diff options
Diffstat (limited to 'lulua/report.py')
| -rw-r--r-- | lulua/report.py | 51 |
1 files changed, 34 insertions, 17 deletions
diff --git a/lulua/report.py b/lulua/report.py index b25201d..0e5ec00 100644 --- a/lulua/report.py +++ b/lulua/report.py @@ -18,7 +18,7 @@ # OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN # THE SOFTWARE. -import sys, argparse, logging, pickle, math +import sys, argparse, logging, pickle, math, unicodedata from gettext import GNUTranslations, NullTranslations from decimal import Decimal from fractions import Fraction @@ -60,21 +60,6 @@ def arabnum (s): m = {'0': '٠', '1': '١', '2': '٢', '3': '٣', '4': '٤', '5': '٥', '6': '٦', '7': '٧', '8': '٨', '9': '٩', ',': '٬', '.': '٫'} return ''.join (map (lambda x: m.get (x, x), s)) -def clamp (v, lower, upper): - return max (min (v, upper), lower) - -def blend (v, a, b): - v = clamp (v, 0, 1) - return (b-a)*v+a - -def blendn (v, *l): - assert 0 <= v <= 1 - n = len (l) - step = 1/(n-1) - i = min (int (math.floor (v/step)), n-2) - stretchedv = (v-i*step)/step - return [blend (stretchedv, x, y) for x, y in zip (l[i], l[i+1])] - def render (): parser = argparse.ArgumentParser(description='Create lulua report.') parser.add_argument('-c', '--corpus', nargs='+', metavar='FILE', help='Corpus metadata files') @@ -88,9 +73,41 @@ def render (): env.filters['approx'] = approx env.filters['numspace'] = numspace env.filters['arabnum'] = arabnum - env.filters['blendn'] = blendn env.filters['fraction'] = fraction + # Map global variables to Arabic letter romanizations, so we can use + # them easily in text. + # Taken from Abu-Chacra’s Arabic – An Essential Grammar. It’s + # too difficult for now to write a general-purpose romanization + # function, because it would need a dictionary. + letterNames = { + 'Hamzah': ('Hamzah', 'ء'), + 'Alif': ('ᵓAlif', 'ا'), + 'Alifhamzah': ('ᵓAlif-hamzah', 'أ'), + 'Wawhamzah': ('Wa\u0304w-hamzah', 'ؤ'), + 'Yahamzah': ('Ya\u0304ᵓ-hamzah', 'ئ'), + 'Ba': ('Baᵓ', 'ب'), + 'Ta': ('Taᵓ', 'ت'), + 'Tha': ('T\u0331aᵓ', 'ث'), + 'Ra': ('Raᵓ', 'ر'), + 'Dal': ('Da\u0304l', 'د'), + 'Dhal': ('D\u0331a\u0304l', 'ذ'), + 'Qaf': ('Qa\u0304f', 'ق'), + 'Lam': ('La\u0304m', 'ل'), + 'Lamalif': ('La\u0304m-ᵓalif', 'لا'), + 'Mim': ('Mi\u0304m', 'م'), + 'Nun': ('Nu\u0304n', 'ن'), + 'Waw': ('Wa\u0304w', 'و'), + 'Ya': ('Ya\u0304ᵓ', 'ي'), + 'Tamarbutah': ('Ta\u0304ᵓ marbu\u0304t\u0323ah', 'ة'), + 'Alifmaqsurah': ('ᵓAlif maqs\u0323u\u0304rah', 'ى'), + } + for k, (romanized, arabic) in letterNames.items (): + env.globals[k] = f'{romanized} <bdo lang="ar">({arabic})</bdo>' + env.globals[k.lower ()] = env.globals[k].lower () + env.globals[k + '_'] = romanized + env.globals[k.lower () + '_'] = romanized.lower () + corpus = [] for x in args.corpus: with open (x) as fd: |