90 Commits

Author	SHA1	Message	Date
fxsjy	04eb4f08cf	fix a bug of changing dictionary	2013-04-26 16:48:46 +08:00
fxsjy	bc049090a5	make lazy load thread safe	2013-04-26 12:54:05 +08:00
fxsjy	d2460029d5	merge lazy load	2013-04-26 09:57:06 +08:00
Herman Schaaf	c6098a8657	Add initialize function and lazy initialization	2013-04-25 21:04:56 +09:00
fxsjy	47d94a13e6	log(1)==0, since we have changed from PRODUCT to sum of LOG	2013-04-25 10:11:04 +08:00
fxsjy	c350fab2b9	fix wrong line number	2013-04-25 09:28:00 +08:00
fxsjy	65b78b2b4d	read() and then split -- faster; from __future__ import with	2013-04-24 22:14:10 +08:00
Neuron Teckid	166c2ca7a5	auto close file; locate error when failing to parse	2013-04-24 19:01:08 +08:00
fxsjy	3f003e2f29	new method: jieba.disable_parallel, which is the inverse operation of jieba.enable_parallel	2013-04-22 12:35:17 +08:00
fxsjy	b46166f768	use CRLF as seperator to make chunks in parallel mode	2013-04-20 18:46:04 +08:00
fxsjy	62cf22121f	new feature: parallel segment with multiprocessing	2013-04-20 14:11:31 +08:00
Sun Junyi	6da857b554	merge changes from master branch	2013-04-19 10:21:34 +08:00
Sun Junyi	012fddf13f	ignore white space	2013-04-12 22:37:53 +08:00
fxsjy	45591bb9ab	support flag '_'; ignore white space	2013-04-12 21:53:03 +08:00
Sun Junyi	c77823aa1d	merge improvement to Py3k branch	2013-04-12 14:58:25 +08:00
Sun Junyi	a383f035ba	support decimal point: example PI=3.141569 = > PI / = / 3.14159	2013-04-08 09:38:49 +08:00
Sun Junyi	659326c4e1	punctuation; improve keywords extraction	2013-04-06 14:02:11 +08:00
Sun Junyi	8e49199993	keep punctuation marks	2013-04-05 21:48:36 +08:00
Sun Junyi	58c363655c	support user defined word tag	2013-03-25 17:28:37 +08:00
Sun Junyi	44e19a2e27	fix bug in pypy	2013-03-22 15:20:19 +08:00
Sun Junyi	0f4f9067c3	fix bugs in jieba for py3k	2013-03-21 11:10:57 +08:00
Sun Junyi	d58402c8f6	for issue 26	2013-02-18 10:31:20 +08:00
Sun Junyi	981d58e106	for issue 26	2013-02-18 10:20:17 +08:00
Sun Junyi	1edc1651ee	try to fix this issue: https://github.com/fxsjy/jieba/issues/26	2013-02-17 16:04:51 +08:00
Sun Junyi	fd20cbbd4b	use logarithmic addition instead of multiplication, to avoid bad case in issue19	2012-12-28 11:29:51 +08:00
Sun Junyi	379cd4933a	support en-chn mixed words, like B超	2012-12-12 11:03:29 +08:00
Sun Junyi	9c07d80edb	first py3k version of jieba	2012-11-28 10:50:40 +08:00
Sun Junyi	5ce72e76b1	add new method: cut_for_search(sentence), which can get better recall rate for search engine's reverse index	2012-11-27 13:37:40 +08:00
Sun Junyi	80bf2fec30	Merge branch 'master' of https://github.com/fxsjy/jieba	2012-11-23 16:01:25 +08:00
Sun Junyi	400889b25c	enhance cut_all=True mode	2012-11-23 15:59:15 +08:00
Felix Yan	085b09c3ea	add file-like object support	2012-11-21 18:07:19 +08:00
Sun Junyi	193bfee1d4	use only one dictionary	2012-11-06 11:01:31 +08:00
fxsjy	90cd4b3014	improve POS tagging	2012-11-06 07:17:26 +08:00
Sun Junyi	d040e92987	new interface: load_userdict(file_name)	2012-10-25 17:06:39 +08:00
Sun Junyi	14faea710b	use file cache to improve the loading speed after the first time of importing	2012-10-25 12:18:33 +08:00
fxsjy	ef0c0284ff	improve speed	2012-10-09 06:37:01 +08:00
fxsjy	9180b90ae3	make model loading more faster	2012-10-06 18:28:52 +08:00
fxsjy	164b782c4e	improve the speed	2012-10-04 13:10:56 +08:00
fxsjy	51765aa6dd	first commit	2012-10-01 15:25:06 +08:00
Sun Junyi	6f6e812afb	first commit	2012-09-29 15:54:04 +08:00