added demo

.streamlit/config.toml

@@ -0,0 +1,5 @@
+[theme]
+primaryColor = "#F3C65B"
+backgroundColor = "#222222"
+secondaryBackgroundColor = "#333333"
+textColor = "#FFFFFF"

streamlit/.streamlit/config.toml

@@ -0,0 +1,5 @@
+[theme]
+primaryColor = "#F3C65B"
+backgroundColor = "#222222"
+secondaryBackgroundColor = "#333333"
+textColor = "#FFFFFF"

streamlit/utils.py

@@ -0,0 +1,138 @@
+import streamlit as st
+import pandas as pd
+from collections import defaultdict
+import re
+import json
+import ast
+import geopandas as gpd
+import io
+import os
+import pydeck as pdk
+import plotly.express as px
+from datetime import datetime
+import glob
+
+def clean_and_split(industries):
+ if isinstance(industries, list):
+ return [industry.strip("[]' ") for industry in industries]
+ elif isinstance(industries, str):
+ clean_str = industries.strip("[]' ")
+ return clean_str
+ else:
+ return industries
+
+def format_description(descriptions):
+ if descriptions:
+ bullet_point = "•"
+ return "\n".join(f"{bullet_point} {desc}" for desc in descriptions.split(','))
+ else:
+ return ""
+
+def calculate_fill_color(count, max_count):
+ return [
+ 255 - 52 * count / max_count,
+ 191 - 167 * count / max_count,
+ 0 + 29 * count / max_count
+ ]
+
+def rewrite_date(date_str):
+ date_obj = datetime.strptime(date_str, "%Y-%m-%d")
+ ordinal = lambda n: "%d%s" % (n, "th" if 4 <= n <= 20 or 24 <= n <= 30 else {1: "st", 2: "nd", 3: "rd"}.get(n % 10, "th"))
+ formatted_date = date_obj.strftime(f"{ordinal(date_obj.day)} %B %Y")
+ return formatted_date
+
+def clean_and_split_str(industries_str):
+ clean_str = industries_str.strip("[]' ")
+ if clean_str != 'None':
+ return [item.strip() for item in clean_str.split("', '")]
+ else:
+ return None
+
+def clean_risk_df(vis_df):
+ vis_df['Industry'] = vis_df['Industry'].apply(clean_and_split_str)
+ vis_df = vis_df[vis_df['Risk Description'].notna() & vis_df['Risk Impact'].astype(str).apply(lambda x: x.strip("[]' ") != '') & vis_df["Pass"]]
+ vis_df = vis_df.drop('Pass', axis=1).reset_index(drop=True)
+ return vis_df
+
+def standardize_country_names(countries):
+ standardized_countries = []
+ for country in countries:
+ if country in ['UK', 'United Kingdom', 'Great Britain', 'England', 'Wales', 'Scotland', 'Ireland', 'global', 'La Serra', 'European', 'Europe', 'European Union' ]:
+ pass
+ elif country == 'USA' or country == 'United States' or country == 'US':
+ standardized_countries.append('United States of America')
+ elif country == 'Czech Republic':
+ standardized_countries.append('Czechia')
+ elif country == 'Republic of Cyprus':
+ standardized_countries.append('Cyprus')
+ elif country == 'Gaza':
+ standardized_countries.append('Palestine')
+ else:
+ standardized_countries.append(country)
+ return standardized_countries
+
+def get_risk_count_countries(df):
+ df['Risk Countries'] = df['Risk Countries'].apply(ast.literal_eval)
+ df['Risk Countries'] = df['Risk Countries'].apply(standardize_country_names)
+ df = df.explode('Risk Countries').reset_index(drop=True)
+ risk_counts = df.groupby(['Risk Countries', 'Risk Type']).size().reset_index(name='Count')
+ risk_counts_pivot = risk_counts.pivot_table(index='Risk Countries', columns='Risk Type', values='Count', fill_value=0).reset_index()
+ for col in risk_counts_pivot.columns[1:]: # Exclude the 'Risk Countries' column
+ risk_counts_pivot[col] = risk_counts_pivot[col].astype(int)
+ risk_counts_pivot['Total Count'] = risk_counts_pivot.iloc[:, 1:].sum(axis=1)
+ risk_counts_pivot['Risk Descriptions'] = risk_counts_pivot.apply(lambda row: ", ".join([f"{col} ({row[col]})" for col in risk_counts_pivot.columns[1:-1] if row[col] > 0]), axis=1)
+ final_df = risk_counts_pivot[['Risk Countries', 'Risk Descriptions', 'Total Count']]
+ return final_df
+
+def get_risk_count_regions(df):
+ region_risk_counts = df.groupby(['European Electoral Region', 'Risk Type']).size().reset_index(name='Count')
+ region_risk_counts_pivot = region_risk_counts.pivot_table(index='European Electoral Region', columns='Risk Type', values='Count', fill_value=0).reset_index()
+ for col in region_risk_counts_pivot.columns[1:]:
+ region_risk_counts_pivot[col] = region_risk_counts_pivot[col].astype(int)
+ region_risk_counts_pivot['Total Count'] = region_risk_counts_pivot.iloc[:, 1:].sum(axis=1)
+ region_risk_counts_pivot['Risk Descriptions'] = region_risk_counts_pivot.apply(lambda row: ", ".join([f"{col} ({row[col]})" for col in region_risk_counts_pivot.columns[1:-1] if row[col] > 0]), axis=1)
+ final_df = region_risk_counts_pivot[['European Electoral Region', 'Risk Descriptions', 'Total Count']]
+ return final_df
+
+def find_most_recent_folder(folders):
+ most_recent_date = None
+ most_recent_folder = None
+ for folder in folders:
+ try:
+ folder_date = datetime.strptime(folder, '%Y-%m-%d')
+ except ValueError:
+ continue
+ if most_recent_date is None or folder_date > most_recent_date:
+ most_recent_date = folder_date
+ most_recent_folder = folder
+ return most_recent_folder
+
+def combine_csv_files(file_pattern):
+ file_list = glob.glob(file_pattern, recursive=True)
+ dfs = []
+ for file in file_list:
+ df = pd.read_csv(file)
+ dfs.append(df)
+ combined_df = pd.concat(dfs, ignore_index=True)
+ return combined_df
+
+def find_date_range(df):
+ df['Datetime'] = pd.to_datetime(df['Date'], format='mixed')
+ most_recent_date = df['Datetime'].max().strftime('%Y-%m-%d')
+ oldest_date = df['Datetime'].min().strftime('%Y-%m-%d')
+
+ return most_recent_date, oldest_date
+
+def preprocess_df(vis_df):
+ vis_df['Industry'] = vis_df['Industry'].apply(clean_and_split)
+ vis_df = vis_df[vis_df['Risk Description'].notna() & vis_df['Risk Impact'].astype(str).apply(lambda x: x.strip("[]' ") != '')]
+ industry_counts = defaultdict(int)
+ for industries in vis_df['Industry'].dropna():
+ for industry in industries:
+ if len(industry.split(' ')) > 3:
+ industry = f"{' '.join(industry.split(' ')[:3])}"
+ if industry[-1] == ";":
+ industry = industry[0:-1]
+ industry = f'{industry}...'
+ industry_counts[industry] += 1
+ return vis_df, industry_counts