def load_terms(df: DataFrame) -> int:
df.drop_duplicates(inplace=True)
df.loc[df["Case Sensitive"] == False, "Term"] = df.loc[
df["Case Sensitive"] == False, "Term"].str.lower()
df = df.drop_duplicates(subset="Term").dropna(subset=["Term"])
terms = []
for row_id, row in df.iterrows():
term = row["Term"].strip()
if not Term.objects.filter(term=term).exists():
lt = Term()
lt.term = term
lt.source = row["Term Category"]
lt.definition_url = row["Term Locale"]
terms.append(lt)
# cache "global" term stems step - should be cached here via model manager
Term.objects.bulk_create(terms)
# update existing ProjectTermConfiguration objects for all projects across loaded terms
from apps.extract.dict_data_cache import cache_term_stems
from apps.project.models import ProjectTermConfiguration
for config in ProjectTermConfiguration.objects.all():
cache_term_stems(config.project_id)
config.add(terms)
return len(df)
def _prepare_term(
dataframe_row: Series,
term_creation_mode: _TermCreationMode) -> Union[Term, None]:
"""Instantiates and returns a new Term objects based on input data.
Args:
dataframe_row (pandas.Series): Input data for Term creation.
term_creation_mode (_TermCreationMode): A logical flag; determines return behavior.
Returns:
new_db_term (Term): A new Term object from the input row.
None: if term_creation_mode is IGNORE.
"""
term = dataframe_row['Term'].strip()
qs_terms_of_this_term = Term.objects.filter(term=term)
# instantiate a new Term
new_db_term = Term(term=term,
source=dataframe_row['Term Category'],
definition_url=dataframe_row['Term Locale'])
# handle term creation modes
if qs_terms_of_this_term.exists():
if term_creation_mode == _TermCreationMode.REPLACE:
qs_terms_of_this_term.delete()
elif term_creation_mode == _TermCreationMode.IGNORE:
return None
return new_db_term
def load_terms(df: DataFrame) -> int:
df.drop_duplicates(inplace=True)
df.loc[df["Case Sensitive"] == False,
"Term"] = df.loc[df["Case Sensitive"] == False, "Term"].str.lower()
df = df.drop_duplicates(subset="Term").dropna(subset=["Term"])
terms = []
for row_id, row in df.iterrows():
term = row["Term"].strip()
if not Term.objects.filter(term=term).exists():
lt = Term()
lt.term = term
lt.source = row["Term Category"]
lt.definition_url = row["Term Locale"]
terms.append(lt)
Term.objects.bulk_create(terms)
return len(df)