third_codes/jieba
1
0
Fork 0
mirror of https://github.com/fxsjy/jieba.git synced 2025-07-10 00:01:33 +08:00
Code Issues Packages Projects Releases Wiki Activity

use logging instead of print

Browse Source
This commit is contained in:
Mozillazg 2013-09-19 10:31:44 +08:00
parent fd96527f71
commit 1cf3f0d00b
1 changed files with 12 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

23
jieba/__init__.py
View File

@ -13,6 +13,7 @@ from math import log
import random import random
import threading import threading
from functools import wraps from functools import wraps
import logging
DICTIONARY = "dict.txt" DICTIONARY = "dict.txt"
DICT_LOCK = threading.RLock() DICT_LOCK = threading.RLock()
@ -22,6 +23,7 @@ min_freq = 0.0
total =0.0 total =0.0
user_word_tag_tab={} user_word_tag_tab={}
initialized = False initialized = False
logger = logging.getLogger(__name__)
def gen_trie(f_name): def gen_trie(f_name):
lfreq = {} lfreq = {}
@ -43,8 +45,8 @@ def gen_trie(f_name):
p = p[c] p = p[c]
p['']='' #ending flag p['']='' #ending flag
except ValueError, e: except ValueError, e:
print >> sys.stderr, f_name, ' at line', lineno, line logger.debug('%s at line %s %s' % (f_name, lineno, line))
raise e raise ValueError, e
return trie, lfreq,ltotal return trie, lfreq,ltotal
def initialize(*args): def initialize(*args):
@ -62,7 +64,7 @@ def initialize(*args):
_curpath=os.path.normpath( os.path.join( os.getcwd(), os.path.dirname(__file__) ) ) _curpath=os.path.normpath( os.path.join( os.getcwd(), os.path.dirname(__file__) ) )
abs_path = os.path.join(_curpath,dictionary) abs_path = os.path.join(_curpath,dictionary)
print >> sys.stderr, "Building Trie..., from " + abs_path logger.debug("Building Trie..., from %s" % abs_path)
t1 = time.time() t1 = time.time()
if abs_path == os.path.join(_curpath,"dict.txt"): #defautl dictionary if abs_path == os.path.join(_curpath,"dict.txt"): #defautl dictionary
cache_file = os.path.join(tempfile.gettempdir(),"jieba.cache") cache_file = os.path.join(tempfile.gettempdir(),"jieba.cache")
@ -71,7 +73,7 @@ def initialize(*args):
load_from_cache_fail = True load_from_cache_fail = True
if os.path.exists(cache_file) and os.path.getmtime(cache_file)>os.path.getmtime(abs_path): if os.path.exists(cache_file) and os.path.getmtime(cache_file)>os.path.getmtime(abs_path):
print >> sys.stderr, "loading model from cache " + cache_file logger.debug("loading model from cache %s" % cache_file)
try: try:
trie,FREQ,total,min_freq = marshal.load(open(cache_file,'rb')) trie,FREQ,total,min_freq = marshal.load(open(cache_file,'rb'))
load_from_cache_fail = False load_from_cache_fail = False
@ -82,7 +84,7 @@ def initialize(*args):
trie,FREQ,total = gen_trie(abs_path) trie,FREQ,total = gen_trie(abs_path)
FREQ = dict([(k,log(float(v)/total)) for k,v in FREQ.iteritems()]) #normalize FREQ = dict([(k,log(float(v)/total)) for k,v in FREQ.iteritems()]) #normalize
min_freq = min(FREQ.itervalues()) min_freq = min(FREQ.itervalues())
print >> sys.stderr, "dumping model to file cache " + cache_file logger.debug("dumping model to file cache %s" % cache_file)
try: try:
tmp_suffix = "."+str(random.random()) tmp_suffix = "."+str(random.random())
with open(cache_file+tmp_suffix,'wb') as temp_cache_file: with open(cache_file+tmp_suffix,'wb') as temp_cache_file:
@ -94,21 +96,20 @@ def initialize(*args):
replace_file = os.rename replace_file = os.rename
replace_file(cache_file+tmp_suffix,cache_file) replace_file(cache_file+tmp_suffix,cache_file)
except: except:
print >> sys.stderr, "dump cache file failed." logger.error("dump cache file failed.")
import traceback logger.exception("")
print >> sys.stderr, traceback.format_exc()
initialized = True initialized = True
print >> sys.stderr, "loading model cost ", time.time() - t1, "seconds." logger.debug("loading model cost %s seconds." % (time.time() - t1))
print >> sys.stderr, "Trie has been built succesfully." logger.debug("Trie has been built succesfully.")
def require_initialized(fn): def require_initialized(fn):
@wraps(fn) @wraps(fn)
def wrapped(*args, **kwargs): def wrapped(*args, **kwargs):
global initialized global initialized
if initialized: if initialized:
return fn(*args, **kwargs) return fn(*args, **kwargs)
else: else: