diff --git a/WhereDoIHaveAnAccount/scraper.py b/WhereDoIHaveAnAccount/scraper.py index e634238..4e8555f 100644 --- a/WhereDoIHaveAnAccount/scraper.py +++ b/WhereDoIHaveAnAccount/scraper.py @@ -124,6 +124,7 @@ def preprocess_data(email_header_df): email_header_df['subject'] = email_header_df['subject'].apply(lambda x: re.sub("ß", "ss", x)) # tokenize + nltk.download('punkt') email_header_df['subject'] = email_header_df['subject'].apply(lambda x: nltk.word_tokenize(x)) # remove stopwords diff --git a/pyproject.toml b/pyproject.toml index 3f20733..d98a991 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -8,7 +8,7 @@ readme = "README.md" authors = [ {name = "Timo Kühne"}, ] -version = "1.4.0" +version = "1.4.1" description = "Simple tool to find out where you have accounts." requires-python = ">=3.6.1" classifiers = [