supported in the ISO-8859-1 character
# set, meaning reportlab can generate them and they are compatible with hocr,
# assuming Tesseract has the necessary languages installed. Note that there may
# not be language packs for them.
HOCR_OK_LANGS = frozenset(
[
# Languages fully covered by Latin-1:
'afr', # Afrikaans
'alb', # Albanian
'ast', # Leonese
'baq', # Basque
'bre', # Breton
'cos', # Corsican
'eng', # English
'eus', # Basque
'fao', # Faoese
'gla', # Scottish Gaelic
'glg', # Galician
'glv', # Manx
'ice', # Icelandic
'ind', # Indonesian
'isl', # Icelandic
'ita', # Italian
'ltz', # Luxembourgish
'mal', # Malay Rumi
'mga', # Irish
'nor', # Norwegian
'oci', # Occitan
'por', # Portugeuse
'roh', # Romansh
'sco', # Scots
'sma', # Sami
'spa', # Spanish
'sqi', # Albanian
'swa', # Swahili
'swe', # Swedish
'tgl', # Tagalog
'wln', # Walloon
# Languages supported by Latin-1 except for a few rare characters that OCR
# is probably not trained to recognize anyway:
'cat', # Catalan
'cym', # Welsh
'dan', # Danish
'deu', # German
'dut', # Dutch
'est', # Estonian
'fin', # Finnish
'fra', # French
'hun', # Hungarian
'kur', # Kurdish
'nld', # Dutch
'wel', # Welsh
]
)
DEFAULT_LANGUAGE = 'eng' # Enforce English hegemony
log = logging.getLogger(__name__)