Structured tables: query columns by name

Name a region as a Table and formulas read like English — =SUM(Sales[Amount]) instead of =SUM(B2:B5).

You'll need: abax only.

Run it

cd docs/examples/data/structured-tables
python run.py

What you should see

 Total amount: 4650.0
  Average qty: 2.75
         Rows: 4.0
    West only: 1600.0

How it works

• detect_table(sheet, r1, c1, r2, c2, "Sales", headers) builds a Table whose first row is the header; wb.tables.add(table) registers it.
• From then on any formula can use structured references: =SUM(Sales[Amount]), =SUMIF(Sales[Region], "West", Sales[Amount]).
• The reference survives edits that would break B2:B5 — insert a row above the table and the label still points at the right data.
• Inside the table, [@Amount] means "this row's Amount", and Sales[#Headers] / Sales[#Totals] address the special rows.

Next steps

• In the GUI, select a region and use the command palette (Ctrl+Shift+P) → Create table to do the same interactively.
• Dynamic arrays pair well with tables: =SORT(UNIQUE(Sales[Region])).
• The automation surface is in Automation API.