I Built a SQLite Editor in 180 Lines, Then Rebuilt It in 730 for the Browser

Background I work with a lot of SQLite databases — patent data, court rulings, government records, scraped datasets. I wanted a quick way to open a .db file, pick columns, filter, and explore. Nothing fancy. I built it first with Streamlit. 180 lines, done in an hour. Used it daily. Then I wanted to share it and hit a wall. Streamlit: 180 Lines, Done Streamlit + pandas + sqlite3. st.data_editor handles virtual scrolling, sorting, inline editing, wide table support — all for free. Key features: Sidebar table list with row counts Column selection (multiselect) + pin columns (sticky left) Raw WHERE clause input as filter Sort, LIMIT, CSV export Cell editing with PK-based UPDATE generation Query execution timing Tables without primary keys are read-only Core query builder: [bash] display = sel_cols or col_names q = f'SELECT {", ".join(f\'"{c}"\' for c in display)} FROM "{S.tbl}"' if where.strip(): q += f" WHERE {where.strip()}" if order: q += f' ORDER BY "{order}" {"DESC" if desc else "ASC