Released v1.3.4，解決 MP_ArticutAPI 封裝後 Process Runtime 出錯的問題

Author: EclairCJ

ArticutAPI/MP_ArticutAPI.py

@@ -1,15 +1,6 @@
 #!/usr/bin/env python3
 # -*- coding:utf-8 -*-
 
-import os
-import re
-import time
-
-try:
-    import rapidjson as json
-except:
-    import json
-
 try:
     import requests
 except Exception as e:
@@ -23,9 +14,14 @@
     else:
         print(e)
 
-from multiprocessing import cpu_count, Pool
+from pathlib import os, sys
 from pprint import pprint
 
+if getattr(sys, 'frozen', False):
+    from multiprocessing.dummy import Pool, freeze_support
+else:
+    from multiprocessing import Pool, freeze_support
+
 try:
     from Toolkit.analyse import AnalyseManager
     from Toolkit.localRE import TaiwanAddressAnalizer
@@ -37,18 +33,28 @@
     from .Toolkit.toolkits import *
     from .Toolkit.NER import GenericNER
 
+import time
+try:
+    import rapidjson as json
+except:
+    import json
 
 class MP_Articut:
-    def __init__(self, url="http://127.0.0.1", port="8964", bulkSize=20, userDefinedDictFILE=None):
+    def __init__(self, url="http://127.0.0.1", port="8964", bulkSize=20, userDefinedDictFILE=None, processes=-1):
+        freeze_support()
         self.port = port
         if url.startswith("http"):
             self.url = "{}:{}".format(url, port)
         else:
             self.url = "http://{}:{}".format(url, port)
         self.bulkSize = bulkSize
-        self.processes = cpu_count()
-        if self.processes > 8:
-            self.processes = 8
+        defaultProcesses = os.cpu_count()
+        if processes > 0 and processes <= defaultProcesses:
+            self.processes = processes
+        else:
+            self.processes = defaultProcesses
+            if self.processes > 8:
+                self.processes = 8
 
         self.userDefinedDictFILE = None
         self.openDataPlaceAccessBOOL=False

ArticutAPI/Toolkit/toolkits.py

@@ -62,7 +62,6 @@ def _segIndexConverter(self, parseResultDICT, posIndexLIST):
                 if posLIST:
                     tmpLIST = []
                     for start, end, seg in posLIST:
-
                         posEndSTR = parseResultDICT["result_pos"][i][:start]
                         segEndSTR = "".join([x.group() for x in self.stripPat.finditer(posEndSTR)])
                         tmpLIST.append((len(segEndSTR), len(segEndSTR)+len(seg), seg))

setup.py

@@ -5,7 +5,7 @@
 
 setuptools.setup(
     name="ArticutAPI",
-    version="1.3.3",
+    version="1.3.4",
     author="Droidtown Linguistic Tech. Co. Ltd.",
     author_email="info@droidtown.co",
     description="Articut NLP system provides not only finest results on Chinese word segmentaion (CWS), Part-of-Speech tagging (POS) and Named Entity Recogintion tagging (NER), but also the fastest online API service in the NLP industry.",