The import or update function for a catalog allows you to import/update entries for a new or existing Mix'n'match catalog.
You can specify the catalog ID for an existing catalog, or add the required information for a new one. If you fill in the property value and switch to another form field, the available information will be loaded from the Wikidata property.
You can import CSV (comma-separated) or TSV (tab-separated) data. You can either upload a data file, or give an URL pointing to such a file. Using the URL has the advantage that the catalog can be automatically updated from that URL again, and new/changed data can be imported into the catalog.
- Please make sure the selection of tab- or comma-separated value is correct!
Every source must have a single header row, with headers from a keyword/pattern list (see below).
To use Google Sheets as a source:
- use menu "File/Publish to the web"
- select the correct sheet
- check "Automatically republish when changes are made" (under "Published content and settings") to have changes in the sheet reflect in the CSV file
- copy the URL from that dialog (not the main sheet URL!) into the URL box
Once you are done with both sections, check the "looks good" checkbox.
The header row of your file/URL will be analysed. There are some column headers that are recognized (case-insensitive):
- id; the ID of the entry in the catalog (mandatory)
- name; the name of the entry in the catalog (mandatory)
- desc (or description)
- type (as "Qxxx"); this will override the default type from the Data section for an entry
- born, died (as "YYYY", "YYYY-MM", or "YYYY-MM-DD")
- autoq (as "Qxxx") for when you already have a good guess at the matching Wikidata item; this will become an "auto-match"
- q (as "Qxxx") for when you already know for certain that this is the matching Wikidata item; this will become a "manual match"
- Pxxx a property (currently, text or item values are allowed; also, P625 (as "-1.23/0.345") is recognized as coordinates
All other column headers will be unrecognized ("?") and ignored.
If all headers are recognized to your satisfaction, click on the "Test" button. If you want to alter your data, re-upload (for files) or update the file at the given URL, then click "Reload".
Running the test will process up to 1000 rows from your data source, and either alert you with an error, or show you the resulting statistics; no changes to Mix'n'Match will have been made at this stage. If you are satisfied with the test statistics, click "Looks good, queue import/update". This will:
- create the new catalog if necessary
- add the import/update information for the catalog
- queue a job to run the import
If everything works as expected, you will see a confirmation message, and a link to the catalog.