Adding a Cache check & a deletion button

col_compare.py

@@ -1,86 +1,175 @@
 import streamlit as st
 import pandas as pd
 import io
+import shutil
+import os
 
-st.title("🔍 Excel Comparator (Full Rows from Multi-Column Match)")
+st.set_page_config(page_title="Excel Comparator", layout="wide")
+st.title("🔍 Excel Comparator — Multi-Column Match + Unique Viewer")
 
-file1 = st.file_uploader("📄 Upload First Excel File", type=["xlsx", "xls"], key="file1")
-file2 = st.file_uploader("📄 Upload Second Excel File", type=["xlsx", "xls"], key="file2")
+# Clear Streamlit Cache Button
+cache_path = os.path.expanduser("~/.cache/streamlit")
+
+def get_dir_size(path):
+    total_size = 0
+    for dirpath, dirnames, filenames in os.walk(path):
+        for f in filenames:
+            fp = os.path.join(dirpath, f)
+            if os.path.isfile(fp):
+                total_size += os.path.getsize(fp)
+    return total_size / (1024 * 1024)  # size in MB
+
+cache_size_mb = get_dir_size(cache_path) if os.path.exists(cache_path) else 0
+st.sidebar.write(f"🗂️ Cache Size: {cache_size_mb:.2f} MB")
+
+if st.sidebar.button("🧹 Clear Streamlit Cache"):
+    try:
+        shutil.rmtree(cache_path)
+        st.sidebar.success("✅ Cache cleared successfully.")
+    except Exception as e:
+        st.sidebar.error(f"⚠️ Error clearing cache: {e}")
+
+file1 = st.file_uploader("📄 Upload First Excel File", type=["xlsx", "xls"])
+file2 = st.file_uploader("📄 Upload Second Excel File", type=["xlsx", "xls"])
 
 if file1 and file2:
     xls1 = pd.ExcelFile(file1)
     xls2 = pd.ExcelFile(file2)
 
-    sheet1 = st.selectbox("📑 Select sheet from File 1", xls1.sheet_names, key="sheet1")
-    sheet2 = st.selectbox("📑 Select sheet from File 2", xls2.sheet_names, key="sheet2")
+    sheet1 = st.selectbox("📑 Select Sheet from File 1", xls1.sheet_names)
+    sheet2 = st.selectbox("📑 Select Sheet from File 2", xls2.sheet_names)
 
     df1 = pd.read_excel(xls1, sheet_name=sheet1)
     df2 = pd.read_excel(xls2, sheet_name=sheet2)
 
-    st.write("**File 1 Preview:**")
+    st.write("### 🔍 Preview — File 1")
     st.dataframe(df1.head())
 
-    st.write("**File 2 Preview:**")
+    st.write("### 🔍 Preview — File 2")
     st.dataframe(df2.head())
 
-    st.subheader("🔧 Select columns to match by")
-    cols1 = st.multiselect("File 1 columns", df1.columns, key="cols1")
-    cols2 = st.multiselect("File 2 columns", df2.columns, key="cols2")
+    st.subheader("📌 Unique Values Finder")
+    with st.expander("🔎 Find Unique Values from Any Column"):
+        file_choice = st.radio("Choose file", ["File 1", "File 2"], horizontal=True)
+
+        if file_choice == "File 1":
+            uniq_col = st.selectbox("Select column from File 1", df1.columns, key="u1")
+            if uniq_col:
+                uniq_vals = sorted(df1[uniq_col].dropna().astype(str).unique())
+                df_unique = pd.DataFrame(uniq_vals, columns=[f"Unique in {uniq_col}"])
+                st.write(f"Found **{len(df_unique)}** unique values in **{uniq_col}** (File 1)")
+                st.dataframe(df_unique)
+                st.download_button("⬇️ Download CSV", df_unique.to_csv(index=False),
+                                   file_name=f"unique_{uniq_col}_file1.csv", mime="text/csv")
+
+        else:
+            uniq_col = st.selectbox("Select column from File 2", df2.columns, key="u2")
+            if uniq_col:
+                uniq_vals = sorted(df2[uniq_col].dropna().astype(str).unique())
+                df_unique = pd.DataFrame(uniq_vals, columns=[f"Unique in {uniq_col}"])
+                st.write(f"Found **{len(df_unique)}** unique values in **{uniq_col}** (File 2)")
+                st.dataframe(df_unique)
+                st.download_button("⬇️ Download CSV", df_unique.to_csv(index=False),
+                                   file_name=f"unique_{uniq_col}_file2.csv", mime="text/csv")
+
+    st.subheader("🔧 Select Columns for Matching")
+    cols1 = st.multiselect("Matching columns from File 1", df1.columns, key="cols1")
+    cols2 = st.multiselect("Matching columns from File 2", df2.columns, key="cols2")
+
+    show_merge = st.checkbox("🔀 Show Merged Matched Rows (Side-by-Side)", value=True)
+    highlight_diffs = st.checkbox("🎨 Highlight Differences in Merged View", value=True)
+
+    def detect_column_types(df, columns):
+        return {col: str(df[col].dropna().map(type).mode()[0]).split("'")[-2] for col in columns if col in df}
+
+    if cols1 and cols2:
+        col_types1 = detect_column_types(df1, cols1)
+        col_types2 = detect_column_types(df2, cols2)
+        with st.expander("🧪 Detected Column Types"):
+            st.write("**File 1 Column Types:**", col_types1)
+            st.write("**File 2 Column Types:**", col_types2)
+
+    def normalize_columns(df, columns):
+        def clean(val):
+            if pd.isna(val):
+                return ""
+            if isinstance(val, float) and val.is_integer():
+                return str(int(val))
+            return str(val).strip()
+        return df[columns].applymap(clean).agg(" | ".join, axis=1)
 
     if len(cols1) != len(cols2):
         st.warning("⚠️ Please select the same number of columns from both files.")
     elif cols1 and cols2 and st.button("🔍 Compare Now"):
-        # Create composite key in both dataframes
-        df1["__key__"] = df1[cols1].astype(str).agg(" | ".join, axis=1)
-        df2["__key__"] = df2[cols2].astype(str).agg(" | ".join, axis=1)
+        df1["__key__"] = normalize_columns(df1, cols1)
+        df2["__key__"] = normalize_columns(df2, cols2)
 
-        # Identify match and mismatch sets
-        keys1 = set(df1["__key__"].dropna())
-        keys2 = set(df2["__key__"].dropna())
+        keys1 = set(df1["__key__"])
+        keys2 = set(df2["__key__"])
 
         match_keys = keys1 & keys2
         only1_keys = keys1 - keys2
         only2_keys = keys2 - keys1
 
-        # Filter full rows
-        df_match1 = df1[df1["__key__"].isin(match_keys)].drop(columns="__key__")
+        df_match1 = df1[df1["__key__"].isin(match_keys)].copy()
+        df_match2 = df2[df2["__key__"].isin(match_keys)].copy()
         df_only1 = df1[df1["__key__"].isin(only1_keys)].drop(columns="__key__")
-        df_match2 = df2[df2["__key__"].isin(match_keys)].drop(columns="__key__")
         df_only2 = df2[df2["__key__"].isin(only2_keys)].drop(columns="__key__")
 
-        st.success(f"✅ Matches: {len(df_match1)} rows")
+        st.success(f"✅ Matches: {len(match_keys)} rows")
         st.info(f"📁 Only in File 1: {len(df_only1)} rows")
         st.info(f"📁 Only in File 2: {len(df_only2)} rows")
 
-        # Show expandable results
-        with st.expander("🎯 Matched Rows from File 1"):
-            st.dataframe(df_match1)
-        with st.expander("🎯 Matched Rows from File 2"):
-            st.dataframe(df_match2)
+        merged = pd.DataFrame()
+
+        if show_merge:
+            df_match1.set_index("__key__", inplace=True)
+            df_match2.set_index("__key__", inplace=True)
+
+            merged = pd.merge(df_match1.add_prefix("F1_"), df_match2.add_prefix("F2_"),
+                              left_index=True, right_index=True, how="inner").reset_index()
+            merged.rename(columns={"__key__": "Match_Key"}, inplace=True)
+
+            with st.expander("🎯 Merged Matched Rows"):
+                st.dataframe(merged)
+
+            if highlight_diffs:
+                def highlight_diff(row):
+                    styles = []
+                    for col in row.index:
+                        if col.startswith("F1_"):
+                            match_col = col.replace("F1_", "F2_")
+                            if match_col in row:
+                                if row[col] != row[match_col]:
+                                    styles.append("background-color: #ffcccc")
+                                else:
+                                    styles.append("background-color: #ccffcc")
+                            else:
+                                styles.append("")
+                        else:
+                            styles.append("")
+                    return styles
+
+                st.subheader("🎨 Highlighted Differences")
+                st.dataframe(merged.style.apply(highlight_diff, axis=1))
+
         with st.expander("❌ Only in File 1"):
             st.dataframe(df_only1)
+
         with st.expander("❌ Only in File 2"):
             st.dataframe(df_only2)
 
-        # Export as Excel
-        def create_excel_file():
-            output = io.BytesIO()
-            with pd.ExcelWriter(output, engine="openpyxl") as writer:
-                df_match1.to_excel(writer, sheet_name="Matched_File1", index=False)
-                df_match2.to_excel(writer, sheet_name="Matched_File2", index=False)
+        def create_excel():
+            buffer = io.BytesIO()
+            with pd.ExcelWriter(buffer, engine='openpyxl') as writer:
+                if show_merge:
+                    merged.to_excel(writer, sheet_name="Matched_SideBySide", index=False)
                 df_only1.to_excel(writer, sheet_name="Only_in_File1", index=False)
                 df_only2.to_excel(writer, sheet_name="Only_in_File2", index=False)
-            output.seek(0)
-            return output
+            buffer.seek(0)
+            return buffer
 
-        excel_output = create_excel_file()
-
-        st.download_button("⬇️ Download Results as Excel", excel_output,
-                           file_name="comparison_full_rows.xlsx",
+        st.download_button("⬇️ Download All Results (Excel)",
+                           create_excel(),
+                           file_name="comparison_results.xlsx",
                            mime="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet")
-
-        # Optional CSVs
-        st.download_button("⬇️ CSV: Only in File 1", df_only1.to_csv(index=False), "only_in_file1.csv", "text/csv")
-        st.download_button("⬇️ CSV: Only in File 2", df_only2.to_csv(index=False), "only_in_file2.csv", "text/csv")
-        st.download_button("⬇️ CSV: Matched Rows File 1", df_match1.to_csv(index=False), "matched_file1.csv", "text/csv")
-        st.download_button("⬇️ CSV: Matched Rows File 2", df_match2.to_csv(index=False), "matched_file2.csv", "text/csv")