from logline_leviathan.exporter.export_constructor import generate_dataframe
import re
import pandas as pd
def create_regex_pattern_from_entity(entity):
words = entity.split()
regex_pattern = "|".join(re.escape(word) for word in words)
return re.compile(regex_pattern, re.IGNORECASE)
def highlight_entities_in_context(context, entity_regex):
def replace_match(match):
return f"{match.group()}"
return re.sub(entity_regex, replace_match, context)
def generate_html_file(output_file_path, db_session, checkboxes, files, context_selection, only_crossmatches, start_date=None, end_date=None, include_flagged=False, only_flagged=False, only_unflagged=False):
# Fetch data using the new DataFrame constructor
df = generate_dataframe(db_session, checkboxes, files, context_selection, only_crossmatches, start_date, end_date, include_flagged, only_flagged, only_unflagged)
# Add line breaks for HTML formatting where needed
if context_selection == 'Kompakte Zusammenfassung ohne Kontext':
df['Sources'] = df['Sources'].apply(lambda x: x.replace(' // ', ' //
'))
df['Timestamp'] = df['Timestamp'].apply(lambda x: x.replace(' // ', ' //
'))
# Iterate over the DataFrame to apply regex-based highlighting
for index, row in df.iterrows():
entity_regex = create_regex_pattern_from_entity(row['Entity'])
df.at[index, 'Context'] = highlight_entities_in_context(row['Context'], entity_regex)
# Replace newline characters with HTML line breaks in the 'Context' column
df['Context'] = df['Context'].apply(lambda x: x.replace('\n', '
') if x else x)
# Convert DataFrame to HTML table
html_table = df.to_html(classes="table table-bordered", escape=False, index=False)
html_template = f"""
Logline Leviathan Report
{html_table}
"""
# Write the HTML template to the file
with open(output_file_path, 'w', encoding='utf-8') as file:
file.write(html_template)