AP Family Trees logo

How can we help you?

View Categories

OPS03: Importing CSV Files

2 min read

Table of Contents

One-place studies arrive with spreadsheets. The Import screen takes a CSV export from any spreadsheet program and files it as a record set - the unit of inventory and of reversal.

The steps #

  1. Upload the CSV. The first row is read as the column headings.
  2. Match the columns. Each of your columns is matched to a record field. Start from one of the prebuilt layouts - FreeBMD births, marriages and deaths, census transcription, parish register (baptisms, marriages, burials) and monumental inscriptions / burial list - or match by hand. Give the record set a name and a record type.
  3. Date all rows as: for a source with a single date, such as a census year, enter it once here and every row carries it. Without this, rows with no date column come in undated and stay off the public pages.
  4. Excel date numbers: tick this if a date column holds bare serial numbers such as 20000, and they are converted to dates.
  5. Preview shows the first rows as they will be stored. Confirm to run the import; large files are processed in chunks, so a 100,000-row transcription completes without timing out.

Templates #

A matching you will use again - the same society transcription layout every quarter - can be saved as a template from the matching step and chosen from Start from a template next time. Templates are stored in the shared core, so they are available to the other AncestorPress plugins on the site.

Undoing an import #

The Imports and record sets list on the Import screen shows every import with its row count. Undo import removes that record set's rows and nothing else. A wrong file or a wrong matching therefore costs nothing: undo, fix, import again.

What the import never does #

Imported values are stored exactly as they appear in the file. Names, dates and places are never corrected or standardised in place; the standardised forms used for sorting are kept separately and stay correctable. Importing never merges or overwrites existing records - importing the same file twice gives you two record sets, by design, and the undo is the recovery.