From 696d47875550c0ea04d98dbd870b7afd633693bb Mon Sep 17 00:00:00 2001
From: Fize Jacques <jacques.fize@cirad.fr>
Date: Tue, 15 Oct 2019 10:54:58 +0200
Subject: [PATCH] simplify wrapper
---
biotex/biotex_wrapper.py | 99 +++++++++++++++++++++++++---------------
1 file changed, 62 insertions(+), 37 deletions(-)
diff --git a/biotex/biotex_wrapper.py b/biotex/biotex_wrapper.py
index 9a5de7c..6aa15d4 100644
--- a/biotex/biotex_wrapper.py
+++ b/biotex/biotex_wrapper.py
@@ -1,53 +1,68 @@
+
+
import os,glob,shutil,sys,time
from pathlib import Path
+
+import pandas as pd
+
home = str(Path.home())
+
+
class BiotexWrapper():
"""
Wrapper to execute and returned the result from the Biotex program
See more about Biotex at: http://tubo.lirmm.fr:8080/biotex/index.jsp
"""
- def __init__(self,**kwargs):
- self.biotexJarPath=kwargs.get("biotex_jar_path",os.path.join(os.path.dirname(__file__),"Biotex.jar"))
-
- self.configuration=kwargs.get("configuration",{
- "patternsSrc":kwargs.get("pattern_path",os.path.join(os.path.dirname(__file__),"patterns")),
- "datasetSrc":kwargs.get("datasetRef_path",os.path.join(os.path.dirname(__file__),"dataSetReference")),
- "stopwordsSrc":kwargs.get("stop_words_path",os.path.join(os.path.dirname(__file__),"stopWords")),
- "treetaggerSrc":kwargs.get("treetagger_bin_path",os.path.join(home,".tree-tagger/")),
- "typeOfTerms":kwargs.get("type_of_terms","all"),
- "language":kwargs.get("lang","french"),
- "score":kwargs.get("score","F-TFIDF-C_M"),
- "patronNumber":kwargs.get("patronNumber",3)
- })
-
- self.write_conf(self.configuration)
-
- self.output_data=None
-
-
- def create_corpus_from_txt_files(self,list_of_text_filename : list, outputfn="output.txt"):
+ def __init__(self,
+ biotex_jar_path = os.path.join(os.path.dirname(__file__),"Biotex.jar"),
+ pattern_path = os.path.join(os.path.dirname(__file__),"patterns"),
+ dataset_src = os.path.join(os.path.dirname(__file__),"dataSetReference") ,
+ stopwords_src = os.path.join(os.path.dirname(__file__),"stopWords"),
+ treetagger_src = os.path.join(home,".tree-tagger/"),
+ type_of_terms = "all",
+ language = "french",
+ score = "F-TFIDF-C_M",
+ patron_number = "3"):
"""
- Format corpus made of multiple file into Biotex Corpus Format
+ Constructor
Parameters
----------
- list_of_text_filename : list
- list of filenames ot the corpus
- outputfn : str, optional
- output filename (the default is "output.txt")
-
- Returns
- -------
- str
- output filename
+ biotex_jar_path : str, optional
+ Filepath of Biotex jar, by default os.path.join(os.path.dirname(__file__),"Biotex.jar")
+ pattern_path : str, optional
+ Directory that contains pre-defined patterns, by default os.path.join(os.path.dirname(__file__),"patterns")
+ dataset_src : src, optional
+ FilePath of datasets used by Biotex, by default os.path.join(os.path.dirname(__file__),"dataSetReference")
+ stopwords_src : str, optional
+ Path of the directory that contains stopwords for each language, by default os.path.join(os.path.dirname(__file__),"stopWords")
+ treetagger_src : str, optional
+ Path of the directory that contains TreeTagger, by default os.path.join(home,".tree-tagger/")
+ type_of_terms : str, optional
+ number of terms you want to extract, by default "all"
+ language : str, optional
+ language of the data, by default "french"
+ score : str, optional
+ score used to sort the extracted term, by default "F-TFIDF-C_M"
+ patron_number : str, optional
+ number of pattern used to extract terms, by default "3"
"""
- corpus=[]
- sep="\n##########END##########\n"
- for file in list_of_text_filename:
- corpus.append(open(file,'r').read())
- open(outputfn,'w').write((sep.join(corpus)+sep).strip())
- return outputfn
+
+ self.biotexJarPath=biotex_jar_path
+ self.configuration={
+ "patternsSrc":pattern_path,
+ "datasetSrc":dataset_src,
+ "stopwordsSrc":stopwords_src,
+ "treetaggerSrc":treetagger_src,
+ "typeOfTerms":type_of_terms,
+ "language":language,
+ "score":score,
+ "patronNumber":patron_number
+ }
+
+ self.write_conf(self.configuration)
+ self.output_data=None
def create_corpus_from_txt(self,list_of_text : list, outputfn="output.txt"):
"""
@@ -115,14 +130,24 @@ class BiotexWrapper():
parsed[2]=float(parsed[2])
data.append(parsed)
shutil.rmtree('output')
+ for f in glob.glob("to_tag_*.txt"):
+ os.remove(f)
self.output_data=data
return self.output_data
+ def terminology(self, corpus):
+ try:
+ self.create_corpus_from_txt(corpus)
+ except:
+ raise Exception("Error while creating file !")
+ return pd.DataFrame(self.extract_terminology("output.txt"), columns = "term in_umls rank".split())
+
+
if __name__ == '__main__':
import argparse
parser= argparse.ArgumentParser()
parser.add_argument("input",help="Your Biotex input filename")
- parser.add_argument('-s',"--sizeOfGram",help="Gram size of the term you want to extreact")
+ parser.add_argument('-s',"--sizeOfGram",help="Gram size of the term you want to extract")
parser.add_argument('-o',"--output",help="Output filename")
parser.add_argument('-d',"--debug",action="store_true",help="debug activated")
--
GitLab