#!/usr/bin/env python3 # -*- coding: utf-8 -*- """ .py """ __authors__ = "Stephane Schneider" __contact__ = "stephane.schneider@inis.fr" from cso_classifier import CSOClassifier import sys import json import os # Version EZ_MASTER de l indexeur memoire #nombre de workers workers="8" # parametrage de l indexeur # Additional parameters: # (i) workers, => nbre de process en // # (ii) modules, => module d indexation à executer : "syntactic", "semantic", "both" # (iii) enhancement, => relevant super-areas, "first", "all", or "no". # (iv) explanation, => chunks of text that allowed the classifier to infer a given topic. # (v) delete_outliers, => emoving erroneous topics that were conceptually distant from the others. # (vi) fast_classification, => determines whether the semantic module should use the full model or the cached one. # (vii) silent. => prints its progress in the console. indexer = CSOClassifier( modules="both", enhancement="first", explanation=True, delete_outliers=False, # False pour que ca marche sinon error : silent=True, fast_classification=True, workers=workers ) # lecture ligne a ligne sur input standard # 1 ligne = un json {} compteur = 0 for json_line in sys.stdin: # deserialisation json compteur += 1 try: data = json.loads(json_line) # trace # print("******** FORMAT OK : id:{}".format(data["id"])) except json.decoder.JSONDecodeError: logging.error("Input format problem line :{} : String could not be converted to JSON".format(compteur)) exit(1) data["value"] = indexer.run(data["value"]) sys.stdout.write(json.dumps(data)) sys.stdout.write("\n")