Screens / Knowledge and setup

Import data

Take a spreadsheet in, match its columns to the product's fields and check the rows before anything lands.

For: Admins and operators bringing in records.

Markdown

2 ways to lay it out

1. Drop, map, review

Pick it when: The default: a file arrives, columns are matched, rows are checked, then it lands.

A step bar across the top and the current step below it: the dropped file, the column mapping, the row check.

Data importProDrop a spreadsheet, then see what was read: rows, columns with sample values, and rows that couldn't be read.
Column mapperProMatch file columns to the object's fields.
Import reviewProHow many rows are ready, which need fixing and why, grouped by kind of problem.
StepperProThe steps of the import, with where the person is.
Reconcile tableProRows that would update records already there, matched against them: conflicts show both values, and the person resolves each.

Install drop, map, review

Pro and Premium items need a licence key; free ones install as they are.

2. Large file with progress

Pick it when: The file is big and the import takes time.

A measured progress bar with the rows done, and the problems found so far listed beneath.

ProgressFreeA bar for work whose progress is measured: an upload, an import, an index.
Import reviewProRows that need fixing, so far.
Task trayProBackground work as one pill, so the person can leave the page.

Install large file with progress

Pro and Premium items need a licence key; free ones install as they are.

Rules that matter most here

Unknown stays unknown
A cell that couldn't be read is shown as unread, with why; it isn't filled with a guess.
Say what can and can't be undone
Importing says how many rows will be created or changed, and whether the import can be undone.
Nothing is pre-selected in a decision
A column the mapper isn't sure about starts unmapped.

Brief for a coding agent

Paste this into the agent that builds the screen. The docs MCP server returns the same text from get_screen.

Build a data import with Candor UI.
DataImport takes the file and shows what was read: rows, columns with sample values and rows that couldn't be read, with why.
ColumnMapper matches file columns to the object's fields. A column it can't match starts unmapped; don't preselect a guess.
ImportReview groups rows by kind of problem and says how many are ready.
Show Progress only where the figure is measured. For a long import, move it to a TaskTray pill.
Before the import lands, say how many rows are created or changed and whether it can be undone.

Start from the "Drop, map, review" layout: A step bar across the top and the current step below it: the dropped file, the column mapping, the row check.
Candor items: data-import (https://candor.themethodic.com/docs/data-import.md), column-mapper (https://candor.themethodic.com/docs/column-mapper.md), import-review (https://candor.themethodic.com/docs/import-review.md), stepper (https://candor.themethodic.com/docs/stepper.md), reconcile-table (https://candor.themethodic.com/docs/reconcile-table.md).
Install: npx candor-ui add data-import column-mapper import-review stepper reconcile-table
Read each item's docs before using it and take its props from there; don't guess them. Pro and Premium items need the person's licence key, which they add with `npx candor-ui key <key>`; don't ask for it.