From bb08d696f9a3eeeff21f547a97279443f5ffa677 Mon Sep 17 00:00:00 2001 From: Jim Geovedi Date: Sun, 30 Jul 2017 21:23:28 +0700 Subject: [PATCH] added hashtag rule and fixed currency rules --- spacy/lang/id/punctuation.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/spacy/lang/id/punctuation.py b/spacy/lang/id/punctuation.py index b72109ba6..d7ee767f7 100644 --- a/spacy/lang/id/punctuation.py +++ b/spacy/lang/id/punctuation.py @@ -11,7 +11,7 @@ _units = (_units + 's bit Gbps Mbps mbps Kbps kbps ƒ ppi px ' 'ratus rb ribu ribuan ' 'juta jt jutaan mill?iar million bil[l]?iun bilyun billion ' ) -_currency = (_currency + r'Rp IDR RMB SGD S\$ ') +_currency = (_currency + r' Rp IDR RMB SGD S\$') _months = ('Januari Februari Maret April Mei Juni Juli Agustus September ' 'Oktober November Desember January February March May June ' 'July August October December Jan Feb Mar Jun Jul Aug Sept ' @@ -25,7 +25,7 @@ HTML_SUFFIX = r'' MONTHS = merge_chars(_months) LIST_CURRENCY = split_chars(_currency) - +TOKENIZER_PREFIXES.remove('#') # hashtag _prefixes = TOKENIZER_PREFIXES + LIST_CURRENCY + [HTML_PREFIX] + ['/', '—'] _suffixes = TOKENIZER_SUFFIXES + [r'\-[Nn]ya', '-[KkMm]u', '[—-]'] + [