From 8fbc11da7a68316f2a4ad3ef19681355ab5e4be4 Mon Sep 17 00:00:00 2001
From: Luigi Maiorano <luigi.maiorano@qumo.io>
Date: Tue, 16 Dec 2025 23:42:25 -0800
Subject: [PATCH] Inline removal of keywords

---
 02-B_Thematic-Processing.py | 55 ++++++++++++++++++++++++++++++-------
 1 file changed, 45 insertions(+), 10 deletions(-)

diff --git a/02-B_Thematic-Processing.py b/02-B_Thematic-Processing.py
index f3411c8..b9ebe04 100644
--- a/02-B_Thematic-Processing.py
+++ b/02-B_Thematic-Processing.py
@@ -345,25 +345,53 @@ def _(mo, tag_select):
     return
 
 
-@app.cell(hide_code=True)
+@app.cell
 def _(frequency_df, min_freq_select, mo):
     mo.stop('keyword' not in frequency_df.columns, "Waiting for keyword extraction to finish")
 
     MIN_FREQ = min_freq_select.value
 
-    freq_df_filtered = frequency_df.loc[frequency_df['frequency'] >= MIN_FREQ].copy()
+    _freq_df_filtered = frequency_df.loc[frequency_df['frequency'] >= MIN_FREQ].copy()
 
-    freq_df_filtered.reset_index(drop=True, inplace=True)
-
-    keyword_freq_filtered = freq_df_filtered.set_index('keyword')['frequency'].to_dict()
-
-    table_selection = mo.ui.table(freq_df_filtered, page_size=50)
+    table_selection = mo.ui.table(_freq_df_filtered, page_size=50)
     table_selection
 
-    return (keyword_freq_filtered,)
+    return MIN_FREQ, table_selection
 
 
-@app.cell
+@app.cell(hide_code=True)
+def _(mo, table_selection):
+    remove_rows_btn = None
+    if len(table_selection.value) >0 :
+        remove_rows_btn = mo.ui.run_button(label="Click to remove selected keywords and update xlsx")
+
+    remove_rows_btn
+    return (remove_rows_btn,)
+
+
+@app.cell(hide_code=True)
+def _(KEYWORD_FREQ_FPATH, frequency_df, remove_rows_btn, table_selection):
+    if remove_rows_btn is not None and remove_rows_btn.value:
+        # get selected rows
+        selected_rows = table_selection.value
+        if len(selected_rows) >0 :
+            rows_to_drop = table_selection.value.index.tolist()
+
+            frequency_df.drop(index=rows_to_drop, inplace=True, axis=0)
+        
+            # Save updated frequencies back to xlsx
+            frequency_df.to_excel(
+                KEYWORD_FREQ_FPATH,
+                index=False
+            )
+        
+            print(f"Updated keyword frequencies saved to: `{KEYWORD_FREQ_FPATH}`")
+
+            print("GO TO STEP 4b) and reload data to continue refining the dataset.")
+    return
+
+
+@app.cell(hide_code=True)
 def _():
     IGNORE_WORDS = {
         'chase as a brand': [
@@ -423,11 +451,12 @@ def _(
     IGNORE_WORDS,
     Image,
     ImageDraw,
+    MIN_FREQ,
     WordCloud,
     blue_color_func,
     buffer,
     canvas_size,
-    keyword_freq_filtered,
+    frequency_df,
     logo_switch,
     mo,
     np,
@@ -438,6 +467,12 @@ def _(
     if run_wordcloud_btn.value:
         pass
 
+    freq_df_filtered = frequency_df.loc[frequency_df['frequency'] >= MIN_FREQ].copy()
+
+    # freq_df_filtered.reset_index(drop=True, inplace=True)
+
+    keyword_freq_filtered = freq_df_filtered.set_index('keyword')['frequency'].to_dict()
+
     # remove specific keywords depending on selected tag
     if IGNORE_WORDS.get(tag_select.value.lower()):
         for word in IGNORE_WORDS[tag_select.value.lower()]: