def __init__(self, max_term_count=None, dict_path=None):
if max_term_count:
self.m_max_term_count = max_term_count
else:
self.m_max_term_count = default_max_term_count
if not WordSeg.m_dict_loaded:
if dict_path:
WordSeg.m_dict_path = dict_path
else:
WordSeg.m_dict_path = default_dict_path
print WordSeg.m_dict_path
WordSeg.m_dict_handle = wordseg.scw_load_worddict(
WordSeg.m_dict_path)
if WordSeg.m_dict_handle:
WordSeg.m_dict_loaded = True
self.m_result_handle = wordseg.scw_create_out(self.m_max_term_count *
10)
self.m_token_handle = wordseg.create_tokens(self.m_max_term_count)
self.m_token_handle = wordseg.init_tokens(self.m_token_handle,
self.m_max_term_count)
self.m_mode['WPCOMP'] = wordseg.SCW_WPCOMP
self.m_mode['BASIC'] = wordseg.SCW_BASIC
self.m_mode['SUBPH'] = wordseg.SCW_SUBPH
self.m_mode['NEWWORD'] = wordseg.SCW_NEWWORD
self.m_mode['HUMAN'] = wordseg.SCW_HUMANNAME
self.m_mode['BOOK'] = wordseg.SCW_BOOKNAME
self.m_mode['DISAMB'] = wordseg.SCW_DISAMB
def __init__(self, dict_path):
print >> sys.stderr, 'WordSegUtil constructed'
self.MAX_TERM_CNT = 2048
self.scw_worddict = wordseg.scw_load_worddict(
os.path.join(dict_path, 'wordseg/chinese_gbk'))
self.scw_tagdict = postag.tag_create(os.path.join(dict_path, 'postag'))
self.scw_out = wordseg.scw_create_out(self.MAX_TERM_CNT * 10)
# token
self.tokens = wordseg.create_tokens(self.MAX_TERM_CNT)
self.tokens = wordseg.init_tokens(self.tokens, self.MAX_TERM_CNT)
def __init__(self, segdict_conf_path, segdict_path):
'''
@brief 构造函数
@param segdict_conf_path 配置文件位置
@param segdict_path 切词库位置
@param tagdict_path 词性库位置
'''
self.max_term_count = 512
try:
self.conf_handle = wordseg.scw_load_conf(segdict_conf_path)
self.dict_handle = wordseg.scw_load_worddict(segdict_path)
self.result_handle = wordseg.scw_create_out(self.max_term_count*10)
self.token_handle = wordseg.create_tokens(self.max_term_count)
self.token_handle = wordseg.init_tokens(self.token_handle, self.max_term_count)
except Exception as e:
log.warning("SegDict Load Error! error=%s",e);
def __init__(self):
print >> sys.stderr, "WordSegUtil constructed"
self.maxTermCount = 2048
dict_ab_url = (os.path.dirname(os.path.abspath(__file__))) + "/dict"
#print dict_ab_url
# 加载词典
#print os.path.join(dict_ab_url, "worddict")
self.hWordDict = wordseg.scw_load_worddict(
os.path.join(dict_ab_url, "worddict"))
self.hTagDict = postag.tag_create(os.path.join(dict_ab_url, "tagdict"))
# hNerDict = wordner.ner_dict_load(os.path.join(dict_ab_url, "nerdict"))
self.hRankDict = wordrank.wdr_create(
os.path.join(dict_ab_url, "rankdict"))
self.hScwOut = wordseg.scw_create_out(self.maxTermCount * 10)
# hNerOut = wordner.ner_out_create(hNerDict, self.maxTermCount)
self.hRanks = wordrank.create_ranks(self.maxTermCount)
# token
self.hTokens = wordseg.create_tokens(self.maxTermCount)
self.hTokens = wordseg.init_tokens(self.hTokens, self.maxTermCount)
# 专名过滤
self.nerWhiteTags = set([
"PER", # 人名
#"LOC", # 地名
#"ORG", # 机构
#"SFT", # 软件
"GME", # 游戏
"SNG", # 歌曲
#"NVL", # 小说
"VDO", # 视频
"BRD", # 品牌
"CTN", # 动漫
"VDO_MVE", # 电影
"VDO_TV", # 电视剧
"VDO_TVSHOW" # 电视节目
])
def __init__(self, max_term_count = None, dict_path = None):
if max_term_count:
self.m_max_term_count = max_term_count
else:
self.m_max_term_count = default_max_term_count
if not WordSeg.m_dict_loaded:
if dict_path:
WordSeg.m_dict_path = dict_path
else:
WordSeg.m_dict_path = default_dict_path
print WordSeg.m_dict_path
WordSeg.m_dict_handle = wordseg.scw_load_worddict(WordSeg.m_dict_path)
if WordSeg.m_dict_handle:
WordSeg.m_dict_loaded = True
self.m_result_handle = wordseg.scw_create_out(self.m_max_term_count*10)
self.m_token_handle = wordseg.create_tokens(self.m_max_term_count)
self.m_token_handle = wordseg.init_tokens(self.m_token_handle, self.m_max_term_count)
self.m_mode['WPCOMP'] = wordseg.SCW_WPCOMP
self.m_mode['BASIC'] = wordseg.SCW_BASIC
self.m_mode['SUBPH'] = wordseg.SCW_SUBPH
self.m_mode['NEWWORD'] = wordseg.SCW_NEWWORD
self.m_mode['HUMAN'] = wordseg.SCW_HUMANNAME
self.m_mode['BOOK'] = wordseg.SCW_BOOKNAME
self.m_mode['DISAMB'] = wordseg.SCW_DISAMB
if query == "" or title == "":
continue
query = urllib.unquote(query)
title = urllib.unquote(title)
label = int(line[7])
pos.append([query, title, label])
strip_chars = [".", ",", "-", "_", ":"]
final = []
MAX_TERM_COUNT = 1024
dict_handle = wordseg.scw_load_worddict("./dict/wordseg_dict/")
result_handle = wordseg.scw_create_out(MAX_TERM_COUNT)
token_handle = wordseg.create_tokens(MAX_TERM_COUNT)
token_handle = wordseg.init_tokens(token_handle, MAX_TERM_COUNT)
for query, title, label in pos:
for char in strip_chars:
query = query.strip(char)
title = title.strip(char)
query_title = []
for line in [query, title]:
wordseg.scw_segment_words(dict_handle, result_handle, line, 1)
token_count = wordseg.scw_get_token_1(result_handle,
wordseg.SCW_WPCOMP, token_handle,
MAX_TERM_COUNT)
query_title.append([
token[7]