Gramps 4.2 Wiki Manual - Manage Family Trees: CSV Import and Export
Gramps Spreadsheet Import/Export
This format allows you to import/export a spreadsheet of data all at once. The spreadsheet must be in the Comma Separated Value (CSV) format. Most spreadsheet programs can read and write this format. It is also easy to write by hand. This is the only Gramps import format which allows for merging with existing data.
New feature since Gramps 3.3
Previously, sources were not exported, but now they are. Sources are referred to by their title text. You can add further details to a source after importing.
New feature since Gramps 4.2.1
Now you can export and import place information. Also, you can refer to places in marriage and person sections by using a "place id". See below for more details.
There are three main uses for this format:
- You can export your core Gramps data into a spreadsheet format, edit it with a text or spreadsheet program, and import the changes and additions back into Gramps. This is handy for sending to others to fill in, or for taking on the road when you don't have your full Gramps application.
- You can import new data into your Gramps database. For example, if you have a set of new people to add to your database, but don't want to hunt and peck your way to finding where they go, you might find it easier to type them into a spreadsheet, and then quickly bring all of them in at once. This is handy if you have a large amount of data that you are cutting and pasting from another application or the web. An example of this is restoring your Gramps database by loading the Narrative Website into a spreadsheet.
- You can also import a set of corrections and additions. Say that you have printed out a report, and you are going through it marking corrections. If you make each correction a section of a spreadsheet, you can "script the edits" and then execute them all at once.
To export your database:
- Start Gramps
- Select "Export" from the Family Trees menu
- Select "Comma Separated Values Spreadsheet (CSV)"
- Select which items to export (people, marriages, families, and/or places)
A selected set of fields of your genealogy data will be saved to a .csv file in the format described below. In addition, the people and families are referenced so that the data can be edited and read back in, thereby updating the database.
There are some columns that will be blank, specifically note and source columns. These are listed in the spreadsheet so that you can make notes for the import, but notes are never exported with this tool. From Gramps 3.3 you can now export source titles; previously no source data was exported.
Your data is broken up into four sections representing places, individuals, marriages, and children. The exported fields and column names are:
- Place, Title, Name, Type, Latitude, Longitude, Code, Enclosed_by, Date
- Person, Lastname, Firstname, Callname, Suffix, Prefix, Title, Gender, Birthdate, Birthplace (or Birthplaceid), Birthsource, Baptismdate, Baptismplace (or Baptismplaceid), Baptismsource, Deathdate, Deathplace (or Deathplaceid), Deathsource, Burialdate, Burialplace (or Burialplaceid), Burialsource, Note
- Marriage, Husband, Wife, Date, Place (or Placeid), Source, Note
- Family, Child
The first column in each area is the Gramps ID. That is what will tie your edits back to the correct data, so don't alter those data. Load this file into your favorite spreadsheet using comma separated, double-quote text delimited, and Text format (any encoding for now). Then you can add or correct data, and save it back out, keeping the same format. You can then import the data back on top of your old data and it will be corrected.
To import your data:
- use the file from above, or create a spreadsheet (described below) with genealogical data
- start up Gramps
- import the file into your current database
The merge part of this code will only add or update information to your database, and it always assume that the spreadsheet data is the correct version.
If you load this spreadsheet into LibreOffice, make sure you select each column as type Text rather than Standard. Standard will reformat your dates and numbers. Also, if you use Excel, you will probably want to select all cells once opened, and change the format of the cells to Text.
The spreadsheet is data made up of columns. Each column should have at the top of it the name of what type of data is in the column. You must use special names for the columns. Currently they are:
place - a reference to this place title - title of place name - name of place type - type of place (eg, City, County, State, etc.) latitude - latitude of place longitude - longitude of place code - postal code, etc. enclosed_by - the reference to another place that encloses this one date - date that the enclosed_by place was in effect
person - a reference to be used for families (marriages, and children) grampsid - to assign a gramps id to the person firstname - a person's first name surname/lastname - a person's last name callname - a common name (nickname) for the person prefix - surname prefix (von, de, etc) suffix - a suffix of a person's name (Jr., Sr.) title - a person's title (Dr., Mr.) gender - male or female (you should use the translation for your language) note - a note for the person's record birthdate - date of birth birthplace - place of birth birthplaceid - place id of birth birthsource - source title for birth baptismdate - date of baptism baptismplace - place of baptism baptismplaceid - place id of baptism baptismsource - source title of baptism deathdate - date of death deathplace - place of death deathplaceid - place id of death deathsource - source title for death deathcause - cause of death burialdate - date of burial burialplace - place of burial burialplaceid - place id of burial burialsource - source title of baptism
marriage - if you want to reference this from a family, you'll need a matching name here husband/father/parent1 - the reference of the person above who is the husband (for female parent1, you'll need to put gender in the person area, or edit it later in gramps) wife/mother/parent2 - the reference of the person above who is the wife (for male parent2, you'll need to put gender in the person area, or edit it later in gramps) date - the date of the marriage place - the place of the marriage placeid - the place id of the marriage source - source title of the marriage note - a note about the marriage/wedding
family - a reference to tie this to a marriage above (required) child - the reference of the person above who is a child source - source title of the marriage note - a note about the family gender - male or female (you should use the translation for your language) [You can put gender here, or in person above]
Case doesn't matter. You may use any combination of these. (Actually, you have to at least have a surname and a given name when defining a person, you have to have a marriage and child columns when defining children, and places need a place reference, but that is it.) The column names are the English names given (for now) but the data should be in your language (including the words "male" and "female").
Order is important in that if you want to reference something in one area to another, the definition MUST come first. For example, if you want to define families of people, the individuals must be defined before the families. The same applies to places. So it is usually best to put the Places data first, people next, then marriages and families.
Each of these can go in its own area in a spreadsheet. There is no limit to the number of areas in a sheet, and each area can have any number of rows. Leave a blank row between "areas". Just make sure that areas are not next to each other; they must be above and below one another.
You can have multiple areas of each kind on a spreadsheet. The only limitation is that if you refer to a person, you must do that in a row lower than where that person is described. Likewise, if you refer to a marriage, you must do that in a row lower than where the marriage is described. References to enclosed_by places must already exist in the database, or be defined in rows above in the spreadsheet.
If you use the 'grampsid' as a way to assign specific ids, be very careful when importing to a current database. Any data you enter will overwrite the data assigned to that grampsid. If you use ids in the place, person, marriage, or family columns that are surrounded by brackets (for example '[i0001]'), the values you use will be interpreted as grampsids as well. If you are adding new items, you are encouraged to avoid use of the bracket format or grampsid columns, so as to avoid accidentally overwriting your data. If you are mixing the bracket (or grampsid) methods with plain references (no brackets), put the plain referenced data after the bracket referenced data.
If you are entering the data in a text file, and if you wish to have a comma inside one of the values, like "Clinton, Co., MO" then you need place the entire value in double-quotes and put the first double-quote right after the preceding comma. For example:
marriage, parent1, parent2, place m1, p1, p2,"Clinton, Co., MO" m2, p3, p4,"Havertown, PA"
A spreadsheet program will do this automatically for you.
Here is an example spreadsheet in LibreOffice, but any spreadsheet program should work.
Notice that the data need not begin in the first column, nor in the first row.
And here is the resulting data in Gramps:
Example CSV with multiple areas
Here is an example of a CSV text spreadsheet with multiple areas:
Place, Title, Name, Type [P0001], Michigan, Michigan, State L1, Canada, Canada, Country L2, USA, USA, Country Firstname, Surname, Birthdate, Birth place id John, Tester, 11/11/1965, L1 Sally, Tester, 01/26/1973, L1 Person, Firstname, Surname, Birthdate, Birth place id p1, Tom, Smith, 22 Jan, 1970, [P0001] p2, Mary, Jones p3, Jonnie, Smith p5, James, Loucher p6, Penny, Armbruster [P0002],Tim, Sparklet Marriage, Husband, Wife m1, p1, p2 m2, p5, p6 Family, Child m1, p3 m1, p6 m2, [P0002]
If you cut and paste that into a file (or use the Import Gramplet), you can import it directly.
A date can be any valid Gramps date, including dates formats like "26 JAN 1973" or "26.1.1973".
If you make your references be Gramps IDs inside square brackets, then you can refer to people already in the database, like this:
Person, Firstname, Lastname joe's boy, Harry, Smith Family, Child [F1524], joe's boy Husband, Wife [I0123], [I0562] firstname, surname Timothy, Jones place, enclosed_by [P0001], [P0002]
This example would create and add Harry Smith to the previously existing family in Gramps, family F1524.
Also, this example would marry two previously existing people, I0123, and I0562.
This also creates a person named Timothy Jones who is not related to anyone.
Finally, this also make place P0001 be enclosed by place P0002.
Real world example
In this example, I had an entire generation to enter, 16 names with marriage dates. The children I already had in the database. I entered them into LibreOffice:
Notice that you can use numbers or strings as the reference names between areas. In the person area, I used the numbers 1 through 16. That made it easy to refer to them in the second area of marriages. The marriages are labeled with the letters A through H.
Also note that the children in the third area are existing people as indicated by the brackets around the Gramps IDs.
Saving as CSV and importing into Gramps produces the far right-hand column in the tree:
|Special copyright notice: All edits to this page need to be under two different copyright licenses:
These licenses allow the Gramps project to maximally use this wiki manual as free content in future Gramps versions. If you do not agree with this dual license, then do not edit this page. You may only link to other pages within the wiki which fall only under the GFDL license via external links (using the syntax: [https://www.gramps-project.org/...]), not via internal links.