dhaturupa 0.1.0

A Python package for Sanskrit verb conjugation (dhāturūpa) - generates conjugation tables for Sanskrit verb roots across all tenses and moods

dhaturupa

Sanskrit verb conjugation. Give it a root, get back the full paradigm.

pip install -e .

Quick start

from dhaturupa import conjugate

result = conjugate('bhū')
print(result['forms']['prathama']['ekavacana'])  # bhavati
print(result['forms']['uttama']['ekavacana'])     # bhavāmi

result = conjugate('bhū', lakara='loṭ')           # imperative
result = conjugate('bhū', pada='ātmanepada')       # override voice

conjugate() returns the 3×3 grid (puruṣa × vacana). get_all_lakaras() does it for all 10 tenses at once.

Search and lookup

from dhaturupa import get_dhatu_info, search_dhatus, list_dhatus

get_dhatu_info('gam')
# {'dhatu': 'gam', 'gana': 1, 'gana_name': 'bhvādi',
#  'pada': 'parasmaipada', 'meaning': 'to go'}

search_dhatus('to know')
# [{'dhatu': 'budh', ...}, {'dhatu': 'vid', ...}, {'dhatu': 'jñā', ...}]

list_dhatus(gana=4)  # all divādi roots

Transliteration

from dhaturupa.transliteration import iast_to_devanagari, devanagari_to_iast

iast_to_devanagari('bhavati')  # भवति
devanagari_to_iast('भवति')     # bhavati

CLI

dhaturupa conjugate bhū
# ============================================================
#   धातु (Dhātu): bhū
#   Stem: bhava
# ============================================================
#   Prathama (3rd)    bhavati    bhavataḥ    bhavanti
#   Madhyama (2nd)    bhavasi    bhavathaḥ   bhavatha
#   Uttama (1st)      bhavāmi    bhavāvaḥ    bhavāmaḥ

dhaturupa conjugate bhū --lakara loṭ
dhaturupa conjugate bhū --all-lakaras
dhaturupa conjugate bhū --json

dhaturupa info gam
dhaturupa search "to know"
dhaturupa list --gana 1

What's inside

85+ roots across all 10 gaṇas. Stems are generated with guṇa/vṛddhi gradation and vikaraṇa suffixes, with pre-computed overrides for irregulars like √gam → gaccha, √kṛ → karo, √jñā → jānā.

10 lakāras:

• laṭ (present), loṭ (imperative), laṅ (imperfect), vidhiliṅ (optative)
• lṛṭ (future), lṛṅ (conditional)
• liṭ (perfect), luṭ (periphrastic future), luṅ (aorist), āśīrliṅ (benedictive)

Both parasmaipada and ātmanepada endings. Zero dependencies.

Formatter

from dhaturupa import conjugate
from dhaturupa.formatter import format_table

table = format_table(conjugate('bhū'))
print(table)

table_deva = format_table(conjugate('bhū'), script='devanagari')
print(table_deva)  # भवति, भवतः, भवन्ति...

JSON API

from dhaturupa import conjugate_json, get_dhatu_info_json, search_dhatus_json

conjugate_json('bhū')
# '{"dhatu":"bhū","gana":1,"forms":{"prathama":{"ekavacana":"bhavati",...},...}}'

conjugate_json('bhū', pretty=True)
# pretty-printed with indentation

get_dhatu_info_json('gam')
# '{"dhatu":"gam","gana":1,"gana_name":"bhvādi","pada":"parasmaipada","meaning":"to go"}'

search_dhatus_json('to know')
# '[{"dhatu":"budh",...},{"dhatu":"vid",...},{"dhatu":"jñā",...}]'

Every public method has a _json twin: conjugate_json(), get_all_lakaras_json(), get_dhatu_info_json(), list_dhatus_json(), search_dhatus_json(). All take pretty=True for indented output. Unicode is preserved, not escaped.

License

MIT