second forms are saved as separate verbs at the end of the dataframe

main.py

@@ -9,19 +9,19 @@ from tqdm import tqdm
 print("getting verbs...")
 print("getting verbs...")
 verbs = get_verbs_gf()
 verbs = get_verbs_gf()
 print(verbs)
 print(verbs)
-# retrieve liks to postimees articles
-# print("getting postimees urls...")
-# postimees_urls = get_postimees_urls()
+retrieve liks to postimees articles
+print("getting postimees urls...")
+postimees_urls = get_postimees_urls()
 
 
-# print("extracting text from the urls...")
-# articles = []
-# for i in tqdm(range(len(postimees_urls))):
-# 	url = postimees_urls[i]
-# 	articles.append(get_text(url))
+print("extracting text from the urls...")
+articles = []
+for i in tqdm(range(len(postimees_urls))):
+	url = postimees_urls[i]
+	articles.append(get_text(url))
 
 
-# # try to find a verb in an article
-# for column in verbs:
-# 	verb_form = verbs.iloc[2][column]
-# 	if type(verb_form) is str:
-# 		print(verb_form)
-# 		print(str(articles[0].find(verb_form)))
+# try to find a verb in an article
+for column in verbs:
+	verb_form = verbs.iloc[2][column]
+	if type(verb_form) is str:
+		print(verb_form)
+		print(str(articles[0].find(verb_form)))
\ No newline at end of file

preprocessing.py

@@ -11,18 +11,14 @@ def read_csv():
 	df = pd.read_csv("verbs_gf.csv", sep=", ", encoding='utf8', header=None, engine='python')
 	df = pd.read_csv("verbs_gf.csv", sep=", ", encoding='utf8', header=None, engine='python')
 	return df
 	return df
 
 
-
-def double(column):
-	new_column = column.str.split('|', expand=True)
+def split_double(df):
+	for i in range(len(df.index)):
+		row = df.iloc[i]
+		split_row = row.str.split('|', expand=True)
 		try:
 		try:
-		column = new_column[0]
-		new_column = new_column[1]
+			second_form = split_row[1]
+			second_form[second_form.isnull()] = split_row[0]
+			df = df.append(second_form, ignore_index=True)
 		except Exception as e:
 		except Exception as e:
-		new_column = [None]*len(new_column)
-	return column, new_column
-
-def split_double(df):
-	for column_name in df.columns:
-		second_column_name = str(column_name) + "double"
-		df[column_name], df[second_column_name] = double(df[column_name])
+			pass
 	return df
 	return df
\ No newline at end of file