The product catalog should be encoded in UTF-8.
We support CSV, TSV and XML catalogs. We can handle these column delimiters : comma, semicolon, and vertical bar for CSV, tab for TSV.
The link doesn’t need to end with .csv or .tsv though. If the catalog comes from an URL, we can run a catalog without HTTPS.
We can run catalogs which include login and password. We must respect a precise syntax for this, as we should put the credentials formated as login:password@ right after the protocol (http or https) when we upload the catalog.
Example : https://login:password@catalog_url.csv
For the XML files, there are few more requirements to respect :
- we need a header which precises the version of the XML standard
- we need one or several overall markups
- we need an item list delimited by a recurring markup
- we need, for each product, recurring attributes which have a static markup name
<?xml version="1.0" encoding="UTF-8"?>
<description>Your fancy product name</description>
A header with column names in the catalog is necessary to run a product feed ad.
To correctly map the catalog with our system, we need at least these three columns : Product ID, Image URL, Description. Please keep in mind that without these columns, we are totally unable to run the product feed.
Column names must contain text to be understandable. They can be written in any format as long as they don’t contain the column delimiter of the file. If this is the case, the column names should be quoted with double quotation marks.
The structure needs to stay the same in the whole file : same number of columns for every line. Please note that a field can be empty.
A text field must be quoted with double quotation marks.
Refreshing the Catalog
We are able to handle an automatic refresh of the product catalog if you need it. We have to make sure that the catalog structure won’t change, and we have to know the refresh frequency that you need (we can do it hourly, daily or monthly).