update

GGNoWayBack · Feb 13, 2024 · 9a3b364 · 9a3b364
1 parent 17cbf42
commit 9a3b364
Show file tree

Hide file tree

Showing 6 changed files with 5 additions and 152 deletions.
diff --git a/cathodedataextractor/cathodetext2chem/__init__.py b/cathodedataextractor/cathodetext2chem/__init__.py
@@ -1,13 +1,4 @@
 # coding=utf-8
-import os
-
-if os.name == "nt":
-    import _locale
-    _locale._gdl_bak = _locale._getdefaultlocale
-    _locale._getdefaultlocale = (lambda *args: (_locale._gdl_bak()[0], 'utf8'))
-
-from .core.formula_parser import *
-
 from .parser_pipeline import CathodeParserPipelineBuilder
 from .regex_parser import CathodeRegExParser
 from .postprocessing_tools import CathodeStoichiometricVariablesProcessing
diff --git a/cathodedataextractor/cathodetext2chem/core/__init__.py b/cathodedataextractor/cathodetext2chem/core/__init__.py
diff --git a/cathodedataextractor/cathodetext2chem/core/formula_parser.py b/cathodedataextractor/cathodetext2chem/core/formula_parser.py
diff --git a/cathodedataextractor/information_extraction_pipe.py b/cathodedataextractor/information_extraction_pipe.py
@@ -103,7 +103,8 @@ def extract(self, path: str):
                     layer3=pp.results,
                     layer4=final_records)
 
-    def _collect_corpus(self, head, tail) -> PipelineData:
+    @staticmethod
+    def _collect_corpus(head, tail) -> PipelineData:
 
         TC = TagClassificationPar2Text()
         TC.get_abstract(head=head, tail=tail)
@@ -119,7 +120,7 @@ def _collect_corpus(self, head, tail) -> PipelineData:
     @staticmethod
     def _preprocess_csie(data: PipelineData) -> PipelineData:
         # preprocess
-        year, doi, _intro, _experi, _partial_text = data[:6]
+        year, doi, _intro, _experi, _partial_text = data[:5]
 
         intro = _intro.strip(PARAGRAPH_SEPARATOR)
 

diff --git a/requirements.txt b/requirements.txt
diff --git a/setup.py b/setup.py
@@ -5,7 +5,7 @@
 long_description = open('README.md', encoding="utf8").read()
 setup(
     name='cathodedataextractor',
-    version='0.0.2',
+    version='0.0.3',
     description='A document-level information extraction pipeline for layered cathode materials for sodium-ion batteries.',
     long_description=long_description,
     long_description_content_type='text/markdown',
@@ -17,7 +17,7 @@
         'word2number',
         'pdfminer.six >=20160614, <=20221105 ; python_version < "3.8"',
         'chemdataextractor2==2.2.2',
-        'text2chem==0.0.2',
+        'text2chem==0.0.3',
         'pymatgen',
     ],
     license='MIT',