Does Visidata support direct cell reference just like Excel?

[algorist-ahmad]

Hello, I was exploring the idea of Visidata as a replacement to Microsoft Excel or Libreoffice spreadsheets. One of their most useful feature is the ability to reference data from another arbitrary cell. I couldn't find any information about referencing another cell in the official documentation or in the discussions. Is this improper use of Visidata or am I just bad at python expressions? Is there no way to reference one specific column of one specific row?

[WaylonDarien]

just wanted to add i also want this :), i like live totals that update as i add numbers to a column, or i like to see various live totals change as i change the categories of certain records (like a live frequency analysis per category, but on the same sheet where i can see it).

i fully recognize i might be trying to shove my excel way of thinking into a data tool that was designed with very different thinking in mind, and if there is a far more graceful, visidata-spirited way of accomplishing this i would love to learn that and change the way i think about these kinds of tasks.

i've gotten part of the way there with a split screen where the frequency analysis is on top and the data sheet is on bottom. I can get closer to a live update with "reload-every" but it keeps crashing my visidata on windows (might do an issue for that later), so i just quickly TAB -> CTRL+R -> TAB every time i want to update the frequency analysis.

this is the closet i've gotten to replicating my old excel/libreoffice workflow

budget

A	Expenses	Amount	Income
1	rent	1300.00	4000.00
1	discretionary	730.00
1	electricity	150.00
1	coffee	150.00
1	gas (overestimate)	100.00
1	internet	92.00
1	phone	45.00

SPLIT

---

budget_A_freq

A	count	Amount_sum	Income_sum
1	7	2567.00	4000.00

ideally would like persistent "remainder" column calculated as (Income_sum) - (Amount_sum) which auto updates every time i refresh the frequency analysis sheet. I would also like the names and hidden columns to persist as well. Ideally it would look and stay like this and be live:

---

better frequency analysis with persistent changes across refreshes

Total_Expenses	Monthly_Income	Remainder
2567.00	4000.00	1433.0

---

even more ideally i wouldn't need a column like column A at all and I could have my budget summary exist in the same sheet as the budget itself.

Expenses	Amount	Monthly_Income	Total_Expenses	Notes	Remainder	Notes
rent	1300.00	4000.00	2567.00	< -- this is NOT live 😔, i would like it to be in the future 🤠👍	1433.00	< -- this is live 👍(¬‿¬)👍
discretionary	730.00				#ERR	< -- these wouldn't be errors
electricity	150.00				#ERR	< -- these wouldn't be errors
coffee	150.00				#ERR	< -- these wouldn't be errors
gas (overestimate)	100.00				#ERR	< -- these wouldn't be errors
internet	92.00				#ERR	< -- these wouldn't be errors
phone	45.00				#ERR	< -- these wouldn't be errors

---

or even better

Expenses	Amount	Summary	Value	ideally
rent	1300.00	Monthly_Income	4000.00	< -- static
discretionary	730.00	Total_Expenses	2567.00	< -- sum(Amount) (ideally live)
electricity	150.00	Remainder	1433.00	< -- Monthly_Income - Total_Expenses (ideally live)
coffee	150.00
gas (overestimate)	100.00
internet	92.00
phone	45.00

[algorist-ahmad]

Thanks for the answer, you've clearly put lots of thought and hours into Visidata. The more I use it, the more I like it but also I realized that it probably was never meant to be an alternative to traditional GUI spreadsheets. Maybe it was a bad assumption on my part.

[WaylonDarien]

you're right, it's not "excel but in the terminal", but it's the closet thing i've found, and the better i get at using visidata the more i prefer it and learn how to do things the "visidata way". i find visidata to be WAY faster and more responsive than excel, i try to shoehorn it into my workflow as much as i possibly can.

i'm reading through the api docs and learning a little python for the exclusive purpose of seeing if i can get a column sum to exist in a single cell. if i learn how to accomplish this i'll lyk

[algorist-ahmad]

Thanks, I'd also like to learn how to use vd to its full potential, but I'm limited by my knowledge of Python, so I could use the help.

Though I already decided to move on to other tools. My case requires that I pipe aggregation values like sums and averages from a dataset in plaintext into scripts to calculate income and expenses for example. My top contenders are mlr (Miller), jq, and q.