Create CSV_UI.py

CSV_UI.py

@@ -0,0 +1,88 @@
+import streamlit as st
+import pandas as pd
+import numpy as np
+
+def highlight_diff1(row, df1, df2, primary_key):
+    styles = []
+    name = row.name
+    if name in df1.index and name in df2.index:
+        index1 = df1.index.get_loc(name)
+        index2 = df2.index.get_loc(name)
+        for col in df1.columns:
+            if col != primary_key and df1.iloc[index1][col] != df2.iloc[index2][col]:
+                styles = ['background-color: green'] * len(row)
+                break
+    elif name in df1.index:
+        styles = ['background-color: yellow'] * len(row)
+    return styles or [''] * len(row)
+
+def highlight_diff2(row, df1, df2, primary_key):
+    styles = []
+    name = row.name
+    if name in df1.index and name in df2.index:
+        index1 = df1.index.get_loc(name)
+        index2 = df2.index.get_loc(name)
+        for col in df2.columns:
+            if col != primary_key and df1.iloc[index1][col] != df2.iloc[index2][col]:
+                styles = ['background-color: red'] * len(row)
+                break
+    elif name in df2.index:
+        styles = ['background-color: yellow'] * len(row)
+    return styles or [''] * len(row)
+
+def main():
+    df1 = pd.DataFrame({
+        'Name': ['John', 'Alice', 'Bob', 'Mad'],
+        'Age': [25, 30, 22, 77],
+        'Location': ['New York', 'Los Angeles', 'Chicago', 'Noida']
+    })
+
+    df2 = pd.DataFrame({
+        'Name': ['John', 'Alice', 'Bob', 'Panda'],
+        'Age': [25, 30, 43, 27],
+        'Location': ['New York', 'Delhi', 'Chicago', 'London'],
+        'Loc': ['NewYork', 'bkjk', 'Chicdfsdago', 'Lofsfndon']
+    })
+
+    st.set_page_config(layout="wide")
+    st.write("### Input Paths")
+    primary_column = 'Name'
+    common_columns = list(set(df1.columns).intersection(df2.columns))
+    common_cols = [x for x in common_columns if x != primary_column]
+   
+    cols_to_display = st.multiselect("Select Columns to Display", common_cols)
+    columns_to_display = cols_to_display + [primary_column]
+    col1_width = st.sidebar.slider("Width of First Column", 0.1, 10.0, 6.5, 0.1)
+    col2_width = st.sidebar.slider("Width of Second Column", 0.1, 10.0, 6.5, 0.1)
+
+    col1, col2 = st.columns([col1_width, col2_width])
+
+    with col1:
+        st.write("### Display DataFrame 1")
+        if df1 is not None:
+            if len(columns_to_display) > 0:
+                df1_display = df1[columns_to_display].set_index(primary_column)
+                df1_display_styled = df1_display.style.apply(
+                    lambda x: highlight_diff1(x, df1_display, df2.set_index(primary_column), primary_column), axis=1)
+                st.dataframe(df1_display_styled)
+            else:
+                st.write("No columns selected")
+        else:
+            st.write("DataFrame not found")
+
+
+    with col2:
+        st.write("### Display DataFrame 2")
+        if df2 is not None:
+            if len(columns_to_display) > 0:
+                df2_display = df2[columns_to_display].set_index(primary_column)
+                df2_display_styled = df2_display.style.apply(
+                    lambda x: highlight_diff2(x, df1.set_index(primary_column), df2_display, primary_column), axis=1)
+                st.dataframe(df2_display_styled)
+            else:
+                st.write("No columns selected")
+        else:
+            st.write("DataFrame not found")
+
+if __name__ == "__main__":
+    main()